If you're having trouble viewing this email, you may see it online.

April 2006

Numismatic Authentication through Non-destructive Surface Science Tests

The following is an adaptation of a presentation made to the members of the Roxbury (NJ) Coin Club on April 5, 2006, and represents a unique use of surface science methods not currently employed in the numismatic field.

Between 1732 and 1772 Spain minted in the New World a series of coins commonly referred by numismatists as columnarios. (They are also often referred to as Pillar Dollars, Pieces of Eight, and Spanish Milled Dollars.) The first coin listed in the popular Red Book of Coins is the columnario -- it is in fact the predecessor of the dollar and circulated on par with the dollar through the first half of the 18th century.

Prior to 1732 the majority of coins minted by Spain were crudely hammered coins referred to as cobs. A specimen dated 1669 is shown above. One of the primary problems with was shaving. Unscrupulous merchants and individuals would shave a small piece of silver from the edge of each coin that came through their hands. A small shaving from many coins could ultimately amount to a valuable hoard. While this was illegal (and punishable by death) it was very difficult to catch the perpetrators as the uneven edges did not lend to detection and in most cases people did not have precise tools for measuring the weight of the coins they exchanged.

This lead to the columnario in 1732 -- a revolutionary concept at the time and a major change to the way coins were minted in the New World. Columnarios were minted using a screw press and then and a secondary process added a milled edge. This new coin immediately stopped the practice of shaving as it was impossible to tamper with the edge of a columnario without damaging the milled floral edge.

234 years after the last columnario was minted, these beautiful coins have become very desirable by numismatists and are highly collectible. The majority of columnarios were minted in Mexico -- such as the 1749 specimen shown above. They were also minted in Guatemala, New Granada (present-day Columbia), Lima (Peru), Potosí (Bolivia), and Santiago (Chile). They are found in 8 reales denominations (as shown above) as well as 4 reales, 2 reales, 1 real, and 1/2 real.

Currently one of the biggest distractions to the collector is the wide proliferation of fake columnarios. While most common date varieties in average grade will sell for under $200 USD, there are many varieties with uncommon dates and from non-Mexican mints that can command values in the thousands. Consequently, there is an underground market for fake columnarios. Unscrupulous individuals trade in these fake coins with many changing hands on eBay. It's currently estimated that between 20 and 30% of columnarios that are listed on eBay are not authentic. Many collectors don't find out their coins are fakes until years later when they try to resell them.

Some fakes are genuine coins that have been altered -- perhaps the date is changed or some other modification is made in order to change a $150 coin into $750 coin. In most cases alterations can be detected by inspection and by comparison.

People have been making fake coins for centuries. Some are quite crude and most specimens from previous generations are quite easy to detect based on visual diagnostics. Currently there is a new class of fakes entering the market which are expertly made using advanced technologies. These fakes are the subject of this study.

Currently, there are myriad diagnostics used in authenticating columnarios. These include the weight test: the coin is weighed and if it is outside of an acceptable range, points are deducted. This test alone is not definitive: many authentic coins are sea salvaged and, after sitting in salt water for over 200 years, a thin volume of surface has been eaten away and so these coins weigh in low. Likewise, there are many fakes that pass the weight test with flying colors. The diameter test is also easy to administer. Unlike today's coins, columnarios vary in diameter slightly due to the edge-milling process which was done by hand on an edge-milling machine. Points are deducted if the coin is outside of an acceptable range. The edge test is also useful but not definitive. Some authentic columnarios have a floral edge in one direction while others change direction midway. The number of floral patterns can range from the high 30's to the high 40's. There are also a variety of visual diagnostics that are used to detect fakes which results can add or remove points.

The point system (still being developed by the author) helps raise red flags and detect expertly made fakes which otherwise might go undetected.

One diagnostic that is especially difficult to test is the metallurgical evaluation. Prior to this study, we relied almost exclusively on what's known as the ring test. The coin is lightly spun on a table top and the ring sound is compared with the ring of columnarios that are known to be authentic. Fakes made from lead or other non-precious metals and alloys are quite easy to detect based on the ring and often can flunk a coin even when the other tests pass. Alas many of the new fakes will pass the ring test.

This brings us to the need to better understand both the metal content and the surface qualities that result from the method of manufacture of the coin we're testing. If we allowed to drill a small hole in the side of the coin and extract a few small pieces of metal, it would be easy to send this to a lab for diagnostic evaluation. Unfortunately, if the coin proves to be authentic, no one is interested in it if has a hole drilled in it -- or is otherwise altered by a destructive test. Hence this approach is unacceptable.

This brings us to a simple surface analysis. For our study we used a ramé-hart Model 200 Contact Angle Goniometer. We began by assembling a collection of 20 columnarios which are known to be authentic based on all of the above diagnostics outlined above and which have come from reputable sources. First, we selected the field above the reverse (side with date on it) crown. It's the largest field anywhere on the coin. Next, we used acetone on a cotton swab to remove any residual surface contaminants on this field. Generally it's poor practice to "clean" coins as it can result in hairline scratches which lower the value of the coin. Consequently, we were very careful to roll, not swipe, the swab in order to reduce the chance scratching the coin.

We dispensed 5uL of deionized distilled water on the field and measured the contact angle. The range for this lot was between 35.1 and 59.1 degrees. This turned out to be a wider range than we anticipated but upon further study, we noticed that the more hydrophilic drops were on coins that had been previously cleaned (and thus had a polished surface) while the drops with larger contact angles were on coins that under magnification had porosity issues -- typically the result of sea damage but even non-sea salvage coins can develop environmentally-induced surface damage. These coins exhibited lower surface energy.

With this comparison data collected and tabulated, we were now ready to test our questionable coins. We have two columnarios dated 1732 -- one of which is shown above. These coins appear to be sea salvage. Using the weight test, ring test, diameter test, and all the visual diagnostic tests we currently use (including comparing the test coin feature for feature with an authentic coin of the same variety), we conclude that these are authentic coins. Still, there is something about these coins that shake your confidence and which leaves you questioning your opinion and method. I should point out that authenticating coins is both an art and a science. The empirical results (science-side) say it's okay while your gut (art-side) says it's not.

We tested both of these coins using our contact angle test using the exact same procedure we used on sample of 20 known authentic coins. We were quite surprised when the contact angles on these two coins ranged between 77.1 and 78.3 -- significantly higher than the range of the good coins. This test alone flunked our coins and rendered them fake.

This has been a very important study as the coins in question are dated 1732 which is the rarest year of all -- aka as the key date. An authentic sea salvaged 1732 similar to the coin shown above sold recently at auction (Ponterio Sale 138, Lot 170) for $7,000USD. If this pair of coins proved to be authentic they could collectively fetch in excess of $14,000USD at auction. As fakes they are virtually worthless.

Ultimately, despite all of the scientific tests, it's always worth getting an objective second opinion. We showed these coins to a noted numismatist, Daniel Sedwick, a noted expert on Spanish American coinage, and Numismatist and Dealer Mike Dunigan of Texas. Both agreed that they were dubious. The latter suggested that the coins were manufactured using a method called spark-erosion. In this process an authentic coin is submerged in an electrolytic bath and then electrically charged. This charge causes sparks to jump across the shortest distance between the authentic coin and a die. The die is then used to manufacture a fake coin. The edges are milled using a secondary process and the result is a coin that passes all visual and empirical diagnostics...all but one that is, the contact angle test.

If you found this article interesting, let us know.

Product of the Month
This month we feature one of our most popular spare parts: the micro-syringe assembly, p/n 100-10-20.

This micro-syringe is included with all ramé-hart goniometers. The part (when ordered as a spare) includes the syringe, one o-ring, and the glass barrel. Needles are available separately. (For a complete list of our needles, see http://ramehart.com/pdf/needles.pdf.)

The volume capacity is 2mL and graduated in 2uL increments. Accuracy is 0.50%. The plunger and shell are constructed of PTFE (Teflon). The o-ring is made from VITON flouroelastomer and can be ordered separately (p/n 100-10-110 for pack of 6 o-rings). The glass barrel is simple to clean and reuse. It can also be ordered separately (p/n 100-10-108). The test liquid comes in contact with with glass, PTFE, and VITON only. The entire assembly can be autoclaved and easily disassembled and reassembled for cleaning. The glass barrel features a LUER joint which allows for myriad needle choices. The plunger is driven by a micrometer style dial for maximum control. It's possible to use the micro-syringe for advancing and receding studies (where volume is added and then removed while measurements are being taken). For more precise or advanced dynamic studies, we recommend our software-driven Automated Dispensing System. The micro-syringe and all related parts can be ordered on our website at http://www.ramehart.com/shop/. On the main menu, click on Manual Dispensing Options.


If you have a white paper or other information relating to your use of a ramé-hart instrument that you would like to contribute to this newsletter for publication, we would invite you to send it to us. We will provide appropriate attribution.

We appreciate your comments, your continued business and look forward to working with you in the near future. If you would like more information on any of our products, or a copy of our current price list, please don't hesitate to contact me personally.  



Carl Clegg
Director of Sales
Phone 973-448-0305
Contact us