Fort Vancouver Historic Structures Report

History and location

In 1841 Lieutenant Wilkes of the United States Exploring Expedition noted that the water in "the well" at Fort Vancouver rose and fell with the level of the Columbia River. [1] By implication these words would seem to indicate that there was only one well within the pickets at that time. This impression is confirmed by the ground plan of the fort drawn by Lieutenant Emmons of the same party. According to this drawing the only well indicated was situated in the northwest section of the fort, west of the granary and south of the large storehouse (No. 18) which stood along the north wall. [2]

This well was of considerable interest to the American explorers. The expedition's geologist, James Dwight Dana, noted that in digging a well at Fort Vancouver, the excavators first encountered two or three feet of soil, then 30 feet of gravel, and then "light quicksand" too "mobile" for further digging. He assumed that water from the river percolated laterally through the sand to supply the well. [3] Wilkes observed that the inhabitants of the post used river water in preference to well water, though they did not "consider the latter as unwholesome." [4]

When this well of 1841 is next shown on a map -- that drawn by Lieutenant Vavasour late in 1845 -- it appears to have shifted position, being much closer to the north stockade wall than indicated by Emmons (see plates VI and VII). The location given by Vavasour was confirmed by archeological excavations in 1952. [5]

This discrepancy brings up a question. Did Emmons, who freely admitted that he could not vouch for his ground plan being correct in every particular, make an error in showing the position of the 1841 well, or was the 1841 well abandoned and another dug farther to the north when the Beef Store was apparently built over the old site between 1841 and September, 1844? [6] Future archeological surveys should provide a definite answer.

An entry in Clerk Thomas Lowe's journal for February 27, 1845, probably would also throw light on the matter if we could be sure of its exact meaning. "The men," wrote Lowe, "are busily employed in sinking the old well near the granary, and in digging another one in the opposite end of the Fort, near the new Bake House." [7] Unfortunately, the words "near the granary" could apply with equal appropriateness to the well shown by Emmons and to that on Vavasour s plan.

At any rate, it is apparent that a well in the northwest quarter of the stockade enclosure underwent some type of renovation during February, 1845. This was seemingly the well shown by Vavasour since one reconstructed in February probably would not have been replaced by the end of the year. This well, indicated on the Vavasour plan as being just north of the Beef Store and about 45 feet west of the granary, is termed "Well No. 1" in the balance of this report.

Lowe's words also pinpoint the date of the second well shown on Vavasour's plan. This feature appears as a small circle about midway between the northeast corner of the Bachelors' Quarters and the southwest angle of the bakery, near the northeast stockade corner (see plates VI and VII). This well is termed "Well No. 2" in this report. Its remains were found during archeological excavations in 1952 only slightly removed from the location as given by Vavasour (see plate I). [8]

This round, boulder-lined well has been left uncovered and now is a primary interpretive feature at Fort Vancouver National Historic Site. [9] It is the only visible surviving original structure of the old fur-trading post. The water level in this well fluctuates with that of the Columbia River exactly as it did during the Hudson's Bay Company period. [10]

Although it seems quite clear that there were only two wells in the fort enclosure at the end of 1845, there is some evidence that a third may have been constructed at a later date. During 1854 a board of Army officers examined the improvements on the Fort Vancouver Military Reservation and made a fairly complete list of the Company's buildings. [11] This document noted that the firm's structures included three wells. There is no proof that the third well was within the palisade -- there seems to have been at least one in the nearby village [12] -- but photographs of 1860 show what appears to be a fire-fighting apparatus in the courtyard near the bell tower, and this device may have been placed over a well (see plates XXVII and XXVIII).

Construction details

a. Well No. 1. The Vavasour map shows this well as a rectangle which according to the scales on two different versions of the plan measures between 25 and 28 feet on one side and between 18 and 20 feet on the other (see plates VI and VIII). The inventory of the Hudson's Bay Company property made during the winter of 1846-1847 listed "one well house, 24 x 18 feet." [13] Almost certainly these data refer to the same structure. Unfortunately, there seems to be no known picture of this building which stood over Well No. 1, nor has any archeological evidence concerning its dimensions or construction yet been uncovered.

Much more information is available concerning the well itself. It was partially excavated by the National Park Service in 1952. Under the site of the well house as shown by the Vavasour plan was found a pit about 15 feet deep and roughly 15 feet square, with rounded corners. In the lower third of the pit were found many artifacts dating from the Hudson's Bay Company period. The upper portion of the pit was filled with dirt, rocks, and trash. Objects recovered here seemed to date from about 1870 to 1900. There was no evidence of walls or other structural elements on the sides of the pit.

At the bottom of the pit, 15 feet below present ground level, the well shaft was discovered. It measured slightly less than five feet square and was cribbed with 6-inch by 8-inch timbers which interlocked at the corners. The shaft was opened to a depth of about eight feet but, due to the danger of cave-in, the exploration was abandoned before the bottom of the well was reached. The well was then backfilled, but an excellent diagram in Mr. Caywood's report preserves a record of the cribbing construction (see plate LVI). [14]

As Mr. Caywood so aptly states, if the cribbing had originally extended to the ground surface, there is now no way of knowing it, since the large 15' x 15' pit, if dug at a date after the well had been constructed, would have obliterated all evidence. Mr. Caywood suggests that the space between the ground surface and the cribbing -- the 15' x 15' pit, 15' deep — may have been used for the storage of perishable foods, but no archeological evidence of its use, "other than as well and deposit for refuse," was discovered. [15]

One object found during the excavation tends to support the view that the cribbing did not extend to the ground surface. Mr. Caywood's description of this discovery is as follows:

During the excavation of Well No. 1, a windlass, complete with iron ferrules and crank handles, was found across the top of the cribbed well shaft. The drum had been made of a fir log and was still in a good state of preservation. The windlass was found in place and would substantiate the theory that water was drawn from the lower level, some twelve to fourteen feet below the [1845?] ground level. The larger excavation above the well shaft must have been used as a "spring house" or cooling room for perishables. [16]

b. Well No. 2. During the 1952 excavations a search was made for this well by running a trench over the site as shown on the Vavasour ground plan. At a depth of about four feet an area of gravel fill was encountered which, on exploration, proved to measure about 24' x 21'. Near the center of this fill area a number of large bounders were found. When these were cleared away a circle of stones forming the top of the well was revealed. There were 15 rocks, averaging about 13 inches in thickness, in this circle. The shaft opening was 5.2 feet in diameter.

The well was filled with large stream boulders, some of which were believed to have weighed between 300 and 400 pounds. These were removed and the well cleared down to water level, 25.6 feet below the collar and 29.8 feet below the present ground surface. As far as it was excavated, the circular shaft was lined with boulders.

Mr. Caywood, who directed the uncovering of the well, made some interesting speculations concerning the method of its original construction:

When the Hudson's Bay Company dug the well, they probably made an excavation nearly 17 feet square. This is indicated by the gravel fill area round the well. Then, from the bottom of this excavation they began to lay the stone well lining, thus building it up towards ground level. At the same time, they refilled behind the lining with clean gravel. When they were through, they had stone lined well shaft only 5.2 feet across and over 25 feet deep. The lining is still as sturdy today as it was then. [17]

Early in 1971 additional archeological testing was conducted in the vicinity of Well No. 2. The complete report on this work is not yet available, but Mr. J. J. Hoffman, in charge of the excavations, believes that earlier conclusions concerning the structural details of this well will have to be altered. "We now know," he reported on February 1, 1971, "that strangely shaped masonry structures lay deeply buried around the well shaft." [18]

No picture has yet been found that shows the aboveground structure of Well No. 2; neither were any remains reported as the result of the archeological explorations. Any reconstruction, therefore, will have to be based upon typical wells at other Hudson's Bay Company posts. Unfortunately, information on this subject seems scanty.

It was noted that in 1876 Fort Ellice possessed a "fine well" which was located near the chief factor's house. The only structural information given about this well, however, was that it had "the proverbial oaken bucket attached to a rope and chain." [19]

Recommendations

Although the archeological excavations of 1952 probably destroyed any traces which might have remained of the well house at Well No. 1, it is suggested that additional searching be done. The most promising areas appear to lie to the south and east of the 1952 excavations, which may not have extended far enough at this site to reach the footings.

It is also recommended that a search be made on and near the site of the Beef Store for possible remains of the well shown in the Emmons ground plan. It would also be desirable to explore the site of the device which seems to be a fire-fighting pump shown in the courtyard in the 1860 photographs. It is possible that a well might also be found in that location.

There seems little chance that further excavation at Well No. 2 will reveal important information about a structure which must have at least roofed that feature. On the other hand, such work seems to offer the only hope of learning anything about the aboveground appearance of the well.

1. Wilkes, Narrative, IV, 378.

2. Emmons, Journal, MS, III, entry for July 25, 1841. See plate III. It will be noted that this location was convenient to the Big House as it must have been placed when first moved down to the new fort site in 1829.

3. James Dwight Dana, Notebooks kept while Serving as Geologist on the U. S. Exploring Expedition, 1838-1842, MS, [III], [n.p.], in Beinecke Rare Book and Manuscript Library, Yale University Library.

4. Wilkes, op. cit., IV, 378.

5. Caywood, Final Report, 21.

6. Mr. Caywood supported the latter view and suggested that the abandoned 1841 well might be found by archeological testing at the site shown by Emmons. Ibid., 21-22.

7. Lowe, Private Journal, MS, 13.

8. Caywood, Final Report, 22. Vavasour definitely stated that the fort contained two wells, thus implying that there were no more at the time of his visit. Schafer, "Documents Relative to Warre and Vavasour's Military Reconnoissance," in OHQ, X (March, 1909), 85-86.

9. See plate LV.

10. Schafer, "Documents Relative to Warre and Vavasour's Military Reconnoissance, in OHQ, X (March, 1909), 85-86.

11. Report of a board of officers, January 23, 1854, in Br. & Am. Joint Comm., Papers, [IX], 104-106.

12. Susan Kardas, 1969 Excavations at the Kanaka Village Site, Fort Vancouver, Washington (typescript, Bryn Mawr College, May, 1970), 97-100.

13. Br. & Am. Joint Comm., Papers, [II], 118-119; Elliott, "British Values in Oregon, 1847," in OHQ, XXXII (March, 1931), 32-35.

14. Caywood, Final Report, 20-21.

15. Ibid., 21.

16. Caywood, Final Report, 44-45.

17. Caywood, Final Report, 22.

18. J. J. Hoffman, Memorandum to Chief, Archeological Investigations, Western Service Center, [Fort Vancouver National Historic Site], February 1, 1971, MS.

19. The Beaver, vol. I, no. 5 (February 1921), 16.