The results of the ACE testing, in which each analyst received the same set of 60 fingerprint comparisons, showed 100-percent accuracy for all trials where an opinion of “identification” was reported (n=268) and an 86-percent accuracy for all trials where an opinion of “exclusion” was reported (n=14). The precision tests for the four categorical opinions reported (i.e., identification, exclusion, inconclusive, and no value) all passed threshold criteria that were determined before the test was administered. Reproducibility (the ability of all the experts to reach the same result independently) and repeatability (the ability of the test to provide the same answer in re-analysis of the material) were assessed. The results varied, depending on the amount of information present in the friction ridge impressions and generally how the images were presented to the participants. The ACE-V trials yielded 100-percent accuracy for all trials in which an opinion of identification was reported (n=271) and a 67-percent accuracy for all trials where an opinion of exclusion was reported (n=18). The precision tests for the four categorical opinions during the ACE-V trials showed higher precision than ACE trials during the verification stage. The number of erroneous exclusions doubled when a verifier was present to verify exclusions; however, the verifiers detected every false positive presented to them (n=9). No erroneous exclusions were detected by verifiers. This showed a strong resistance to bias regarding identification but a high degree of susceptibility to bias toward exclusions during a nonblind verification procedure. Methods and materials are described in detail. 11 tables, 3 figures, 26 references, and 1 appendix