Then comes the matter of time. How soon after the explosion occurred did the pressure begin to be felt? How long did it take to reach its maximum and how long to die out again? These questions need answers which the apparatus so far described does not give. True, the speed of the paper may be known approximately, but all that I have described will occur within the space of a fraction of a second, and it is difficult to tell the speed of the paper with sufficient accuracy. Therein we see the purpose of the third style. It is attached electrically to the "tenth-of-a-second time-marker." This consists of a weight suspended at a height. The force of the explosion lets it drop. The moment it starts to fall it causes the style to make a mark on the paper. When it has fallen a certain distance the style makes another mark. And the distance that the weight falls between the making of the two marks is so adjusted that the space between them on the chart represents exactly a tenth of a second. Thus a scale is formed upon the chart by which the other times can be measured. There is the line terminating at the moment of explosion; the straight line changing into an up-and-down curve, representing the time and the variation of the pressure; finally there are the two marks representing a tenth of a second by which the other marks recorded upon the chart can be interpreted. But the mere pressure and velocity of the explosion form but a part of the knowledge desired. How the explosion is formed, whether or not the coal-dust is burnt up entirely, whether, indeed, it be the dust itself which burns or coal-gas given off by the dust under the heat of the preliminary explosion, what the gas is which is left by the explosion at various stages--these are important things to be known, and they can only be ascertained by taking samples of the gases in the gallery at different moments during and after the explosion. To obtain these samples bottles are used, but the question is how to get them filled at just the right time. Into the shell of the gallery holes are drilled, and to these the metal bottles or flasks are screwed, a pipe leading from the mouth of each bottle well in towards the centre of the gallery. The end of this tube is closed by a cap of glass above which there stands poised a little hammer. Controlling the hammer is an electrical device called a "contact-maker," so arranged that just at the desired moment the hammer falls, breaking the glass, and admitting a sample of the gas in the gallery, the bottle and its tube having previously had the air exhausted from them, so that on the glass being broken the gas is sucked in. At the same moment a weight falls, attached to the end of a cord, and this, on reaching the end of its tether, closes the end of the tube, and the sample is imprisoned until such time as the bottle can be disconnected and taken away to the laboratory for its contents to be analysed.