MIGRATION OF SALMON IN COLUMBIA RIVER. 1 43 
value this evidence second only to actually capturing the salmon out toward the sea 
from the state trap. 
Marked Chinooks were not recaptured above Millers Sands. Whether they got 
through during the closed season from August 25 to September 12, or from what other 
reason they were not retaken, is wholly a matter of conjecture. Sharp lookout was 
kept for them all along the river at the United States hatcheries, and especially at the 
Ontario (Oreg.) state hatchery, where I collected in early September. No marked fish 
appeared at the Ontario station up to the close of the fishing about November 1, and 
none were taken at the government stations. 
SILVER SALMON. 
The silver salmon, with a single exception, were all retaken by Seufert Brothers 
Company on the Celilo rapids at the Tumwater seining grounds. One, number 87, 
was taken at- Cottonwood Island, by Mr. Ed Le Roy. This last fish was out the longest 
of all the fish retaken — 57 days. 
An examination of table III and of figures 3 and 4 will show that great diversity 
exists as to the degree of etching by corrosion shown by the buttons of these silver salmon. 
The button of number 79 was smooth and clean on both sides. This salmon was out 33 
days, but evidently did not spend much if any of its time in brackish water after it was 
marked. 
Table III. — Marked Silver Salmon Retaken and Extent of Corrosion of Marking Buttons 
by Sojourn in Salt Water. 
Number. 
Time out 
in days. 
Distance 
from state 
trap. 
Corrosion of marking buttons. 
“ U. S. Fish” surface. 
Numbered surface. 
Miles . 
7 ^ 
28 
210 
Slightly corroded about head of rivet . . . 
Deeply corroded on head of shaft and about 
rivet. 
87 
57 
70 
about rivet, but not deeply pitted. 
89 
30 
210 
Slightly corroded about one-half the 
Corroded over entire head of shaft, and 
head of rivet. 
deeply pitted about rivet and on inside 
of end of shaft. 
97 
33 
210 
Deeply coiToded and pitted over this 
Corroded on one-third the head of shaft and 
surface of the button except head of 
slightly on end of rivet. 
rivet; most corroded of all the but- 
tons. 
Number 97, which was out the same length of time and retaken at the same place 
as 79, had the most deeply corroded and pitted button of the entire series. It was even 
more corroded than chinook button number 123 which was out 31 days and was retaken 
only 15 miles up the river. Number 89 was also a deeply corroded button. These two 
fish, 89 and 97, bear evidence of a considerable sojourn in salt or strongly brackish water 
after they were tagged. The buttons of the 3 remaining silvers grade between the 
extremes just discussed, number 75 being almost smooth and 79 considerably corroded. 
