During the second half of 1997, JET carried out a broad-based series of experiments in deuterium-tritium (D-T) producing a total of 675 MJ of fusion energy and setting records for fusion power (16MW), ratio of fusion power to plasma input power (0.62, and 0.95 ± 0.17 if a similar plasma could be obtained in steady-state) and fusion power duration (4MW for 4s). A large scale tritium supply and processing plant, the first of its kind, allowed the repeated use of the 20g tritium on site, supplying a total of 99.3g of tritium to the machine. The D-T physics programme allowed the size, heating requirements and operating conditions of the International Thermonuclear Experimental Reactor (ITER) to be defined more precisely. The threshold power required to access the high confinement operating mode foreseen for ITER is significantly lower in D-T than in deuterium (giving increased operational flexibility for ITER) but the global energy confinement time itself is practically unchanged (no isotope effect). ITER demonstration pulses, in which the important dimensionless parameters for confinement were matched to those of ITER, predict ignition for ITER provided the required densities can be reached. Three radio frequency schemes for heating ITER were also tested successfully in D-T. The results agreed well with code calculations giving confidence in the use of these models for predicting ICRF heating in future machines. Finally, the D-T experiments provided the first clear evidence of alpha-particle heating, showing it to be consistent with classical expectations and confirming the process by which ignition and thermonuclear burn will occur in ITER and a fusion reactor.