The net integrated energy flux into a train of internal waves is evaluated in a two-layer model. The nonzero value for this integral results from the difference in the stratification between the initial and final state, similar to the energy supply to a surface bore. We apply this expression to waves measured by Wesson and Gregg <1988> in the Strait of Gibraltar and to waves measured by Farmer and Smith <1980> in Knight Inlet. We find the energy supply to be important to the energetics, but the data do not allow a definitive test of the conjecture that the primary energy balance is between this supply and dissipation. We contrast our conjecture to the solitary-wave considerations of Bogucki and Garrett <1993>.¿ 1997 American Geophysical Union