134 ALASKA FISHERIES AlTD FUR INDUSTRIES IN 1919. 



seen. Bear trails led through the swamp in various directions, 

 coming out to the shore in many places; in fact almost every little 

 point or solid tuft of ground at the edge of the water had been 

 utilized by bears as a feeding place. The tracks were visible in 

 the mud, tind the skeletons or parts of dead fish la}^ in abundance 

 in the trail near the water's edge. While this was the only bear 

 actually seen feeding on the creek, evidence seemed to indicate 

 that a considerable number of these animals were in the habit of 

 visiting the stream and feeding on the salmon. 



Because of the statements made by residents, an I especially by 

 cj'nnory men n Cordova, that the brown l)ear was the greatest 

 enemy of the salmon and contributed "much more than the can- 

 neries" to the destruction of the fish, particular attention was paid 

 to the problem here and at other points visited. It is certainly 

 true that on visiting such a stream and looking over the banks one 

 gets the impression that the bear has exacted a heavy toll from 

 the spawning salmon. 



On the other hand, when an actual count was made of the fish, 

 it appeared that even considering each of the uncertain fragments as 

 representing an entire fish that had been eaten, the total consump- 

 tion was not very large in comparison with the number in the stream. 

 Of even greater importance is the fact that ]^ractically all the fish 

 seen on the shore were such as had been almost, if not entirely, 

 spawned out and seemed to represent the half-dead fish which after 

 spawning had floated up from the nests in the deeper pools and had 

 been carried by the current near the shore where the bears could 

 reach them. This statement should naturally not be interpreted 

 as indicating that the bears never catch fish before spawning or 

 during the period when they are only ])artly s))awned out. Wliat 

 happens earlier in the season could not be definitely told by condi- 

 tions at the time of the investigation, when the stream contained fish 

 not yet ripe as well as those partly spawned out and many that were 

 completely exhausted and were merely floating near the surface, 

 making only an occasional movement and incapacitated to escape, 

 but at that period the fish caught by the bears were almost exclu- 

 sively spawned out. Furthermore, among the carcasses that could 

 be identified, the majority of the fish on which the bears had fed 

 were males. Indeed, many s])ent fish were )>icked up from the 

 water without the least difficulty. They were the only ones that at 

 that time came near the surface or banks of the stream, whereas the 

 spawning fish stayed down on the nests in deeper water and were 

 both shy and quick to evade any attem]>t at approach. 



A large number of dead salmon were also seen fungused and 

 marooned on the bottom in the broader lakelike portions of the out- 

 let. Live salmon were found nest building and spawning at almost 

 every suitable point in the course of the stream. The party walked 

 up along the bank and tlu'ough the bed of the creek for something 

 over a mile. The stream flows over a bottom composed of coarse 

 gravel and water-worn stones in an endless series of short meanders 

 separated by little ripples that left a deeper pool about every 10 

 yards. In each of these pools from 20 to 30 red salmon were grouped 

 together. The fish showed various stages of maturity from such as 

 were hardly colored at all to those that were already half spent. 

 Including those scattered along the shore there were more dead than 



