Biotic mixing is becoming very wide- spread from human activities and is recognized as a potential threat to native biodiversity. In theory, species-area curves can be used to predict the ultimate biodiversity loss and homogenization occurring from faunal mixing. One way to test this empirically at coarse spatial and temporal scales is to quantify past episodes of faunal mixing in the fossil record. Analysis of diversity changes following the origin of the Isthmus of Panama indicates that marine biodiversity change was similar to that predicted by species-area patterns: the isolating effects of the Isthmus produced a roughly 45–60% increase in total biodiversity in the Atlantic and Pacific regions nearby. Conversely, if the isolating effects of the Isthmus were to disappear through the Panama Canal and other human activities, a decrease of about 31–38% of current species diversity might be expected. Extrapolation of these results to global scales indicate that global marine species diversity may decrease by about 58% if worldwide biotic mixing occurs. This supports previous estimates based on terrestrial biotas that predict global decreases from 35 to 71%, with a mean decrease of 55%.