on 2 January 1860 started one of the most embarrassing accidents of the history of planetary astronomy, with Le Verrier's official announcement of the discovery of the "intramercurial" planet Vulcan.
http://en.wikipedia.org/wiki/Vulcan_(hypothetical_planet)
actually, embarrassing though it was, the accident provided the first experimental data to confirm Einstein's General Relativity. But this was not realized until 1915...

Well these things happen all the time in science and not only in astronomy. The more contemporary question is when is enough evidence enough? Will you keep delaying publication of your findings and risk that someone will scoop you or will you publish and risk being wrong? It's a balancing act and it's not easy. Funding agencies do not like scientists who sit on results while they make sure they are "right".