This paper presents results of a case study that addresses many issues surrounding the difficult task of preservation in a digital library. We focus on a subset of these issues as they apply to the preservation of scholarly articles encoded in current web standards. We also describe the two common preservation mechanisms, emulation and migration, as well as our selection of the latter for our particular case. Finally, we compare two approaches to migration, automatic and manual, and discuss their strengths and weaknesses in our context. We show that consistent use of open standards leads to more efficient migration processes and issue a "call to arms" to the digital preservation community to ensure that scholarly material currently on the web can be preserved for future generations.