The current from a simple voltaic cell does not remain at a constant strength, but after a short time it begins to weaken rapidly. The cell is then said to be polarized, and this polarization is caused by bubbles of hydrogen gas which accumulate on the surface of the copper plate during the chemical action. These bubbles of gas weaken the current partly by resisting its flow, for they are bad conductors, and still more by trying to set up another current in the opposite direction. For this reason the simple voltaic cell is unsuitable for long spells of work, and many cells have been devised to avoid the polarization trouble. One of the most successful of these is the Daniell cell. It consists of an outer vessel of copper, which serves as the copper plate, and an inner porous pot containing a zinc rod. Dilute sulphuric acid is put into the porous pot and a strong solution of copper sulphate into the outer jar. When the circuit is closed, the hydrogen liberated by the action of the zinc on the acid passes through the porous pot, and splits up the copper sulphate into copper and sulphuric acid. In this way pure copper, instead of hydrogen, is deposited on the copper plate, no polarization takes place, and the current is constant. Other cells have different combinations of metals, such as silver-zinc, or platinum-zinc, and carbon is also largely used in place of one metal, as in the familiar carbon-zinc Leclanché cell, used for ringing electric bells. This cell consists of an inner porous pot containing a carbon plate packed round with a mixture of crushed carbon and manganese dioxide, and an outer glass jar containing a zinc rod and a solution of sal-ammoniac. Polarization is checked by the oxygen in the manganese dioxide, which seizes the hydrogen on its way to the carbon plate, and combines with it. If the cell is used continuously however this action cannot keep pace with the rate at which the hydrogen is produced, and so the cell becomes polarized; but it soon recovers after a short rest. The so-called “dry” cells so much used at the present time are not really dry at all; if they were they would give no current. They are in fact Leclanché cells, in which the containing vessel is made of zinc to take the place of a zinc rod; and they are dry only in the sense that the liquid is taken up by an absorbent material, so as to form a moist paste. Dry cells are placed inside closely fitting cardboard tubes, and are sealed up at the top. Their chief advantage lies in their portability, for as there is no free liquid to spill they can be carried about and placed in any position.