Motility is the most widely used indicator of sperm quality. Computer-Assisted Semen Analysis (CASA) allows the objective evaluation of sperm motility parameters. CASA technology is a common tool to predict semen doses in farm animal reproduction. The kinds of video cameras used until now for image acquisition have presented limited frame rates (FR), which have a negative influence on the quality of the obtained data. The aim of the present work was to define the optimal frame rate for a correct evaluation of boar sperm motility and its subpopulation structure. Eighteen ejaculates from nine mature boars of the Pietrain breed were used. Using the ISAS v1 CASA-Mot system, with a video camera working up to 200 Hz, six FRs (25, 50, 75, 100, 150 and 200 fps) were compared. ISAS D4C20 counting chambers, warmed to 37°C, were used. FR affected all the kinematic parameters, with curvilinear velocity (VCL) and BCF the most sensitive ones. All the parameters showed differences among animals. Non-linear correlation showed the asymptotic level for VCL at 212 fps, being the highest FR for all the parameters. For future studies based just on progressive motility, almost 100 fps FR for 0.5 s must be used, while when kinematics must be considered, almost 212 fps for one-second should be analysed. Three principal components were obtained (velocity, progressivity and oscillation), being similar at 50 and 200 fps. Cells were grouped in four subpopulations but with different kinematic and cellular distribution at both FRs.

Contents Dogs have undergone an intensive artificial selection process ever since the beginning of their relationship with humans. As a consequence, a wide variety of well‐defined breeds exist today. Due to the enormous variation in dog phenotypes and the unlikely chance of gene exchange between them, the question arises as to whether they should still be regarded as a single species or, perhaps, they be considered as different taxa that possess different reproductive traits. The aim of this study was therefore to characterize some male reproductive traits, focusing on kinematic characteristics of dog spermatozoa from several breeds. Thirty‐seven dogs from the following breeds were used: Staffordshire Bull Terrier, Labrador Retriever, Spanish Mastiff, Valencian Rat Hunting Dog, British Bulldog and Chihuahua. Semen samples were obtained via manual stimulation and diluted to a final sperm concentration of 50 million/ml, and they were subsequently analysed by the computer assisted semen analysis (CASA‐Mot) ISAS®v1 system. Eight kinematic parameters were evaluated automatically. All parameters showed significant different values among breeds and among individuals within each breed. The fastest sperm cells were those of Staffordshire Bull Terriers and the slowest were recorded in Chihuahuas. The intra‐male coefficient of variation (CV) was higher than the inter‐male CV for all breeds with the Staffordshire Bull Terrier showing the lowest values. When taking into consideration the cells by animal and breed, discriminant analyses showed a high capability to predict the breed. Cluster analyses showed a hierarchical classification very close to that obtained after phylogenetic studies with genome markers. In conclusion, future workers on dog spermatozoa should bear in mind major differences between breeds and realize that results cannot be extrapolated from one to another. Because sperm characteristics are associated with breed diversity, dogs may represent a good model to examine changes in reproductive parameters associated with selection processes.