Wexler et al,: Temporal variation in larval growth of Thunnus albacares in the Panama Bight 



13 



the swimming behavior of the larvae within the mixed 

 layer. Passive transport would probably occur only dur- 

 ing the egg, yolk-sac, and first-feeding stages (the first 

 8-10 days after fertilization) because yellowfin tuna 

 larvae are competent swimmers and can hold their posi- 

 tion against strong currents in the laboratory beginning 

 at around 8-10 mm SL (D. Margulies, personal com- 

 mun.). Although the maximum average area of larval 

 yellowfin tuna distribution from the time of hatching is 

 probably our best estimate, the physical and biological 

 processes that occur in such a large area may not be 

 representative of processes occurring on much smaller 

 scales that may be more specific to conditions affecting 

 larval transport, growth, and survival (Owen, 1989). 



Prey abundance 



Our ichthyoplankton data were collected within a 0.5- 

 degree area that included the Frailes Islands where 

 larvae were sampled and may provide an index of prey 

 abundance (at least for the first one or two weeks of 

 feeding until piscivory occurs). Although our data were 

 spatially limited, the measured plankton volumes pro- 

 vide the only available estimates of zooplankton levels 

 for the periods of interest. 



The use of different gear types (i.e., the bongo and 

 Tucker trawl) for ichthyoplankton collections during 

 1990-92 may have affected the amount of microzoo- 

 plankton sampled during the different years. Microzoo- 

 plankton abundance has not been compared between 

 these two types of sampling nets. However, Shima and 

 Bailey (1994) reported that the bongo and 1-m Tucker 

 nets caught similar numbers and size distribution of 

 larval walleye pollock (Theragra chalcogramma). The 

 higher plankton volumes collected by the bongo in 1990 

 may be an underestimate of plankton abundance com- 

 pared to the volumes of water being sampled by the 

 Tucker trawl with a larger mouth opening (McGowan 

 and Fraundorf, 1966). Given that plankton volumes 

 were probably under-represented in the bongo tows, the 

 difference in magnitude between the amounts of plank- 

 ton sampled by each net type may actually be greater. 



Another potential bias in comparing plankton vol- 

 umes among the different years was that more areas 

 (stations) were sampled with the bongo in 1990 than 

 with the Tucker trawl in other years (MSB station on- 

 ly). However, the mean plankton volume would have 

 been similar in 1990 if only the MSB station had been 

 used in the analysis (Table 3j. 



Growth 



Daily growth rates estimated from the exponential 

 models for each of the three years (1990, 1991, and 

 1997) ranged from 0.46 to 2.06 mm/d and were generally 

 greater than those reported for other congeners (Jen- 

 kins and Davis, 1990; Lang et al, 1994). However, the 

 larvae represented in those studies were predominantly 

 younger and in earlier stages of development (and thus 

 would exhibit slower absolute growth) than the flexion 



and postflexion larvae and transitional juvenile stages of 

 yellowfin tuna collected in our sampling area. The slower 

 growth rates observed in southern bluefin tuna (Thun- 

 nus maccoyii) larvae were also associated with density- 

 dependent and oligotrophic conditions in the East Indian 

 Ocean (Rochford, 1962; Jenkins and Davis, 1990; Young 

 and Davis, 1990). Our growth rates, however, were 

 comparable to similar developmental stages of other 

 scombrids that inhabit relatively similar, productive 

 nearshore waters, such as king and Spanish mackerels 

 {Scomberomorus cavalla and S. maculates, respectively; 

 DeVries et al., 1990), black skipjack (Euthynnus lineatus; 

 Wexler, 1993), and little tunny {Euthynnus alletteratus; 

 Allman and Grimes, 1998). 



Distinct differences in the average size-at-age and 

 growth rates were very apparent between our 1990 and 

 September 1991 collections of yellowfin tuna larvae. 

 Size-dependent processes (i.e., predation and starva- 

 tion; Pepin, 1988; Grimes and Isely, 1996) or density- 

 dependent growth and survival (Jenkins et al., 1991) 

 may affect the size-frequency distributions of surviving 

 larvae. A simulation model (Pepin, 1988) demonstrated 

 that with increased food abundance, the mean and vari- 

 ance in larval growth rates increases, but, as predator 

 abundance increases, the variance in growth rates de- 

 creases for any given mean. Instantaneous growth rates 

 for yellowfin tuna larvae of a similar age in 1990 were 

 2 to 3 times higher than those in 1991, and plankton 

 volumes were 2 to 7 times higher than those in 1991. 

 Increases in food availability, such as that during 1990, 

 may also attract predators and result in greater rates 

 of mortality of the slowest-growing individuals, so that 

 they are not represented in the sampled population. 

 Although the larvae in 1991 were growing more slowly 

 than those in 1990, they probably do not represent 

 the slowest-growing larvae of their cohort. Typically, 

 postflexion larval and early-stage juvenile scombrids 

 collected during the reduced upwelling season in our 

 sampling area have exhibited more variable growth 

 (Wexler, 1993), but have been predominantly healthy 

 (Margulies, 1993). Therefore, slower or faster growing 

 survivors at this stage may be independent of their 

 nutritional condition, and larvae collected by the sam- 

 pling method we used represent the survivors and most 

 competent individuals of their cohort. 



Growth may have been slower in 1991 because of 

 higher larval densities, limited food availability, and 

 available prey composition. A strong inverse relation- 

 ship exists between growth rates and stocking densities 

 of yellowfin tuna larvae and early-stage juveniles (up to 

 18 days after hatching) fed a constant food supply in the 

 laboratory (lATTC, lATTC'; Margulies et al.*^). The re- 



'' lATTC (Inter-American Tropical Tuna Commission). 

 2000. Annual report of the Inter-American Tropical Tuna 

 Commission 1998, 357 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 2001, 148 p. lATTC, 8604 La Jolla Shores 

 Drive, La Jolla, CA 92037. 



