Near St Guillaume an anti-aircraft battery opened fire. The Hun pilot then thought it better to leave Archie to deal with us, and he annoyed us no more. Some of the shell-bursts were quite near, but we could not afford to lose height in distance-dodging, with our machine in a dubious condition twenty-five miles on the wrong side of the trenches. Toutprès, to the south-west, was to have been included in the list of towns covered, but under the adverse circumstances V. decided not to battle against the wind more than was necessary to get us home. He therefore veered to the right, and steered due west. The south-west wind cut across and drifted us, so that our actual course was north-west. Our ground speed was now a good deal greater than if we had travelled directly west, and there was no extra distance to be covered, because of a large eastward bend in the lines as they wound north. We skirted the ragged Forêt de Quand-Même, and passed St Guillaume on our left. The behaviour of the engine went from bad to worse, and the vibration became more and more intense. Once again I thought it would peter out before we were within gliding distance of British territory, and I therefore made ready to burn the machine--the last duty of an airman let in for the catastrophe of a landing among enemies. But the engine kept alive, obstinately and unevenly. V. held down the nose of the machine still farther, so as to gain the lines in the quickest possible time. Soon we were treated to a display by the family ghost of the clan Archibald, otherwise an immense pillar of grey-white smoky substance that appeared very suddenly to windward of us. It stretched up vertically from the ground to a height about level with ours, which was then only five and a half thousand feet. We watched it curiously as it stood in an unbending rigidity similar to that of a giant waxwork, cold, unnatural, stupidly implacable, half unbelievable, and wholly ridiculous. At the top it sprayed round, like a stick of asparagus. For two or three months similar apparitions had been exhibited to us at rare intervals, nearly always in the same neighbourhood. At first sight the pillars of smoke seemed not to disperse, but after an interval they apparently faded away as mysteriously as they had appeared. What was meant to be their particular branch of frightfulness I cannot say. One rumour was that they were an experiment in aerial gassing, and another that they were of some phosphorous compound. All I know is that they entertained us from time to time, with no apparent damage.