Looks like this scavenger hunt was already started by the Molten rock mod
people, who have done some spectacular work so far. I think it's best to take this discussion over there.
I ran into the magma safe mechanism problem. What a pain. The only stone with a high enough melting point to resist magma is bauxite. Unfortunately, Bauxite is literally rarer than gold in the game. This has become a real problem for me because I'm playing Goblins, and can't order the stuff in. But it an issue which I know has bitten a lot of people.
There are various suggestions on the modding front, such as adding [ANY_USE] tags for iron, etc. However, I'd like to suggest a better solution which is both reasonable and won't adversely affect gameplay.
The wiki states that magma's temperature is between 1300-1400°C. Bauxite has a melting point of about 2000°C, so it can survive as a mechanism indefinitely. Iron and Nickel also have high enough melting points, but you can only make mechanisms out of stone!
However, the problem here is not that there are no stones which have high enough melting points. Basically, if you take a look at the matgloss_stone_mineral.txt file in the raws, you can see that very few stones actually have a given melting point. Having done a little research, I can see why. It's hard to get your hands on mineral melting points, as they're not definite things. Different components have different melting points.
After a bit of googling, I managed to find a list of melting points for the mineral Olivine. Almost every source states this mineral has a high melting point. Wikipedia (uuhh) gives a ranges of between 1200-1900°C. The encyclopaedia Britannica website give a range of 1550°C to 1300°C. A few more sources confirm these ranges. Personally, I'm willing to put down a figure of 1450°C for the stone in the game (based largely on the fact that I want to use it for magma safe mechanisms). Since the game needs a single number, I think this is a reasonably fair one to put down in the raw file. However, we should try and discuss this to give the file some veneer of respectability.
My proposal is this. A group of volunteers take it upon themselves to find as much information about a particular stone in the game as they can, and report back with a list of melting points, etc, and a range of melting point temperatures, and a recommendation. We take a look at the sources, throw around some guesstimates and all come up with what we hope is a reasonably reflective melting point for the stone in the raw file. For completeness, we should even look at stones with given melting points.
When we're finished, we come up with what we hope will be an accurate and complete matgloss_*.txt files for use in mods or even the main game. Hopefully this process will result in some more usable stones for magma projects, and an improvement for the game/mods in general, but we should try to as accurate as possible regardless of the outcome.
A bit of googleing lead me to, at the very least a range of melting points. With a more work and a lot of hands, I think we can come up with similar ranges and sources in good time.
Here are all the stones listed in the matgloss_stone_gem.txt, matgloss_stone_mineral.txt, matgloss_stone_layer.txt and matgloss_stone_soil.txt text files in the raws. This is from 40d11 on linux.
From matgloss_stone_soil.txt (Not sure should we bother with these?)
sandy clay loam
silty clay loam
From matgloss_stone_gem.txt (This is a long list)
honey yellow beryl
precious fire opal
red flash opal
light yellow diamond
faint yellow diamond
I'll start by proposig the first entry in our list, with sources
Source1: (Wikipedia) 1200-1900°C
Source2: (Encyclopedia Britannica) 1300-1550°C
Hopefully we can get at least a few more minerals, and hopefully scrutinise the sources enough to come up with the best recoomendations.