Capillary loaded chambers are frequently used for semen analysis. Poiseuille flow of specimen into these chambers causes migration of suspended particles or cells in a direction transverse to the flow, which results in their preferential accumulation in the Segre-Silberberg (SS) planes. This SS effect depends on the transverse velocity gradient in the laminar flow. For semen analysis in thin capillary-loaded slides, the SS effect can lead to erroneous estimation of sample sperm-cell concentration. To better understand chamber flow dynamics and SS effect significance, we assessed flow uniformity, inflow cell velocity, and results of concentration measurements under different flow conditions for latex bead and porcine and human sperm suspensions. Overall, a concentration peak was present at the meniscus, which continued through chamber loading. High-velocity SS preferred planes, which channeled particles toward the meniscus, were located at the fractional positions of beta = .27 and beta = .73, where beta is the distance from wall to plane normalized to the chamber depth. In computer-automated semen analysis, a standard 20-microm x 18-mm x 6-mm chamber is commonly used, and these studies supported our previously published fluid-flow theory for this type of chamber. Conversely, the SS effect does not appear to have time to develop in the 100-microm-depth hemacytometer, which is deeper than the standard slide, has lower transverse velocity gradient, and consequently does not exhibit concentration variation due to the SS effect. These findings provide further support that hemacytometry, when performed properly, remains the gold standard. Applicability of our findings to routine semen analyses was then tested in 2 studies performed with independent boar studs. These studies compared diluted boar semen concentrations estimated by standard hemacytometry and in capillary-loaded 20-microm slides, using a computer-automated semen-analysis system designed to compensate for the SS effect. Good numerical agreement for sperm concentration with a high degree of correlation (r(2) = .936) was found between the 2 techniques. These findings reaffirm the need to critically assess new technologies for accuracy, repeatability, and precision.