Discussion and Summary The rabbit has proved useful for studying certain physiologic changes brought about by R. mooseri toxin. Depending on the dose of toxin employed, either rapid or late death was observed. In general, rapid death was accompanied by moderate to massive progressive hemolysis which resulted in maximum hemoglobinemia at the time of death. Hand in hand with such rapid death and massive hemolysis, a progressive increase in plasma potassium concentration was regularly observed and in the course of 30 to 60 minutes reached extremely high levels. In contrast to such rapid death, an entirely different sequence of events was noted in those rabbits which expired after several hours. Here, although moderate hemolysis did occur, it was maximal long before exitus. Hyperkaliemia developed; however, it did not appear until these animals were prostrate and death was close at hand. Thus, although both hemoglobinemia and hyperkaliemia did appear in these rabbits dying late, they did not bear the same close temporal relationships as were observed in the rabbits dying rapidly. In both types of death electrocardiographic abnormalities were usually observed. These abnormalities were compatible with hyperkaliemia and were associated with the presence of elevated potassium concentrations. In most rabbits a fall in arterial blood pressure appeared soon after the injection of toxin. Subsequently, hypotension became more marked and reached similar low levels irrespective of the time of death. Rapid death would appear to result from the rapidly rising plasma potassium concentration which attained terminal levels of 12 to 15 mEq/L. Such plasma potassium levels are known to be incompatible with life in several experimental animals (16, 17, 18) and man (19). Furthermore, the electrocardiographic changes observed in this study bear a close resemblance to those reported in hyperkaliemic death of dogs (16, 17), cats (18), rabbits (20) and man (21). The rapid massive hemolysis is adequate to account in large part for the profound hyperkaliemia. Thus, based upon hematocrit determinations approximately 30% of the total circulating red blood cell mass of these rabbits underwent lysis prior to death. Theoretical calculations for a rabbit weighing 1,000 grams using 70 ml and 42 ml for the total blood and plasma volumes, respectively, (22) and our observed value of 116 mEq of potassium per liter packed rabbit erythrocytes, indicate that the destruction of 30% of the circulating red blood cell mass in the course of 1 hour would liberate potassium into the plasma at a rate approximating 0.02 mEq/Kg/minute. However, in vivo hemolysis does not appear to proceed at a linear rate which is assumed in the above calculation, but rather increases rapidly to a maximal rate at the time of death. Consequently, at this time the liberation of potassium may well exceed the theoretical values. Normal rabbits are reported to succumb when potassium is infused in excess of 0.025 mEq/Kg/minute (23). It is of interest to note that potassium concentrations comparable to those observed here have been reported to occur in rabbits just prior to death following rapid lysis of a comparable red cell mass by saponin (24). Additional factors, such as the rapid onset of shock or the untoward effect of marked hemoglobinemia on renal function described in man (25), may have served to retard rapid elimination of this cation through normal renal channels. Further evidence, suggesting some impairment of renal function, was found in the elevated terminal plasma phosphate levels in these animals. Presumably, a large amount of this anion was derived from the in vivo destruction of erythrocytes, and, because of altered renal mechanisms, it accumulated as did potassium. Whether the increased potassium in the plasma of these rabbits arose entirely through the mechanism of red cell lysis or in part from damaged tissue cells cannot be stated with certainty at this time. However, we suspect that damaged tissue cells also provided a share of the potassium. Whatever its source, large amounts of potassium did accumulate and reached lethal levels. The role of hemolysis in the pathogenesis of late death in the rabbit is difficult to evaluate. Varying degrees of renal failure, manifested by terminal increases in B.U.N. values, were demonstrable in certain of these animals which expired after a period of 60 minutes or more subsequent to the injection of toxin. Plasma phosphate levels were similarly noted to be increased at the time delayed death occurred. A slow liberation of potassium from erythrocytes as well as other damaged cells, accompanied by increasing renal malfunction, may explain the delayed but definite accumulation of potassium observed. The end result, however, was potassium excess and inevitable exitus. The marked hemoconcentration demonstrated in rats and mice by this study and previously observed in the latter by others (1, 2) suggests that rapid plasma exudation does occur in these species and implies that capillary damage may play an important role in the production of toxic tissue damage. We do not wish to suggest that there is absence of capillary damage in rabbits which die from rickettsial toxin. However, our data permit no statement regarding the presence of such damage in this species of experimental animal. Hypotension was demonstrated to varying degrees in the majority of treated rabbits. The exact mechanism of its production is not clear. Regardless of its origin the severe hypotension might be expected to give rise to numerous secondary changes which could easily be misinterpreted as primary effects of R. mooseri toxin. The results of this study concerning R. mooseri toxin clearly indicate the value of employing several different species of susceptible hosts in defining the general physiologic effects of a given toxin.