At the end of May 1997, tritium neutral beams were run for the first time in the JET Octant 8 Neutral Injector Box (NIB8). On 4 June 1997 there were indications of a water leak in the NIB8. The leak was associated with operation of the Fast Shutter [1] system positioned between the Injector and the Rotary Valve. The presence of water was detected at a very early stage by routine operating procedures using the NIB Mass Spectrometer and Penning gauge instrumentation. Operations on NIB8 were suspended whilst the pressure was still below 5.10-7mb with the Cryopumps cold [2]. The pressure was still two orders of magnitude below the interlock trip level from the Fast Beam Interlock system which would have automatically inhibited pulsing. Some 3cm3 of water was estimated to have leaked into the NIB