The instrument which has been found to be the most effective in producing very powerful sound waves is called a _siren_. It consists of a tube or horn, having at the bottom a fixed disc with slits in it. Outside this disc is another movable one which revolves against the first, and which also has slits in it. When the second disc revolves, the passage way into the horn is opened and closed intermittently and suddenly, as the slits in the discs coincide or not. Air or steam under a pressure of 10 to 40 lbs. on the square inch is blown into the horn, and the rapid interruption of this blast by the revolving slits causes it to be cut up into puffs which, when sufficiently frequent, give rise to a very loud sound. The air under pressure is admitted to a back chamber and awaits an opportunity to escape, and this is given to it when the revolving disc moves into such a position that the slits in the fixed and moving disc come opposite each other. In comparative trials of different sound-producing instruments, nothing has yet been found to surpass this siren as a producer of penetrating sounds. It has been found very important that the frequency of the note given by the siren should coincide with the fundamental tone of the trumpet or horn. As will be explained in the next lecture, every column of air in a tube has a particular natural time-period of oscillation. Suppose, for instance, that for a certain length of trumpet-tube this is ¹⁄₁₀₀ second. Then the siren with that trumpet will be most effective if the interruptions of the air-blast are 100 per second. Lord Rayleigh has also shown that the shape of the mouth of the trumpet is important, and that this should not be circular as usual, but elliptical or oval, the shortest diameter of the ellipse being one quarter of the longest one. Also that the mouth should occupy such a position that the longer axis is vertical. Moreover, he considers that the short axis of the oval should not exceed half the wave-length of the sound being emitted. With a trumpet-mouth of such a shape, the sound is prevented to some extent from being projected up and down, but diffused better laterally—a result which is desired in coast sound-signals. The information accumulated as regards the distances at which sounds can be heard is very briefly as follows:— First as regards wind. The direction of the wind has a most remarkable influence on the distance at which a given loud sound can be heard. In one instance, the noise of a siren was heard 20 miles in calm weather; whereas, with an opposing wind, it was not heard more than 1¹⁄₄ mile away. It has been found that for calm weather a low-pitched note is better in carrying power than a high note, but in rough weather the opposite is the case.