484 HANAHAN leased from the leukocytes of sensitized immunized rabbits. This mediator (s) then activated platelets releasing vasoactive amines. Subsequently two groups of investigators, Sirganian & Osler (2) and Benveniste, Henson & Cochrane (3) independently reported a confirmation of that earlier observation. The latter group also coined the term, platelet activating factor, and this led to further studies by Benveniste and collaborators on the chemical identification of the mediator. During this period the Paris group gave substance to the fact that platelet activating factor was potentially a lipid but did not precisely define its chemical nature. However, in 1979, our laboratory in San Antonio (4) reported the semisynthesis of an alkylacetylglycerophosphocholine (AGEPC) which mimicked exactly the biological behavior of the naturally occurring material (5), and very shortly thereafter Benveniste and colleagues (6) reported on an alternate semisynthetic pathway to the same compound. At the same time, Snyder and collaborators in Tennessee (7) published a semisynthetic approach to AGEPC which they considered to be the hypotensive agent under active study in their laboratory. It is now well established that the hypotensive agent and the platelet activating factor are one and the same compound. However it was not until some nine months later that our group in San Antonio was successful in elucidating the structure ofthe naturally occurring platelet activat ing factor generated from sensitized rabbit basophils (8). Its chemical structure is particularly novel as shown in Figure 1. This compound is unique in several ways: 1. It represents the first bona fide example of a biologically active phosphoglyceride. 2. It possesses an O-alkyl ether residue at the sn-I position and a short chain acyl moiety, i. e. acetyl, at the sn-2 position. At the sn-3 position, the polar head group in all naturally formed platelet activating factors is that of an O-phosphocholine group. Although the configuration of the naturally occurring material is presumed to be that of the sn-3 type by reference to the activity of synthetic model compounds, in actual fact the amounts of material isolated in the usual instance are too small to allow a strict confirmation of its stereochemistry. To date, it has been shown that 1-0-alkyl derivative is some 300-fold more active than the analogous fatty acyl derivative (4). Further, all reports on the CH20 (CH2), CH,