There has been a flood of recent work on polarization effects in optical fibers, most of it focused on polarization mode dispersion (PMD). There are good, practical reasons for this recent interest. As the data rate per channel increases, PMD becomes an increas-ingly important limitation in communication systems. However, my own interest in polarization issues was, at least originally, focused on more fundamental questions. Since nearly all the contributions in this volume focus on the more immediate practical issues that are a consequence of polarization effects, it seemed useful to me to focus this contribution on more fundamental issues. The nice thing about fundamental issues is that even though they attract less attention than immediate practical issues, they typically stay relevant longer and can impact practical issues 10-20 years in the future.Throughout the 1970s, a considerable body of work established the basic polarization properties of optical fibers. Much of this work was summarized by Kaminow in 1981 [1]. It was found that the birefringence ∆n/n is in the range 10 −4 -10 −9 , with communication fibers in the range 10 −6 -10 −7 . It was found that the intrinsic