Chapter Eleven: Lead Seals by J. David McMahan
On February 28, 1818, RAC Chief Manager L. A. Hagemeister, seeking to establish more disciplined and businesslike practices in Russian America, presented the following proposal to the New Arkhangel office:
I note from the accounts that there is often inaccuracy and confusion because of variations in terms denoting different receivers of furs. In order to avoid this, and to fulfill orders of the Main Office to my predecessor, the office will distinguish furs by stamping [them] as follows: The seal with the inscription "RAK" [i.e., RAC] must appear on all company furs. To indicate where they originate, depending on where the furs are collected, order to be cut on the seal under the brand [kleimo]: ...[Hagemeister, in Pierce 1984:39-40].
Hagemeister completed the above proposal by illustrating various
symbols, along with their meaning, and stressed the need to keep
the meanings of the symbols secret. The use of lead seals to close
confidential communications, to authenticate documents, and to
tamper-proof packages was well established prior to Hagemeisters
communication. Specimens technologically similar to those of Russian
America are known from the ancient Roman and Byzantine empires,
where they were used to close confidential communications from
leaders in government, church, military, and commerce. Modern
day counterparts, generally made from less toxic substances such
as aluminum or plastic, now are used to tamper-proof electrical
meters and cargo containers although lead is still used
in some instances. The Castle Hill collection includes 30 lead
seals (Figures 11.1-11.4). All but three are considered RAC fur
bale seals. One of the non-inclusive specimens has been identified
as a British wool packers seal. Two others are incomplete,
but are probably also cloth packers seals based on the style
The use of lead seals in the fur trade was an industry standard during the 18th and 19th centuries. They have been documented on British and French sites in the eastern U.S. and on Hudsons Bay Company posts in the Pacific Northwest. Lead seals are known from other Russian-American Company posts, but are relatively rare. Shubin (1990:444) illustrates three lead seals from the RAC Kurilorossiia settlement in the Kurile Islands. Two examples are known from the Erskine House collection at the Baranof Museum, Kodiak, and four examples are known to be in the Sitka National Historical Park collection. Alice Ryser and Sue Thorsen, respectively, supplied photos of the specimens in these collections for comparison. The Castle Hill collection, which includes both blanks and finished examples, presents an opportunity to examine a sizable collection from recorded context. Attributes of the Castle Hill lead seals are presented in Table 11.1.
Prior to analysis, all seals were scanned at high resolution
on a flat-bed scanner to provide images for inventory purposes.
The seals were photographed individually and in groups under controlled
cross-illumination with high intensity carbon arc lamps. The resulting
images, on tungsten balanced slide film, were commercially digitized.
During analysis, each seal was examined under a low power dissecting
microscope with direct and cross-lighting. Dimensions, weights,
details of manufacture, and markings were recorded in a database
that was distilled into Table 11.1 above. One of the seals was
treated in multiple dilute acetic acid baths to reveal surface
details. Before and between treatments, the specimen was photographed,
rinsed in distilled water, dehydrated in isopropynol, and examined
as described above. All other seals were cleaned by gently brushing
in distilled water to remove adhering soil and loose lead oxide,
but not the patina.
|Table 11.1 (pdf)|
Figure 11.1. Lead seals from Castle Hill (1 of 4 plates).
Figure 11.2. Lead seals from Castle Hill (2 of 4 plates).>
Figure 11.3. Lead seals from Castle Hill (3 of 4 plates).
Figure 11.4. Lead seals from Castle Hill (4 of 4 plates).
Results of Analysis
Fur Bale Seals:The Castle Hill bale seals range in diameter from 160mm to
342mm, and in thickness from 3.0mm to 11.7mm. The average bale
seal is the approximate size of a U.S. nickel (22mm), although
somewhat thicker (4.8mm) and considerably heavier (around 11.2
grams)(Figure 11.5). In most cases, ovoid or ellipsoid apertures
on opposing edges provide evidence of a longitudinal channel through
which cordage or wire was passed. Although, compression and occlusion
has rendered many of the apertures barely visible. With the exception
of two textile packer seals, the seals are believed to relate
to the RAC fur trade. Divergent sizes and surface markings suggest
that a few of the seals may be associated with specialized functions
or shipments. Other seals, devoid of markings but technologically
similar to embossed specimens, are believed to be blanks.
Figure 11.5. Representative fur bale seal from Castle Hill (97-6368).
All identifiable bale seals, regardless of size, were manufactured in the same manner. Like those of the ancient Roman and Byzantine empires, blanks were cast in a mold, leaving a hollow channel from edge to edge. The lead blanks varied in size, but were roughly equivalent to that of a coin. The cordage or string used to secure the parcel or bale was passed through the channel and then knotted. The lead blank was then placed between the jaws of a boulloterion, a pincer-like implement with disc-shaped jaws engraved with an inscription and/or image. The boulloterion had a projection above the jaws that was struck with a hammer to compress the lead blank, sealing the channel upon the string and stamping the inscription or image on the lead blank (Whitlow 1996). In developing a typology for the lead seals from Fort Michilimackinac, Stone (1974:281) used the term "series C" to describe seals of this style.
The most common marking on the Russian bale seals was the "PAK" [RAC, or Russian-American Company] described in Hagemeisters letter (Pierce 1984:39-40). Almost half of the bale seals in the collection exhibit this inscription on one surface (Figure 11.6). Five of the "PAK" seals were stamped with the imperial eagle on the opposing side, and nine of the "PAK" seals exhibited additional script on one or both surfaces. Two "PAK" seals (97.35 and 97.3863) included the Russian word meaning "seal" beneath the company logo, while another exhibited incomplete text that appears to translate "Yakut[sk]" (translations courtesy of Katya Solovjova Wessels, U.S. National Park Service). Two others included the apparent source codes "x" and "+" for Kodiak and Sitka, respectively (Pierce 1984:39-40). Two seals were inscribed with "M" to indicate "sea otter, 2nd grade," while three were inscribed with the Russian "g" to indicate "sea otter yearling, 2nd grade" in accordance with Hagemeisters scheme (Pierce 1984:39-40). One incomplete seal (97.6375) exhibits Russian text that translates "... region" and another (98.128) includes combinations of Russian letters that could not be translated.
Stylistic differences in inscriptions and design are apparent in the Castle Hill bale seal collection, and not all markings can be interpreted on the basis of Hagemeisters scheme. We know from subsequent correspondence between Hagemeister and the New Arkhangel office that a prior system for marking bale seals was in place under Baranov:
Supplementing my proposal of 28 February,
No. 63, I request that the office ship old furs surrendered by
[postupivshiia v zdachu] Mr. Baranov which already have the old
stamp as is, while the newly acquired [furs] be stamped with
the new ones which for this [purpose] are already made; in invoices
the record should be, for example: sea otters, marked above by
[PAK +], with N, M, and so forth below.
The Castle Hill seals offer few conclusions on the basis of stratigraphic context, as most (n = 21) are from the upper (mixed) level of the workshop complex (Stratum I). These only can be assigned to the circa 1805-1867 period. Three seals, from a trash deposit (Stratum IIa) associated with buildings in the workshop complex can be assigned to the 1805-1840 period. Four of the seals (two each) were recovered from deposits associated with Buildings 1 and 4, and can be seriated on the basis of mean ceramic (MCD) and terminus post quem (TPQ) dates for the buildings. This suggests with a high level of confidence that specimens 98-129 and 98-5718 (Bldg. 1, circa 1827) predate specimens 98-1294 and 98-7825 (Bldg. 4, circa 1839). Stylistic differences in design and the size/style of lettering are apparent. The sample size is too small, however, to establish whether or not differences are a result of change through time or within the range of diversity for boulloterions in use at any given time.
Eight seals revealed evidence of organic residue within the
line holes when viewed under magnification. Residues from three
of these seals (98-129, 98-8778, and 98-6374) were microscopically
identified as coir (coconut husk fiber)(Figure 11.7). Another
seal (98-127) has a section of ferrous wire in the line hole,
and a second (98-8780) exhibits ferrous staining from a wire.
It is not surprising that coir cordage was used, along with wire,
as a binding for the fur bales. Coir is more resistant to saltwater
than jute, cotton, or hemp. It was readily available, as evidenced
by coconuts and husk fibers in the Castle Hill deposits. Litke
(1827:58) lists "cord made from coconut husks" as a
product obtainable from the Sandwich Island trade.
Figure 11.6. Drawings of select lead seals from Castle Hill (drawings by Margan Grover).
Figure 11.7. Photomicrographs of coir (Cocos nucifera) fibers in lead seal channels.
Cloth Seals: One seal (98-130) has been identified as a cloth packers
seal (Figure 11.8). One side is embossed centrally with the word
"PACKER." Around the top edge of the seal is the embossed,
block-lettered inscription "THOMAS GAN[?]ELL" and around
the bottom edge is an embossed inscription that begins "RO
but is otherwise illegible (alternatively, see Egans translation,
below). The opposite side of the seal bears the hand-scratched
inscription "12935" above "22.3." Two other
specimens (98-24875 and 98-2520), are fragmentary and devoid of
legible markings. They are believed to also be cloth seals on
the basis of style. The latter specimen (not in photo plates)
is the outer washer-shaped disc of a two-part or four-part seal.
Both seals were recovered from disturbed deposits (Level I) which
contained materials dating from 1805-present.
Images of the packers seal were examined by Geoff Egan (British Museum, Department of Medieval and Later Antiquities) and Wiard Krook (Archaeological Department, City of Amsterdam), who have researched European lead seals. Both Egan (e-mail, 4/5/01) and Krook (e-mail, 1/25/01) agree that the seal is British, but neither can identify specimens in their collections with similar markings. From a high-resolution photo, Egan provided the following:
Thanks for the photos - I can now see what is going on. 13925 may well be a consignment number; 22.3 is probably 22 yards and three quarters length; I read (with some certainty) Thomas Daniell, Packer, London - - this individual could as likely as not be listed in contemporary trade directories for London, several of which from the early C19th survive in the Guildhall Library here in the City. Regret I do not at present have the time to sit down and search this source for you, but in the absence of documentation at your end, it is a possible lead to bear in mind [Geoff Egan, e-mail, 4/5/01].
Figure 11.8. Cloth packer seal (98-130) from Castle Hill.
In the U.S., packers seals are known from the 18th and early 19th centuries. Packers' seals have been reported from Fort Michilimackinac in Michigan (Stone 1974:286-289; Adams 1989:28), as well as a British fort on the Niagara River in New York (Calver and Bolton 1950:268). Egan (1995:91-92; Fig. 185, #266), in his discussion of lead cloth seals in the British Museum, describes packers seals as follows:
Packers were those who, in the present context, made sure cloths were packed into proper parcels fit for exportation, and some became international traders in their own right A number of packers seals are known, with a full personal name, as on the one below [Fig. 185, #266]. They appear mostly to be (and may all be) of eighteenth/early nineteenth-century date. Several include London in the stamp (no other placename has so far been recorded on this category of seal).
Cloth seals easily are distinguished from fur bale seals based
on the manner in which they were made. Whereas fur bale seals
were cast as a single disc-shaped blank with line channels extending
from edge-to-edge, cloth seals were cast in stone molds as either
two-disc or four-disc blanks which were joined by narrow strips
(Figure 11.9). This allowed the discs to be folded over on each
side of the cloth before being stamped with one or between two
dies. One of the discs was cast with a rivet that could be pushed
through the fabric and into a corresponding hole in the other
disc, apparently a system that was devised specifically for marking
commercial textiles (Egan 1995:4). The use of lead seals to mark
textiles destined for commercial sale was a widespread practice
in the cloth-producing countries of Europe between the 13th and
19th centuries (Egan 1995:1). They were used within the framework
of industrial regulation as evidence of quality control and/or
payment of taxes.
In developing a typology for the lead seals from Fort Michilimackinac, Stone (1974:281) used the term "series A" to describe seals of this style. Adams (1989) conducted a detailed analysis of 25 lead seals from Fort Michilimackinac, 18 of which were series A, and concluded that most of the Michilimackinac specimens were cloth seals. It should be cautioned, however, that this does not hold true for all series A seals. Some researchers have long held that the lead seals commonly referred to in the archaeological literature as "bale seals" were used to seal and identify the contents of packaged goods (Stone 1974:281; Steer 1977:122). For example, an intact series A lead seal was attached to a bundle of files recovered from the Winnipeg River during an underwater archaeological project (Wheeler et al. 1975:62). The Winnipeg seal is stylistically similar to the Castle Hill specimen (98-130), with a stamped logo ("WA") on one side, hand-inscribed numbers on the opposing side, and intact textile fibers sandwiched between the discs. The hand-inscribed numbers ("5339" over "21 ½") undoubtedly refer to a lot number and 21 ½ yards of cloth, following Egans understanding of lead seals (described in the e-mail cited above). It is conceivable that the use of the Winnipeg seal to secure files was a secondary use of the seal, or perhaps the files were wrapped in trade cloth for transport.
Figure 11.9. Schematic drawing of two- and
four-part cloth seals, adapted from a drawing
by Nick Griffiths
The packers seal from Castle Hill is important in that
a small quantity of wool cloth is preserved in the fold of the
lead strip. The identification of the cloth as wool was confirmed
by microscopic examination of fiber samples from the specimen
(Figure 11.9). This has provided evidence for the origin (i.e.,
the packing location) of some of the wool supplied to the Russian-American
Company, although not the route or series of suppliers through
which it passed. Intensive analyses of wool samples from the Castle
Hill textile collection, including the characterization of fibers
based on scale patterns, diameter, and other attributes may potentially
produce insights on varieties and sources of the raw wool supplied
to Russian America.
Of 30 lead seals recovered at Castle Hill, all but three are
fur bale seals comparable to other examples from Russian America
(i.e., from the Kurilorossiia settlement in the Kurile Islands;
from the Erskine House in Kodiak; and from the Bishops House,
and a burial in the lawn of St. Peters-by-the-Sea church in Sitka).
Common design elements include the imperial eagle and "PAK"
on opposite sides. Other markings include Russian letters that
symbolized the type, quality, and source of the furs in the bale.
Some of these are consistent with instructions given by Hagemeister
in 1818 (Pierce 1984:39-40, 49). Others cannot be matched with
Hagemeisters symbols, and may relate to marking practices
under Baranovs administration. Because stratigraphic associations
are tenuous for most of the seals, their design elements cannot
be seriated on the basis of age. Four of the seals (two each),
however, can be firmly associated with Building 1 (circa 1827)
and Building 4 (circa 1839 or later).
The lead seals constitute an interesting and important part of the overall assemblage. Their analysis has contributed to our understanding of daily work activities and trade relations at Castle Hill, such as the use of coir bindings, along with wire, to secure the bales. The manner in which the seals were manufactured, like the colonial architecture of the workshop buildings, suggest a continuation of traditional material culture in combination with the more cosmopolitan aspects of a seaport settlement.