78 BULLETIN OF THE BUREAU OF FISHERIES 



reduce the catch per trap simply because of the competition between traps. Sup- 

 posing that a single trap were operated and caught 25 per cent of the run, then if 

 another trap were placed in an equally advantageous position, but beyond the first 

 trap, so that the only fish to reach the second trap would have to pass the first trap, 

 we would expect the second trap to catch less than the first. If the second trap 

 were equal in efficiency to the first, it would catch 25 per cent of the fish that passed 

 the first one — that is, 25 per cent of 75 per cent, or approximately 19 per cent of 

 the rim. In the same manner a third trap, equally efficient but located beyond the 

 second, would catch only 25 per cent of the fish that evaded the second trap, or 14 

 per cent of the run. Undoubtedly there is some such competition between units of 

 gear, and undoubtedly tliis will tend to reduce the catch per unit as the number of 

 units is increased, regardless of any depletion due to overfishing. The situation is 

 comphcated further by the fact that the fish do not always pass directly through a 

 given fishing area but move back and forth, often on the tides, and thus repeatedly 

 run the gantlet of the fishing gear. In this way the effect of the competition between 

 gear is reduced and under certain circumstances might be entirely nullified, so that 

 any change in the amount of gear would cause a corresponding change in the size of 

 the catch. 



In the second place, it is quite possible that a given fishing effort will take a dif- 

 ferent percentage of a small run than of a large run. To malce use again of the 

 example given above it is possible that while a single trap would take 25 per cent of 

 an average run it might take only 20 per cent of a large run but would take 30 per 

 cent of a small run, or vice versa. So far as we know, there is no evidence that 

 such is the case, but it is a possibility that should be kept in mmd. 



Again, it is very probable that a present-day trap is, m effect, entirely different 

 from those used in 1906; that the two are by no means comparable units. The 

 one trap fished in 1927 or the two fished in 1925 and 1926 were planned and driven 

 in the light of all the experience gained in nearly two decades of fishing in these 

 waters and undoubtedly were driven in the localities and in the manner which have 

 proved most effective. Certainly, as the niunber of traps has been reduced since 

 1916 it was the less productive ones that were eliminated. 



A fourth complicating factor is the reduction in the intensity of fishing due to 

 the increase in the weekly closed period from 36 to 84 hours, which has been effective 

 for the past four years. This undoubtedly has tended to reduce the annual catch 

 per trap, and it may be assumed that the reduction in catch has been approximately 

 in proportion to the reduction in the time during which fishing is permitted. Such 

 an effect necessarily must be taken into accoimt in any analysis of catch per imit 

 of gear and an adjustment made therefor. 



In spite of the difficulties in the way of getting an accurate measure of the catch 

 per unit of effort, we have felt that it was worth while to make the attempt in the 

 case of the Nelson Lagoon data, inasmuch as the conditions here for the analysis of 

 the trap catches are about as ideal as they are likely to be anywhere in Alaska. The 

 available data do not show the actual number of days or weeks fished per season, 

 so that we could not calculate the number of fish caught per trap per day or per trap 

 per week. 



