The life cycle of large anticyclonic rings in the Gulf of Mexico (GOM) is widely described by pinch off from the Loop Current, migration across the Gulf and eventual spin down along the western slope. Extensive observational and modeling efforts provide a relatively consistent picture of rings pinching off from the Loop Current and of complex interaction between anticyclones and cyclones driven by bathymetry along the western and northwestern shelf. The observational record for Loop Current rings (LCRs) during the intermediate period of westward translation is less clear. A number of studies recognize distinct anomalies in LCR characteristics in deep water as the rings enter the western Gulf near 92-94W. These include abrupt changes in the geometry of observed drifter trajectories and derived eddy parameter fits as well as changes in both ring translation speeds and the estimated rate of ring decay. Such observations are consistent with intense interaction and mass exchange between the rings and other coherent mesoscale features known to be present in the western Gulf. We test the hypothesis that interactions with the ambient mesoscale field can lead to rapid loss of coherence of some LCRs well before they reach the 'eddy graveyard' in the western Gulf. We use the data-assimilating, eddyresolving numerical GOM model described by Kantha et al. (2005) to assess the fates of readily identified LCRs Fourchon, Juggernaut, and Millenium during the period July 1998 to August 2001. Lagrangian metrics, including relative dispersion of small drifter clusters seeded in the ring cores, analysis of evolving blobs seeded in the ring cores, and finite-scale Lyapunov exponents, are used to track model ring evolution. These metrics clearly show that interactions with existing mesoscale cyclones and anticyclones caused Fourchon and Juggernaut to break up near 92W on advective time scales. In addition, Millenium also experienced an intense deformation, stirring, and mixing episode near 92W. Blob studies showed that the core fluid of Millenium was ultimately dispersed over much of the western basin. Our results show that some LCRs may break up through interactions with existing western Gulf cyclones and anticyclones prior to reaching the western slope.