Take an ordinary glass and fill it with tapwater. Next take a metal spoon and dissolve a heaped teaspoon of salt in the glass. Immediately after dropping the salt in the glass ensure that the spoon constantly taps the side of the glass ... making that familiar tink-tink sound. What do you notice?

You should hear the frequency of the sound lower quite dramatically over a period of about 10-15 seconds and then slowly climb back up to the original frequency (before the salt was in the glass). In the spectrogram above, the frequency drops from about 2500Hz at time 0 to about 500Hz after about 6 seconds. One of the first thoughts that springs to mind is that the process is based on a temperature change --- however the glass does not seem to heat up appreciably when the salt drops in.

If you are too lazy to actually try this for yourself then simply listen to the version I prepared earlier.

Interestingly, if you try the same thing with sugar you notice no such effect. This would seem to point towards an ionic-bond versus covalent-bond based explanation.