This study examined the effects of prey exoskeleton characteristics on gastric evacuation patterns in Atlantic cod Gadus morhua. Three distinct stages were highlighted in the gastric evacuation of crustacean prey characterized by a robust exoskeleton. The experiments confirmed that the three shrimp species, Pandalus borealis, Pandalus montagui and Eualus macilentus, and the crab Chionoecetes opilio, were evacuated from the stomach at different rates. The duration of all stages increased with increasing ash (and carbonate) content of the fresh prey. Thickness, chemical composition and morphology of the prey exoskeleton all affected gastric evacuation: duration of initial delay, overall evacuation rate and a decreased evacuation rate at the end of the process. The power exponential function (PEF), with its shape parameter, described the course of evacuation for these prey types well, especially the initial delay. The PEF does not, however, allow describing evacuation by the current stomach content mass independent of meal size, which limits its usefulness in estimating consumption rates of wild G. morhua. To predict and describe gastric evacuation of prey with a robust exoskeleton, it is therefore suggested that the square-root function be expanded with an initial lag phase, coupled to the mechanistically based cylinder model of gastric evacuation.