FISHERY BULLETIN: VOL. 70, NO. 3 



Currents, as is evident from the drift model 

 results, also cause variability in annual landings 

 and in the size-frequency distributions. It was 

 noted that in the western portion of the model, 

 drifting objects were found in a single degree- 

 square area that had been introduced along long 

 120°W during 3 seasons. If the drifting ob- 

 jects are fish schools an age difference of 9 

 months would be reflected in the sizes of fish 

 caught. 



TWZO results have shown that the mean geo- 

 strophic flow in the North Equatorial Current 

 can vary by up to a factor of 2 from year to year 

 and that there are large interyear diff"erences 

 in the wind stress. The time of drift from the 

 eastern to the central North Pacific can there- 

 fore vary more, than the range indicated by the 

 drift model, and it is possible for skipjack schools 

 that entered the North Equatorial Current dur- 

 ing 1 year to catch up with those that entered 

 during the previous year. 



Rothschild (1965) also states that lack of 

 growth, or slow growth, as reflected by size fre- 

 quency distributions, can be due to a movement 

 of fish through the fishery. This movement can 

 be fish schools drifting with the currents. 



The size of fish caught in the Hawaiian fish- 

 ery is also aff'ected by the time, place, and size 

 of fish entering the North Equatorial Current. 

 Fish recovered in Hawaii were tagged in April, 

 June, September, October, and November near 

 Baja California, the Revillagigedo Islands, Clip- 

 perton Island, and near the boundary of the 

 Equatorial Counter Current and South Equa- 

 torial Current at lat 4°N, long 119°W (Table 5) . 

 The size-frequency distributions of skipjack 

 caught in the Baja California and Revillagigedo 

 Islands regions presented by Rothschild showed 

 large variation from season to season and year 

 to year. Williams (1972), in another article of 

 this issue, proposes three alternate migration 

 models that explain the recruitment of skipjack 

 into the eastern North Pacific. He concludes 

 that oceanographic conditions in the central and 

 eastern Pacific have a vital controlling eff"ect on 

 the abundance of skipjack in the eastern Pacific 

 fishery. 



Finally, year-class strength determined by 

 survival of larvae and juveniles, as suggested by 



Rothschild, is not ruled out as contributing to 

 the catch rate variations in the Hawaiian fish- 

 ery. Large interyear diflferences in oceano- 

 graphic conditions as reflected by the sea-surf- 

 ace temperatures at Christmas Island (Seckel 

 and Yong, 1971) and the large interyear dif- 

 ferences in sea-air interactions observed during 

 the TWZO investigation undoubtedly aff'ect the 

 survival of larvae, as suggested previously, and, 

 therefore, the population size. However, vari- 

 ations of year-class strength of medium and large 

 fish in the Hawaiian fishery may be masked by 

 the effects of varying currents in the eastern 

 and central North Pacific on the distribution of 

 skipjack. 



An attractive aspect of the drift hypothesis is 

 its simplicity. Skipjack while in. the North 

 Equatorial Current need not do, know, or remem- 

 ber anything other than to search for food. They 

 need not be able to recognize the concentration 

 of salt in the water or distinguish between water 

 types and then know what corrections to make 

 in order to reach the preferred location. They 

 need not be able to recognize time of warming 

 early in the year and then know whether they 

 should or should not enter the Hawaiian fishery. 

 The salinity and temperature indices correlate 

 with availability of skipjack in Hawaiian waters, 

 because the same water motions that aflfect the 

 distribution of temperature and salinity in the 

 North Equatorial Current (Seckel, 1962) also 

 affect the distribution of skipjack. 



In general, it is important to recognize that 

 what is loosely called migration consists of the 

 two components of travel: one resulting from 

 the mean velocity of the water (Vw) and the 

 other from the mean velocity of the fish or fish 

 school relative to the water (Vf). Extreme sit- 

 uations are those where one component is very 

 much smaller than the other so that it can be 

 neglected. There are probably many cases 

 where Vf and Vw are of the same magnitude. 

 An example of these are the migrations of Pa- 

 cific salmon (Royce, Smith, and Hartt, 1968). 

 During certain times of the oceanic life of salm- 

 on, average travel speeds (Vf + Vw) of about 6 

 to 12 miles per day (13 to 26 cm sec~^) are indi- 

 cated. Ocean currents (Vw) with speeds of only 

 5 to 10 cm see"' are therefore of the same mag- 



784 



