Previously unaccounted cryptic speciation requires revaluating species identification, particularly for fisheries assessment purposes. We describe a rapid and reliable method for distinguishing between phenotypically similar species that utilizes simple otolith morphometry (length, width, thickness and weight) with or without fish length, within a traditional canonical discriminant analysis (CDA). Data were subject to CDA in order to differentiate between the cryptic Etelis carbunculus (ruby snapper) and E. marshi (pygmy ruby snapper). A very high allocation success rate was achieved using otolith morphometry and fork length (99.6% for E. carbunculus and 100% for E. marshi) or otolith morphometry only (98.8% for E. carbunculus and 100% for E. marshi), which indicated the high discriminatory power of this method. The CDA successfully grouped samples of the same species collected from different locations in the eastern central Indian and South Pacific Oceans, indicating the robustness of this technique to discriminate between species, irrespective of their geographic range. This technique can be applied to archived otolith collections to confirm teleost species identification, and likely has broader applications for species identification involving extractive, diet or video-based studies. © 2013 Elsevier B.V.