OBJECTIVE Few have described the relationship between arachnoid protrusions (villi) and adjacent spinal radicular veins, and the descriptions that do exist are conflicting. Some authors have even denied the presence of spinal arachnoid villi, suggesting that they play no role in cerebrospinal fluid (CSF) absorption. METHODS To further elucidate these structures, laminectomies from C-2 inferiorly to S-2 were performed in 10 fresh human adult cadavers. Following removal of the laminae, the dural nerve sleeves were identified and the spinal nerves excised 1 cm lateral and medial to the intervertebral foramina. Samples were submitted for histological and immunohistological analysis. RESULTS The authors identified arachnoid villi in all specimens. The length of these structures was approximately 50 to 170 microm. Regionally, these villi were more concentrated in the lumbar region, but they were not present at every vertebral level, with observed skip zones. Occasionally, more than one villus was identified per vertebral level. The majority of villi were intimately related to an adjacent radicular vein. There was a direct relationship between the size of the adjacent radicular vein, and the presence and number of arachnoid villi. CONCLUSIONS Findings in the present study have demonstrated that arachnoid villi exist and are morphologically associated with radicular veins. These data support the theory that CSF absorption occurs not only intracranially but also along the spinal axis. Further animal studies are necessary to prove that CSF traverses these villi and is absorbed into the spinal venous system.