During the acquisition of seismic data in a marine environment, the positions of the hydrophone groups are derived from data obtained using compass units deployed at intervals along the length of a streamer. Interpretation of the compass data to derive the streamer position has always proved difficult. The data is a set of measurements of spatial gradients at known distances along the streamer. The integration method used to derive the streamer shape requires additional data to resolve the undetermined constant. Thus the derived positions of the hydrophone groups have a translational indeterminacy. The compass measurements are independent and all measured azimuths can he expected to have similar error estimates. However streamer shapes, derived using a least-squares polynomial fitting algorithm are critically dependent on the accuracy of data obtained from the first and last units. These features have lead to innovative schemes to position the two ends of the streamers. Lasers, acoustics, and radio positioning, have all been used to locate accessible points closest to the streamer ends viz the towpoints and the tailbuoy. While this additional data has improved the quality of streamer positioning, these methods still leave the compasses as the only means of generating position information in the region where the hydrophone groups are actually situated.