In a surgical trauma center, programs and workshops have improved the performance on focused abdominal sonogram for trauma (FAST). The purpose of this single-blind study was to prove that a cadaver laboratory competency-based instruction program may be an effective method of FAST training to acquire the skills that would be applied in the trauma room. The study was divided in two parts, laboratory and clinical. Nine surgical residents were divided into two groups: Group I performed the test only once, and Group II performed the training twice. A third "group" was the senior ultrasound technician, whose readings served as our "gold standard" with which to compare the resident readings (Group III). Using cadavers, a 2-cm catheter was introduced into the peritoneal cavity. Sequential aliquots of normal saline were introduced into the abdominal cavity at 0-, 200-, 400-, 600-, and 1000-cc increments in each group tested. The residents were asked to describe their examinations for the presence or absence of fluid in the abdomen. The ultrasound examination was then performed with the cadaver in three different positions to study if there was any difference of fluid detection in varied positions. True positive, true negative, and accuracy were then calculated comparing the three different groups of test sonographers. In the second part of the study, the same residents were then followed in the trauma room, where they performed the FAST in the absence of the ultrasound technician during emergencies. As in the laboratory, the accuracy of their reading compared with that of the ultrasound technician was also evaluated. From 400 cc and upward, Group II began having an overall significantly superior accuracy than the first group and the technician in most quadrants examined. The trend was apparent for more accurate results in all quadrants and positions by all groups as the fluid was increased. Overall, group II was most superior in detection of intra-abdominal fluid in the cadaver. In the clinical scenario, the residents as a whole had similar accuracy (92% vs 96%) in reading FAST as the ultrasound technician. Our results suggest that surgical residents have the ability to detect fluid in the abdomen, there exists a fast learning curve, and the minimum detection level of fluid was between 200 and 400 cc in the peritoneal cavity in the laboratory. Surgical residents were able to detect intra-abdominal fluid in the trauma situation, as shown by the 92 per cent accuracy of the FAST in the emergency situation. We conclude that a cadaver laboratory training program is an important adjunct to improve the skills of the resident in performing and reading FAST.