Giant axons from the squid Dosidicus gigas were internally perfused with rabbit antiaxoplasm antibodies and their effect upon the action potential and the membrane potential was studied. Necessary requirements for the antibodies to affect these parameters in a consistent manner were: (a) removal of the bulk of axoplasm from the perfused zone, accomplished by initially perfusing with a cysteine-rich (400 rnM) solution, and (b) addition of small amounts of cysteine (30 mM) to the antibody-containing solution. When these experimental conditions were met, conduction block ensued generally within 3 hr of the first contact of the axon inner surface with the antibody Antineurofilament antibodies and nonspecific antibodies had no effect. External application of antiaxoplasm antibodies had no effect.Interest in the effects of specific immune sera on nervous system function has been stimulated by the recent studies of L. Mihailovid and collaborators (1, 2). T h e y have shown that antibodies prepared against tissue from well-defined anatomical areas of the brain will alter the electrophysiological properties of that area when injected diffusely into the central nervous system, in certain instances showing a marked specificity (1). These authors also reported that lobster giant axons become inexcitable when immersed in a medium containing antibodies against the lobster ventral cord. T h e y conclude that the antibodies exerted their effects from the inner side of the membrane, penetration being achieved by leakage at the microelectrode impalement site (2).Analytical studies from this laboratory (3) on the macromolecular constituents of uncontaminated axoplasm extruded from Dosidicus gigas giant axons have revealed 15 protein species which are resolved by polyacrilamide discontinuous electrophoresis, and 7 of these were shown to be antigenic as determined by precipitation lines on agar; none of these 7 proteins is present in squid blood. The present communication reports the effects of internal