I put a safety pin (which I falsely thought wasn't stainless, apparently) in a bit of brine in a mug and left it for 2 weeks on my desk. The mug ended up covered in salt crystals from the level of brine up to the edge, and even a bit beyond the edge. My only hypothesis how the salt gets transported to grow this crystal layer is as brine, through the crystal (which must be wet via capillary action). The surprising thing is that I can barely notice the crystals on walls are conductive (which a colleague suggested to me as a test for whether the crystals are moist) -- if I use my multimeter probes stuck <5mm apart, I sometimes get ~15MOhms. I haven't gotten around to measuring conductivity more properly (using contacts of larger surface area, looking at more of the V/I curve, etc.).
Another interesting thing is that the crystallized thing on the walls seems to have at least two layers of different structure. I haven't tried to figure out why yet (one hypothesis that I have is that it's related to the brine being made from iodized salt).
@robryk Yeah, reckon you're right on capillary action. Top layer crystallises as evaporation lets it hit saturation, brine continues to creep up and over those crystals, repeat.
Conductivity is interesting - many crystals have a hydrous and anhydrous form. The hydrous ones aren't "wet", the H2O molecules, individually, are part of the lattice (see concrete etc), which I'm guessing affects conductivity. You should be able to do "gravimetric determination"....
I would be extremely surprised if I could get mass transport through a crystal where all the liquid was bound into hydrates. After all, I can't see how I can get mass transfer without having ions dissolved in the liquid, which can't work then (right?).
I'd not be surprised if there were various hydrates there, but I don't think that alone would explain how the crystal grows on the end that's far away from the liquid, or would it?