Natural communities are often characterized by species rich, reticulate food webs.  Efforts to estimate the strength of species interactions in such systems have been hampered by the multitude of direct and indirect interactions their food webs exhibit and have been limited by an assumption that the strengths of pairwise species interactions display linear functional forms.  Here we present a new method for directly measuring, on a per capita basis, the nonlinear strength of trophic species interactions within such food webs.  The approach of this method is observation-based, requiring three pieces of information obtainable through field surveys and laboratory experiments: (1) species abundances, (2) predator-prey-specific handling-times, and (3) predator-specific feeding-surveys that tally the number of predator individuals observed to be feeding and not feeding on each of the predator's prey species at any given time.  In this talk we describe the advantages and assumptions of the method, and apply it to data from a New Zealand intertidal food web that reveals some interesting implications concerning the way ecologists measure and model the species interactions of empirical food webs.