But the question may be asked, ‘Do not the marine animals and plants utilise a certain amount of the saline matter contained in the salt water?’ The answer to this is certainly in the affirmative, for all sea-weeds require and abstract small proportions of certain salts, the nature of which varies considerably in the case of different species; and, further, all the shelled crustaceans and molluscs require the salts of lime for the development of their external coverings, and fishes for the growth of their bony skeletons. Hence the above suggestion as to the replenishment of loss by evaporation with pure water is not perfectly satisfactory. It will answer quite satisfactorily, however, providing the sea water is _occasionally_ changed for an entirely new supply. Again, since carbonate of lime is removed from sea water more than any other salt, being such an essential constituent of both the external and internal skeletons of so many marine animals, as well as of the calcareous framework of the coralline weeds, we suggest that the aquarium may always contain a clean piece of some variety of carbonate of lime, such as chalk, limestone, or marble, which will slowly dissolve and replace that which has been absorbed. Water is rendered denser, and consequently more buoyant, by the presence of dissolved salts; and, since the density increases with the proportion of dissolved material, we are enabled to determine the degree of salinity by finding the density of the solution. We can apply this principle to the aquarium, as a means of determining whether the water contains the correct amount of sea salt, also for testing any artificial salt water that has been prepared for the aquarium. Probably some of our readers are acquainted with some form of hydrometer--an instrument used for finding the density of any liquid; but we will describe a simple substitute that may be of use to the owner of a marine aquarium, especially if the salt water for the same is artificially prepared. Melt a little bees-wax, and mix it with fine, clean sand. Then, remembering that the wax is lighter than water, and consequently floats, while sand is considerably heavier, and sinks, adjust the above mixture until a solid ball of it is just heavy enough to sink _very slowly_ in sea water. Now make two such balls, and then cover one of them with a light coating of pure wax. We have now two balls, one of which will float in sea water, and the other sink, and these may be used at any time to test the density of the water in, or prepared for, the aquarium. If the water is only a little too salt, both balls will float; while, if not sufficiently rich in saline matter, both will sink.