In order for a hypothesis to be valid, it must be testable.
Experimentation is required
yes, but no.
What experimentation has been done to demonstrate that observations made in the fossil record are the result of an evolutionary mechanism?
That's not how it works. You have a hypothesis. It should make predictions. To test the hypothesis, you write down the predictions before looking at the test. You have someone else collect the data. Then you compare. If the predictions match the data, your hypothesis works.
With evolution, it makes many, many predictions. One of them is this: as oganisms evolve, some will die out. They will be present further down in the geological column, but not further up. New organisms will not be present further down, but will be further up. To test that - or to experiment - one looks at what fossils are found where. Older, extinct species - like dinosaurs - are not present beyond a certain point. Check. Newer species - like , people or modern rabbits - are not found mixed in with dinosaurs. check. Test (experiment) confirmed.
There are other more extensive tests that can be done with genetics, if you look them up.