We have already mentioned the use of a syringe as a means by which an aquarium may be temporarily restored to a satisfactory condition providing it has not been neglected too long, and some recommend forcing air, or, still better, pure oxygen gas, from an india-rubber bag into the water. We have used, for the same purpose, a stream of oxygen from a steel cylinder of the compressed gas with very satisfactory results; and since oxygen may be now obtained, ready compressed, at a very low price--about twopence a cubic foot--there is much to be said in favour of this method as an auxiliary in the hands of the owner of a small tank, though we hardly recommend it as a prime means of aëration to take the place of the fountain. In any case, where a stream of air or oxygen is employed, an exceedingly fine jet should be used, in order that the expelled gas may take the form of a stream of minute bubbles; for, as previously stated, the water can absorb the gas only very slowly, so that there must necessarily be a considerable waste when the gas issues rapidly. Further, the smaller the bubbles passing through the water, the greater is the total surface of gas in contact with the liquid, the volume of the supply being the same, and hence the more effectually will the solution of the gas proceed. Again, another advantage of the fine stream of minute bubbles lies in the fact that the smaller these bubbles are the more slowly they rise to the surface of the water, and thus the longer is the time in which the gas may be absorbed during its ascent. A fine jet, well suited to the purpose here defined, may be made very easily by holding the middle of a piece of glass tubing in a gas flame until it is _very_ soft, and then, immediately on removing it, pulling it out rather quickly. A slight cut made with a small triangular file will then enable the operator to sever the tube at any desired point. Yet another method of maintaining the air supply of aquaria is adopted in the case of some of the large tanks of public aquaria and biological laboratories situated close to the sea, and this consists in renewing the water at every high tide by means of pumps. It must not be supposed that an indoor aquarium, even when well established, and supplied with the best possible system of aëration, requires no further care and attention. In the first place there is a continual loss of water by evaporation, especially in warm and dry weather, and this must be rectified occasionally. Now, when water containing salts in solution evaporates, the water passing away into the air is perfectly free from the saline matter, and thus the percentage of salt in the residue becomes higher than before. It is evident, therefore, that the loss by evaporation in a marine aquarium must be replaced by the addition of _fresh_ water, which should either be distilled, or from the domestic supply, providing it is soft and moderately free from dissolved material.