904 



BUIyLETiN OF THE BUREAU OF FISHERIES. 



larger. As between the two systems, the troughs and perforated pans, the 

 difference is insignificant on the smallest flow (2 liters) and is in favor of the pans. 

 As the flow increases in volume, the advantage of the pans becomes somewhat 

 more appreciable, but even with the 6-liter flow the troughs are a practicable 

 method of correcting water. Unfortunately, larger flows than this were not 

 tested with the troughs. The following two determinations were made on a f^ow 

 of 12.5 liters per minute, delivered May 3 by the electric pump and passed 

 through pans each perforated by 345 holes of irregular size, but all of them 

 considerably larger than i millimeter. The condition of the water before 

 treating is shown by the "untreated" sample, and is substantially in agreement 

 with that shown by table i. 



Sample. 



Untreated 



Through 5 pans 

 Through 6 pans 



2 p. m 

 II a. m. 



I p. m. 



Temperature 

 of water. 



15-5 

 15-5 



15-5 



CO2. 



38.1 

 27. 2 

 16.8 



21. o 

 17.4 

 15-9 



o. 13 

 4.0 



4.8 



The water after passing 6 pans still has about 2 c. c. too much nitrogen 

 and about i c. c. too little oxygen per liter. The correction is less complete 

 than in any other case, on account of the larger flow and the larger holes in 

 the pans. 



From tables 11 and iii it is seen that the pebbles contained in the troughs 

 add considerably to the efficiency. 



The practical application of these experiments may be found in their show- 

 ing that aeration and deaeration sometimes require an extremely thorough or 

 intimate exposure of water to the atmosphere to restore the dissolved air con- 

 tent to the normal. The water must be spread into very thin sheets, as in troughs, 

 or if subdivided into streams, as by perforated pans, must be reunited and sub- 

 divided repeatedly. By increasing the lengths of trough or the number of 

 pans, the correction can finally be made complete. Troughs have practically 

 the efficiency of pans under the conditions of trials described herein. The objec- 

 tion to them lies in their warming of the water if the air temperature is high, 

 their expense and cumbersomeness. One great advantage they possess is that 

 they require little vertical space, and therefore they could be used where the 

 fall from tap to trough is too short to permit the insertion of a sufficient number 

 of pans, provided there is sufficient room laterally. 



When pans are used, the diameter of the holes must be controlled largely 

 by the volume of water, the amount of fall available, and especially by the 

 sediment the water carries. The smaller the holes the better, as far as 



