Like the femme fatale in The Beatles' Rocky Raccoon, antiphospholipid antibodies (aPL) and the syndrome with which they are associated (thrombosis, recurrent fetal deaths, thrombocytopenia) have attracted considerable attention despite a serious nomenclature problem. Along with increasing experience in the optimal management of patients with the aPL syndrome and growing evidence that autoantibodies may play a direct pathophysiologic role, research over the past several years demonstrates that a large proportion of "antiphospholipid" autoantibodies do not, in fact, recognize phospholipids. These new data indicate that the antigenic targets of antibodies detected in conventional anticardiolipin and lupus anticoagulant assays are phospholipid-binding plasma proteins, most notably, &-glycoprotein I (&GPI) and prothrombin or complexes of these proteins with phospholipids. Further, autoantibodies to other phospholipid-binding plasma proteins and certain molecules expressed on vascular endothelium may also be associated with the antiphospholipid antibody syndrome (APS) although they are not detectable in standard aPL assays.This review summarizes recent information on the specificities of autoantibodies associated with APS and discusses the insights these data offer into the potential role of autoantibodies in the syndrome's pathophysiology. Antigenic specificities of aPL: historical perspective Current concepts of aPL date to the early part of this century and the development of nontreponemal serologic tests for syphilis (STS). In the early 1940s, Mary Pangborn identified the antigenic component of the tissue extracts used in these tests as a novel anionic phospholipid, which she named cardiolipin (1). As a result of widespread population screening for syphilis, in the 1950s it was observed that individuals with chronically false-positive STS results, when followed up for many years, often developed systemic lupus erythematosus (SLE) (2). False-positive STS results were also noted to be present in early reports of SLE patients with acquired inhibitors of in vitro phospholipid-dependent coagulation tests, in particular, the activation of prothrombin (3). Laurel1 and Nilsson investigated this association and suggested that the anticoagulant effect and the positive reactions with phospholipid antigen were ascribable to the same causal factor (4). Additional support for the hypothesis that these acquired coagulation inhibitors were directed against phospholipid came from studies demonstrating that the anticoagulant effect was enhanced when the phospholipid in the test system was diluted (5).In contrast, a number of key features of the inhibitors, subsequently termed lupus anticoagulants (6), were not compatible with phospholipid specificity and suggested an important role of plasma proteins. First, certain lupus anticoagulants were species specific, for example, they prolonged coagulation reactions of human plasma but not bovine plasma (7). Second, certain lupus anticoagulants did not inhibit phospholipid-dependent reacti...