Radiographic contrast material in doses commonly used in clinical practice produces major alterations in cardiac rhythm. In order to separate the direct cellular effect of the constituents of contrast material from the indirect actions of neural and hormonal mediators, a tissue culture technique was developed using newborn rat heart cells. The rate of contraction of both individual cells and of a syncytium of cells decreased and became briefly asystolic after contrast material was added to the media, then recovered with varying degrees of arrhythmia. Cells in culture less than one week showed significantly fewer arrhythmias and less recurrent asystole after contract material than older cells (P less than 0.05). Cells pretreated with pharmacologic amounts of atropine (0.08 microgram/ml), ouabain (0.02 microgram/ml), lidocaine (16 microgram/ml), and quinidine (20 microgram/ml) continued to demonstrate the initial asystole, although quinidine and lidocaine diminished the frequency of arrhythmias in the recovery stage. Hyperosmolar dextrose produced asystole, as did Renografin-60, Hypaque-50, and Conray only at concentrations greater than 1000 mOsm, suggesting that an osmolarity threshold may exist for the production of asystole. Less fibrillation occurred when the sodium and calcium ionic concentrations of contrast material were adjusted to that of plasma, although the arrhythmias could not be eliminated. Renografin-60, containing the greatest amount of calcium-chelating agents, produced a significantly greater degree of terminal asystole (P less than 0.05), which was reversible upon addition of CaCl2. Thus, contrast material produces a decrease in the rate of contraction in cultured cardiac cells in the absence of neural and hormonal mediators. The evoked arrhythmias are dependent upon the osmolality and ionic sodium and calcium concentrations.