It must be understood that the weeds are to serve two distinct purposes:—They are to supply at least some of the oxygen required for the respiration of the animal inmates, and also to serve as food for them. Some marine fishes and molluscs feed on the fronds of the weeds, and among these the common periwinkle may be mentioned as one of the most voracious. If many such animals are housed in the aquarium, it will be necessary to replace at intervals those species of weeds that suffer most from their ravages. The zoospores thrown off by the weeds, particularly in the autumn, are also valuable as food for some of the animals.

Notwithstanding the destructive character of the periwinkle just referred to, it has one redeeming feature, for it is certainly useful in the aquarium as a scavenger, as it greedily devours the low forms of vegetable life that cover the glass and rocks, thus helping to keep them clean; and the same is true of the common limpet and other creeping molluscs. Some of these are even more to be valued on account of their partiality for decaying vegetable matter, by devouring which they reduce the amount of the products of decomposition passing into the water.

Other details concerning the selection of animal and vegetable life for the indoor aquarium must be left to the discretion and experience of the keeper, for it is impossible by written instructions and advice to cover all the various sources of loss and trouble that may from time to time arise. If, however, the general hints for the management of the marine aquarium here given be faithfully followed, there ought to be no further losses than must accrue from the injudicious selection of animal species, and these will decrease as experience has been acquired respecting the habits of the creatures introduced.

We must now pass on to matters pertaining to the maintenance of the healthy condition of an aquarium which, we will suppose, has been established with due regard to scientific principles. Under this head we shall consider, (1) the aëration of the water, (2) the repair of loss due to evaporation, and (3) the regulation of light and temperature.

It has already been shown that the marine aquarium can hardly be maintained in a satisfactory condition as regards its air supply by leaving the aëration of the water entirely to the action of plant life; and herein this form of aquarium differs from that employed for the animal and vegetable life derived from ponds and streams. Fresh-water weeds develop and multiply with such rapidity, and are such ready generators of oxygen gas that it is a very easy matter to establish a fresh-water aquarium that will remain in good condition for years with but little attention; it is therefore important that we should point out the difference in treatment necessary to those of our readers who are already acquainted with the comparative ease with which the fresh-water aquarium may be kept in good order, lest they expect the same self-aërating condition in the marine tank.

It is never a good plan to leave the renovation of the water of the aquarium until there are visible signs within that something is going wrong. It is true that an unsatisfactory condition of the water, revealed by a slight taint in the odour, or a general turbidity, or the formation of a slight scum on the surface, may sometimes be rectified by the prompt application of some method of artificial aëration, but the aim of the aquarium-keeper should be not the rectification of unsatisfactory conditions, but the establishment of such a method of aëration that the unsatisfactory condition becomes an impossibility. We do not wish to discourage anyone who has the slightest desire to start a marine aquarium. Our aim is to point out any difficulties that lie in the way in order that the aquarium may be a success; and thus, having stated that the difficulties attending it are somewhat greater than those connected with the management of a fresh-water aquarium, we should like to add that these practically disappear when one is prepared to devote a short time at regular intervals in order to see that the process of aëration is properly carried out.

Some recommend the occasional injection of air by a syringe as one means of aërating the water; but, although this may be all very well as a temporary purifier of the slightly tainted aquarium, it is hardly suitable as a means of maintaining a good, healthy condition. It must be remembered that oxygen gas—the gas of the atmosphere so essential to animal life—is only very slightly soluble in water. By this we mean not only that water dissolves oxygen very slowly, but also that it can never hold a large supply of the gas at any one time. This being the case, it is clear that the use of a syringe for a short time, though it discharges an enormous total volume of air into the water, will result in the actual solution of only a small quantity. No method of aëration is perfect that allows the admission of air for a short time only at comparatively long intervals; the most perfect system is that in which air is slowly but continuously passed into solution.

Since air is slightly soluble in water, it is clear that it must be continuously passing into any body of water that has its surface freely exposed to it; hence a wide and shallow aquarium is much more likely to keep in good order than one that is narrower and deeper. But, with marine aquaria, the simple absorption from the air at the surface is not in itself sufficient, as a rule, to maintain a healthy condition. Yet it will be advisable to remember this matter when constructing a tank for marine life.

One of the prettiest, and certainly one of the most effectual, methods of supplying air to an aquarium is by means of a small fountain with a very fine spray. The water need seldom be changed, but the fountain may be fed by water from the aquarium, and as the fine spray passes through the air it will absorb oxygen and carry it in solution to the tank.

The accompanying diagram illustrates the manner in which this can be accomplished. The aquarium (a) is supplied with an outlet (o) about an inch from the top by means of which the water is prevented from overflowing, and the outlet pipe leads to a vessel (v) of considerable capacity which, for the sake of convenience and appearance, may be concealed beneath the table on which the aquarium stands. Some feet above the level of the aquarium is another vessel (c), supported on a shelf, having about the same capacity as v, and supplied with a small compo pipe that passes down into the aquarium, and then, hidden as much as possible by the rockery, terminates in a very fine jet just above the level of the water in the centre. The upper vessel should also be provided at the top with a loose covering of muslin to serve as a strainer, and this should be replaced at intervals as it becomes clogged with sedimentary matter.