Late last week I reported my puzzlement over an error I encountered using Biostrings with R 2.6. The error was reported as an improper call to nchar. I received good advice from Martin, Herve and Henrik, which helped me to resolve the problem. I report this here in case it is of use to anyone else. Martin & co. suspected that -- even though I was sure I had installed the latest Biostrings to use with R 2.6 -- I might have an older version of Biostrings lurking somewhere in my libPaths. By starting up with R --vanilla the older Biostrings was ignored, and I learned that the new Biostrings is indeed compatible with the new 3-argument 2.6 version of nchar. Thanks to all who helped! - Paul