The linkage was supposedly between the variation in solar cycle *length* (it's not always 11 years) and temperature, which is a rather tentative connection. Entertainingly, it looks like this is yet another case of cherry-picking your data to fit the results you're looking for.

http://stephenschneider.stanford.edu...onLaut2004.pdf

I saw another graph showing the cyclic behaviour over a longer period of time, but naturally I can't find it now. At any rate, it showed that the supposed correlation simply wasn't there. And the presented problems with the CO2-forcing model based their reasoning on old, outdated data. Hmm, something about this whole thing sounds eerily familiar...

This image shows a pretty good breakdown of the various factors influencing climate.
http://en.wikipedia.org/wiki/Image:C...ttribution.png