Anesthesia with diethyl ether significantly alters the course and outcome of experimental infections with the equine encephalomyelitis virus (Eastern or Western type) or with the St. Louis encephalitis virus. No comparable effect is observed in experimental infections produced with rabies or poliomyelitis (Lansing) viruses. The neurotropic virus infections altered by ether anesthesia are those caused by viruses which are destroyed in vitro by this anesthetic, and those infections not affected by ether anesthesia are caused by viruses which apparently are not destroyed by ether in vitro. Another striking difference between these two groups of viruses is their pathogenesis in the animal host; those which are inhibited in vivo by ether anesthesia tend to infect cells of the cortex, basal ganglia, and only occasionally the cervical region of the cord. On the other hand, those which are not inhibited in vivo by ether anesthesia tend to involve cells of the lower central nervous system and in the case of rabies, peripheral nerves. This difference is of considerable importance in view of the fact that anesthetics affect cells of the lower central nervous system only in very high concentrations. It is obvious from the complexity of the problem that no clear-cut statement can be made at this point as to the mechanism of the observed effect of ether anesthesia in reducing the mortality rate in certain of the experimental neurotropic virus infections. Important possibilities include a direct specific effect of diethyl ether upon the virus and a less direct effect of the anesthetic upon the virus through its alteration of the metabolism of the host cell.