Background:Smears prepared by cytocentrifugation, the so-called cytospins, are widely used in human and veterinary cytology. However, the high cost has hampered the availability of commercial cytospin centrifuges in some veterinary clinics and laboratories. Nevertheless, cytospins are important for evaluating fluids with very low cellularity such as cerebrospinal fluid (CSF) or bronchoalveolar lavage fluid (BALF). Objectives: The aim of this study was to devise and test the use of a low-cost, in-house manual cytocentrifuge to obtain cytospin preparations. Methods: Twenty-two fluid samples (including CSF and BALF) were collected from dogs and cats. These were processed in a conventional cytocentrifuge and in an in-house, manual centrifuge (salad spinner). The cytospins obtained by the 2 methods were compared by scoring cellularity, number of cells per field, hemodilution, cell preservation, and proportion of ruptured cells. Additionally, cell number and size were compared by morphometry. Differences between the automated and manual method were statistically assessed. Results: The morphology and cellular detail of cytospin preparations produced by both methods were identical. There was an almost perfect agreement for cellularity, number of cells per HPF, hemodilution and cell preservation (kappa ≥ 0.85), and a moderate agreement for the amount of ruptured cells. Cell recovery was comparable (including in CSF and BALF). Conclusions: The manual cytocentrifuge produced cytospins with similar cell yield as the automated cytocentrifuge. Considering the low cost and portability, this new method should be particularly useful for cytologic diagnosis in small clinics, developing countries, and in field studies.