In 1908, Moreschi (8,9) found that bacteria and red blood cells did not agglutinate when placed in contact with their specific immune serum. Agglutination did occur, however, when an antiglobulin immune serum for the animal species used in the reactions was added to the system. Unfortunately, the finding was not exploited, however, since it did not come to the attention of researchers generally. Coombs et al. (6), in 1945, extended Moreschi's studies without being aware of them. In the years that followed, Coombs made major contributions to the knowledge of blood groups through application of the antiglobulin (AG) procedure, and made many interesting and ingenious modifications of it. Mycoplasma infections do not normally elicit a high agglutinin response. Recently, the AG test has been applied to the detection of antibodies to Mycoplasma gallisepticum (MG) not revealed by saline agglutination. Adler and DaMassa (1) demonstrated that this procedure would enhance saline agglutination reactions with mycoplasma of avian origin. They were able to show that M. synoviae was antigenically related to MG, an observation not made previously. This paper describes facets of the AG procedure that are useful in adapting the method to identification of incomplete antibodies or to enhancement of weak antibodies in completing a serologic reaction. The antigen employed in most instances was MG.