Withdrawals from the Potomac-Raritan-Magothy aquifer system in New Jersey, which includes the Upper, Middle, and Lower Potomac-Raritan-Magothy aquifers, are the principal source of groundwater supply in northern Gloucester and northeastern Salem Counties in the New Jersey Coastal Plain. Water levels in these aquifers have declined in response to pumping. With increased population growth and development expected in Gloucester County and parts of Salem County over the next 2 decades (2005-2025), withdrawals from these aquifers also are expected to increase. A steady-state groundwater-flow model, developed to simulate flow in the Potomac-Raritan-Magothy aquifer system in northern Gloucester and northeastern Salem Counties, was calibrated to withdrawal conditions in 1998, when groundwater withdrawals from the Potomac-Raritan-Magothy aquifer system in the model area were more than 10,100 Mgal/yr (million gallons per year). Withdrawals from water-purveyor wells accounted for about 63 percent of these withdrawals, and withdrawals from industrial self-supply wells accounted for about 32 percent. Withdrawals from agricultural-irrigation, commercial self-supply, and domestic self-supply wells accounted for the remaining 5 percent. Results of the 2000 baseline groundwater-flow simulation, incorporating average annual 1999-2001 groundwater withdrawals, indicate that the average simulated water levels in the Upper, Middle, and Lower Potomac-Raritan-Magothy aquifers are 31, 27, and 30 feet below the National Geodetic Vertical Datum of 1929 (NGVD 29), respectively, and the lowest simulated water levels are 77, 65, and 59 feet below NGVD 29, respectively. In the full-allocation scenario, the maximum State-permitted (allocated) groundwater withdrawals totaled 16,567 Mgal/yr, an increase of 72 percent from the 2000 baseline simulation. Results of the full-allocation simulation indicate that the average simulated water levels in the Upper, Middle, and Lower Potomac-Raritan-Magothy aquifers are 49, 43, and 48 feet below NGVD 29, respectively, which are 18, 16, and 18 feet lower, respectively, than in the 2000 baseline simulation. The lowest simulated water levels are 156, 95, and 69 feet below NGVD 29, respectively, which are 79, 30, and 10 feet lower, respectively, than in the 2000 baseline simulation. Simulated net flow from the Potomac-Raritan-Magothy aquifer system to streams is 8,441 Mgal/yr in the 2000 baseline simulation but is 6,018 Mgal/yr in the full-allocation scenario, a decrease of 29 percent from the 2000 baseline simulation. Simulated net flow in the 2000 baseline simulation is 1,183 Mgal/yr from the aquifer system to the Delaware River but in the full-allocation scenario is 1,816 Mgal/yr from the river to the aquifer system. Four other simulations were conducted that incorporated full-allocation conditions at water-purveyor wells in Critical Area 2 but increased or decreased withdrawals at selected water-purveyor wells outside Critical Area 2 and agricultural-irrigation and industrial-self-supply wells in the study area. The results of the four simulations also indicate net flow from the Delaware River to the Potomac-Raritan-Magothy aquifer system. A growth scenario was developed to simulate future withdrawals in 2025 estimated from population projections for municipalities in the Salem-Gloucester study area. Simulated withdrawals for this scenario totaled 10,261 Mgal/yr, an increase of 6 percent from the 2000 baseline simulation. This total includes about 25 Mgal/yr withdrawn from the Englishtown aquifer system for domestic self-supply. This scenario incorporated full-allocation withdrawals at water-purveyor wells in Critical Area 2, and increased withdrawals at water-purveyor wells outside Critical Area 2. Results of this simulation indicate that the average simulated water levels in the Upper, Middle, and Lower Potomac-Raritan-Magothy aquifers are 32, 29, and 32 feet below NGVD 29, respectively, which are 1, 2, and