278 SALMON OF THE KARLUK RIVER, ALASKA 



two or three fins, so that their presence in the future runs of fish could be noted. The 

 experiments were initiated to determine the rate of survival of the fish during their 

 stay in the ocean. 



Rich and Holmes (1929), in reviewing the results of previous marking experi- 

 ments, pointed out that fish occasionally have one fin, or two fins in close proximity to 

 each other (both ventrals), accidentally missing. In the marking experiments carried 

 on at Karluk the adipose and one or two other fins were amputated, as it was considered 

 that the finding of a fish with two widely separated fins missing as a result of an acci- 

 dent would be an extremely rare occurrence. 



During the marking of seaward migrants at Karluk and the subsequent examina- 

 tion of the run of adult fish, salmon have been found with the following fins missing: 

 adipose, right ventral, left ventral, both ventrals, right pectoral, and left pectoral. 

 Fish with the dorsal, anal, and caudal, or one of the above mentioned fins badly 

 deformed, have also been observed. More than 400,000 seaward migrant red salmon 

 have been examined at Karluk, and in no case bas a fish been found which had both 

 the adipose and some other fin missing or badly deformed. 



The results of other marking experiments, in which data on the percentage return 

 of marked fish from the experiments were obtained, are reviewed for the sake of com- 

 parison with the results obtained at Karluk. It should be noted that in several 

 instances species other than red salmon were marked, and in no instance were the fish 

 marked as large as the seaward migrants marked in the Karluk experiments. 



Rich and Holmes (192S) in their experiemnts in marking chinook salmon on the 

 Columbia River, from 1916 to 1927, had returns ranging from 0.002 to 0.45 percent 

 of the number of fish liberated from a single marking experiment. They pointed out 

 that — 



These figures have very little significance, however, because they represent not the total 

 returns but an unknown and varying proportion of the total. 



In four of their experiments the records are believed to be fairly complete, and in 

 their opinion 



. . . the returns that have not come to our attention certainly would not add enough to make 

 the totals more than 1 or 2 percent of the liberation. 



Snyder (1921, 1922, 1923, 1924) marked chinook salmon on the Klamath and 

 Sacramento rivers in California, and the proportion of marked fish recovered was 

 approximately the same as in the experiments of Rich and Holmes. 



In 1930, Davidson (1934) marked 36,000 seaward migrant pink salmon at Ducka- 

 bush River, Hoods Canal, Wash., by amputating the adipose and dorsal fins. In 

 1931, 50,000 seaward migrant pink salmon were similarly marked at Snake Creek, 

 Olive Cove, Alaska. These fish were approximately 40 mm. long at the time of 

 marking. From the first experiment 10 marked fish were recovered, or 0.028 percent 

 of the number marked. From the second experiment 23 marked fish were recovered, 

 and it was calculated that the total number of marked fish in the escapement was 54, or 

 0.108 percent of the number marked. These data represent only the return of marked 

 fish in the escapement. However, the total return from either experiment could hardly 

 have equaled 1 percent of the number of fish marked. 



Pritchard (1934a) marked 8,741 pink-salmon fingerlings at Cultus Lake, British 

 Columbia, in 1932, by the amputation of both ventral fins. These fish were released 



