Some most important steps in improving our protecting measures were taken as a result of the lessons learned from the attack; in fact, it may be taken that all measures in defence against gas have been learned from bitter experience, and to this extent the sufferings of the victims may be taken as having at any rate some compensating value. In such a new and strange and continually developing kind of warfare very little can be done by _a priori_ argument. This fact we have always tried to impress on the men--that the gas warfare orders, sometimes apparently trivial and frequently wearisome and annoying, have all been made as the result of lessons learned from actual attacks. Among the chief things that were done after the December nineteenth attack was the improvement of our system of alarms. The bells and horns in the front line had been found quite insufficient, especially for warning people in the rear; and the telephone could not be depended on for this purpose owing to the possibilities of the wires being cut by shell fire. To protect them from being cut, all wires would have to be buried at least six feet deep in the ground, and this is practically impossible owing to the work involved. It would consequently be fatal to depend on telephonic communication, especially as a gas attack is nearly always accompanied by a pretty heavy bombardment of rear lines. In one case I knew, during just such a bombardment, the staff captain at a brigade headquarters was talking to one of the battalions when the whole telephone instrument seemed to burst into a sheet of flame in his hands, owing to a cut wire. The battalion concerned was isolated for more than an hour as a result, and anything might have happened in the meantime. For these reasons it was decided to adopt for gas alarms sirens worked by compressed air, which would make a noise sufficiently loud and distinctive to be heard long distances away. The type of siren which was used has been kept in use ever since in continually increasing numbers and has proved extraordinarily useful. It is known as the Strombos horn, and consists of the horn proper and two iron cylinders of compressed air charged to a pressure of one hundred and fifty atmospheres. Only one cylinder at a time is connected to the horn, the other being kept as a reserve.