364 REPORT OF THE COMMISSIONER OF FISHERIES. 
flowed through the hatchery troughs, ponds, ete., and has gathered in 
a waste brook at some distance below the hatchery and pond system, 
save that the process has not been complete enough to remove all the 
excess of nitrogen (‘‘creek water” sample, Table IV). In fact, the 
adjustment begins the instant the water emerges from the wells, and 
in most cases by the time it reaches the fishes it contains somewhat 
more oxygen and less nitrogen than at the well. The effect of the 
deaerating process on the loss of trout fry was shown by passing the 
‘‘reservoir pond water” through a finely perforated metal plate with 
a fall of about 3 feet. During a trial of nine days a trough containing 
6,000-7,000 fry lost 645, against a loss of 2,583 in a similar trough 
containing the same number, but supplied directly from the pond 
without deaeration. The process, which did not completely correct 
the water, reduced the loss 75 per cent. Complete correction would 
probably result from repeating the process or by sufficiently increas- 
ing the fall. The water of hatchery well No. 1 was completely relieved 
of its excess of nitrogen by allowing it to flow drop by drop down an 
inclined wooden plank 10 feet in length. (See Table IV, p. 574.) 
Very few of the Nashua wells delivered free gas, and these only in 
small amounts. From one of these about 500 c. ce. were delivered and 
collected during twenty days and constitute sample 2 of Table I. Only 
air gases were present. Part of the sample was tested for methane, 
unsaturated hydrocarbons, and carbon monoxide, without showing a 
trace of any of these. (Dr. D. A. Morton, Syracuse, N. Y.) The 
sample had no marked odor. The largest pond at the station, used 
chiefly as a reservoir supply and largely spring-fed, had a soft bottom 
from which occasional large bubbles rose. By ramming the mud with 
a stick, large quantities of a gas about 96 per cent nitrogen (Table I) 
could be released. Methane, which might have been expected, was 
absent. This gas seems to be of much the same origin as that from 
the air-bearing spring in Tennessee, though delivered in much smailer 
quantities, and’ may reasonably be supposed to come from a depth 
ereat enough to cause the supersaturation which existed in this pond, 
as in all the sources of water in the vicinity. 
At the Nashua Station the gas symptoms were in evidence, but were 
less marked than in either of the other described cases of the results 
of supersaturated water. Exophthalmia with presence of gas appeared 
in adult trout in ponds, and the general condition of these trout was 
poor. This condition is believed to be secondary to the supersatura- 
tion, which, while not sufficient to kill the adults directly by embolism, 
causes the protrusion of the eye and consequent inflammation. The 
partial or total blindness resulting keeps them from feeding properly, 
and as they fall off in condition and become weaker they are attacked 
by the fungus Saprolegnia. Among the deaths which resulted no 
case of free gas within the vessels was discovered. The fry showed 
occasional gas blisters externally, and in very young fry gas was fre- 
