Fort Vancouver Historic Structures Report

History and location

Little is known about the history of the Powder Magazine at Fort Vancouver. The earliest mentions thus far encountered of the structure date from 1832, when the existence of a "stone" building for the storage of gunpowder was noted by members of the Wyeth party. [1]

During Outfit 1834 (mid-1834 to mid-1835) Amable Arquoitte, one of the trappers attached to the Indian trade conducted from Fort Vancouver, received a "gratuity" of £8 for "rebuilding Powder Magazine." [2] Whether this work involved changing the location of the building is not known. John Dunn, a Company employee during most of the 1830s, could have been describing the magazine before or after reconstruction when he said it was "built of brick and stone." [3]

The first definite knowledge of the location of the Powder Magazine is provided by the Emmons ground plan drawn on July 25, 1841 (Plate III, vol. I). It shows the "Magazine--the only brick building " situated in the extreme southwest corner of the fort enclosure as it was at that time. According to the plan, entry to the magazine was through a door in its north wall. Another visitor of 1841 also stated that the Powder Magazine was made of brick. [4]

To add to the confusion, when Lt. Mervin Vavasour of the Royal Engineers visited Fort Vancouver during late 1845 and early 1846, he reported to his superiors that the post contained one "small stone Powder Magazine." [5] His ground plan indicates beyond a doubt, however, that he was describing the same structure as Emmons's "only brick building" (see Plates VI, VII, VIII, vol. I).

This lack of agreement on the part of witnesses as to the material of which the building was constructed lasted long after the magazine had disappeared. Testifying in connection with the Company's claims for compensation for loss of its Oregon properties as a result of the boundary settlement of 1846, Thomas Nelson said that in 1851-52 the fort contained only one small brick building. [6] Former Company physician H. A. Tuzo, on the other hand, claimed that in 1853 he found the "fire-proof powder magazine" built of brick and stone. [7]

The list of observers who noted that the Powder Magazine was built of brick, stone, or brick and stone could be considerably expanded without shedding reliable light upon the situation. [8] Archeological excavations in 1947 proved that the witnesses who reported both brick and stone were correct. [9] They also demonstrated that the location in the southwest corner of the stockade enclosure as plotted by Vavasour was almost exact. This site is now designated Building No. 6 on the site plan of Fort Vancouver National Historic Site.

The Powder Magazine was still standing when the army assumed control of the Company's buildings at Fort Vancouver on June 14, 1860. The next day a board of officers pronounced the structure "useless to the public service." [10] Its subsequent fate is unknown. [11]

Construction details

General description. No pictures of the Fort Vancouver Powder Magazine are known. After the abandonment of the post by the Company, witnesses testified that the magazine was of "fire-proof" con struction, built of brick and stone, with an "arched roof" of the same material, and "copper doors." [12]

These meager facts, together with what is revealed by the actual foundation remains and by the Emmons ground plan, which shows that the single entrance was on the north side, constitute the sum total of what is known specifically about the Fort Vancouver Powder Magazine. Additional information that will be required before reconstruction drawings can be prepared must be obtained from what is known of similar structures at other Company posts of the same period.

Fortunately, several magazines, which are available for study. still exist at former Hudson's Bay Company establishments. Historic Architect A. Lewis Koue and this writer visited the excellent example at Cumberland House, Saskatchewan, and the large magazine at Lower Fort Garry, Manitoba. Photographs were obtained of the surviving structures at Metabetchouan Post on Lake St. John, Quebec, and at Moose Factory on Hudson Bay (Plates XXIX and XXX). An historic photograph of the wooden powder magazine at Fort Chipewyan is also useful (see Plate XXXI). From all of these specific and comparative sources certain hypotheses may be hazarded concerning details of the Fort Vancouver Powder Magazine.

Dimensions. The Vavasour ground plan of 1845 depicts the Powder Magazine as measuring about twenty feet square. The 1846-47 inventory gives its dimensions as l8 feet square. The stone foundations of the magazine, still partially intact, were uncovered during the archeological explorations at the fort site in 1947. The foundation walls were two feet thick and formed a square, each side of which measured about 19-1/4 feet, exterior measurement, and about 15-1/4 feet, interior measurement. [13]

In 1973 National Park Service archeologists again uncovered the magazine foundations. The foundation thickness, of two feet was confirmed, but this time the outside dimensions were reported to be twenty by twenty feet. [14] It would appear, then, that the historical record and the physical evidence are in reasonable agreement .

No information is available concerning the height of the Powder Magazine. It must have been a rather low structure, however, because it cannot be seen rising behind the pickets in any known view of Fort Vancouver.

Foundations. Archeologists in 1947 reported the Powder Magazine foundations to be composed of stones set in mortar made from coral. Unfortunately, beyond stating that the stones were not rounded boulders, the excavation reports contain no description of the rocks. They did not indicate whether they were shaped, partly shaped, or simply used as they were gathered. The 1947 photographs of the exposed foundations are of little help in this regard (see Plate XXXIII).

The final reports on the 1973 excavation are not yet avail able as this study is being written. However, preliminary descriptions by Project Archeologist J. J. Hoffman are illuminating. He states that the foundation was a "20 ft. square trench filled with rock rubble that was mortared with coral-derived lime." The masonry, he elaborated, was "local stone and British brick mortared with lime derived from Hawaiian coral." [15]

What may have been a piece of coral from that very shipment was found in the Powder Magazine foundation that presumably was rebuilt with the rest of the structure in 1834-35 (see Plate XXXIV).

Coral was still being imported for building purposes at Fort Vancouver in 1845. [18] And even in 1850 much coral and brick was brought to Oregon as ballast. [19]

Wall material. Archeologists in 1947 found pieces of broken brick on the Powder Magazine foundations. Their reports did not state whether this brick was loose or whether it was imbedded in the mortar. [20] This question seems to have been clarified by the findings of the 1973 excavations, already noted, that the foundation was composed of local stone and British brick mortared in lime.

This evidence would appear to support the eyewitnesses who said the magazine was built of brick or of brick and stone. Because stone was difficult to come by in the immediate vicinity of Fort Vancouver, it is probable that the walls (and vaulted roof) above the foundation were largely, if not entirely, of brick.

The archeological reports are undoubtedly correct when they describe the brick found at the magazine site as "British brick." It is known that bricks were imported from England to Fort Vancouver as early as 1825, during the first burst of building activity. [21] Subsequent importation is adequately recorded. [22] On the other hand, no bricks are known to have been made in the Oregon Country until about 1841. Furthermore, British statutes established the dimensions for bricks as 8-1/2 inches by 4 inches by 2-1/2 inches. [23] Presumably the fragments found in the magazine foundations had the same composition as other bricks having exactly those dimensions and recovered during excavations at Fort Vancouver. Examples of these bricks are in the artifact collections at Fort Vancouver National Historic Site.

Walls and vaulted roof. The surviving magazine at Cumberland House, the exterior measurements of which are about fourteen by fourteen feet, is not much smaller than was that at Vancouver. The stone walls at Cumberland House are two feet thick and rise without offset, at least on the outside, from the foundation. There may be an offset on the inside to support the floor joists, in which case the foundation would be an inch or two thicker than the walls.

The Cumberland House magazine has only a wooden roof, and the powder house at Lower Fort Garry, while it has a vaulted brick ceiling, is oblong in shape. Therefore, neither seems to provide a model for designing the "arched roof" of the square magazine at Vancouver .

It will be noted from photographs of fur trade magazines (Plates XXIX, XXX, XXXI, XXXII) that oblong structures seem generally to have had gabled roofs, while square structures appear ordinarily to have had pyramid-shaped roofs. At Lower Fort Garry, even though the oblong magazine has a vaulted brick ceiling, it is covered on the outside with a gabled, wood-frame roof.

If this same condition held true at Fort Vancouver, the vaulted ceiling must have arched inwards from the four walls to a central peak. Only such a construction would permit the installation of an outer pyramid-shaped roof. Of course, it would have been perfectly possible to have placed a gable roof on a square building, in which case the vaulted stone inner ceiling could have been a simple arch closed by brick or stone gables at each end.

Outer roof. Undoubtedly the magazine, despite having an arched brick or stone "roof," or ceiling, was further covered by an outer wood-frame roof. This practice, illustrated by the surviving magazine at Lower Fort Garry, would have assured the exclusion of moisture.

Such an outer roof probably would have had the same general appearance as that on the Cumberland House magazine (Plate XXXII). However, the existing roof on the latter structure is of recent origin and is framed according to present-day building practices. The Fort Vancouver roof undoubtedly was framed in the Canadian manner (see Plate LXXXI, vol. I) and sheathed solidly with planks as described in Chapter XI on the Sale Shop.

The distinctive feature of the roof, however, undoubtedly was the outermost layer of roofing. Because the building was described as "fireproof," it can be safely assumed that the roof was covered with interlocking sheets, or shingles, of tin as was common practice on the magazines at other Company posts (see Plates XXX and XXXI).

It is known that metal roofing was used at Fort Vancouver. During October 1831 Chief Factor McLoughlin made an "additional" requisition by the vessel Ganymede for 100 boxes of "Tin for covering Roofs." [24] During archeological excavations in 1950 and 1952 many tin sheets, twelve inches square with interlocking edges, were found at the sites of Well No. 1, the Wheat Store, and the Root House. Those at the well and Root House seemed to have been piled in as trash, and one later curator, working from the 1952 excavation field notes, gained the impression that at least some of these tin shingles may have come from the Powder Magazine. [25] The sheets at the Wheat Store, on the other hand, were ranged as if in place. Many were interlocked, with the hand-wrought, one-inch nails that held them to the roof still in place. [26] Archeologists in 1973 found a "large amount" of "shingling nails" within the magazine foundations. Presumably they were of the same type as those recovered at the Granary. [27]

The roofing excavated during 1950 and 1952 has since disappeared from the Fort Vancouver National Historic Site collections. It is hoped that future digging will uncover additional samples.

At Cumberland House and Lower Fort Garry the exposed wood surfaces at and under the eaves are likewise covered with metal. At Fort Garry this was accomplished simply by wrapping the tin roofing under the edges of the roof. At Cumberland House 20-gauge metal sheathing (probably lead) was placed over all unshingled wood surfaces in the manner shown in Plates XXXVI and XXXVII. See also Plate XXXVIII.

One other problem concerning the outer roof remains to be considered. It will be noted that the plan of the magazine foundation as excavated in 1947 shows the remains of round Douglas fir posts abutting the outsides of each of the four foundation corners. There were two of these posts at each of the southern corners, but only one each at the northern corners (see Plate XXVIII).

The archeologists who redug this site in 1973 reported that "large, rectangular wooden posts were set outside and flush to the foundation." [28] Obviously these were, in whole or in part, the same posts Mr. Caywood had uncovered in 1947, but the shape was interpreted differently. Project Archeologist J. J. Hoffman, who supervised the 1973 excavations, speculated that these timbers were '' remains of a wooden safety roof that rested on the posts rather than the brick walls." [29]

Such may well have been the case, though the use of exterior wooden posts to support the roof would not seem to have been a usual Company practice. Also, a wooden safety roof would not be compatible with the historical evidence concerning an arched roof; and unless the posts were clad in metal, they would have been contrary to the description of the building as "fire-proof."

This writer cannot offer any other logical explanation for the presence of the posts, which of course may not have extended above ground level. It is suggested that architects, when the final report on the 1973 excavations is available, explore all possible alternate uses before deciding to design a roof with exterior wooden supports.

Door and door frame. It has been seen that the door of the magazine was in the north wall. If it was like the one at Cumberland House, its threshold was of stone and practically at ground level. The opening in the wall at Cumberland House is two feet nine inches wide, and around it, except on the bottom, flush with the exterior wall. surface, is fitted a frame of 3-1/2-inch by 8-inch timbers, completely clad in metal sheathing. On the exterior surface of the frame this sheathing is tacked down at the outer edges by hand-forged spikes, spaced unevenly between three and four inches apart. The heads of the spikes vary in shape, some being roundish and some almost square. The shape of the frame and the means by which it is fastened to the walls are shown in Plate XXXVIII.

The door proper is composed of heavy wooden planks and is covered on all four sides and top and bottom by 20-gauge metal, seemingly iron. The metal is applied in overlapping sheets. Two joints where three sheets lap are visible on the front of the door. These joints are fastened by rows of spikes with larger heads than those on the door frame. The heads are about one-half inch in diameter and are spaced about three inches apart. There is a row of similar spikes down each side edge of the door (see drawing of door in Plate XXXVIII and photographs in Plates XXXIX and XL).

Most of the inner surface of the door is covered by a single sheet of the metal sheathing. Lapped over this sheet at the sides and top for perhaps eight to ten inches are the edges of sheets that have been wrapped around the top and sides from the front. [30]

The iron hinges and other iron hardware are well illustrated by Plates XXXVIII, XXXIX, XL, and by Plate XCII in volume I. The pintles for the powder magazine door hinges at Lower Fort Garry are of brass. [31] The door very probably was originally locked by a padlock of the type shown in Plate CIV, volume I.

In all likelihood the powder house door at Fort Vancouver was not unlike that at Cumberland House, except that the metal covering was of copper rather than iron. Brass pintles would have aided in eliminating sparks.

Floors. The floor of the Cumberland House magazine is made of two-inch by eight-inch wood plank flooring on wood sleepers (see Plate XXXVIII). It is possible that the powder house at Vancouver had a similar arrangement.

Furnishings

Probably the only "furnishings" in the magazine were the barrels and kegs of gunpowder stored there. Various indents and inventories from Company posts scattered across present-day Canada show that at different times the firm imported a fairly wide range of types of gunpowder. If we confine our attention to the Columbia District, however, the matter is somewhat simplified.

For Outfit 1838, for instance, the indent, or requisition, for the district called for only:

15 lbs. [barrels?] battle Powder
50 Bls. TPF Powder ea. 100 lbs.
44 half Bar. Cannon Powder ea 50 lbs. [32]

The inventory of goods on hand at the Fort Vancouver depot in the spring of 1844 listed the following types and amounts of gunpowder:

20/100 bbls. Canister Gunpowder
13-1/2 bbls. Cannon Gunpowder
97-15/100 bbls. TPF Gunpowder
27 Kegs TPF Gunpowder each 66-2/3 lbs. [33]

Thus, it will be seen that powder was received and stored in 100-pound barrels 50-pound half barrels, and 66-2/3-pound kegs. During November 1840 John Lee Lewes, in charge of Fort Simpson on the Mackenzie River, complained of certain defective powder kegs that had been shipped to him from Norway House. Other kegs, he said, those "bearing the Tower mark on them and copper fastened," were satisfactory, and he asked for only those in the future. [34] Probably the "TPF" powder kegs at Fort Vancouver were of this latter type.

Because, on the average, 62.4 pounds of gunpowder occupy only one cubic foot of space, it will be seen that none of the kegs and barrels stored in the Vancouver magazine were very large. Powder barrels made for the British Army around the 1840s had the following dimensions:

Undoubtedly the structure, small as it was, could accommodate the stock of powder kept on hand for use in the Columbia District, although it seems to have been Company practice to keep some powder in the sale shops and warehouses. [36]

Recommendations

a. When issued, the report or the 1973 excavations at the site of the magazine should be carefully studied for additional information on the dimensions of the structure and the types and shapes of stones used. The wooden posts outside the foundations also require careful examination to determine their possible function.

b. Unless new evidence to the contrary should be uncovered, the walls and arched ceiling of the reconstructed magazine should be of brick. The bricks should be reproductions of the original British bricks now in the collections at Fort Vancouver National Historic Site.

c. The outer roof should be covered with interlocking tin roofing of the same pattern as that used at Lower Fort Garry, unless further archeological excavations should produce samples of the actual tin roofing used at Fort Vancouver, in which case the samples should be copied.

d. The outer roof should have the form and appearance of that on the surviving magazine at Cumberland House, but the framing should be of the old Canadian type, not the recent form now on the Cumberland structure. The underside of the eaves should be covered with metal.

e. The door should be patterned on that at Cumberland House, except that it should be covered with copper instead of iron. Hard ware should be of brass where practicable and fastened with spikes or rivets and not screws. The padlock should be a copy of an actual specimen recovered at the fort site or of an original one of similar type.

f. It is suggested that archeologists remain alert for samples of the tin roofing used at Fort Vancouver and that further efforts be made to determine the fate of those reported by Mr. Caywood.

1. F. G. Young, ed., The Correspondence and Journals of Captain Nathaniel J. Wyeth, 1831-6, Sources of the History of Oregon, vol. 1, pts. 3-6 (Eugene, Oregon, 1899), p. 176.

2. H.B.C., District Statements, York Factory, 1832-1835, H.B.C.A., B.239/1/5, MS, p. 141. Arquoitte may have been a free trapper, because he was not credited with any wages during the year.

3. Dunn, Oregon Territory, p. 102.

4. Nellie Bowden Pipes, ed. and trans., "Translation of Extract from Exploration of Oregon Territory . . . Undertaken During the Years 1840, 1841 and 1842 by Eugene Duflot de Mofras," Oregon Historical Quarterly 26 (June, 1925): 153.

5. Joseph Schafer, ed., "Documents Relative to Warre and Vavasour's Military Reconnaissance in Oregon, 1845-6," Oregon Historical Quarterly 10 (March, 1909): 85.

6. Br. & Am. Joint Comm., Papers, [9]:90.

7. Ibid., [2]:176—77.

8. For further examples, see ibid., [11]:74; Thomas Jefferson Farnham, Travels in the Great Western Prairies, the Anahuac and Rocky Mountains, and in the Oregon Territory (Poughkeepsie, N. Y., 1841), p. 194.

9. Louis R. Caywood, Exploratory Excavations at Fort Vancouver, 1947, mimeographed ([San Francisco:] United States Department of the Interior, National Park Service, [1947]), Plate 3 (Plate XXVIII in this vol).

10. Br. & Am. Joint Comm., Papers, [9:]75-77.

11. During archeological excavations in 1952 a number of tin shingles were found at the Root House site. It was believed that these may have been from the roof of the Powder Magazine and that they may have been dumped as trash when the army demolished that structure. Alan Cherney, "Cataloguing Artifacts, Vancouver Vault, January 2, 1966-April 19, 1968," typescript (Vancouver, Washington: National Park Service (Fort Vancouver National Historic Site), 1968, p. 19.

12. Testimony of H. A. Tuzo , in Br. & Am. Joint Comm., Papers, [2:]176-77; Testimony of D. Mactavish, in Br. & Am. Joint Comm., Papers, [11:]74.

13. Caywood, Exploratory Excavations, p. 9 and Plate 3.

14. J. J. Hoffman, Memorandums to Regional Archeologist, Pacific Northwest Region, National Park Service, Fort Vancouver National Historic Site, July 2 and August 1, 1973, MSS, in files of Pacific Northwest Regional Office, Seattle.

15. Ibid.

It was many years after the establishment of Fort Vancouver before a convenient local source of lime was discovered. [16] At least as early as 1832 the Company was importing coral from the Hawaiian Islands for building purposes. In that year Duncan Finlayson shipped McLoughlin enough coral from Oahu to make 300 barrels of lime , and with the shipment he sent instructions for burning the coral in a kiln. [17]

16. A specimen of limestone from the Willamette Valley was brought to Fort Vancouver in 1839, but evidently the source was not deemed satisfactory, because the post continued to import coral from the Hawaiian Islands. William Fraser Tolmie, "Diary," Washington Historical Quarterly 23 (July, 1932): 214.

17. H.B.C.A., B.2 23/b/8, MS, fol. 42d.

18. Henry J. Warre, "Travel and Sport in North America, 1839-1846, by General Sir Henry J. Warre, K.C.B," ed. Felix W. Warre, MS, in Public Archives of Canada, Ottawa, p. 109.

19. D. H. Vinton to Maj. Gen. T. S. Jesup, Washington, March 29, 1850, in U.S., Congress, House, 31st Cong., 2d sess., Exec. Doc. No. 1, pt. 2 (Serial 595), p. 251.

20. Caywood, Exploratory Excavations, Plate 3; Caywood, Final Report, p. 11.

21. H.B.S., 4:1.

22. For example, see H.B.C.A., B.239/n/71, MS, fols. 155-155d.

23. Hoffman and Ross, Fort Vancouver Excavations--I, pp. 58-65, 75.

24. Barker, Letters of Dr. John McLoughlin, p. 233.

25. Cherney, "Cataloguing Artifacts," p. 19.

26. Caywood, Final Report, pp. 41-42.

27. J. J. Hoffman, Memorandums to Regional Archeologist, July 2 and August 1, 1973.

28. J. J. Hoffman, Memorandum to Regional Archeologist, August 1, 1973.

29. Ibid.

30. Cumberland House magazine details are based on a field visit, September 16, 1967.

31. Visit to Lower Fort Garry National Historic Park, September 20, 1967. The door itself is missing.

32. H.B.C., York Factory Indent Book, 1823-1838, H.B.C.A., B.239/n/71, MS, fol. 158.

33. H.B.C.A., B.223/d/155, MS, p. 102.

34. Glazebrook, Hargrave Correspondence, p. 325.

35. F. A. Griffiths , The Artillerist's Manual, and British Soldier's Compendium, 2d ed. (Woolwich, England: Printed by E. Jones for the author, 1840), p. 85.

36. According to a report of 1874, powder was stored in the old stone store at Fort William. [National Heritage Limited,] Fort William, Hinge of a Nation ([Toronto, 1970]), p. 39; Gunpowder was stored in the warehouse of Fort Simpson on the Northwest Coast in 1868. Teichmann, A Journey to Alaska, p. 108.