THEILACKER: MORTALITY OF SEA-CAUGHT JACK MACKEREL 



ing to each size and health category was divided by 

 the duration— the number of days jack mackerel are 

 expected to remain in each category (Table 4). Dura- 

 tions spanned 1 to 6 d depending on age and condi- 

 tion. For healthy fish, duration is simply the size-class 

 interval divided by the growth rate Healthy fish 

 belonging to the smallest size group (<3.5 mm) grow 

 at 0.05 mm/d (Theilacker 1978) and begin to eat at 

 3.2 mm SL. Thus duration for this size interval (0.3 

 mm) was 6 d. Growth rate for older fish was 0.15 

 mm/d; the rate was determined for this study by 

 counting daily growth increments in otoliths (Hewitt 

 et al. in press). The duration that a larva remains 

 in one of the starvation states is a function of the 

 persistence of the histological criteria. Young jack 

 mackerel deprived of food in the laboratory show 

 signs of starvation for 3 d before dying, and larvae 

 recovering from a period of starvation show these 

 signs for 2 d (Theilacker 1978, 1981). Older fish may 

 be more resistant to starvation, but as I had no in- 

 formation for older jack mackerel, I used the dura- 

 tions for younger larvae 



For the smallest jack mackerel living 350 km off- 

 shore, 45% were dying of starvation per day. Daily 

 mortality dropped rapidly to 5% to zero for older 

 larvae (Table 4). Increasing the durations for the 

 older larvae in the starving and recovering 

 categories (Table 4) decreases this estimate of daily 

 mortality. 



Results of the histological examination of jack 

 mackerel collected near islands and banks allow a 

 preliminary assessment of the effects of different 



100 r 



o 



z 



> 

 cc 



< 



</) 



»- 

 z 



UJ 



o 



DC 

 1U 



a 



80 - 



60 



40 - 



20 



50 100 150 200 250 300 350 

 NUMBER OF JACK MACKEREL COLLECTED 



Figure 15— Percentage of starving Trachurus symmetricus lar- 

 vae (number starving/number analyzed) related to the number of 

 larvae collected at each offshore station; station number is 

 indicated. 



habitats on starvation (Table 5; Fig. 16). A large dif- 

 ference existed in the daily larval mortality caused 

 by starvation between the open ocean and island and 

 bank habitats. In areas near the isl ands, none of the 



lying of starvation whereas 



first-feeding larvae were 

 45% from the open ocean were dying of starvation. 

 /&7 In addition, healthy larvae taken near islands were 

 \s apparently more fit than healthy larvae captured in 

 the open ocean, as the larvae from the island habitats 

 had abundant quantities of glycogen in the liver (Fig. 

 10), whereas livers of larvae from the open ocean 

 rarely contained glycogen stores (Fig. 9). This in- 

 dicates that food must have been much more abun- 

 dant in the island habitat because not only were 

 fewer fish starving but the healthy fish were able 

 to store glycogen. The healthy fish from the open 

 sea may have been just able to meet their daily 

 metabolic requirements. 



The morphological data gave essentially the same 

 results as did the histological method. On the basis 

 of morphometric evidence, 70% of the first-feeding 

 jack mackerel (<3.5 mm SL) were starving and the 

 number decreased to zero for older jack mackerel 

 (Table 6). Although the results were similar, the mor- 

 phological categories used to classify the fish were 

 different from the histological ones. In particular, 

 there was no morphological category for dying fish. 

 For the morphometric SWDA, larvae were grouped 

 by feeding treatment (Table 6), and the histological 

 categories (Table 4) were based on the dominant lar- 

 val tissue conditions determined to characterize a 

 nutritional state. Thus the morphometric SWDA 

 cannot be used to estimate the number of larvae 

 dying per day due to starvation. 



DISCUSSION 



Larval Starvation and Recruitment 



Both histological and morphological criteria in- 

 dicate that starvation is probably a major source of 

 larval jack mackerel mortality at the time of first- 

 feeding but that the survivors of this 6-d period are 

 much less vulnerable to starvation. Prey (mainly 

 young stages of copepods) are more abundant at the 

 nearshore islands and banks off the coast of Califor- 

 nia than offshore (Beers and Stewart 1967, 1970; Ar- 

 thur 1976, 1977; Devonald 1983), and survival of 

 first-feeding jack mackerel was higher in the near- 

 shore habitats than offshore Thus selection of 

 spawning sites may have a great effect on survival. 

 Eggs and larvae of jack mackerel are very widely 

 distributed; they occur from Baja California to 

 British Columbia and up to 400 mi off the coast of 



© 



ii 



