FISHERY BULLETIN: VOL. 87. NO. 3, 1989 



niles. The mortality curves based on Z2«) fitted 

 the larval and juvenile data better than Ziit), 

 and therefore, we chose Z2it) to describe the fate 

 of young sauries. The mortality curve with a 

 constant IMR takes the form of the exponential 

 mortality curve, 



P, = Poexp(-ai) 



where Pt is the daily fish production at age t, tis 

 the age in days from hatching, Pq is the daily 

 larval production at hatching, and a is the daily 

 IMR. 



For the computation of larval production and 

 mortahty rate averages of 15 years, we used the 

 data of the NIT category. For the computations 

 of individual years, however, the number of NIT 

 samples were not sufficient; therefore, we had to 

 use the data from all the time periods. The data 

 sets of individual years were classified into 10 

 time periods of 11 size classes and were cor- 

 rected by dividing numbers of fish by their spe- 

 cific retention rates (Table 1). The corrected 

 numbers of fish were summed up in each size 

 class and divided by the total number of tows of 

 the year to calculate catch/tow values as the 

 abundance index of each size class. 



We used the catch data of the early, main, late, 

 and off spawning seasons (from September to 

 August of the following year) as a unit of a 

 spawning year instead of the calendar year. Ac- 

 cording to the recent growth model by Watanabe 

 et al. (1988), the saury becomes adult within a 

 year, so, we supposed that the larvae produced 

 in a spawning year constitute the major part of 

 the year class of the current year. 



The catch/tow value of each size class was 

 further divided by the duration of the corre- 

 sponding size class. This provided the production 

 rate (catch/tow/day), which was used for the 

 computation of larval production at hatching and 

 mortality. The parameters, larval production at 

 hatching (Pq) and daily IMR, are estimated by 

 using a nonlinear program. (Par (Dixon et al. 

 1985)). 



RESULTS 



The Overall Mortality Curve for 

 1972-86 



To understand the fate of young sauries from 

 1972 to 1986, the average catch/tow/day from the 

 NIT samples (Table 2) was used to estimate the 

 larval production at hatching and the IMR. All 

 the fish were first grouped into 11 size classes, 

 ranging from 7.5 mm (5.0-9.9 mm) to 57.5 mm 

 (55.0-59.9 mm), and the midpoint of each size 

 classes was converted to age. The data from the 

 first two size classes (7.5 mm and 12.5 mm) were 

 excluded from the analysis because the catch/ 

 tow/day in these two size classes was lower than 

 the next older larvae. The downward bias could 

 be due to extrusion of the larvae through the 

 mesh. The estimates of the larval production at 

 hatching (Pq) and daily IMR were 1.255 larvae/ 

 tow/day (SE = 0.111) and 0.078 (SE = 0.004), 

 respectively (Fig. 4). The total daily mortahty 

 rate {{P,-i - Pt)/Pt-i = 1 - (exp(-a))) was 

 7.5% from the newly hatched larvae to the 52 d 

 old juvenile. 



Table 2.— Daily production (catch/tow/day) at age (day) of Pacific saury from NIGHTTIME samples, 1972-86. 



'Fish knob length after preservation. 



^Fisti knob length after capture (before preservation). 



^Not used for the computation 



606 



