AbstractWater uptake in the crab, Rhithropanopeus harrisi, was studied using D2O (deuterium oxide) as a label. Results were expressed as per cent saturation, or the D2O concentration in the blood, attained in a 30‐minute exposure to D2O at 20°C, expressed as a percentage of the D2O concentration in the bathing medium. Crabs acclimated to 75% sea water (SW), to which the blood is about isosmotic, had a mean per cent saturation of 51.8, and in 10% SW‐acclimated crabs (blood hyperosmotic by about 40% SW) the mean per cent saturation was 33.9. Thus, water uptake, or exchange, is slower when the osmotic gradient favors net water uptake than when no osmotic gradient exists. In gill chamber perfusion experiments at least 87% of the total water influx in 10% SW and 92% of the total in 75% SW are taken up in the gill chambers, the gills probably being responsible for most, if not all, of the uptake. The time course of acclimation between 10% SW and 75% SW was followed in both directions. Measurable changes in water uptake occurred in 30 minutes; half the change from 75% SW to 10% SW occurred in two hours (seven hours for the transfer from 10% SW to 75% SW), and over 90% of full acclimation was reached in two days.Gill chamber perfusion experiments showed that the receptor or receptors which perceive the change between 10% and 75% SW and mediate the change in water uptake are not in the gill chamber. Thus the system responsible for the change in water uptake is composed of at least a receptor with some means of communication (presumably nervous or hormonal) with the effectors (probably the gills).The use of Na+‐free and Cl−‐free D2O media showed that the absence of these ions did not affect water uptake, and that the change in water uptake was not caused by a change in influx of these ions.