The shore crab, Carcinus maenas L. (Portunidae), is a coastal and estuarine species, which can live and reproduce under brackish water conditions; freshly hatched larvae have been observed in the field at salinities below 15 parts per thousand. In the present laboratory study, the tolerance of hypo-osmotic stress was experimentally investigated in early larvae of a marine (North Sea) population of C. maenas reared at four different salinities (15, 20, 25, 32 parts per thousand). Two and 4 days after hatching, the Zoea I larvae were moult-staged microscopically, and their rates of respiration and growth (changes in dry weight, W, carbon, C, nitrogen, N, and hydrogen, H) were measured. Survival and development were monitored until the megalopa was reached: 15 parts per thousand did not allow for development beyond the first zoeal stage, while metamorphosis to the megalopa was reached at salinities greater than or equal to 20 parts per thousand. At 20 parts per thousand, development was significantly delayed and mortality enhanced as compared with 25 and 32 parts per thousand. Rates of growth and respiration decreased during exposure to reduced salinities less than or equal to 25 parts per thousand. Hence, the suppression of growth could not be explained as a consequence of enhanced metabolic losses per larva. Instead, a partial C budget indicates that the Zoea I larvae suffered from decreased capabilities of assimilating ingested and subsequently converting assimilated matter to tissue growth. Net growth efficiency (K-2, C-based) was at 25 and 32 parts per thousand initially high (> 60% during the postmoult and intermoult stages of the Zoea I moult cycle), but decreased during the later stages (down to less than or equal to 30% in premoult). An inverse pattern of C partitioning was observed at less than or equal to 20 parts per thousand, with initially low K-2 values (less than or equal to 21% during the first 2 days of the moult cycle), and a later increase (up to greater than or equal to 46% in premoult). Thus, larval growth was initially suppressed under conditions of reduced salinity, but this was later (during premoult) partially compensated for by an increase in C assimilation and K-2. Our observations indicate that Zoea I shore crab larvae react during the late stages of their moulting cycle less sensitively against reduced salinities than during postmoult and intermoult. This suggests that the transition between moult cycle stages C and D-0 may be a critical point for effects of hypo-osmotic stress, similarly as already known in relation to effects of nutritional stress. Negative effects were found also when freshly hatched Zoea I shore crab larvae were exposed only transitorily (for 24-72 h) to 20 parts per thousand, with significantly lower rates of survival, development, growth, respiration, and K-2. These effects increased with increasing duration of initial exposure to reduced salinity.