The evolutionary origin of the tritocerebral neuromere, which is a brain segment located at the junction between the supra- and subesophageal ganglia in most mandibulates (arthropods such as crustaceans and insects), is a subject rich in contentious debate. Various models have argued that the tritocerebrum came from a segmental nerve cord ganglia that was recruited into the head during the course of arthropod evolution. However, despite much thought on the subject, the origin of the tritocerebrum remains obscure. Here I describe the development of the tritocerebral commissure in Drosophila and demonstrate that the tritocerebral and mandibular commissures actually form as one commissure and then separate in a manner very similar to how the anterior and posterior commissures of a ventral nerve cord neuromere form. I propose that the tritocerebral neuromere originated from the splitting of an ancestral neuromere located in the anterior subesophageal ganglion into distinct tritocerebral and mandibular neuromeres. Also, I discuss the problem of arthropod brain neuromere homology in reference to this hypothesis.