40 A Brief History of a River Tank.. 



purpose of any particular vessel, it must be prepared in the 

 first instance so as to contain within itself all the conditions 

 essential to success, and when it is completely stocked it must 

 be left to take care of itself; and if it becomes unsightly, or 

 the animals become diseased and perish, then in that particular 

 instance the natural system has failed through some error or 

 oversight of the practitioner. Of course in all this there is 

 much that is artificial ; for instance, it is artificial to put fishes 

 into glass vessels of any kind, and very artificial to build minia- 

 ture caves and grottoes of coke or broken bricks ; but the term 

 natural system is nevertheless quite legitimate, because the 

 endeavour at every step is to illustrate the operation of natural 

 laws ; whereas on the other system the endeavour may be 

 simply to carry out a fanciful device, or a scheme which in its 

 first elements sets the laws of nature at defiance. 



The tank I have to describe was fitted and furnished more 

 than seven years since. It is a simple rectangular vessel, in 

 form nearly a double cube, and its position is in the entrance- 

 hall adjoining the rear wall, where it is sufficiently illumi- 

 nated to render every part of its contents agreeably visible to 

 the eye, yet it receives scarcely any direct light whatever. It 

 is impossible there should be anything more simple than the 

 furnishing of this vessel, yet it does not lack certain features 

 that render it attractive to unscientific eyes. I fitted it myself, 

 and proceeded as follows : — I procured from the cellar a 

 number of large pieces of coke, some of which were broken, 

 to afford ample choice of blocks of various forms and sizes. 

 Those were preferred which had a somewhat rugged face, but 

 which nevertheless were large and fiat, giving plenty of linear 

 surface, so as to be best adapted for building a , wall of very 

 moderate thickness. These were two or three, times dipped 

 into a thick batter of Portland cement, and were finally 

 cemented together in blocks of suitable width for convenient 

 lifting, with one or two empty flowerpots at the base and within 

 every block. The flower pots were used to create a flat basis, so 

 that each block would rest on the bottom of the tank securely, 

 and for the additional reason of the increased lightness of the 

 blocks so constructed, the fronts being made solid and massive, 

 yet hollow within, and the pots not at all or only slightly covered 

 with rock on the sides intended to form the backs of the 

 blocks. Thus a rugged wall of rock was formed in a series 

 of detached pieces, massive in appearance, yet of no greater 

 weight than admitted of convenient lifting. Such a wall 

 made of mica schist — which is the best of rocks for aquaria 

 where its weight is of no consequence — would soon have ruined 

 the tank, and perhaps have brought down tank and stand 

 within a few hours of the first fitting. With coke and flower- 



