ABSTRACTSTwo possible sources of error in the measurement and interpretation of the specific gravities of gold coins are investigated. First, the effect of the presence of copper was measured and found to be more significant than previously thought. It is shown that the presence of 5% of copper in a gold‐silver alloy results in a measured gold content which is 3.3% too low when the coin is analysed by the specific gravity method. Secondly, the importance of cleaning the coins before measuring the specific gravity is also demonstrate