14 



Fishery Bulletin 105(1) 



lationship may be even more pronounced under limited 

 food conditions (Jenkins et al., 1991). Although we have 

 only indirect evidence for density-dependent growth of 

 the 1991 cohorts, the occurrence of more yellowfin tuna 

 larvae sampled at the surface with the night light and 

 the large numbers of other fish larvae collected in the 

 ichthyoplankton tows coincident with the lower plank- 

 ton volumes may indicate growth-limited conditions 

 during this period. The slower growth of the late-stage 

 larvae in 1991, when plankton abundance was much 

 lower, may also be indicative of the types or species of 

 preferred prey (zooplankters and fish larvae) that were 

 available in the area that the larvae occupied. In the 

 laboratory, yellowfin tuna larvae have predominantly 

 selected all stages of cyclopoids over other types of co- 

 pepods when offered a mixed assemblage of zooplankton 

 prey (Margulies et al., 2001) and have become piscivo- 

 rous beginning at approximately 6-7 mm in SL (Margu- 

 lies et al.''). During this transitional stage in their diet, 

 growth becomes much more rapid and variable (Kaji 

 et al., 1999; Margulies et al.^), and the availability of 

 specific types of fish larvae may influence their ability 

 to switch to piscivory in the ocean. Yellowfin tuna lar- 

 vae readily consume other, smaller conspecifics in the 

 laboratory, but it is not known if there is a preference or 

 growth advantage for consuming certain species of fish 

 larvae during the transition to a more piscivorous diet. 

 Although the available prey composition could affect the 

 growth of late-stage yellowfin tuna larvae, intra- and 

 interspecific competition for limited food resources dur- 

 ing the 1991 period may have been the principal cause 

 of slower growth. 



The temporal variation in size-at-age within the same 

 season and year (1991) may be related to the physical 

 and biological characteristics of the area occupied by 

 each group of larvae since hatching. Larval distribu- 

 tion could be determined by the location and timing 

 of yellowfin tuna spawning and the small- and large- 

 scale dynamics of physical oceanographic processes. 

 Average sizes of the larvae and their otoliths of the 

 1991 August-September group were distinctly and sig- 

 nificantly smaller than those of all other groups. The 

 back-calculated first-feeding dates of this group coin- 

 cided with the lowest plankton volumes measured in 

 our local sampling area and with the only collection 

 of first-feeding yellowfin tuna larvae from our ichthyo- 

 plankton tows. In contrast, the mean sizes of larvae and 

 otoliths of the 1991 July group were similar to those of 

 the fastest-growing group of 1990, despite low plankton 

 volumes similar to those of the September 1991 period. 

 We believe that the September 1991 group may have 

 been spawned nearer to our local sampling area (Fig. 2) 

 and that they were more exposed to feeding conditions 

 in the vicinity of the Frailes Islands than were the 

 faster-growing larvae of the July 1991 group. 



Physical effects on growth 



The probability of feeding success in marine larvae, and 

 subsequent growth rates and survival, may increase 



with moderate levels of wind-induced microscale tur- 

 bulence in their feeding environment (Rothschild and 

 Osborn, 1988; Cury and Roy, 1989; Ware and Thomson, 

 1991; MacKenzie et al., 1994; lATTC*, lATTC^). Pre- 

 liminary estimates of wind speeds that produce optimal 

 turbulent velocities and maximum survival of first-feed- 

 ing yellowfin tuna larvae in the laboratory are moderate 

 to high (D. Margulies, personal commun.) compared to 

 wind speeds measured in the Panama Bight during this 

 study. This estimate of optimal wind speeds is based on 

 the assumption that maximum abundances of yellowfin 

 tuna larvae in the EPO occur at depths of to 20 m. 

 However, the data on wind stress and velocities in the 

 estimated area of larval distribution may not represent 

 the frequency of optimal wind speeds associated with 

 areas where first feeding of each cohort occurred. Addi- 

 tionally, wind event durations and frequencies, for which 

 data were not available, may also play a significant role 

 in the optimal survival of marine larvae (Wroblewski 

 et al., 1989). Although moderate to high wind-induced 

 turbulence may enhance early larval survival, growth 

 rates may actually be slower during a portion of the 

 larval phase as a result of higher larval densities and 

 increased competition for limited resources. 



Temperature-limited or -enhanced growth was not 

 clear from our analyses of the yellowfin tuna larvae 

 collected. Although a parabolic relationship between 

 growth rates and SSTs was evident, it is unlikely that 

 the two slowest-growing groups represented by the up- 

 per (29.1°C; group V) and lower (27.6°C; group III) 

 mean temperatures in our study (Table 2) are approach- 

 ing thermal tolerance limits for yellowfin tuna larvae. 

 In the laboratory, successful hatching of yellowfin tuna 

 larvae still occurs at upper temperatures of 32-34°C 

 and first-feeding larvae are able to survive and feed be- 

 tween temperatures of 21° and 32°C (Margulies et al.^^). 

 Lower temperatures during the August-September 

 1991 (group III) period (Table 2) may have resulted in 

 slower growth than that of the other collection periods, 

 but we have only found significantly slower growth 

 rates and differences in the mean sizes at first feeding 

 when mean temperatures were less than 27°C in the 

 laboratory (senior author, personal commun.). In con- 

 trast to the most rapid growth rate in 1990, the larvae 

 in 1997 were growing more slowly when a strong ENSO 

 event and the highest SSTs occurred. The optimum 

 temperature range for growth of yellowfin tuna larvae 

 in the Gulf of Mexico was 29-29.5°C (Lang et al., 1994), 

 which was a similar temperature range for larvae of 

 the 1997 period in our study. We do not, however, have 

 information on relative food abundances during this 



* lATTC (Inter-American Tropical Tuna Commission). 



2001. Annual report of the Inter-American Tropical Tuna 

 Commission 1999, 183 p. lATTC, 8604 La Jolla Shores 

 Drive, La Jolla, CA 92037. 



^ lATTC (Inter-American Tropical Tuna Commission). 



2002. Annual report of the Inter-American Tropical Tuna 

 Commission 2000, 171 p. lATTC, 8604 La Jolla Shores 

 Drive, La Jolla, CA 92037. 



