and a supply of air from the surface is out of the question, recourse is still had as before to some form of storage power for propulsion. At present this is always electric.
The problem of diving demands attention next. For surface sailing a submarine boat, like any other, needs considerable buoyancy, so as to float with a considerable fraction of its bulk free above water. For diving, on the other hand, her buoyancy must be very small. These conditions are met by varying the amount of ballast carried. This is universally done by admitting water into, or expelling it from, suitable air-tight tanks distributed through the bottom of the boat. The filling of these tanks requiers only the opening of a valve. To empty them
requires power. Formerly this was done by means of pumps. But pumping is slow work. A much more expeditious method of emptying the water tanks is to blow out the water by admitting compressed air from the reservoirs. The air so used is finally delivered into the living rooms for breathing, and the pressure in the reservoirs is restored again win n at the surface. By thus varying the quantity of ballast a boat may be caused to sink, or, if already beneath the water, be caused to rise to the surface either slowly or rapidly as may be desired. It is easy to imagine circumstances, either accidental or otherwise, where a very sudden return to the surface might be imperative. To provide for this in emergencies the most practical boats are furnished with
