One can't help but feel that there is something odd about the fuel-ice scenario. I don't mean that it can't happen if long-sector temps are so low as to cold-soak the fuel to an incipient stage. But how come so many virtually identical flights re. time and temperature have been flown over that and similar routes without reported problems? Or perhaps there have been other knife-edge situations which were disguised or mis-diagnosed by aircrew or engineers? The AAIB must certainly have wondered about that conundrum. A chum of mine routinely flies an A-340 from South America to Auckland, New Zealand, and I know that they very carefully monitor outside air temps and, from memory, limit their great-circle track vertix to 72 deg. south lat.; even though it does increase air distance. Half jokingly, some months ago I suggested in Forums and to some chums who fly ultra-long haul, to make sure they test fuel-flow integrity by reverting to manual power control for a period and making some power changes just to check that all was well. Just maybe, it is not such a silly idea after all???