In the course of investigation on precipitin and complement fixing antibodies produced by injections of edestin, it was observed that, while the serum of a rabbit immunized with this protein showed the presence of precipitin antibodies, it did not show any complement fixing antibodies when employing the usual 2 units of complement in the fixation tests. It appeared reasonable at first to accept this finding as evidence of the lack of relation between these two types of antibodies. It seemed, however, that possibly the employment of 2 units of complement in the tests might give a sufficient excess of this ingredient to render a serum negative for complement fixing antibodies, although a reasonable number of such antibodies might still be present in the serum. With this possibility in mind, a series of complement fixation tests were carried out, using complement gradations of 1¼ units, 1½ units, 1¾ units and 2 units. The quantity of serum employed was 0.01 c.c.It was observed that the same serum which gave ...