THE AQUARIUM AT THE WORLD'S COLUMBIAN EXPOSITION. 151 



These conclusions are still further reinforced by the experience of the Illinois 

 commissioners in a live fish exhibit maintained in the Illinois State Building. Here 

 still shallower pools with white enameled bottoms were used, supplied, as in our own 

 central basin, by a dripping', splashing inflow of water falling from a considerable 

 height. The miscellaneous collections in these pools were almost free from parasitic 

 fungus attacks throughout the entire season. 



The interior decoration of tanks of all descriptions with rock work and other 

 naturalistic material is a matter of especial practical importance. These decorative 

 materials should cover the interior surfaces, especially the backs and ends, but should 

 not be so constructed as to occupy the interior of the tanks and thus obstruct a view 

 of the contents. Especial care should be taken that no holes or crevices are left in 

 which dead fish or other noxious objects may remain concealed. The best materials for 

 naturalistic decoration are native rock, clean sand and gravel, and Portland cement. 

 With proper cleaning down, repeated soaking, and thorough washing out, no contain 

 ination of the water need be feared, even with a closed circulation. Mud and clay, 

 water-soaked wood, and the like, are of course inadmissible, and cement must be used 

 for the imitation of these materials. 



I need hardly add that in any well-constructed aquarium suitable storage and 

 work rooms and offices of administration will be provided. It would, of course, be a 

 great error in a permanent establisment to omit ample provision of laboratory rooms 

 and appliances for systematic scientific study and experiment. These could be fur- 

 nished at a slight addition to the first cost and could be maintained at a comparatively 

 trifling expense. They would greatly increase the usefulness of any aquarium by 

 making its resources available for the advancement of knowledge, thus enlisting the 

 sympathy, aid. and support of scientific men and of the educated classes generally. 



Our circulating apparatus for the salt water was, when supplemented and com- 

 pleted according to the plans of Engineer Reeves and the Commissioner, in every way 

 satisfactory for summer maintenance, with the single exception of a lack of sufficient 

 means for temperature control — a difficulty due to the character of the building and 

 not to the circulating pla r, t.* It is possible that a better plan of securing pressure 

 might be devised than that of the use of the elevated tank. A closed underground 

 cistern of air-tight construction, into which the water should be forced by the pump, 

 depending upon air pressure within the cistern to send the water up to the aquarium 



*The temperature of the salt water, beginning May 1 at 50°, reached 70° F. — regarded by us as the 

 danger point for susceptible marine species — on June 19, but the second day afterwards fell off a 

 degree. The 7th of July it came up to that point again, and remained between 70- and 74°, with only 

 an occasional drop to 69A°, until August 12; 7(K ami 70A° were again repeatedly reached during the 

 latter half of August, but the mouth closed with a temperature record of 67°. In September, 71A° was 

 leached at the middle of the month, but the record fell to 61 u and 62 c for Sentember 30. During 

 October the thermometer readings varied from 67° on October 12 to 57° on the last day of the month, 

 when the exposition closed. The highest temperature uoted (74°) was reached July 25, with an 

 average air temperature in the aquarium for the day (6 a. m. to 6 p. in.) of only 83°. These tempcra- 

 tiues were reached notwithstanding the constant use of the cooler, the shading of the windows in the 

 wall by cheese-cloth curtains, and the covering of the roof lights with an outside canvass screen. 



The fresh-water record runs from May 1 to September 30, three daily observations being taken for 

 the air and water. May 1 the lake water, as received in the aquarium tanks, stood without variation 

 nM2 u F., the air averaging 46 1 -. The rise in temperature for May was gradual and steady, reaching 

 hi J May 15, 50° on the 21st, and 55° on the 30th. It remained practically at this point until June 13, 

 when it began to rise again, going somewhat irregularly to 65° on June 30. This rise continued to 

 '0 , first readied July 11, and to our maximum record of 74°, made July 24, 31, and August 1. The 

 'all from this highest point was gradual and very slow, the last observation (.September 30) being 59 



