/images/quoteimage/arthur-eddington-fb-63872.jpg

It is also a good rule not to put overmuch confidence in the observational results that are put forward until they are confirmed by theory.

Arthur Eddington