DISSOIvVED CONTENT OF WATER. 



901 



is SO great that the fishes are immediately suffocated before the nitrogen has 

 time to cause any symptoms. When the oxygen is increased by aeration, the 

 nitrogen, of course, is at the same time decreased. 



The oxygen, as shown by boiling determinations, agrees approximately 

 with that shown by titration, in no case rises to i c. c. per liter, and varies from 

 less than o.i c. c. to 0.45 c. c. (excluding first sample). This deficiency is prob- 

 ably practically as bad, as far as fishes are concerned, as if the water were 

 absolutely lacking in dissolved oxygen. 



COMPARISON OF MEANS OF CORRECTION. 



Having thus a water in which the coexisting deficiency of oxygen and the 

 excess of nitrogen were each more extreme in degree than in any case yet met 

 with, it was thought desirable to experiment in correcting it by exposing it to the 

 atmosphere, and to compare methods or devices for accompUshing a thorough 

 exposure. The question of what was the best general method was submitted 

 to Mr. H. von Bayer, the architect and engineer of the Bureau, who recom- 

 mended flow along sanded and pebbled troughs on a gentle incline as theoretic- 

 ally best adapted to expose and correct small or moderate volumes of water. 

 Wooden troughs were accordingly made under his direction and when finished 

 were substantially as follows: Total trough length 44 feet, 6 to 8 inches wide. 

 The water flowed 22 feet through two joined troughs with a fall of 2 inches, 

 then fell 10 inches to the second set of two troughs and returned through 22 

 feet, with a fall of 2 inches, and was delivered into an aquarium. The troughs 

 were painted with asphalt and sanded on the moist asphalt. When dry, peb- 

 bles of various sizes were strewn along the bed of the troughs, thus imitating 

 natural flow in pebbly brooks. 



The troughs were suspended one set above the other near the ceiling in the 

 aquarium grotto and the water delivered into the head of the upper from a 

 rubber tube. Two trials were made in the sanded troughs, but without pebbles, 

 with a flow of 2 liters per minute. The following table shows the correcting 

 effect of such troughs. 



Table II. — Water from Electric Pump before and after Passing Through Sanded Troughs 



Without Pebbles. 



