At present, the van Huijzen zoning method [1] is mostly used at home and abroad, and the area between the infraorbital fissure and the rock-occipital fissure below the skull base is classified as the lateral cranial base, above which mainly corresponds to the middle cranial base. There are numerous pores between the two, which can communicate with each other through natural foramina such as rupture foramen, circular foramen, oval foramen, and sphenoid foramen, and the bone at the base of the middle cranial fossa is thin, and the bone on the lateral side of the circular foramen and oval foramen is especially obvious, adjacent to the mandibular fossa, where the thickness is only about 1 mm [2]. Therefore, tumors in the middle skull base and lateral skull base area are easy to grow along the natural foramina between the two and enlarge and destroy these foramina, or directly destroy the bone of the middle skull fossa base, causing intracranial and extracranial communication. The tumor in this area is deeply located and adjacent to important nerves and blood vessels, which is difficult and risky to operate, and involves multiple disciplines such as neurosurgery, maxillofacial surgery, otorhinolaryngology-head and neck surgery, plastic surgery, etc. It has been a difficult point in skull base surgery. We used microanatomy combined with three-dimensional imaging studies to propose a new bony surgical landmark and to explore its significance in surgery in this region. MATERIALS AND METHODS 1. Study materials: 15 cases (30 sides) of national adult cranial wet specimens adequately fixed by 10% formalin. 2. Imaging equipment: PHILIPS Brilliance 64-row CT, Destroscope virtual reality system (Volume Interaction, Singapore, software: RadioDexter TM 1.0). 3.Image scanning and 3D reconstruction method: CT scan of the head (FOV 26, Rotation Time 0.75s, Matrix 512×512, window position: 40, window width: 300, layer thickness: 0.67mm). fusion, observation, and measurement. 4. Microanatomical method: The frontotemporozygomatic-anterior infratemporal-inferior fossa approach was simulated, and a combination of naked eye and surgical microscope (4-25x) was used to dissect layer by layer and observe and measure the relevant structures and data. 5. Statistical analysis: SPSS13.0 statistical analysis software was applied to describe the measurement data with mean ± standard deviation; paired t-test was used to compare the microdissection and 3D image measurements; to verify the approximate equality of relevant data, ANOVA with repeated measurements was used; P