Thanks! And sorry, I should have been clearer. My ultimate goal is to convert the endnotes into footnotes and then the footnotes into inline citations (but I do have some code—actually, *your* code!) to perform that task. But I do also find it useful to have a version of the document with actual footnotes so my goal would have been to process the files twice (once for the conversion to footnotes and once to get them into inline citations). Does this clarify the problem? Thanks again!