different results of fisher.test function in R2.8.1 and R2.6.0

Hi;
I use the function fisher.test to compute in R2.8.1 and R2.6.0,and the
results are not identical.the last number is different.  why?
thank you !

Can you show us the data?  The only change listed in the NEWS file since 2.6.0 was with the option simulate.p.values=TRUE - you don't say whether you used that option.

      -thomas


Thomas Lumley			Assoc. Professor, Biostatistics
tlumley at u.washington.edu	University of Washington, Seattle