R help - May 2012 - Performing negative binomial regression on data, using stepwise selection

hello all,
I've been trying for some time to try and model a big dataset using
generalized linear models, or more precisely negative binomial regression.
Unfortunately i have too many explanatory variables and i'd like to cut
some of them off using a stepwise selection method.
i've been struggeling with the code for quite a long period and can't
find
a way to do this....any ideas?

Thanks,
Eli.

	[[alternative HTML version deleted]]