Embody. — Trout Feeding Experiments 27 



cover which could be locked securely. The water always con- 

 tained oxygen to the point of saturation. It was also rendered free 

 from all food organisms by passing it through a settling basin 

 and series of screens. Thus the conditions in each trough were 

 identical with those of every other trough insofar as it was possible 

 to make them. 



Preliminary tests indicated that the most accurate measure- 

 ments and observations could be taken only in comparatively 

 small basins and with a small number of trout. Consequently, the 

 troughs just described were made to hold a volume of water 

 available to the fish, 3 feet long, 16 inches wide and 5 inches deep 

 and the number of trout placed in each trough varied from 46 to 

 100 depending upon their size. A regular hatching trough 8 feet 

 long and 14 inches wide was used for the chinook salmQn finger- 

 lings, while in the case of 2-year-old brook trout, a cement basin 

 8 feet long, 3 feet wide and 14 inches deep was found convenient. 



All the trout were weighed in a glass vessel of water which had 

 previously been balanced upon a set of standard laboratory 

 scales. The trout were first taken out of the trough with a small 

 dip net, allowed to drain one minute and then carefully placed in 

 the glass vessel of water. The net was then lifted carefully out of 

 vessel allowed to drain one minute into the same, and then the 

 process was repeated until a sufficient number of trout were 

 ready to be weighed. With this precaution the error due to adding 

 or removing water was reduced to a minimum. Weights were 

 always taken from 12 to 15 hours after feeding. 



The experimental procedure consisted in (1) weighing trout at 

 the beginning and end of each experiment ; (2) feeding them once 

 daily in the case of yearlings and twice daily in the case of finger- 

 lings. The food was given slowly and in sufficient quantities only 

 to insure entire consumption. Hence there was no waste to be 

 deducted from the total weights as recorded. (3) Records were 

 kept of the total food consumed during the experimental period; 

 of the mean daily water temperature, and the mortality. In this 

 manner a rather large series of data was obtained, the more sig- 

 nificant of which are included in a table which follows. 



