We investigate hybridization and introgression between ecologically distinct sister species of silverside fish in the Gulf of California through combined analysis of morphological, sequence, and genotypic data. Water diversions in the past century turned the Colorado River Delta from a normal estuary to a hypersaline inverse estuary, raising concerns for the local fauna, much of which is endangered. Salinity differences are known to generate ecological species pairs and we anticipated that loss of the fresher-water historic salinity regime could alter the adaptive factors maintaining distinction between the broadly distributed Gulf-endemic Colpichthys regis and the narrowly restricted Delta-endemic Colpichthys hubbsi, the species that experienced dramatic environmental change. In this altered environmental context, these long-isolated species (as revealed by Cytochrome b sequences) show genotypic (RAG1, microsatellites) evidence of active hybridization where the species ranges abut, as well as directional introgression from C. regis into the range center of C. hubbsi. Bayesian group assignment (STRUCTURE) on six microsatellite loci and multivariate analyses (DAPC) on both microsatellites and phenotypic data further support substantial recent admixture between the sister species. Although we find no evidence for recent population decline in C. hubbsi based on mitochondrial sequence, introgression may be placing an ancient ecological species at risk of extinction. Such introgressive extinction risk should also pertain to other ecological species historically sustained by the now changing Delta environment. More broadly, salinity gradient associated ecological speciation is evident in silverside species pairs in many estuarine systems around the world. Ecological species pairs among other taxa in such systems are likely poorly understood or cryptic. As water extraction accelerates in river systems worldwide, salinity gradients will necessarily be altered, impacting many more estuary and delta systems. Such alteration of habitats will place biodiversity at risk not only from direct effects of habitat destruction, but also from the potential for the breakdown of ecological species. Thus, evolutionary response to the anthropogenic alteration of salinity gradients in estuaries merits investigation as the number of impacted systems increases around the globe, permitting parallel study of multiple systems, while also permitting a conservation management response to help preserve this little championed component of biodiversity.