The bibliography check:

Is each entry complete? Any obvious typos? Check against the actual book/article if you have it on hand. Does anything look wrong? If it does recheck it. But don't assume that if it looks right it is.

Are all entries actually cited your own book / article? Maybe you put something in the bibliography at an earlier stage and didn't use it.

Are all sources referenced in the text listed in the bibliography? Check each reference in chapter against the master bibliography.

Do all entries reflect the format of style you are using? (MLA, Chicago, etc...) Do you distinguish continuous numbering in journals?

Can a reference be ambiguous? For example, "Honig, 33," if there are three books by Honig in the bibliography... If I use three separate translations of Gypsy Ballads, I need a way of disambiguating them in the references as well as in the bibliography.

Can the reader find each entry in the bibliography easily? For example, I have Lorca/Blackburn under García Lorca, with a separate note under Blackburn to check García Lorca. I also have an entry "Lorca, Federico García" with a note to check under the more correct García Lorca, Federico.

What other issues specific to a particular project arise? Many citations of electronic sources?

***

Have you protected yourself against plagiarism? For example, giving credit to a critic who turned you on to another, third source can be a nice touch. There's a secondary plagiarism in simply taking someone else's bibliography as your own. You sometime want to say, "XX also cites this same text by YY," as a preemptive move, if you've read both XX and YY. Even if you read YY before you read XX quoting YY. It's not absolutely necessary in all cases, but if you carry citation practices this far then you'll also protect yourself against more primary forms of plagiarism. It's a truism that plagiarism is usualy sloppiness, laziness, rather than downright stealing, so the way to protect yourself is not to say, "I'm a good person, I would never STEAL anything," but to have a good understanding of scholarly work practices and the larger issues involved. (I am indebted to Thomas Basboll and his blog "Research as Second Language" for stimulating me to think about these issues. He makes some of these points in different ways in recent posts.)