A physical model was built to study the properties of the feeding strike of the mantis shrimp that are responsible for drag reduction and cavitation control. The model had three goals: 1) The model was to be outfitted with a method to collect kinematic, force and cavitation data. 2) The velocity and acceleration profile of the model were to be predicted with a mathematical model of the mechanism. 3) The model was to match as many drag and cavitation sensitive properties of the mantis shrimp strike as feasible and have a means to control the rest. The first iteration of the model met the first goal but not the second or third. It matched the strike in maximum velocity, appendage size and shape and environmental temperature and salinity but did not control acceleration profile, water quality or pressure. Data collected with high-speed video of strikes of the model and Gonodactylus smithii showed the model to cavitate at speeds at which no cavitation was seen in animal strikes. The model was redesigned to be driven by the stored elastic energy in the deflection of a beam spring. The redesigned model reached the animals maximum accelerations but not velocities. Environmental variation was found to not substantially contribute to the variation in cavitation onset velocity between the model and animal experiments.