Countermarks are intriguing but when they are on cut portions of the original host coin then many additional questions arise.
This note is to explore more about the cutting and countermarking of the 8 reales and what quantity of silver loss was experienced during the cutting process.
BNS Research Blog 
Eric’s Analysis of cut 8 Reales countermarked by Rothsay Mills provides an interesting examination of these relatively rare coins. It would appear, in these times where the value of a coin denomination and the value of its silver content were closely aligned, that the underweight element of these cut segments relative to their denominational value allowed Rothsay Mills a bit of scope as the value of silver on the bullion market increased or decreased. But was this scope by calculation or by chance?
However, of more interest to me is the exploration in the blog of the cutting process and the loss of silver resulting in the average weight loss of 9.6% for the cut half segments and 8.2% for the third cut segments.
Examination of these cut segments clearly shows that they were not cut by a hammer blow on a blade, as was the case in many of the cut segments in the West Indies. I was surprised to read that the segments were cut by shears (scissors), as my feeling was that the clean straight cut edges were the result of a press action consisting of a fixed platform and a movable element that came down vertically on the coin. A brief email exchange with Dave Greenhalgh indicated that a scissors type shear was fully up to the job, and in due course I look forward to a demonstration of this process.
So I ask what is the explanation for the fairly significant percentage of silver loss? Firstly the nominal weight Eric used of 27.06g is probably high, but I understand why he used this figure, as it is a published value that is specified in Spanish references. I looked at a range of 8 Reales (52 coins) ranging from fine to good very fine and a typical weight is more in the order of 26.60g, but using this weight against the averages Eric calculates only reduces the silver loss by 1 to 1.5%. The light filing to ‘round’ the corners would account for a very small reduction, but this would be difficult to calculate. So how can the rather significant loss of silver be accounted for?
Referring to Table I: Numbers 3 and 6 are in my collection. Number 3 is a ‘right hand’ half and Number 6 is a ‘left hand’ half (these obtained from separate sources many years apart). Amazingly the original coin features of both halves match up on the obverse and reverse suggesting they are (or could be) the two halves of the original coin. The original coin was cut about 8 degrees off the vertical line of the shield. There does not seem to be a ‘bump’ as a result of the later application of the countermark shifting metal, as when the two halves are put together the line of light showing through is very small and uniform. When the two halves are put together and the diameter measured from 3 o’clock to 9 o’clock the measurement is 41.2mm, which is at the top end for the diameter of an 8 Reale coin. Yet the combined weight of these two segments is only 24.15g; 2.91g light of the nominal official weight and 2.45g light of a typical circulated coin.
Number 6 shows a line from the top edge to about the centre of the coin. Could this be evidence of the cutting tool, but the coin being slightly repositioned before the full cut was made? This line is only 0.5mm or slightly less and if caused by the cutting instrument could not account for the significant percentage of silver loss as noted in Table I.
So for the moment the question of what is the cause of the missing silver remains unanswered.
It appears from what Ken writes that the half cut dollars in his possession were countermarked and then cut, as there is no ‘bump’ on the cut edge from subsequent countermarking. If this is the case then it seems there was no fixed rule as to the order of cutting and countermarking.
The extra cut mentioned by Ken could have been made after the coin was cut in half, perhaps in an attempt to steal silver from the cut half.
It will be interesting to read of Ken’s views on cutting after his visit to Dave Greenhalgh.
Test message only. Please ignore.
Second test message – again please ignore.
Evidence would indicate the coin was countermarked prior to cutting. While on the left hand ‘half’ the countermark is situated several millimetres from the edge is inconclusive on the right hand ‘half’ the countermark is right at the cut edge. Therefore if stamped after cutting an edge bump should have occurred.
Eric raises the point that the partial cut near the edge present on the left hand half might be evidence of someone attempting to shave off some silver for personal profit. Perhaps after seeing the difficulty of trying to shave off such a thin slice the effort was aborted.
The meeting with Dave Greenhalgh was very interesting and informative. Dave was very generous with the time he gave me to demonstrate his shearing tool (a very large pair of scissors). He cut through some silver half-crowns including one that he had annealed. I am convinced that with a large shearing tool of this type, properly mounted to a workbench it would cut an annealed dollar coin provided the blades were sharp and the fitting between the two parts of the tool were tight. However, I am not convinced that a straight up and down cut edge would be achieved or that the coin parts would not suffer some distortion. I also question the process to cut a dollar into virtually equal thirds using this type of tooling. I believe this process would not afford the precision to obtain consistent results, and Eric has shown that the consistency of the Rothsay 1s 8d cut thirds is very uniform. I also think that evidence of cutting the dollars in this manner would leave a telling feature at the junction-angle of the two cut edges.
Examination of the Rothsay Mills cut 2s 6d and 1s 8d indicate to me that a press type tool with a placement fixture to accurately hold the dollars to be cut. In ‘simple terms’ this tooling would consist of a fixed platform that incorporates the placement fixture and a vertically moveable block that would descend on half of the coin with great pressure thus ‘shearing’ the coin in half. This would produce a clean straight up and down cut edge. A similar tool with an angled face would be used for the cut thirds. The first cut would remove one third of the dollar and the remaining segment would be repositioned and cut in half resulting in the dollar being cut into three equal parts. Rothsay Mills would have had a comprehensive machine shop and skilled tool makers. This would have been necessary for repair and maintenance of the mills equipment. The type of tooling mentioned above would be well within the capabilities of experienced tool makers.
In theory the type of tooling described above should result in virtually no silver loss, however, as Eric has shown the loss of silver amounts to a significant percentage; far more than I would have expected. I believe there is more to learn and we have not yet identified where or how this loss of silver has occurred. I hope someone can shed some additional light on this matter.
Though I have no engineering background or experience, Ken’s idea re cutting the 8 reales, as set out in his last paragraph, seem eminently sensible to me (certainly for the halves). He also mentions that ‘evidence of cutting the dollars in this manner (scissor type tool) would leave a telling feature at the junction-angle of the two cut edges’. This, too, I can agree with regarding the cut thirds. He finishes by raising the issue of the silver loss, which must now be the centre of investigation.
I have recently discussed the above cutting problems with an engineer. He told me that when cutting metal, to prevent over-cutting or splitting, it was common practice to drill a hole at the end of the cut. He then proposed that a hole could have been drilled in the centre of the 8 reales before cutting commenced. This would prevent further cutting or splitting and would also be a cutting guide to the centre. A further result of this hole-drilling would be the loss of silver. Add this to the final filing of the cut sharp edges and perhaps the result is the loss of silver that has been calculated. More work needs to be done and any engineering input would be gratefully received.
David Wolfer is an American researcher into West Indian cut coins. The following details regarding Rothsay have been extracted from a longer email covering cut West Indian coins dated 13 August 2019.
Regarding Eric Hodge’s article on the cut coins of Rothsay Mills, I think there is a relatively simple explanation for his weight dilemma. It appears Eric uses the colonial 8 reales official weight of 27.06g as the basis for his calculations without taking wear into consideration. After all, the host coins from which his segments were derived from had seen plenty of service. In the Gurney book,(note 1) Bob provides weight-range estimates for worn 8 reales; VF- 26.8g, VG- 26g, Good- 25g.
Eric’s images show well worn segments. If Gurney’s 25g is the approximate weight for a Good-graded coin, that’s an 8% reduction off the official weight (.076 rounded up), just from wear alone. With additional smoothing of corners and alignment inconsistencies from cutting with a shear (for halves at least), additional weight is lost. Multiply Eric’s #14 times two and it comes to 24.98g, just under Gurney’s estimate for a Good coin. In my opinion, host coin wear and edge softening explains the loss.
We can use the third-cut segments for comparison, but let’s talk about their cutting style first. Mr. Greenhalgh believes these pieces were cut with shears and I agree. However, Eric’s photographic evidence suggests this was no ordinary set of shears, but a special cutting tool capable of whole coin segmentation in a single pass. In other words, three segments were created simultaneously. As to whether they were guillotined or press-cut is anyone’s guess when viewing from the obverse side of the cut, but reverse edges undoubtedly hold the tell-tale evidence needed to make that determination. If I were to take a guess, the pieces appear to be press cut.
Working from this assumption, we can now compare machine-generated third-cut weights with the 1/2-cut segments. Eric’s average weight for the thirds is 8.28g; multiply by three gives us 24.84g, only slightly lower in weight than the #14 x 2, 1/2-cut segment. Add up weights for the three specimens in Eric’s photo and the total comes to 24.99g. Again, this is awfully close to Gurney’s wear-based weight estimate for heavily worn 8 reales, of which all Eric’s third-cut photographic examples qualify.
For all practical purposes, the Rothsay Mills coins went through the same process and, as a result, the combined segments from any particular host should show no measurable reduction in silver.
This leads me to several conclusions. [1] A silver segment cut with a shear of some sort is underweight either from prior host wear, inaccurate cutting, additional illicit removal, or any combination of these three factors. [2] A silver segment that has been machine-cut with a multi-segment cutting tool was created this way not only for the sake of efficiency but also for consistency in weight and segment size. As such, the sum total of segments excised from the host should add up to the same weight of the host coin prior to cutting.
It is interesting to me that a tri-form cutter had been designed and was proving functional in England in c1811.
The Rothsay Mills tokens move a considerably more complicated cutter head technology up seven years, much closer to the 1807-1812 time period.
Note 1.) Counterfeit Portrait Eight-Reales. The Un-real Reales, by Robert Gurney, 2014, CreateSpace Independent Publishing Platform.