Michele, My suggestion is that you generate id's for the unittitles of these components, then generate an index from the dsc with the appropriate targets. The generated index (if it is serving properly), should result in the same functionality. With the back of the book material such as indexes, the intent of the index and nexus of information with the dsc should be ascertained and replicated in your new index. You could first mark up by hand what the component and corresponding index are supposed to look like. This could serve as a model for a stylesheet that generates the index from the dsc. Example: http://hdl.loc.gov/loc.music/eadmus.mu002007.3 Sincerely, Mike Ferrando IT Specialist Library of Congress Washington, DC (202) 707-4454 ________________________________ From: Michele R Combs <[log in to unmask]> To: [log in to unmask] Sent: Thu, May 20, 2010 1:00:29 PM Subject: Index to correspondence? Some of our older collections have their correspondence arranged chronologically; the finding aid then includes, at the end, an index to correspondence, in this form: Acevedo, Cristobal de 1962 Feb 5 1962 Mar 16 1962 Sep 10 American Oxonian 1947 Mar 13 (Denham Sutcliffe) Amis, Kingsley, 1922- 1957 Nov 9 1958 Jan 6 1958 Jan 21 1958 Aug 22 ...etc. We want this included in our finding aid, and we want "Index to correspondence" to appear in the auto-generated table of contents. Unfortunately, the <index> construction in EAD requires the <ref target=""> element, more suitable for traditional indexes with page numbers; it doesn't really work for something like this. So I'm wondering if, and if so how, others have chosen to encode such indexes to correspondence. As a simple nested list? In a separate file as a related finding aid? Other options? Thanks -- Michele (be green - don't print this email!) ~~~~~~~~~~~~~~~~~~ Michele Combs Manuscripts Librarian Special Collections Research Center Syracuse University Libraries 222 Waverly Ave. Syracuse, NY 13244 315-443-2081 [log in to unmask] ~~~~~~~~~~~~~~~~~~