4 



ball-tap arrangement fitted to the water-motor, so that when- 

 ever the water in the lower tank gets below a certain level, 

 the pump is started, and a fresh supply pumped up. This 

 automatic arrangement works day and night, with only an 

 occasional visit from the attendant. 



Three years' experience has shown by this arrangement on 

 an average 1,250 gallons of fresh water have to pass through 

 the motor to pump 850 gallons of sea water into the tower. 

 This is about our daily circulation, and costs in water J%d,* 

 per day. The salt water is all made artificially, except such 

 as is used to convey live stock from the coast. I used the 

 formula given by Mr. W. H. Jones, who has been so suc- 

 cessful at the Aston Park Aquarium, and found the cost to 

 be about one-third of a penny per gallon. After all the 

 chemicals have been added to the fresh water in the proper 

 proportions, the whole was well stirred up every day for ten 

 days or so. The water was then allowed to settle for a couple 

 of days, and then circulated in the tanks for two days more 

 before any live stock was added. By this means the water 

 obtained a brilliancy which is seldom seen in large aquaria. 

 We have never had any trouble with it, and year by year the 

 crystal appearance gets more pronounced, and objects which 

 two years ago could not be kept alive a week, live now quite 

 healthily month after month. I attribute a great part of this 

 success to the constant circulation of the water day and 

 night. The vulcanite jets I had suppled at first I have 

 altogether discarded. The holes, varying from one-sixteenth 

 to three-sixteenths of an inch, were too large and not of the 

 proper shape. Too much water was used by each jet, and 

 the air taken into the tanks by the force of the water was in 

 such large bubbles, that the greater part immediately rose to 

 the surface again without imparting its oxygen to the water. 

 Again, the force of water had to be so strong as to stir up the 

 sediment in each tank, and thus thicken the water. The 

 system I have now adopted gives very different results. A 

 small glass tube of suitable length is drawn out to a fine point 

 for the jet and adapted to the supply pipe by means ot an 

 indiarubber stopper. There is a stop-cock to each jet, so 

 that I can regulate the amount of water passing through. 

 The top of the glass tube is within an inch of the surface of 

 the water, and myriads of tiny air bubbles are taken quite to 

 the bottom of the tank. They are so small that they do not 

 rise quickly to the surface, and they can be traced floating 

 gently along six feet from the jet. There are two jets in each 

 tank, but only one is used excepting in very hot weather. 

 The difference in results between this system and the old one 



* This has since been reduced to Shd. to 4d. by using the 

 arrangements described in Nos. V. and VI. 



