Statistically comparing samples with only means and SD


I'm looking for some help with some data I'm trying to analyse.
I have data consisting of three samples, all three of which are the means of a gaussian data set, for which I have the standard deviations (I don't have access to the raw data, unfortunately). I would like to know if it is possible to run an ANOVA or ANCOVA test on the data that I have, and if so how I would go about it? Is there another parametric test I could use?

Any thoughts would be greatly appreciated; this has been driving me crazy for some time now!!