The cochlear aqueduct (CA), otherwise known as the cochlear canaliculus, connects between the scala tympani of the cochlea (a part of perilymphatic space) and the subarachnoid space near the glossopharyngeal nerve in the posterior cranial fossa. Thus, the arachnoid connects to an otolithic membrane via the CA. The CA usually accompanies a cochlear vein draining to the inferior petrosal sinus. A bony canal that contains the CA is called the periotic canal or duct, although it is difficult to distinguish from the CA itself when the term “duct” is used (Bast, 1946, Spector et al., 1980, Ishikawa et al., 2018). In adults, the bony canal can be demonstrated by gross dissection (Rask-Anderson et al., 1977). According to a micro-CT analysis (Atturo et al., 2018), multiple periotic canals were present, but most appear to be canals for veins. The development of CA is characterized by the “anlage,” which maintains a simple morphology (a linear mesenchymal condensation without lumen) for a long period during intrauterine life (stages of 27-84 mm CRL; approximately 8-12 weeks). Waltner (1925) and Bast (1946) described the anlage was as a syncytial tissue and a reticulated tissue, respectively.

In embryos, a postero-infero-medial part of the otic capsule opens to the posterior cranial fossa and, from this site, a loose mesenchymal structure, the Hyrtl fissure, extends to the round window (reviewed by Spector et al., 1980). Although the fissure is filled with tissues (Rodríguez-Vázquez et al., 2022), some researchers believe that the Hyrtl fissure is a “space or duct” to allow fluid communication (reviewed by Mudry (2009) and Alalade et al. (2022); details, see the Discussion). Mukherji et al. (1998) stated as such, and Jacker and Hwang (1993) described the initial CA appearing as an extension of the subarachnoid space. However, Jacker and Hwang’s (1993) observations were based on CT images of adult temporal bones, not histological sections of embryos. Spector et al. (1980) also introduced the arachnoid origin of the CA, but their materials were limited to late-stage human fetuses. Strangely, these research groups did not use a term “anlage of CA” for the initial fissure (see a paragraph above).

The CA anlage is markedly different from loose mesenchymal tissue of the cochlear scala tympani before cavitation. However, previous researchers considered that the scala’s cavitation extends to and provides an initial lumen of the CA appearing in the mesenchymal condensation (Waltner, 1925, Bast, 1946). However, before cavitation, in contrast to the tympanic cavity, the scala does not contain vessels (Hayashi et al., 2022). Our group does not consider the scala’s cavitation as an active event that includes the differentiation and development of epithelial-like structures, such as otolithic membranes (Kim et al., 2011). Consequently, the major aim of this study was to re-examine a developmental process of the CA cavitation without epithelium, in particularly, whether the CA lumen develops from the cochlear side or meningeal side. In addition, we also examined 1) whether a vein develops independent of the CA and 2) prenatal growth or degeneration of the CA.