This paper presents preliminary results of an on-going research project aimed at identifying a potential network of biological corridors that would link existing and proposed protected natural areas in Central America. Based on current theories from the disciplines of landscape ecology and conservation biology, maintaining or restoring landscape linkages would contribute to the protection of biological diversity in the region. Funding was provided by USAID in collaboration with the Paseo Pantera Consortium. The specific results presented here are those which were made possible by the spatial analysis capabilities of ArcInfo GRID software. The first step in our analysis was the development of a weighted-criteria suitability model which forms the basis for the creation of alternative versions of a 'relative cost surface.' In the next step, a resulting cost surface is used with the GRID 'corridor' function to produce a corridor potential surface. Finally, a feasability surface is generated by 'clipping' the suitability surface with those areas that have the greatest corridor potential. This project demonstrates that although there is great potential for using the GRID 'corridor' function for this type of research, continued work on the thoretical basis for the creation of valid 'relative cost surfaces' is necessary.