The increased sophistication of imaging techniques in veterinary medicine allows the detection of a wide variety of intracranial and intraspinal lesions; however, imaging often does not provide a definitive diagnosis for nervous system (NS) lesions. Cytology is emerging as a useful diagnostic tool for obtaining a fast and accurate assessment of NS lesions, but little information is available for dogs and cats. The purpose of this study was to assess the accuracy of cytologic evaluation of squash samples from NS lesions in dogs and cats and to consider cytology-based diagnostic guidelines and sources of misdiagnosis. Cytologic specimens from masses localized in the central and peripheral NS taken during surgery or postmortem examination were classified into 3 groups according to the final histopathologic diagnosis: Group 1 = completely correct diagnosis, when the cytologic diagnosis and final histologic diagnosis were exactly correlated; Group 2 = partial correlation, when the cytologic diagnosis only partially correlated with the final histologic diagnosis, and Group 3 = no correlation, when the cytologic diagnosis was incorrect and there was no correlation with the general histologic type of lesion. The diagnostic accuracy of cytopathology was calculated by considering the histopathologic diagnosis as the "gold standard," and calculating a 95% confidence interval (CI). A total of 42 animals (33 dogs and 9 cats) were included in the study. The cytologic diagnoses were classified in Group 1 for 32 cases (76%; 95% CI 0.63-0.89), in Group 2 for 6 cases (14%; 95% CI 0.04-0.25), and in Group 3 for 4 cases (10%; 95% CI 0.006-0.18). Considering both complete and partial correlation as an adequate result, cytologic diagnosis was satisfactory in 90% of biopsies. Although the current series of cases is relatively small, cytologic evaluation of squash preparations can be considered a fairly accurate and reliable tool in the diagnosis of NS lesions.