Problem: The facial nerve synkinesis is one of the common complaints after facial palsy. However, the precise mechanism of synkinesis is still unknown. In this study we tried to establish an animal model of synkinesis. Methods: Extratemporal portion of the facial nerve (group 1, n = 6) and the intratemporal portion (group 2, n = 6) were compressed by using a micro needle holder. After the compression, the grade of facial nerve injury was evaluated by electroneurography. The existence of synkinesis was judged by the presence of R1 in the orbicularis oris muscle by the electrophysiolgical blink reflex. For evaluating the misdirection of nerve fibers, double labeling was performed by injecting retrograde fluorescence tracer in the orbicularis oculi muscle and the orbicularis oris muscle, and then facial nucleus was investigated for the presence of the neurons with aberrant fluorescence. Results: In both groups, all animals showed the ENoG threshold under 10%. In group 1, none of animals developed synkinesis. On the other hand, in group 2, all animal developed synkinesis. Furthermore, disrupted neurons in the facial nucleus, representing the existence of misdirection, were observed in all animals of group 2. Conclusion: In this study we succeeded in making a synkinesis model only in group 2 and reported the existence of misdirection in these animals. Sunderland found that the facial nerve consisted of a single bundle in the intratemporal portion, even though it had a mulitifuniculated plexiform structure in the extratemporal portion. Because of this anatomical feature, when the facial nerve was injured in the intra- and extra-temporal portions, the misdirection may occur only in the intra-temporal portion. This is thought as a reason for the appearance of synkinesis only in group 2. Significance: It is important to know the pathogenesis of synkinesis for diagnosing and treating facial nerve disorders. Support: None reported.