Introduction: The aim of this study was to propose the optimal methodology for stallion semen morphology analysis while taking into consideration the staining method, the microscopic techniques, and the workload generated by a number of samples. Material and Methods: Ejaculates from eight pure-bred Arabian horses were tested microscopically for the incidence of morphological defects in the spermatozoa. Two different staining methods (eosin-nigrosin and eosin-gentian dye), two different techniques of microscopic analysis (1000× and 400× magnifications), and two sample sizes (200 and 500 spermatozoa) were used. Results: Well-formed spermatozoa and those with major and minor defects according to Blom’s classification were identified. The applied staining methods gave similar results and could be used in stallion sperm morphology analysis. However, the eosin-nigrosin method was more recommendable, because it allowed to limit the number of visible artefacts without hindering the identification of protoplasm drops and enables the differentiation of living and dead spermatozoa. Conclusion: The applied microscopic techniques proved to be equally efficacious. Therefore, it is practically possible to opt for the simpler and faster 400x technique of analysing sperm morphology to examine stallion semen. We also found that the number of spermatozoa clearly affects the results of sperm morphology evaluation. Reducing the number of spermatozoa from 500 to 200 causes a decrease in the percentage of spermatozoa identified as normal and an increase in the percentage of spermatozoa determined as morphologically defective.