Hi Paul,
The whole world of grep and regular expressions is a completely new one to me, but I'm discovering that it is fantastically powerful. It had been my intention to strip out most of the email header material in the Studer List so that just "From," "To," and "Subject" were left, and I was gradually performing hundreds of search and replace operations to do this. What I didn't realize at the outset was how variable the email record structure was, and how many damaged/incomplete/anomalous records there were. So I lost a couple of days of work when I discovered that I had these weird floating fragments at the end of the process. On this second iteration, I'm processing the records much more slowly and gently, working with each of the seven segments individually and dealing with the problem records more systematically. I'm very aware that this is not only an important source of information for those of us going forward into the future with open-reel recorders, but the historical record of a large number of prominent people, and it's terribly important that I not distort it in any way. At this point, I'm fairly certain that every character ? some 12 million of them! ? will be exactly where they were originally, or were supposed to be.
When the TextWrangler documentation wasn't adequate to what I needed to do, I ordered a copy of Jeffrey Friedl's _Mastering Regular Expressions_ on the recommendation of a friend. I'll have to look up your son's book, as that may come in handy too ? thanks for the recommendation.