Hallmarking day at LAO

Excellent day at the assay office on Monday where I went for one of their info days - many thanks to Steve and the others involved. It was a good mix of history, theory and hands-on, a great lunch, and fantastic just seeing around inside the Hall. In the labs we were lucky enough to see and handle a couple of the most valuable coins produced by the Royal Mint, 1 kilo fine gold with a face value of £1000 (though obviously a gold content currently worth nearly 30x that). I also got to test my great-grandfather's Australian 1870s gold rush nugget in the x-ray spectrometer thingie - which showed it to be 99% pure with a trace of silver, mounted on a 10 carat bar brooch. I'd recommend one of these events to anyone new to assaying/hallmarking. (I finally completed my registration with LAO about a week ago, and plan to send my first packet off before the end of the month.)

By the way, Steve, I carried on puzzling for a while over the question I put to you and David. I think I found a satisfactory explanation in Mark Grimwade's metallurgy book which I was reading on the train - and in fact Wikipedia makes it quite simple - but I'll see what people here reckon... So, the question was this: when we are talking about millesimal fineness (ie 925 or 375 or whatever) it's clear that we mean parts per thousand of a precious metal, the remainder being made up of the other elements in the alloy, but what exactly are the "parts"? Take for the sake of argument a 50:50 mixture of gold and aluminium: you'd get very different results depending on whether it was 50% of each by mass, or by volume, or by number of individual atoms...

Alan