97-532 CbGu-5
97-285 CbGu-5
Jonathan Ferguson
This section is intended to describe and analyze the ceramics recovered from the historic components of the LaVase River excavations conducted by the Corporation of the City of North Bay in 1997 and Settlement Surveys Ltd. in 1992. Only fineware is examined here, as no other ceramic types were recovered.
In 1992 and 1997, the LaVase Island Site yielded 24 ceramic sherds, 16 of which belong to identifiable types. For a small trading post such as the LaRonde Post, large numbers of fineware are not to be expected. High cost and the chance of breakage during transportation made fineware vessels valuable yet impractical personal items. Such is the case at the North West Company and Hudson's Bay Company post at Rocky Mountain House, Alberta, which dates to the nineteenth century. On that site, the recovered ceramic assemblage included only nine sherds (Noble 1973: 117-18).
The dates provided for the various fineware types must be considered while keeping certain cautions in mind. A range of dates of manufacture, while providing a terminus post quem for the introduction of that type, cannot be seen to give a rigid terminus ante quem. The 'heirloom factor' (usually around fifty years) must never be discounted; that is, early ceramics are sometimes seen at later sites. The dates of greatest popularity are derived through the cumulative seriation of archaeological collections recovered from a selection of eastern North American historic sites (Grange 1977: 15, 34).
Reference will here be made to similar ceramics from late eighteenth and nineteenth century sites, which may serve to further place the LaVase ceramics in their historical context. Special attention should be paid to parallels from the Garden Island Trading Post Site (CbGx-9), which was operated by the Hudson's Bay Company between 1821 and 1848 (Dibb and Sweetman 1995: 6). As the immediate successor to the LaRonde Post, this fur trading post is of particular relevance to this discussion. Unfortunately, little information concerning this site is available.
Finds from the other LaVase River excavations, conducted by Archaeological Services Inc in 1995 and Laurentian University in 1996, will be cross-referenced where it appears that sherds are derived from the same vessel or share significant similarities.
BANDED WARE
Cat. Nos. CbGu-5 97-285, 97-532
97-532 CbGu-5 |
97-285 CbGu-5 |
Two blue banded pearlware base sherds are among the artifacts recovered from the LaVase Island Site in 1997. It seems most likely that these sherds come from the same vessel, which was apparently a ring-footed saucer. The paste, design, glaze and blue ring decoration of these sherds are macroscopically identical. The only irregularity is that sherd 97-285 has only one blue ring, but extends to a height at which sherd 97-532 exhibits a second band. This dissimilarity can be explained by a slight deviation in the brush-stroke while the outer band was being painted.
Blue banded pearlware was manufactured from 1780 to 1830, giving a mid-range production date of 1805 (Grange 1977: 70). This range coincides well with the expected period of occupation by the LaRondes on Bothwell Island.
Possible parallels are found on other contemporary Northern Ontario sites, but this type of decoration too often receives only cursory examination. It is often not clear, for example, whether researchers are referring to bands applied by painting or slipping. One bowl from Lower Fort Garry, operated by the Hudson's Bay Company between 1830 and 1911, appears to have the same type of painted blue banding as was found at the La Vase Island Site (Sussman 1979: 154, cf. Figure 129). A similarly decorated blue banded rim sherd was found at the 1782 to 1923 HBC post at Martin's Falls, Ontario (Vyvyan 1980: 146, cf. Figure 71d). Banded ware recovered from the Garden Island Trading Post Site include examples painted in brown or black, or both (Dibb and Sweetman 1995: 25-27).
One hand-painted, blue banded body sherd (Cat. No. 4115) was recovered on the LaVase Island Site during the 1995 field season (Archaeological Services Inc. 1996: Appendix 4). No further information is available to determine if this sherd might have belonged to the same vessel.
SHELL-EDGE WARE
Cat. Nos. CbGu-5 5-92-52, 5-92-54
 5-92-52 CbGu-5 |
Five sherds, found on Bothwell Island during the 1992 Settlement Surveys Ltd. test excavation, have been identified as blue shell-edge pearlware. Catalogue number 5-92-52 includes two rim sherds which almost certainly belong to the same plate. The rim of this plate was straight (i.e. unscalloped) with shallow, slightly curved impressions about 10 mm long. Blue paint was applied to the lip and bled c. 4 mm down the grooves. Three exfoliated body sherds were collectively catalogued as 5-92-54. One of these sherds mends with one of the rim sherds; further similarities in paste and glaze strongly suggest that all five sherds derive from the same plate.
Blue shell-edge pearlware was produced between 1780 and 1830, with a mid-range date of 1805. The popularity of this design peaked around the year 1800 (Grange 1977: 27-28, 70). Because of the demand for this style, measures were taken to increase productivity. The earlier technique of applying the paint downward with the brush in order to produce a feathered appearance is seen on samples from 1780 to about 1795. After 1795, however, this time-consuming practice was abandoned in favour of the more efficient (if less aesthetic) method of sweeping the brush around the rim of the vessel, perhaps on a lathe (Reid 1977: 46-49). As the plate in question appears to have been produced with this later technique, it may be more accurately dated to between 1795 and 1830.
The popularity of this design ensures that it is often found on archaeological sites and sometimes in considerable quantities and varieties, as at Signal Hill National Historic Site (1800 to 1860) in St. John's, Newfoundland (Jelks 1973: 63-66). However, because decoration is found only around the rim, body sherds are often misidentified as plain ware. This appears to be the case at the early nineteenth century Ermatinger House in Sault Ste. Marie, Ontario. All edge ware specimens from this site belong to the post-1795 type (Reid 1977: 46-49). At least one unscalloped blue shell-edge ware rim sherd was also recovered from the LaVase Park Site (CbGu-4) in 1996. The paint of this sherd (Cat. No. CbGu-4 96-1) had been applied in the 1780 to 1795 downward-brushing technique (Patrick Julig, Laurentian University 1998: personal communication). The historic component of this site has been interpreted as a farmstead; bricks recovered from the surface have been dated to after 1855 to 1860. This site, located on Lake Nipissing on a point 2 km south of the LaVase River, is not known to have been associated with the LaRonde Post (Archaeological Services Inc. 1996: 84). Both blue and green edge ware sherds were recovered from the Garden Island Trading Post Site (Dibb and Sweetman 1995: 25-27, cf. Plate 10). None of these sherds, however, appear to belong to the same set as those recovered from the LaVase Island Site.
Three blue edge ware rim sherds were found on Bothwell Island in 1995: one straight edge (Cat. No. 4022), one straight edge with 'chicken-claw' impressions (4116) and one possibly scalloped edge (4321) (Archaeological Services Inc. 1996: Appendix 4). More information on these sherds is needed before any comparison can be carried out with the 1992 finds.
TRANSFER-PRINTED WARE
Cat. Nos. CbGu-5 5-92-53, 97-470, 97-481
 97-481 CbGu-5 |
Transfer-printed wares are represented by three examples recovered from the LaVase Island Site. Two blue and one red-purple transfer-printed sherds were found during the 1997 and 1992 seasons, respectively.
Catalogue number 97-481 represents a blue transfer-printed pearlware rim sherd from a plate. The decoration includes a roundel along the shoulder, which appears to contain a landscape scene. The extant half of this scene includes a terrace or balustrade upon which stands an antique vase. In the background can be seen a rooftop and treetops. This roundel is framed by sinuous floral tendrils, which extend into the geometric band around the inside of the plate. This geometric band includes, from top to bottom, a thin band, stippling, a row of dots, a dotted diamond motif, the tendrils, more dotted diamonds, another row of dots, stippling and another thin band. Below this last band is found a stylized floral motif, extending into the white field of the base. The dotted diamond pattern is often seen with the popular willow pattern (cf. Jelks 1973: Figure 72b-d), but otherwise, this sherd does not appear to belong to that style. The under surface of the sherd is undecorated, but includes a ring footed base.
 97-470 CbGu-5 |
 |
Sherd 97-470 (included in the catalogue along with the brown stamped and banded sherd, discussed below) is a blue transfer-printed pearlware base sherd, most likely from a plate. The upper surface depicts a figural or landscape scene (i.e. not exclusively floral), the general content of which is, unfortunately, impossible to ascertain because of the very small size of the sherd (maximum length is 22 mm). A bare forearm and hand extend into one side of the sherd. On the other is depicted what appears to be a truncated Ionic column, complete with voluted capital and base, and fluted drums. The under side of this sherd is of more interest, for it includes an impressed maker's mark which may help identify its period of manufacture. Central to this design is an stamped anchor, without the intertwined rope which is occasionally encountered. Above the anchor is found a double-crescent shaped impression which may be an illegible inscription. Below the anchor is another inscription; under magnification and oblique lighting, its surviving portion appears to read "GRANI". Most likely, this once formed part of the word "GRANITE", which would have been properly centred with the anchor. Despite this interpretation, it must be noted that the material of this sherd is pearlware, not granite ware. The presence of the anchor design suggests (but does not prove) that this vessel was the product of the Davenport potteries in Longport, Staffordshire, England (Godden 1964: 189).
These two examples of blue transfer-printed pearlware can be dated to between 1795 and 1840, the period of manufacture for this type. The popularity of this decorative technique reached its highest peak around 1813, somewhat earlier than its mid-range date of production in 1818 (Grange 1977: 28, 70). This date range can perhaps be somewhat narrowed for the impressed sherd (97-470), because Davenport marks were inscribed in upper-case letters only after 1805 (Godden 1964: 189). However, because it is not certain that this sherd represents a Davenport, this narrower range will not be used for dating purposes.
 5-92-53 CbGu-5 |  5-92-53 Reverse |
Catalogue number 5-92-53 is a red-purple transfer-printed rim sherd, probably belonging to a bowl. The inside surface includes, from top to bottom, stippling just under the rim, a leafed tendril, an open white space and then another area of stippling. The outer surface of the bowl includes a band of stippling along the rim, with a large leafy floral motif underneath.
Coloured transfer-printed pearlware is generally dated later than its blue counterpart. The manufacturing date range for miscellaneous light blue (which is distinct from the dark blue seen above) and coloured transfer-printed pearlware is from 1818 to 1864. The midpoint for this date range is 1841 (Grange 1977: 70).
Transfer-printed ware, especially blue, is commonly found on nineteenth century sites and often comprises the majority of decorated sherds recovered. This pattern is true at both the 1800 to 1860 component of Signal Hill National Historic Park in St. John's Newfoundland (Jelks 1973: 66) and at the early nineteenth century Ermatinger House in Sault Ste. Marie, Ontario (Reid 1977: 43-46). The Garden Island Trading Post Site yielded blue, green and brown transfer-printed sherds, including possible flow blue patterns. Of particular interest is a printed maker's mark from the Copeland and Garrett potters, which can be dated to between 1833 and 1847 (Dibb and Sweetman 1995: 25-27 cf. Plate 10).
IRONSTONE
Cat. No. CbGu-5 97-517
 |
 |
Only one example of moulded ware was recovered from the LaVase Island Site during the 1997 field season, the two mended sherds of 97-517. Judging by its matrix, these sherds appear to be ironstone. In form, this specimen appears to be a deep saucer or small bowl, surviving in part from the rim to the base. On the inside surface, two thin, moulded rings are found immediately below the rim, while a moulded floral tendril pattern extends between the rings and the base. This motif continues around the rim, in a repeatedly mirrored pattern. The base includes a simple ring foot.
As a general class, ironstone was manufactured over a relatively long period, between 1813 and 1900. The mid-point of this range is 1857 (Grange 1977: 70).
The problems in distinguishing between stoneware, ironstone and refined white earthenware have already been discussed by researchers, but no absolute criteria have been identified (Jelks 1973: 58-59). As a result, comparisons are often difficult to draw between different sites. The 1800 to 1860 component at Signal Hill National Historic Park in St. John's, Newfoundland, yielded a number of ironstone plates, bowls, cups and saucers, of which some bore moulded geometric or floral motifs (Jelks 1973: 69). Three late ironstone vessels were identified at the Ermatinger House in Sault Ste. Marie, Ontario (Reid 1977: 49-51). Late nineteenth century examples of moulded white hard-paste earthenware were found at Lower Fort Garry, Alberta (Sussman 1979: 12-15, cf. Figures 1-3).
Two sherds were found on Bothwell Island during the 1995 excavations which may correspond to the bowl described above. Catalogue number 3801 represents a moulded refined white earthenware rim sherd from a plate, which also includes turquoise transfer-printed decoration. The moulded handle from a refined white earthenware vessel (3802) was also unearthed (Archaeological Services Inc. 1996: Appendix D). The latter sherd, if it should prove to be better described as ironstone, might have belonged to the same set as the saucer found in 1997; the same could be true if 97-517 is in fact refined white earthenware.
SPONGED WARE
Cat. Nos. CbGu-5 97-512, 97-523
 |
Three sponged ware sherds (2 rim and 1 body) were recovered during the 1997 field season, all from Bothwell Island (CbGu-5). While these sherds do not mend, their consistency suggests that they likely belonged to the same vessel, which appears to have been a bowl.
One sponged ware rim sherd (CbGu-5 96-130) found in 1996, also on the island, exhibits identical decoration (Patrick Julig, Laurentian University 1998: personal communication). It seems almost certain that the 1996 find represents part of the same bowl postulated for the above sherds.
All the sponged ware is imprinted with a blue specked pattern and both rim sherds have a blue band painted around the rim. Decoration is only found on the interior surface. Such a combination of blue sponging and banding is also found on most sponged vessels recovered from the Ermatinger House, an early nineteenth century mansion in Sault Ste. Marie, Ontario. This decorative pattern was first produced by Scottish potters, but was also manufactured in Turnstall England by William Adams, from about 1820 to 1850 (Reid 1977: 49).
One blue sponged sherd (Cat. No. 4021) was also recovered on the island during the 1995 season (Archaeological Services Inc. 1996: 137, Appendix D). Unfortunately, a more detailed description of this sherd is not available.
STAMPED WARE
Cat. No. CbGu-5 97-470
 97-470 CbGu-5 |
Only one stamped ware sherd was unearthed during the 1997 field season, again on Bothwell Island. The interior surface of this rim sherd shows a brown floral stamped motif immediately below a brown painted band near the lip. This sherd probably belonged to a saucer.
This stamped pattern appears to be identical to that of a pair of sherds found during the 1995 excavations. Of particular interest is the fact that while one of those sherds (Cat. No. 3131) comes from the island, the other (3881) was found at the LaVase North Bank Site. Other brown stamped sherds found in 1995, from both the island (3911 and 4104) and the north bank (3132), may also belong to this same vessel; one blue banded and stamped specimen (3762) was also recovered (Archaeological Services Inc. 1996: 106-7, 108, 137, Appendix D, cf. Plate 10). It appears probable that the brown stamped sherds from the 1995 and 1997 seasons once belonged to the same saucer.
While stamped wares are difficult to date precisely, such finds are, in Ontario, normally attributed to the second half of the nineteenth century (Archaeological Services Inc. 1996: 106). A stamped and banded design very similar to the one described above was unearthed in the 1800 to 1860 colonial component of Signal Hill National Historic Park, St. John's, Newfoundland (Jelks 1973: 10, 66, cf. Figure 71j). Another example of this pattern (albeit yellowish red in colour) on a saucer comes from the Hudson's Bay Company Post at Lower Fort Garry, occupied between 1830 and 1911 (Sussman 1979: 7, 113, cf. Figure 107a).
UNIDENTIFIED SHERDS
Cat. Nos. CbGu-5 97-414, 97-511, 97-512, WS-45, WS-138
One pearlware base sherd (97-511) was recovered from the LaVase Island Site. Its ring footed base is quite similar to those seen on the blue transfer-printed (97-481) and blue banded (97-285, 532) plates, but does not appear to match, because of either decoration or thickness. The period of production for undecorated pearlware was from 1780 to 1830, which has a mid-range date of 1805. The height of popularity for this type occurred around 1800 (Grange 1977: 70). This sherd, however, likely once formed part of a decorated vessel, as no undecorated pearlware rim sherds have been found. It has, therefore, been classified as unidentified rather than undecorated.
One sherd included in catalogue number 97-512 is a refined white earthenware body sherd. Two exfoliated sherds are collectively catalogued as number 97-414. The flat shape of their surfaces indicates that these were once base sherds from one or two vessels. Sherd WS-45 is another fragment of refined white earthenware with clear glaze. One surface is flat, suggesting that this sherd formed part of a base. The sherd is quite thick, measuring 6 mm from surface to surface. Catalogue number WS-138 includes three exfoliated microsherds of clear glazed, refined white earthenware. Some of the unidentified refined white earthenware ceramic sherds above might be pearlware. The small size of these fragments, however, prevents any further analysis.
CONCLUSIONS
The ceramic assemblages from the 1997 and 1992 field seasons at the LaVase River sites are fairly small, being comprised of only 26 sherds, of which 16, all from Bothwell Island, are of identifiable types. Nevertheless, some general conclusions can be drawn from them in order to better understand the chronological setting for the Euro-Canadian components of the LaVase Island Site.
Two overlapping horizon complexes are spanned by the finewares recovered from Bothwell Island. The earlier period is characterized by types which were most popular between 1780 and 1830. While some creamware may be encountered, pearlware is dominant during this time, and is the most common ceramic ware recovered from the island (ten identified sherds). The later horizon cluster, most common in contexts dated from 1820 to 1875, includes whiteware, ironstone and brown stoneware bottles (Grange 1977: 35, 84). Both ironstone and whiteware were found on Bothwell Island, but are represented by only six sherds.
Because of the different number of sherds from different time periods, it is useful to weight dating considerations for the frequencies of the various types. Table X summarizes the calculation of the mean date for this assemblage, known as formula dating. Production date ranges and mid-range dates are as described above in the type descriptions. Mid-range dates are used here rather than modal dates (i.e. dates of highest popularity as seen archaeologically) because the latter are not available for all types. Moreover, it has been shown that using either dating system has no significant difference on formula dating (Grange 1977: 44, 50-53).
Table 23. Formula dating of fineware from the LaVase Island Site.
| Fineware Types | Production Date Range | Mid-Range Date | Sherd Frequency | Date X Freq. |
| Blue banded pearlware | 1780-1830 | 1805 | 2 | 3610 |
| Blue shell edge pearlware | 1795-1830 | 1813 | 5 | 9065 |
| Blue transfer-printed pearlware | 1795-1840 | 1818 | 2 | 3636 |
| Red-purple transfer-printed pearlware | 1818-1864 | 1841 | 1 | 1841 |
| Moulded ironstone | 1813-1900 | 1857 | 2 | 3714 |
| Blue sponged and banded | 1820-1850 | 1835 | 3 | 5505 |
| Brown stamped and banded | 1850-1900 | 1875 | 1 | 1875 |
| Totals | 16 | 29246 | ||
Mean Date =  (Date X Freq.) / (Freq.) |
1828 | |||
The mean date of 1828, which results from formula dating, should only be used in the most general of frameworks when considering the chronological setting of the LaVase Island Site. The expected period for the La Ronde Post component of the island is from about 1795 to 1821. This expected phase is earlier than the mean date derived through formula dating. However, the range of dates of production must here be kept in mind. Table 24 shows the ranges of dates of manufacture for the different types of ceramics recovered from the site, as well as the expected period of occupation by the La Ronde family.
Table 24. Ceramic chronology of the LaVase Island Site, showing the ranges of type production and the period of expected occupation.
Key:
Banded = Blue banded pearlware
Edge = Blue shell edge pearlware
Blue Trn = Blue transfer-printed pearlware
Ironstone = Moulded Ironstone
Purp Trn = Red-purple transfer-printed pearlware
Sp = Blue sponged and banded
Stamped = Brown stamped and banded
It may be seen from Table 24 that, with the exception of stamped ware, the periods of manufacture for all ceramic types overlap the expected dates for the LaRonde Post component of the site. The problem of stamped ware might be solved with a more precisely defined period of production; such vessels are only "generally dated to the last half of the nineteenth century in Ontario" (Archaeological Services Inc. 1996: 106). Perhaps earlier examples might also be encountered. It is questionable whether later types such as red-purple transfer-printed pearlware and blue sponged and banded ware could have reached such an isolated outpost so soon after the inception of their manufacture. Nonetheless, such a scenario cannot be ruled out: the later sherds may have been deposited during the terminal phase of occupation. Moreover, because the site is located at the foot of the then busy LaVase portages, it continued to be a popular campsite for voyageurs. For example, Nicholas Garry camped at the river mouth in 1921 without mentioning the post (Archaeological Services Inc. 1996: 28). Late ceramic types could have been deposited by such visitors.
Regardless of later depositions, the high frequency of pearlware suggests that the LaVase Island Site was indeed occupied during the expected date range of 1795 to 1821. The fineware ceramics unearthed at the LaVase Island Site in 1992 and 1997 are, generally speaking, compatible with the expected period of occupation by the LaRonde family.
Back to the Lavase River/Fort Laronde