Methods for intranasal inoculation of viruses are often described poorly and the effects of variations in the technique on the outcome are unknown. Standardization of protocols is key to compare studies and minimize animal use. The clinical and virological outcome of infection with rabies virus (genotypes 1 and 5) upon administration of different inoculum volumes (25, 50 and 100μl) and different anesthetic regimens were examined. Administration of 25μl of virus as a drop on both nostrils under brief superficial isoflurane anesthesia (92μl/dm(3), recovery after 85 ± 1 0s) was the most effective to infect the brain and induced 100% lethal infection 9 days later. Increasing the inoculum volume reduced infectivity significantly, with decreased viral loads in the brain and only 40% mortality. Increasing the depth of isoflurane anesthesia (230μl/dm(3)) improved the infectivity of the large-volume inoculum (90% mortality), probably because of suppression of swallow and sneeze reflexes. Compared to isoflurane anesthesia, xylazine-ketamine anesthesia reduced the infectivity of the inoculum significantly. Thus, administration of a small volume of virus on the nostrils under brief gas anesthesia is a safe and reproducible technique to induce infection of the brain. Since needles are not required, this helps to preserve the integrity of the physical barriers, animal welfare and the manipulator's safety.