The kinetics of protein synthesis inhibition in a cell-free system from rabbit reticulocyte lysate was studied after addition of abrin and ricin and the isolated A chains. The toxin A chains inhibited protein synthesis at a rate proportional to the amount added. When intact toxins were added to the reticulocyte lysate, the kinetics of protein synthesis inhibition indicated that the A chains must be liberated before ribosome inactivation can take place. The splitting of the toxin in the lysate was directly demonstrated by the use of labeled toxins. The amount of abrin and ricin bound to HeLa cells under different experimental conditions was correlated to the concomitant inhibition of cellular protein synthesis. In the presence of lactose, which inhibits toxin binding, much higher concentrations of toxins were required to inhibit protein synthesis than in the absence of lactose. A linear relationship was found between the lactose concentration in the medium and the toxin concentration required to give 50% reduction in protein synthesis after 3 hours. The amount of toxin bound to the cell surfaces in the presence of lactose was either determined directly or calculated from the apparent association constant between toxins and surface receptors at the various lactose concentrations. Under different conditions involving a 300-fold variation in the concentration of toxin required to reduce protein synthesis by 50% after 3 hours, the amount of toxin bound to the cell surface was found to be the same. The toxicity thus appears to be determined by the number of toxin molecules bound to the cell surface. Purified ricin B chain was used to compete with the toxins for the receptor sites. Only after addition of high amounts of B chain was the toxicity of abrin and ricin appreciably reduced. The data do not support the view that receptors with especially high affinity are involved in the uptake of the toxins. When the time required for 50% inhibition was plotted versus the inverse value of the square root of the number of toxin molecules bound per cell, a straight line was obtained, intercepting at about 30 min. The data indicate that the observed lag time cannot be due entirely to the fact that the A chains must be liberated before they can act.