Tut Sprep or Micratinc SALMON IN THE CotumBtA RIver. 
$y CHAS. W. GREENE. 
(Abstract.)* 
In the solution of this problem I devised a scheme whereby individual 
fishes could be given individual tags that would render identification ab- 
solutely certain if the fish should be recaptured. This plan was nothing 
more or less than the use of the conventional stock-marking aluminum 
buttons. These buttons are light and cannot be torn apart and they carry 
serial numbers on one face; on the other can be placed such special 
marks as one may select. 
On August 14, 1908, I marked fifty-nine fish at Sand Island, just with- 
in the mouth of the Columbia River. 'These fish were liberated in the 
river in the hope that some would be retaken, and thus we might glean 
the story of their migration. ‘The fish were marked by inserting num- 
bered buttons through the caudal fin. 
Seventeen of the fifty-nine fish liberated were retaken and reported 
to me; sixteen buttons were also returned to me. The fish were retaken 
along the river from a point four miles below where they were liberated 
up to the Dalles of the Columbia, just below Celilo Falls, a total dis- 
tance of two hundred and fourteen miles. Near the upper limit quite a 
number of fish were taken and six of these had traveled a distance which, 
when rated, gives an average individual speed of from six and one-third 
to seven and one-half miles a day. 
The following table is constructed to show the actual time from lib- 
eration to recapture, the distance covered, the probable time consumed 
in the straight-away run on a basis of the speed of number 76 (seven and 
one-half miles), and the days unaccounted for. My view is that these un- 
accounted days are chiefly spent in the lower estuary of the river in be- 
coming acclimated to the fresh water. 
17This investigation was undertaken in cooperation with the United States 
Bureau of Fisheries. This abstract is published by the consent of and with the 
approval of the U. S. Commissioner of Fish and Fisheries. 
