Replicating the Bloomsbury Tin
Condensed version of a presentation for the Middle Atlantic Archaeological Conference, Friday, March 7, 1997, Ocean City, Md.
Tinplating of sheet iron originated in Bohemia in the middle ages. By the seventeenth century, tinware was a popular cheap substitute for pewter throughout Europe (Brown 1988). The English tinplate industry began in 1667 when Andrew Yarranton led a party of industrial spies to visit plate mills in Saxony, where they learned the process. For the rest of the century, various entrepreneurs attempted to establish an English tinplate industry.
Finally, in 1697, the first English rolling mill at Pontypool began producing a uniform and thin iron plate that was well suited to tinplating. A few years later, they added a plating shop and the English tinplate industry was under way. The industry spread to other parts of the West Midlands and South Wales within a very few years.
Up to 1740, the small amount of tinware used in America was imported from England. After 1750, tinware manufactured in Connecticut from English plate were distributed throughout America. By the end of the century, a native American tinware industry had developed and the industry was becoming mechanized (Mulholland 1981:96).
Replication is a time-honored method for analysing archæological tinware. Tinplate recovered from archæological sites frequently is smashed and distorted when it is removed from the ground. To the untrained eye, a typical tinplate artifact is nondescript .
At Fort Ligonier, Pennsylvania, replication was used effectively to reconstruct the shapes and construction methods of smashed cups and buckets (Grimm 1970: 168-169).
A replicator not only reproduces the shape of the artifact in its original state, but can interpret the technology of its maker. Knowing the maker's level of skill, technological awareness, and attention to style, can help interpret the cultural context of the site.
Richard Haddick in his tinsmith shop
The east well at Bloomsbury yielded two moderately well preserved tin cups and some flattened tin. A wire bail handle appeared to belong with the tinware. As the pieces of tinplate were sorted, it became apparent that the flattened tin included a damaged bottom and part of a side of a tapered basin. Other tin items were beyond identification.
The flattened side and bottom of the basin
The cups and fragments were taken to Richard Haddick, a historical tinsmith in Wyoming, Delaware, for interpretation and replication. Haddick concluded that there were three fully reproduceable vessels: two cups and a basin. As nearly as possible, the construction methods of the eighteenth-century original were duplicated, using hot-dipped tinplate similar to the type that was available to the original maker. The sheet was cut to 10" by 14", the size shipped by English makers to the American market (Audels 1962).
In 1780,. Cornish tinplate manufacturers standardized the two principal sizes of plate in terms of ranges of dimensions:
Singles: 12 3/4" by 9 1/2" to 13 3/4" by 10"
Doubles: 15" by 11" to 16 3/4" by 12 1/2"
ÒDoubles" plate was heavier, and was used for larger products. Cornish tinplate was shipped thereafter in boxes of 225 sheets (Phil Kelley, personal communication).
The 10" by 14" sheet has exerted a longstanding hold on tinware design. Alongside plans for air-conditioner ductwork and refrigerator pans, Haddick's copy of the Audel sheet-metal manual published in 1962 includes plans for vessels virtually identical to the eighteenth-century originals. Using the manual and the specimens, Haddick was able to reconstruct copies of the original three vessels.
The smallest vessel was a cup with a 9" circumferance and 2 7/8" diameter. The finish height is 2". No handle survived, but solder encrustations provided dimensions and locations for the handles. The edge was rolled over 18-guage wire.
With allowances for overlaps, Haddick cut a side piece 2 7/32" by 9 1/4" and a circular bottom 3 3/16" in diameter. He used a burring machine to embed the wire in the edge before forming the cylinder. Because the side seam overlaps, the wire edge is shorter than the whole sheet. A notch in one upper corner of the sheet accommodated this difference.
Haddick finishes the edge with a burring machine
The side was then rolled into a cylindrical shape and its bottom edge was turned out on the burring machine. Burring machines had not been introduced at the time the cups were made, but a careful worker could turn the edge with hand tools. A simple soldered lap joint formed the cylinder. The bottom was then turned up on the burring machine. It was snapped over the cylinder and then the lip was turned down to lock the bottom.
The cup has been formed into a cylinder by rolling
While it might have been watertight with only a locked joint, it had been soldered. Using 50/50 solder, Haddick made a sweat joint from the inside of the vessel. Entry and exit points of the soldering iron are visible on both the original and the replica. Haddick noted that the original maker probably was right-handed, judging from the direction of the lap joint.
Using solder joint remnants as a guide, Haddick made a handle with a finished width at the bottom of 3/8" and 3/4" at the top. The length of the handle blank was the same as the height of the side blank, 2 7/32", since a piece of this size would have been left over if the side of the cup had been cut from a piece of tin 10" wide. The handle was hemmed and then butt soldered to the cup along its seam. The butt solder joints would have been a weak spot in the design, and may explain why no handle was found with either cup.
Soldering the lap joint with a modern electric soldering iron.
Originally a copper soldering "iron" was used for this purpose. Capacity is about 200 cc, or three quarters of a modern measuring cup. The replica cup is remarkably similar to the one illustrated in the 1962 Audel manual.
Elapsed time to make the cup was about an hour. In the mid-nineteenth-century, such a cup would sell for 6¢.
The second cup, with a capacity of about a pint, was made the same way, probably by the same tinsmith. It is 4" in diameter, 3" tall, and 12 1/2" in circumference. The disk for the bottom was originally 4 1/2" in diameter. The blank for the side was 3 7/32" by 12 3/4". Haddick noted that it had been roughly used before being discarded. The handle is 1" wide at the top and 5/8" at the bottom, also with butt joints. The base is soldered inside and outside, as opposed to the inside-only soldering of the smaller cup.
The two replicated tinplate cups based on specimens found at the French and Indian War Fort Ligonier in Pennsylvania were 3" high and 5.1" high respectively. Tinware was not uncommon in the fort remains (Grimm 1970: 168, 151).
Uncannily close to the Audel's manual was the tapered two-quart basin. The two pieces for the side of the basin fit comfortably in the 10" by 14" sheet size. The basin tapers from 6 1/4" on the bottom to 8 1/2" at the top. Once a sheet metal pattern had been made, the tapered basin went together almost as quickly as the smaller cups. The bottom seam of the original had not survived, but Haddick soldered it inside and out, in order to ensure that it would hold water under rough use.
Also in the same part of the well, but fragmented beyond recognition, were pieces of a tinware vessel with a locked seam. Haddick explained that locked seams can be used on a vessel that would be heated, since it did not require solder. A soldered lap seam, such as on the cups and basin, might come apart if heated without containing a liquid. Coffee pots and kettles, for example, would be jointed without solder.
This is an abbreviated version of the article that appeared in the final report of the Bloomsbury excavation, to be published by the Delaware Department of Transportation.
