THE AQUARIUM AT THE WORLD'S COLUMBIAN EXPOSITION. 147 



material conveying- the water from the aquaria to the underground cistern, and hard- 

 rubber electric pumps of a capacity of 3,000 gallons per hour, by which the water was 

 lifted from the underground cistern to the tank in the roof. Between the aquaria 

 and the cistern was inserted a sand and gravel filter, and between the roof tank and 

 the aquaria a heating apparatus like that provided for the fresh water. This heater 

 was used, however, chiefly as a cooler by passing a current of lake water through 

 the inner coil. The pumps were in duplicate, only one being in service at a time. 

 These and the rubber tubing connected with them were made to order by the 

 Rubber Comb Company of Rochester, N. Y. 



The fresh water entered the tanks by means of numerous jets, from which it escaped 

 with sufficient force to carry the air well towards the bottom. We found it desir- 

 able, however, to maintain a more efficient bottom circulation by attaching rubber 

 tubes to some of these jets. In the marine aquaria the salt water was similarly intro- 

 duced to the tanks, but an additional aerating apparatus was provided in the form 

 of an automatic air-pump working by water pressure obtained from one of the fresh- 

 water supply pipes. From this pump air was distributed to all the salt-water tanks, at 

 the bottom of which it escaped in the form of clouds of minute bubbles forced through 

 plugs of bass wood inserted in the sides of rubber tubes. This highly successful 

 device for the subdivision of the air is the result of a series of experiments made by 

 Mr. Harron, of the U. S. Fish Commission, under the instruction of the Commissioner. 



The tanks were decorated internally with great skill and ingenuity by Mr. J. B. 

 Mora, of Chicago. For the salt-water aquaria a calcareous tufa had been obtained, 

 at a cost of about $30 a carload, at Toledo, Ohio, where large deposits of it are quar 

 ried for use in the manufacture of paper. It came in selected shapes admirably 

 adapted to decorative work. As we could not foresee the kinds of marine animals 

 which it would be possible to secure, no especial naturalistic effects were attempted in 

 the decoration of the marine aquaria, but generalized grotto-like and other ornamental 

 designs were worked out with charming result. This same material was used for a 

 massive central rockwork in the pool, from the summit of which the water fell in 

 cascades and trickling streams. 



The interior finish of the fresh-water tanks was carefully planned in advance with 

 reference to the favorite haunts and native surroundings of the principal classes of 

 fishes which we mighc expect to obtain. Muddy river banks, bluffs, and islets of strat- 

 ified rock, ponds with sunken logs, pebbly lake shores, margins of pools, submerged 

 trees and stumps, the water-logged rubbish of an old boat landing, fallen brushwood 

 and the like, and rocky beds of streams with mossy incrustations were all admirably 

 reproduced, in the only manner practicable with the time at our disposal, in cement, 

 sand, and gravel. For the purely ornamental fish, like golden ides, goldfish, and the 

 more brilliant minnows, aquaria were decorated in a purely fantastic manner, with a 

 view simply to heightening the artistic attractiveness of the collections. 



Most of our internal decorations were originally designed in expectation of adding 

 to them effects to be derived from aquatic vegetation appropriate to the habits and 

 environment of the various kinds of fish. For the purpose of getting an earlier start 

 of aquatic plants than could be had from the vicinity, collections were made through- 

 out January and February at Eustis, Fla., and shipped to the aquarium early in 

 March. This part of our preparations failed, however, in part because of the unsea- 

 sonable weather which froze or severely chilled the plants in transit, and partly 



