excluding NAs in data frame without deleting rows

Hi all,

I have a binary matrix with NAs included.  Each row and column
includes at least one NA, so I don't want to omit them.  Is there a
way to sum across rows and columns, ignoring the NAs but not deleting
the row or column?  If not, I suppose I can write a loop function, but
I have learned that it is best to stay away from loops if possible.

Thanks for any help,

Wade