Mammalian embryos experience not only hormonal stimuli, but also mechanical stimuli (MS), such as shear stress (SS), compression and friction force, in the fallopian tube before nidation. Embryo development performed using previously described in vitro dynamic culture systems is significantly better than that performed using conventional static culture systems. Previously, we found that thawed human embryos showed developmental improvement in the blastocyst stage following a tilting embryo culture system (TECS) culture compared with static culture. However, a disadvantage of the system is the need to use electric devices inside the incubator under humidified conditions. To solve the problem, we developed a dynamic embryo culture system using air actuation and evaluated the applied MS and embryo culture results. We developed an air actuation system with microfluidic channels to apply MS by deforming a 0.1-mm-thick poly(dimethylsiloxane) (PDMS) membrane. The PDMS microfluidic device was placed in a humidified incubator and the mechanical actuator was placed outside the incubator. The embryos' motion in the microfluidic channel was recorded using an inverted microscope and a colour CCD camera with a frame rate of 30 frames s–1. Syringe velocity (VS) was controlled using a software model of the actuation system. The observed maximum velocity of the embryos (VE) and fluid velocity (VF) were calculated by tracking the images of the embryos and the particles in the medium, respectively. The experiments were repeated 3 times. Frozen 2-cell-stage embryos of imprinting control region (ICR) mouse were thawed. 10 to 13 embryos were applied into the microfluidic channel and cultured in ∼200 μL of potassium simplex optimized embryo culture medium covered with mineral oil for 3 days in a humidified environment of 5% CO2 in air at 37°C. The experiments were repeated 5 times. Chi-squared test and Student's t-test were used to determine differences in the blastocyst development rate and in the number of cells in the blastocysts between the groups, respectively. A P-value <0.05 was considered significant. Results: When syringe velocity (VS) was 0.5 mm over a period of seconds, the embryos rotated and did not slide. When VS and fluid velocity (VF) increased, the embryos slipped, did not come in contact with the floor. We conclude that different types and amounts of MS can be applied to the embryos by changing VS. We compared embryo development from the 2-cell stage to the blastocyst stage between static and dynamic cultures in the medium channel. Dynamic culture significantly improved the rate of development to the blastocyst stage (dynamic, 74% (n = 126); static, 62% (n = 118); P < 0.05). The average number of cells (mean ± standard error of the mean) in blastocysts obtained in dynamic and static cultures was 83 ± 3 (n = 54) and 76 ± 3 (n = 51) (P < 0.05), respectively. When the mouse embryos moved at VE of 0.2 mm s–1, there were significant differences in both blastocyst development rate and the average cell number of blastocysts between the 2 groups.