It has been several years since I first drafted this website page. I am revising it for two reasons: (1) to make additions and/or corrections, and (2) to revisit previous theories and conclusions as I continue seeking new information that allows me to inch closer to the "ultimate truth." I will not always be correct. My hope is that others will continue to expand on my effort to make certain that the actual, factual truth is maintained for future generations.
In order to better understand David Thompson's trip to view Flathead Lake on March 1, 1812, some brief background is necessary. During his final weeks of trading from Saleesh House, located near Thompson Falls, MT, he and his men concentrated their efforts on picking up as many furs, and as much pemmican, and dried meat as was available, before heading back across the Rocky Mountains for the last time. His efforts were focused on the Salish and Pend d'Oreille bands wintering near Dixon, MT, and on a Kootenai band camped on Camas Prairie, about ten miles south of the town of Hot Springs.
Shown below is the Indian road between the towns of Plains and Dixon. Thompson called it the "Saleesh Road to the Buffalo" (RttB). The Kootenai camp was about halfway between the Saleesh House trading post and the Salish camps near Dixon.
The next image, looking east, shows the route Thompson followed from Plains to where the Kootenai camp was located on Camas Prairie. The "long lake" noted in his journal is shown and Highway 38 between the towns of Plains and Hot Springs is indicated by the yellow line from right center to upper left. Highway 200 runs from bottom center to right center. The RttB was located in the draw on the north side of the Plains cemetary.
The white pin near the upper left, where the highway turns at Rainbow (Dog) Lake, is where the RttB turned to the south and followed Camas Creek to the prairie floor near where the Kootenai camp is shown.
Thompson never mentioned anything in his journal about leaving the Columbia Department of the North West Company (NWCo) and retiring from the fur trade. However, I believe his intention to do so played a significant role in his interest in seeing a few places he had not seen in his three seasons of trading in this region.
On February 25, 1812, Thompson set out from Saleesh House with a Métis guide named Pierre Gaucher, aka Le Gauche (Left-hand), to view, and write extensive notes about what is now the Missoula Valley and the Hell Gate Canyon. Thompson called the canyon where the Indian road into the valley was located, Defile of Courter after an American fur trader killed near there by the Blackfeet. Two days later he was back at his camp near the mouth if Mission Creek. Business was good, so on March 1, 1812, he decided to take an up-close look at Flathead Lake (his "Saleesh Lake") and the surrounding country.
Fast forward. On July 23, 1926, Thompson historians, T.C. Elliott and J.B. Tyrrell, arrived in Kalispell, MT. Their purpose in being there was to pin down the location of the "high Knowl" which Thompson described as the place from where he viewed the lake and the surrounding country on his March trek. That evening they gave a presentation on the subject at the Elk's Temple.
On July 26, the two men, along with an entourage, motored to a small knoll on the south shore of the lake about three miles east of Polson. It is located about a quarter of a mile southeast of Bird Point on the line between sections 1 and 6 at an elevation of 3,023 feet, at 47° 41' 59" N, 114° 5' 31" W. In the remainder of this OBServation (OBS) we will refer to this spot as the E/T (Elliott/Tyrrell) location
For almost 100 years, this spot has been accepted as the place where David Thompson first viewed the lake. However, based on available evidence, we feel confident that the E/T location is not the knoll described and visited by Thompson.
At approximately 1:30 pm on March 1, 2012, 200 years ago almost to the hour when Thompson arrived at the top of "Knowl" described in his journal, Linda and I stood on top of the E/T knoll with about 30 other people to commemorate Thompson's visit. The event was sponsored by the Polson-Flathead Historical Museum, and I was asked to say a few words about the famed explorer/surveyor who supposedly stood there 200 years before. The group had a great on-site discussion about Thompson's visit. That evening, I gave a PowerPoint presentation revealing for the first time the contrarian information provided here.
From the top of the E/T knoll, you cannot see Goose Hill—the knoll from where Thompson and Le Gauche viewed Flathead Lake and the surrounding country in 1812. This is a significant consideration. Two miles west of the E/T location a steep rise in elevation blocks almost the entire view-shed of the Flathead River course, with the exception of the lake outlet, located at the Highway 93 bridge at the west end of downtown Polson.
That's me in the beaver hat and red blanket-shirt. Linda was taking pictures.
In October 1809, Thompson hired a Salish guide to lead him from Kullyspel House back to the Big Bend in Kootenai (his McGillivray's) River. The bend is located just south of the dam on Lake Koocanusa a few miles east of Libby. MT. That was where he was to meet his clerk, James McMillan, who was bringing in a new supply of goods for trade along the Clark’s Fork (Saleesh) River.
He also hired the guide's son as a hunter to provide food along the way. As part of that deal, Thompson provided the young man with a blanket, his personal rifle, and accoutrements that likely included powder, ball, and patching.
This was one business agreement that did not work out well for Thompson. The guide did not know the way back to the big bend. Consequently, he got them lost and ended up leading Thompson miles to the southeast of where they should have turned north.
On Monday October 16, 1809, Thompson took a noon OBS to determine his location. He was only a few miles from where the Little Bitterroot River emptied into the Flathead River. Likely using previous OBS points that he had recorded in his journal, he became acutely aware that his guide had no idea how to get back to the Kootenai River. The following day, he fired the man and set out on his own. To make matters even worse, the man's son turned out to be both a poor hunter and a thief. In the end, he deserted the party taking Thompson's belongings with him.
Le Gauche was nothing like the earlier guide. Thompson respected him and trusted in his knowledge of the country. Engaging Le Gauche to lead him to a good vantage point from which he could see the lake and surrounding country, Thompson "...got a loan" of some horses from the Indians for the trip.
On March 1, 1812, Thompson wrote the following in his journal:
"March 1, 1812, Sunday – A fine day but very sharp Night & Morning Northly Wind. Went off to the upper Saleesh Camp of abt 20 Tents the lower Camp is abt 60 tents & 20 of Widows, old Women,&c - The old Chief has a Wier made, they catch a few grey Carp, with small Scales, like the red Carp in form & Scales, but the Colour is silver grey with small speckled Salmon Trout - At 9..10 AM set off & in abt 1/8 M crossed the Wier Brook of abt 15 yds (across) now only 18 In deep - we held on for abt 3/4 M up along the Brook, which comes from the NE defile then turned off behind the range of Meadow Knowls & held on abt N8E/18M - our straight Co from the Brook of Saleesh Camp is N15E 2M"
Thompson's journal is filled with compass bearings, descriptions of places, and observations to determine longitude and latitude coordinates used to pinpoint his location. They were later used to draw extremely accurate, detailed maps. However, the old adage about the devil being in the details certainly applies to Thompson's notes. Having a very limited supply of both paper and ink, he was extremely frugal with both. Substituting symbols for words helped conserve both.
Since my book was published in 2008, Linda and I have continued our efforts to track Thompson's movements in the region during his three winter, trading seasons here. Part of that effort has been to revisit and refine earlier work as we have become more familiar with routes he used and places he visited.
The following image provides information about his day-trip to have a close-up look at Flathead Lake. Two camps of Salish wintered in this area providing a brisk trade. They were located a couple of miles north of the town of Dixon along a trail that closely followed what is now Highway 212.
Thompson first established a winter trading camp here in 1809. It was about a three-day ride up-river from Saleesh House. It was located near the Flathead River about 2.5 miles north of Dixon and two miles south of the upper Salish camp at, or very near, 47° 20' 49" N, 114° 17' 7" W. The furs and meat that he obtained were shipped downriver to the post by canoe and on horseback where they were sorted and pressed for shipping back across the Rockies.
My original interpretation of his March 1 entry was to track him directly from the mouth of Mission Creek to the knoll. I spent a lot of time on that effort. I was wrong. I recognized at the time there were a few details that did not fit as nicely as I had anticipated. However, I am now more convinced it ended up leading to what is, and has been all along, the right place.
The important fact I missed was that his direct travel route to the top of the "bare Knowl" did not start at the mouth Mission Creek as I originally thought. Instead, it started from a point on upper Mission Creek, which he described perfectly. His point of beginning (POB) is shown in the white circle on this image.
On March 1, 1812, Thompson and his guide began their journey to the "...bare Knowl" from the upper Salish camp. The second part of his journal entry provided details about their route upon leaving the upper Salish camp:
"At 9..10 AM set off & in about 1/8 M crossed the Weir Brook of abt 15 yds [across] now only 18 In deep - we held on for abt 3/4 M up along the Brook, which comes from the NE defile then turned off behind the range of Meadow Knowls held on abt N8E 18M - our straight Co[urse] from the Brook of Saleesh Camp, is N15E 2 M - we came smartly on Trot & hard Gallop to 1..25 PM when we alighted on the top of bare Knowl, commanding a very extensive View of the Lake & Country far around..."
As shown in the following image, a short distance above the upper Salish camp, they crossed to the north side of Mission Creek. Thompson wrote: "... in about 1/8 M [660 feet] crossed the Weir Brook." This part of his route is shown as the green line from the upper camp. The yellow line is explained below.
The yellow line along Mission Creek to the POB in the next image is the distance he noted as "...abt 3/4 M up along the Brook." Note the old stream channels, now often dry.
At that point, they were on Mission Creek near 47° 22' 27" N, 114° 15' 54" W on today's National Bison Range. The blue line on the right shows Thompson's compass bearing to what he called the "NE defile" in his journal. This is where they began their course "...abt N8E 18M", the red line running north from the POB.
The following elevation profile shows that within a few miles north of the POB, Thompson would likely have been able to see Goose Hill looming on the horizon. For reference, the first "dip" to the right of the circle is Crow Creek; the next is Mud Creek.
He was probably checking his compass bearing for accuracy along the way as well, since he recorded it to the nearest two degrees. Based on his journal, this two-degree bearing was to his most accurate standard.
Two hours and five minutes after turning to the N8E heading, Thompson noted they crossed a brook about 30 feet wide. This could only describe his crossing of Crow Creek.
"At 11..15 AM crossed a Brook of 10 yds comes from where I was at the Kootanae Camp [in] 1810..."
This is a very important reference, since it identifies the general location of a Kootenai camp he had personally visited. The blue line from the POB in the previous Google image shows Thompson's compass bearing to what he called the "NE defile" in his journal. This image shows that area.
The most likely location for the 1810 Kootenai camp is shown by the circle. The reason I believe this is due to the fact that the land to the south and east is pockmarked by hundreds of small potholes. Many are filled with water much of the year. They are visible in this image as small, black dots.
Seven minutes later, they crossed another brook about 10 feet wide.
"... at 11..22 a deep little Brook of 3 yds [across] - 2 1/2 ft deep from the Mountains to the northd of the latter, carries no Wood from hence, runs off to the Wtd & does not join the other Brook..."
This had to be Mud Creek. Their ground speed (on horseback) between the two streams would probably have been about four to five miles per hour, implying it was perhaps a half-mile between the points where they forded the streams.
As shown on this next image, the two streams converge only a short distance west, but the spot was obviously not visible to him.
The next image is from Thompson's 84° Map showing two separate streams flowing from the foothills of the Mission Mountains and emptying into the Flathead River. Thompson's journal entry was very clear about that.
After crossing the two creeks, the men continued along the N8E course to the top of the knoll. The following image shows the north end of their route.
At 1:25 PM, Thompson and LeGauche reached the top of the high knoll. The view was impressive as noted by Thompson in his journal:
"...we came smartly on Trot & hard Gallop to 1..25 PM when we alighted on the top of bare Knowl, commanding a very extensive View of the Lake & Country far around "...Sketched off the Lake &c - the Sortie of the Saleesh River bears from me S80W 6M the River then makes a Curve to the Westd & bends round to the bitter Root Rivulet..."
Thompson's journal entry provided valuable information about his view from the knoll (Goose Hill). He estimated the general direction (S80W) and distance (six miles) the river flowed from its outlet at the lake until it turned to a more westerly direction. From there he described it bending around to a course that eventually led south and west to near the mouth of the Little Bitterroot River.
Depending on his concern for accuracy, Thompson recorded compass bearings in 2°, 5°, or 10° increments. On that day, he was obviously trying to be as accurate as possible. From the top of the knoll, Thompson likely took a back-shot along the route they had followed from the meadow hills, easily seen from the summit. Back-shot bearings were commonly used to check, or to improve, accuracy.
The entry provides the bearing from his position on the knoll to the easily identifiable Big Bend in the Flathead River. The location is identifiable since it is very near the place where the Little Bitterroot River turns sharply to the east before emptying into the Flathead River about five miles downstream. The OBS site was located on the Magera ranch that was originally homesteaded by the Melton family in 1910.
Following is the view to the south that Thompson would have had from the top of the Goose Hill knoll, also known as Pilot Knob. This photo was taken looking almost due south. This ridge, now cut by a gravel pit, would have provided horsemen approaching from that direction an easy path to the summit. The black circle marks the vicinity of the Saleesh camp at the mouth of the "Weir Brook" near the Bison Range.
Goose Hill is located on Back Road immediately west of Pablo Reservoir. It is visible from Highway 93 in the vicinity of Mile Marker 56 located just north of the town of Pablo. The next photo was taken from a spot just north of the knoll looking due south.
The next photo shows the prominence of the knoll, in the white circle, as part of the surrounding landscape. It was taken from the front-center deck of a Flathead Lake dinner cruise Linda and I enjoyed while celebrating my 71st birthday. In the early 1900s, boat captains used it as a reference for landing at the Polson dock and referred to it as "Pilot Knob."
The captain of our cruise boat kept Goose Hill pretty much right on the bow to within a few hundred yards of the dock when it disappeared behind the ridge. The town of Polson is on the left. Note a second knoll to the right of the trees. We will discuss it later.
The red line on the left in the next image is the line of sight from the knoll to the trees shown off the bow of the cruise boat in the previous photo, very close to the spot where Linda snapped the above photo. The red line on the right indicates the direction from the knoll to the E/T location shown in the white oval. The white circle shows the location of the cruise dock, located very near where early 19th century docks were located.
The following is an elevation profile from the top of Thompson's Goose Hill perch to the E/T location. I include this to show, very clearly, that neither Goose Hill nor a single point of interest to the west noted in Thompson's Journal can be observed from the E/T location.
The southwest, line-of-sight view to the southwest from the E/T location is obstructed about 2.5 miles west by higher ground. There is essentially no view to the south and west. Consequently, Thompson could not have been at the E/T site when he described several places within his view on March 1, 1812. We know from his journal he had been east of this site to a Kootenai camp 1810.
The following painting by noted Montana artist Frank Hagel hangs in Sykes Cafe in Kalispell, MT. It depicts Thompson and Le Gauche viewing the Flathead Valley and Lake from atop the knoll. In it the Flathead River is shown for a short distance where it flows out of the lake. Hagel's location was almost certainly based on Thompson's Narrative or Journal. The E/T location is almost four miles east of the river outlet.
Following is the view from the top of Rocky Butte today using Google Earth 3-D. Note that the mouth of the Flathead River, as shown in the Hagel painting, is visible, as is the city of Polson in the upper right.
This is Doug Crosby's photo from the top of Goose Hill.
The difference between the view shown in Hagel's painting and that in the photo taken by Doug Crosby from the top of Goose Hill (elev. 3,538) is very slight. When comparing the two, it is easy to recognize their similarities, especially the background detail. The only major conflict between the two is that the outlet of the Flathead River is not visible from the Goose Hill location.
I have never been to the top of Rocky Butte (elev. 3,430) but am convinced Hagel's view was from there. So, let's explore that point for a moment. Following is a topographic map showing both the Hagel site (Rocky Butte) and the Crosby site (Goose Hill).
I believe the key to resolving the question is Thompson's notation that he and LeGauche then "...held on abt N8E 18M..." That was his observation regarding the most direct route from where they were on Mission Creek to the knoll. Goose Hill fits the reference. Rocky Butte does not. Another clue may be the following part of his journal entry for the day:
"...we came smartly on Trot & hard Gallop to 1..25 PM when we alighted on the top of bare Knowl..."
Thompson and LeGauche were pushing their horses pretty hard to get to the viewpoint. The ridge on the south side of Goose Hill would present very little challenge of them riding at a trot, or even a gallop, up what was any easy slope. That is not the case with Rocky Butte. The easy access there on horseback would have been from the east side.
In the lower left of the following Google image is a red circle marked "Thompson 1809 OBS." This is near the place where, on Monday October 16, 1809, Thompson had fired the guide who had gotten the party lost before finally admitting he was not familiar with the short cut between the Clark’s Fork and Kootenai Rivers.
From that point, Thompson proceeded up the Little Bitterroot River, referred to in his journal as the bitter Root Rivulet. His OBS was determined from a spot about five miles from where it empties into the Flathead, marked "Mouth of the Little Bitterroot" in the above image. The high hills west of the OBS location reach an elevation of 5,000 feet within about three miles, and consequently loom in one's view to the southwest from the knoll.
Although Thompson never again set foot along the lower reaches of the Little Bitterroot River after 1809, he recognized the drainage and it's relative location from his 1812 perch on the high knoll.
During his years in the fur trade, Thompson often made sketches of what he regarded as special places. Unfortunately, most have been lost. Only a handful of sheets have survived. One of those is a sketch of the Mission Mountains was done while he and Le Gauche were on top of the high knoll on March 1, 1812. The following short entry appeared in his journal that day: "Sketched off the Lake &c..." Further down in the same entry he notes:
"...the Scenery on the east Side of the Lake is not striking, a sameness of Hills appears to the Scenery I drew off 2 years ago - to the southd of which last a bold wide Gap formed one common Road to the Buffaloe a few years ago, but from its vicinity to the Peagans, disused for many Years - the line of Hills "c is sketched off as per Map - marked NE View of the NE defile &c..."
His reference here to the "Road to the Buffaloe" was an old trail up Mission Creek to St. Marys Lake then across a divide to the upper Jocko River. It crossed the south end of the Mission Range near the headwaters of the Jocko. From there, it passed along the west side of Placid Lake and a little southwest of the lower end of Salmon Lake. At Clearwater Junction it intersected with the Indian road along the Blackfoot River between the towns of Bonner and Lincoln, MT.
As Thompson learned during his time with the Flatheads, traveling that route became deadly. For protection from the Blackfeet, the Flatheads and other tribes going east to hunt buffalo gathered their bands in the Missoula Valley then traveled together to the hunting grounds. More warriors against a common enemy made sense.
Thompson was very careful to capture detail such as the prominent, identifiable peaks of the Mission Mountains. Shown below is sheet No4 - South of the Saleesh Lake, one of the five sheets that he sketched to illustrate the Mission Range. The original is archived at the Thomas Fisher Rare Book Library at the University of Toronto.
I am aware of only three panels that have ever been found; two of the Mission Mountains in MT, and one of Mount Nelson in Jasper National Park, Alberta.
Thompson produced several sketches while he was in this area, probably while he was trading for furs in the Dixon area. According to his journal, he also made sketches of the area around Saleesh House located near Thompson Falls, and of the Missoula Valley during his short visit there with Le Gauche on February 25, 1812, just two days before his trip to see the lake.
The Rest of the Story
David Thompson left detailed notes in his journal about how he and Le Gauche got the knoll. So let's take a look at the rest of the story with a relatively short synopsis and comparison of its location compared with that of the E/T location. I have already provided several thoughts about why I think the E/T location is incorrect.
This image shows the two routes based on Thompson's recorded bearing (N8E) and distance (eighteen miles) from his starting point on the north side of Mission Creek.
Assuming Thompson recorded his bearing and distance from the POB correctly, there is no way the E/T location can be correct. Thompson's bearing and distance from the POB simply do not support the E/T location. In addition, the elevation profile discussion should put to rest once and for all that, any view of the Little Bitterroot to the southwest of Goose Hill would have been impossible from the E/T location.
To summarize, note that the POB of the projected E/T route would have been within a half-mile of what is now the namesake reservoir of the Ninepipes National Wildlife Refuge. The area for several miles around is largely swampy and filled with potholes. In 1812, traveling along a N8E bearing would have been an extremely nasty experience.
The upper Crow Creek channel is visible between the red lines. Also shown is the location of Fort Connah, aka the 3rd Flathead House, the last Hudson's Bay Company trading post allowed to operate in territory claimed by the United States.
In his journal, Thompson provided three courses and distances from the upper Salish camp to the POB. The first was "...in about 1/8 M crossed the Weir Brook." The second, "...held on for abt 3/4 M up along the Brook", was the distance from where he crossed Mission Creek to his POB located on the Bison Range. The third was, "...our straight Co from the Brook of Saleesh Camp [Pauline Creek] is N15E 2 M."
He then provided his bearing to the top of the knoll as "abt N8E 18M." It is important to note his bearing was stated in terms of 2° of accuracy. When he noted a bearing to that accuracy, he was being as precise as his compass allowed. To me, including the term "abt" in this reference is another indication of the care he was taking to make it as precise as possible.
Finally, he provided the direct distance and bearing from near the lower Salish camp to the POB when he noted: "...our straight Co from the Brook of Saleesh Camp, is N15E 2 M." In my opinion, that bearing and distance were from the vicinity of the lower camp on Pauline Creek to the POB.
All things considered, Linda and I are convinced that the red line on the left best represents Thompson's route on March 1, 1812. His mileage estimates, the description of the surrounding country within his view, and his identification of the Little Bitterroot drainage all support the Goose Hill location on Back Road, at the west edge of Pablo Reservoir, as being the place where he and Le Gauche stood—not the E/T location.