As paper currencies lose what luster they never really had, precious metals become more valuable and susceptible to counterfeiting.  The incentive is simple; give someone less of what is valuable, while convincing them they got what they bargained for, and keep the difference for yourself.   Since the earliest times, gold coins were "shaved" to remove metal from the coin, lowering the weight.  This needs to be distinguished from normal circulating wear.  Everyone has seen old coins that have been worn down from circulating in the public.  Very tiny amounts of the metal are left on desk tops, fingers, pockets and carrying containers. To deal with this situation in the past, the mint would accept worn coins within a certain range at full value and replace them with a newly minted coin.  This is where the term "Legal Tender" originated.  

The first and best way to defend against a fake gold coin is to deal only with a trusted, reputable dealer or trader.  People who have a long track record of fair dealing didn't earn their position of trust by intentionally cheating customers.  In spite of this, even trustworthy dealers and traders sometimes make honest mistakes.  In the end, the burden of determining the value and authenticity of the gold falls on the receiving party.  This article is by no means exhaustive or highly technical --  it's purpose is to give some basic information and practical ways to safeguard your gold dealings.  Given the rumors of tungsten filled gold bars and the rapidly escalating price of gold, it is now imperative to trade with care.

Coins and Bars

The most common way you will receive gold today is in the form of coins or bars.  This is an improvement over the naturally occurring nuggets and dust, which were the way gold was first traded.  Historically, frontier gold producing areas did not have a mint, so this raw form of gold formed a useful, but inconvenient medium of exchange for miners to buy supplies.  A merchant weighed the nuggets or dust on a balance and accepted the gold as payment for the miner's purchase, a time consuming and tedious process.  Assay offices also sprang up in busy mining communities to purchase the gold found by miners; this gold would then be sold to the mint to be made into coins.   

For these reasons, coins were created to make it easier to conduct trade -- there is no need for a balance to check the weight, purity is assured and with the invention of the reeded edge (the markings on the edge of a coin), it's somewhat tamperproof because edge shaving is easily detectable (though by no means perfect).  Coins are easier to handle and trade because they carry a recognizable "brand", usually that of a country or mint and many are legal tender in the country of issue.  Up to an ounce, coins are relatively easy to stamp, carry around, count and store.  People trust coins, especially those from a legitimate government and they have been used for centuries.  

Bars come in two forms; cast, and stamped or machined.  Cast bars are made exactly like the name suggests, the gold is melted and poured into a mold.  Once it cools, the maker stamps their mark, the purity and a serial number on the bar.  Typically, the weight varies slightly for larger, cast bars and there is often an extra payment or refund due the buyer, depending upon the actual weight of the bar.  There may be grinding marks, where the manufacturer took off the extra gold.  Today, large bars are used mainly by government banks to store their reserve gold and by the commodity exchanges, where the standard "contract" size bar is 400 ounces and the bars are known as "good delivery" bars.  They are frequently traded in 1 kilo amounts, as well.  Smaller bars in 10 ounce and 1 ounce size are often stamped or machined, which makes it more difficult to tamper with and they can be traded like coins, since they are in effect large coins.  Cast bars are the easiest to counterfeit.

A rumor that has been circulating for some time now is the faking of 400 ounce good delivery bars by filling them with tungsten.  These would be very difficult to detect because tungsten has very nearly the identical density.  A bar cast with 1/4 inch of 24K gold on the outside and filled with tungsten would look and feel identical to the real article.  This topic is covered here.  For most consumers, bars of this size are out of the question for cost reasons alone.  If you are considering the purchase of these kinds of bars, be sure to ship them directly to a legitimate assay office to be sure you get what you paid for.

It's All About the Physics

All the ways to determine the authenticity of a coin or bar of precious metal are based on differences between physical properties of the valuable metal and the lesser metals used to swindle the buyer.  Each metal is an element, meaning it can't be further broken down into another element (without very heavy equipment like a particle accelerator, which is not practical).  An element has its own unique density, conductivity, reaction to acids, melting point, conductivity of sound through the metal, etc.  A fake gold coin will necessarily have an exterior of gold to help fool the buyer, regardless of the contents of the interior and whether the coin is created from an alloy (mix of metals) or a slug of base metal with a cladding or plating of gold on the outside.

Visual Inspection

The easiest place to start is the visual appearance of each metal.  Does the coin look gold in color?  Gold has a dull golden shine. Once you've spent time handling gold, this is the first give away.  Keep in mind that some coins are 99.99 percent pure gold.  Others are designed for circulation and have some copper and silver alloyed with the gold.  These coins have a reddish tint.  Jewelers use these properties to make many different shades of gold for jewelry.

The color difference is exaggerated in these photos, but it gives an idea of the difference between 99.99% pure gold coins (left) and 90% pure gold coins with an alloy of copper and silver that make the coin harder and able to circulate (right).

Density

The physical characteristics of metals are immutable and this is the counterfeiter's first decision; which metal(s) to use for the fake. Each metal has characteristics like density, malleability, the speed of sound in the metal and so on.  One of the best and easiest places to start is with density of the various metals (this is NOT the same as the atomic weight, density is the amount a fixed volume weighs):

Summary Table of Metal Densities (g/cm³)

From this table, it is apparent that there are a very limited number of metals that are suitable to use in a high quality counterfeit gold coin with a similar density and much lower value that fit the bill.  Most of the metals on the list are too light and you will be able to spot the difference just by holding the fake coin in your hand because they won't have that heavy gold feeling.  Some of these metals can be melted together with the gold to form an alloy, or they can be used as the core of the coin, sandwiched or plated with real gold to give the illusion of solid gold.  

As we saw above with the gold coin photos, alloys with gold change the color ranging from pure gold (yellow with a slight orange tinge) to pink/rose, white and pale green.  These colors can be created with the addition of silver and copper and many other gold alloys are possible.  The greater the concentration of metals besides gold, the greater the shift in color.  The more copper and silver in the alloy, the lower the density, e.g. a gold alloy made with 5% silver and 20% copper will look red with a density of 15.15 g/cm³, a significant difference from that of pure gold (19.32 g/cm³). Therefore, these types of counterfeits are less likely to turn up because they are easy to spot by just looking at the appearance of the coin, compared to a known good specimen.

Aluminum through copper are all too light and mercury is liquid at room temperature, so it's immediately off the list.  It would be very easy for anyone to detect a coin made with these metals because the first thing people who have never held a gold coin say when they hold a real one is "wow, that's really heavy".  Light weight is a dead give-a-way to a fake made with these metals.  Looking at cost and availability, Osmium, Iridium, Platinum, Rhenium, Neptunium and Plutonium are off the list of potential counterfeiting metals as they are difficult to obtain and expensive.  Plutonium is more valuable for use in nuclear reactors or weapons and it is highly radioactive. Depleted uranium is a possibility, but availability is tough because it is carefully controlled by governments and it can be radioactive.  Platinum could work but it's more valuable than gold (it is today, in the past it was used in gold fake coins).

For practical purposes this leaves silver, lead and Tungsten.  Silver and lead are actually much less dense than real gold, so the best test to rule out these metals is density.  How do you measure the density of a metal?  You can use a scale, because if the coins are the correct size and they are made from lead or silver, they will weigh about half the right amount.  Another very accurate test is specific gravity, the ratio of the weight in air and the weight in water requires a precise scale and placing the object in a container of water.   The coin is weighed in air then weighed in water and a ratio is calculated.  Each metal has an exact specific gravity.  

An easier and more practical way to test the density of a coin is to use a tool called a Fisch, a plastic tool that first checks the physical diameter and thickness and finally serves as a simple tip balance to test the weight.  This tool will kick out the lead and silver fakes, since it's physically impossible to get the proper weight into a coin that is the same diameter and thickness of a gold coin.  It's possible to make a coin with lesser metals of the correct weight, but the size will be much larger, a give away to a fake gold coin.   This tool is highly recommended by gold dealers and bankers.  Currently, the Fisch can test the Krugerrand, American Eagle (1, 1/2, 1/4, 1/10  ounce sizes), Maple Leaf, Mexican 50 Peso, Mexican 20 Peso, British Sovereign, 100 Corona, US $20 Double Eagle & US $10 Eagle as well as medallions and small bars. The Fisch also checks the Platinum Noble and the U.S. Silver Dollar.  I highly recommend it.

The Fisch, a practical gold coin testing tool.

Sound

Another classic test is the "ring test".  Drop the coin on a hard countertop to hear it's "ring".  Gold and silver have very distinctive tones and qualities of sound, which are fairly easy to distinguish.  The saying "it rings true" and the song Bing Crosby sang "Silver Bells" testify to this quality.  Metals like lead have a dull thud.  Even the sound of a tungsten slugged gold coin will sound different.  It's best to have a known genuine coin around for comparison (and be sure to note the issue, date and mint mark to avoid confusion with the coin being tested).

Electronic Testing

Electronic testing runs a current through the coin to test the resistance and determine the purity of the metal.  High quality testers like this can work and they are mainly used for buying jewelry, not coins.

Destructive Testing

The easiest destructive test is the scratch test, which does what the name suggests, scratches the surface of the coin to see if the color is the same underneath.  This identifies those fakes that are stamped coins plated with gold.  Gold plated coins also have a tendency for the details of the coin to fill in with the plating.  Look carefully at the detail of coin with a 10x magnifier and compare it with a genuine coin, if you have one.

The touchstone is the other classic way to test gold.  The edge of the coin is run along a stone so a very small amount of the gold leaves a "trail" on the stone.   Acid of various strengths is dripped on this trail to determine the purity of the gold.

Melting is the final and most destructive way to test the coin and for pure gold it is 1947.52F.  Tungsten does not melt until its temperature reaches 6192F.  If you melted a tungsten filled gold bar, you would have liquid gold with an "ice cube" of tungsten on the bottom (it's the same density, so it wouldn't float).

Caveat Emptor!