Excellent article, Bob. Probably the most complete writing that I've seen on the subject.
The "red acid", for testing silver, that he speaks of, is a combination of nitric acid and potassium dichromate.
He says that there can be cadmium and/or cyanide on silver contact points and that these make it very hazardous to melt. Most do contain cadmium. None, however, contain cyanide, unless the seller was trying to refine them himself, which is unlikely. In any case, don't melt silver contact points, because of the cadmium.
His discussion of silver flake, plated out of photo fixer, is very poor and incomplete. When dry, the lowest Ag content I've seen was about 80%. The best ran about 98% Ag. The 80% variety was black, powdery, and smelled strongly of sulfur (silver sulfide). The 98% stuff was hard, white, metallic, brittle, a little bit nodular, and usually about 1/8" thick, more or less. Usually, the Ag is somewhat between these limits and shares some of both sets of characteristics. It is usually wet, especially the lower quality stuff, and must be dried before weighing - I usually put it in a steel pan on a hotplate - it really stinks, so do this outside. It can be easily melted with very little loss in the slag. While melting with borax, several pieces of rebar, about 24" long, are placed in the crucible. They are then removed before pouring. The iron combines with the sulfur and goes into the slag. The resulting silver can be quite pure.
Concerning gold plate. There is usually a thin layer of nickel plate underneath most all gold plate. The gold is so thin that it only takes a few strokes with a pencil eraser to rub a spot of it off. This exposes the white nickel below, which is easily seen in contrast to the yellow gold around it. Go slow so you don't erase the nickel - you want to be able to see the white color of the nickel. If you do erase the nickel, the exposed metal will likely be yellow (copper alloy) and that will confuse everything. I have used this test 1000s of times, mainly on electronic stuff. I got so good at it that I could get a pretty good estimate of the thickness of the gold plating. Of course, this won't work for gold filled.
In the early 80s, there were lots of guys going around selling plated chains. Their pitch was that the stuff was hot and was stolen by looters during the Detroit riots. They were selling it cheap since they couldn't sell it through legitimate channels. The clasps were usually marked 10K or 14K. When I got approached by these guys, I asked if I could use an eraser on it. After all, if it was solid gold, the eraser wouldn't hurt it. The white nickel usually showed after only a stroke, or two, of the eraser - very thin! I was probably lucky that I didn't get shot for ruining their merchandise.
