PACIFIC SALMON 



81 



"in the midst of their migration in the sea, 

 far from the spawning streams toward 

 which they are directing their course. ' ' 



Salmon "... associated in the same 

 trap on Unga Island . . . scattered to more 

 than 20 different streams, ranging from 

 Cook Inlet on the south to Goodnews Bay, 

 the Kuskokwim, and the mouth of the 

 Yukon in the far north." It is true that 

 "The physical and chemical conditions of 

 their immediate environment assuredly 

 were identical for all the salmon tagged 

 and released at a given time from a single 

 trap. ' ' The guiding influence is a response 

 to factors associated with a fresh-salt water 

 mixture. Dall (1882) states, "the Kusko- 

 kwim and Bristol Bay rivers discharge more 

 to the southward, the result being a south- 

 west or southeast set of the waters near 

 them, according to the prevailing wind and 

 tide." He mentions "this general south- 

 erly set of the southeast part of Bering 

 Sea, "as having been "observed by various 

 navigators. ' ' These observations show that 

 there are gradients from the tagging sta- 

 tion to the spawning grounds to which 

 salmon were directed. 



If salmon move to sea, as it has been 

 abundantly proved that they do, why do 

 they reverse their response and return to 

 thC' ijpawning grounds ? Much excellent 

 work lias been done on both sides of the 

 Atlantic. Again the migratory movements 

 of the Pacific salmon are most easily ana- 

 lyzable. The tagging experiments with the 

 Pacific salmon seem to prove that the feed- 

 ing salmon drift with the ocean current 

 since the returns from outside the imme- 

 diate influence of a stream is southward, 

 that is, against the ocean current. The 

 extent of this drift at the present time must 

 be purely speculative. However, there has 

 been no evidence that a salmon which does 

 return has drifted beyond the influence of 

 the inshore waters that mingle with the 

 offshore waters to form the ocean drift 

 waters. This offshore water has two char- 

 acteristics. First, the sea water is mixed 

 with fresh water. Second, this offshore 

 water must have a lower carbon dioxide 

 tension during spring and summer than 



the ocean water due to the photosynthesis 

 of the more intense population of phyto- 

 plankton and to the sea weeds of the con- 

 tinental shelf. These two factors alone 

 might keep the salmon within their influ- 

 ence. Why do salmon approaching sexual 

 maturity return along definite paths to the 

 spawning grounds? The ova and sperm 

 heads contain a protein (protamine) con- 

 taining large percentages of arginine. It 

 requires a vast protein metabolism to 

 obtain the necessary arginine. Protein 

 metabolism and especially fasting — ^both 

 necessary for the liberation of arginine 

 from the muscular tissue of the salmon — 

 tends to produce acidosis of the blood, i.e., 

 lower the alkali reserve of the blood. This 

 is common knowledge. A salmon with a 

 low alkali reserve blood would find low car- 

 bon dioxide tension water more advantage- 

 ous. We have presented evidence if not 

 proof (see bibliography) that fish do have 

 receptors in their gills sensitive to carbon 

 dioxide ten^sion. Due to photosynthesis 

 there must be a gradient of carbon dioxide 

 tension from inshore to offshore waters. 

 The salmon in its return is responding to 

 the carbon dioxide tension gradient or a 

 fresh-salt water gradient. There is again 

 evidence that the salmon is reacting to a 

 carbon dioxide gradient at least inshore. 

 Ward (1927) states, ". . . the coastal 

 streams which are utilized by the Pacific 

 salmon come out of the mountains near the 

 shore. They are characterised by rapid 

 flow of water and considerable drop within 

 a short distance. In the main, they are 

 fed by glaciers or snow fields of the high 

 mountains so that the water temperature 

 is relatively low and the water flow in sum- 

 mer high. ' ' Streams of this type have low 

 carbon dioxide tension water. It has been 

 shown by actual observation (Powers, 1928, 

 and Powers and Hickman, 1928) that the 

 waters of the Columbia and Fraser River 

 systems have relatively low carbon dioxide 

 tensions. 



In order that there be a directive re- 

 sponse there must of necessity be a gradient 

 of the factor or factors to which the salmon 

 are responding. It is not the contention of 



