The presumptive identification of saliva has traditionally been performed by the identification of amylases found in human body fluids, but which are also present in micro-organisms, plants, and many animals. This has been done with methods such as the Phadebas assay or radial diffusion in a starch/agarose gel; however, these traditional methods cannot differentiate between amylases. A method has been developed that can differentiate human salivary amylase from other types of amylases. This indirect ELISA technique uses a monoclonal salivary amylase antiserum. The current study tested this method by using 8 saliva swabs, 8 semen swabs, 15 urine swabs, 5 blood swabs, 7 fecal swabs, 14 perspiration swabs, and 5 swabs of postmortem stomach contents. In addition, individual swabs were made from whole saliva diluted with phosphate buffered saline. This report describes in detail the collection and preparation of samples, the ELISA procedure, and background absorbance and plate threshold determination. Study results show that the ELISA method is highly specific for human salivary amylase and can be used as a confirmatory test for human saliva. All saliva-containing swabs and 13 percent of non-saliva human body fluid swabs showed a net absorbance with the method. Of the eight non-saliva swabs that yielded a net absorbance, none exceeded a salivary amylase activity of 0.003 units. By comparison, only 3 of the 16 saliva-containing swabs showed an activity below 0.2 units. 3 tables, 1 figure, and 13 references