We cannot obtain the other product, the caustic soda, by the same simple means. In our little experiment we succeeded in manufacturing some of it in the region around the cathode, and had we drawn off some of the liquid from there we would have been able to detect its presence. But it would have been mixed up with much ordinary salt, and for commercial purposes we need the caustic soda separate from the salt. The principle is, however, just the same, as you will see. Imagine a large oblong vat divided by vertical partitions into three separate chambers. These partitions do not quite reach the bottom of the vessel, so that there is a means of communication between all three chambers. This is closed, however, by filling the lower part of the vat with mercury up to a level a little higher than the lower ends of the partitions. Thus we have three separate chambers with communication between them but that communication is sealed up by the mercury. The two end chambers are filled with salt water, or brine, while the centre one is filled with a solution of caustic soda. In each end compartment is a stick of graphite, both being electrically joined together and so connected up that they form anodes, while in the centre compartment is the cathode. When the current flows from the anodes it carries the sodium ions with it, just as it did in our little experiment. But its course, this time, is not straight, since in order to travel from anode to cathode it has to pass through the openings in the partitions, in other words through the mercury. On arrival at or near the cathode the ions of sodium cause the caustic soda to be formed just as in our experiment, but in this case, you will notice, the formation takes place in a chamber from which the salt brine is completely excluded by the mercury. Brine is continually fed into the outer chambers and the solution of caustic soda is drawn from the centre one, while the chlorine is collected over the anodes. And now we can go a step further on our progress from common salt to explosive. In the soap works there are enormous coppers in which are boiled various kinds of fat. The source of the fat may be either animal or vegetable, many kinds of beans, nuts and seeds furnishing fats practically identical with that which can be got from the fat flesh of a sheep, for instance. To this fat is added some caustic soda solution and the whole is kept boiling for some considerable time. This protracted boiling is to enable the soda thoroughly to attack the fat and combine with it, whereby two entirely new substances are formed.