FISHERY BULLETIN: VOL. 74, NO. 1 



Table l. — For the years 1964-71 the data are presented for the catch in short tons by semestral cohort in the three areas (N, 5, S) 

 within the CYRA. Also given are the percent of the total catch (Sp^ + Sg + Big) by cohort within the areas. The category, Big, 

 represents the fish of length / greater than 145 cm which we feel are not ageable under the present system. The percent of the 

 individual semestral cohorts (S^ or Sg) caught in the three areas is also given. Note the erratic shifting of the cohort dominance 

 (S^ or Sg) in the catch as well as the distribution of the cohorts between areas. 



Year 



North 

 A 



5 

 A 



South 

 A 



Total 

 A 



North 

 B 



5 

 B 



South 

 B 



Total 

 B 



Total 

 A + B 



Big 



1964 

 % total A + B 

 % total A or B 



1965 



1966 



1967 



1968 



1969 



1970 



1971 



99,019 



85,499 



87,530 



87,848 



113,011 



121,623 



133,526 



103,896 



2,921 

 2.9 



4,543 

 5.0 



3,626 

 4.0 



1,802 

 2.0 



1,602 



1.4 



4,888 

 3.9 



9,176 

 6.4 



9,277 

 8.2 



the mean lengths and modes of the two semes- 

 teral cohorts are separated by approximately 16 

 cm (Tomlinson and Sharp work in progress). A 

 significant number of animals may shift from the 

 leading edge of one labeled distribution into the 

 trailing edge of the other, but we are assuming 

 that countershifts are equally as probable and 

 both are irreversible. An effect of shortening the 

 sampling "season," since the implementation of 

 regulations, has been to distort the apparent 

 abundance of the two groups and merge the 

 modal distributions into a single amorphous dis- 

 tribution (Figure 2). 



The cohorts are treated independently by the 

 model. Each cohort is considered to have a unique 

 effect in the analysis of the net biomass and 

 numbers estimates for a given fishing year. Dif- 

 ferential exploitation of these cohorts can be 

 determined from the catch-effort length- 

 frequency data and as such warrants this disin- 

 tegration technique as opposed to treating the 

 year class as a single unit. We have, however, 

 decided not to present in this report the area 

 breakdown results in the simulations. When the 

 cohorts are separated, it is possible to construct a 

 catch table for each from the length-frequency 

 sample data from the fishery. With this catch 

 table and the catch data (yield) it is possible to 

 determine the relative mortality (F) attributable 



to fishing, by assuming a constant natural mor- 

 tality (M), a necessary, but perhaps poor assump- 

 tion in the case of tunas due to the inherent rapid 

 changes in ecological status as they grow. The 

 Murphy cohort analysis procedure (Murphy 1965; 

 Tomlinson 1970) was used for estimation of re- 

 cruitment at first availability to the fishery (A'40). 

 Using this approach we have generated the un- 

 derlying population structure for the historical 

 series we wish to represent. 



Energetics 



The energetics parameters for free-swimming 

 predatory species such as the tunas must be 

 size-related functions due to the broad range of 

 sizes commonly encountered in the fishery; 1.3 kg 

 to greater than 62 kg, or 40 cm to greater than 

 145 cm. In no case for fish has anyone measured 

 physiological parameters from such a range of 

 sizes. 



Magnuson (1973) discussed the effect of gas 

 bladders and lift surfaces on the velocity of ob- 

 ligatory swimmers such as the tunas. He deter- 

 mined the relationships between size and mini- 

 mum velocity for maintenance of hydrostatic 

 equilibrium for several scombrid species, includ- 

 ing skipjack tuna, Katsuwonus pelamis, and 

 Thunnus albacares. This work has provided a 



38 



