Mild strains of the virus of herpes are described the action of which tends to be confined and local. Unless, therefore, these mild strains are injected intracranially they do not tend to produce virus encephalitis in the rabbit.
Recovery from infection with the mild strains confers immunity to virulent strains of the herpes and allied viruses.
Long glycerolation reduces the number of viable organisms. This loss among the mild strains may reduce the virus below the strength required for an effective extracranial although not below the strength needed for an intracranial inoculation.
Herpes virus carriage in man, even under highly favorable conditions, is difficult of detection by means of rabbit inoculation. The detection may be achieved by intracranial when it cannot be accomplished by intracorneal inoculation.
The virus producing encephalitis in the rabbit attaches itself chiefly to and multiplies in the substance of the central nervous system. Hence its detection in the cerebrospinal fluid is rarely accomplished. When the inoculation of the virus is made intracranially and especially when the inoculum is composed of active brain tissue, the virus is discoverable in the cerebrospinal fluid by rabbit inoculation much more frequently than when the virus encephalitis follows an extracranial variety of infection.
The herpes virus is capable of excretion by the kidney of the rabbit and of being detected in the urine by rabbit inoculation.
Among the rarer symptoms of virus encephalitis is excessive lacrimation. While salivation is frequent, lacrimation is exceptional.
A comparison of the Levaditi, Doerr, and Goodpasture strains of virus indicates the first to be of medium, the second of mild, and the third of high degree of neurotropic activity. The Doerr strain resembles the mild herpes strains described in this paper. The Goodpasture virus, while exceeding the Levaditi strain in affinity for the central nervous system, falls below the H. F. strain in this regard.
Neutralization of virus by the serum of infected and recovered rabbits takes place regularly within certain quantitative limits. Neutralization with human serum is inconstant and capricious and without demonstrable relation to previous attacks of epidemic encephalitis.
Comparison of the clinical types of encephalitis as presented by the epidemic variety in man and the experimental virus variety in rabbits brings out certain correspondences and certain differences. It is only in partial and essentially superficial aspects that the two diseases can be identified one with the other.