ROTHLISBERG: VERTICAL MIGRATION OF PENAEID SHRIMP LARVAE 



can be obtained. The broad picture based on per- 

 centages indicated that more postlarvae occurred 

 at the surface than near the bottom during day- 

 light while more were near the bottom at night. 

 The actual locations were not given but were 

 probably pooled observations of a number of 

 nearshore stations. Thus, the possibility exists 

 that these postlarvae, close to nearshore nursery 

 grounds and sampled mostly on flood tides, were 

 in a tidally induced behavior pattern that has 

 been shown for the postlarvae of P. duorarum 

 (Hughes 1969a, b), P. aztecus (St. Amant et al. 

 1966), P. plebejus (Young and Carpenter 1977), 

 and P. merguiensis (Munro 1975; Staples 1980). 

 The zoeal and mysis-stage larvae of Trachypen- 

 aeus and Sicyonia were also sampled in the 

 Eldred et al. (1965) study. These early larvae (all 

 stages pooled) showed a marked increase in 

 abundance near the bottom during the day and 

 slightly increased abundances at the surface at 

 night. In their summaries, no larval distribu- 

 tions with intermediate depths and times were 

 shown. In the Gulf of Carpentaria there are at 

 least eight genera represented: Penaeus, Meta- 

 penaeus, Atypopenaeus, Parapenaeopsis, Tra- 

 chypenaeus, Metapenaeopsis, Solenocera, and 

 Eusicyonia. When generic resolution was applied 

 to the present series of samples, no changes in the 

 patterns emerged, but the numbers of larvae 

 available for each analysis decreased. It there- 

 fore seemed reasonable to analyze the pooled 

 samples since it appeared the behavioral pat- 

 terns were family-wide. 



Temple and Fischer (1965) were the first to 

 show clear patterns of vertical distribution by 

 penaeid larvae. They too were sampling mixed 

 genera of penaeids (Penaeus, Trachypenaeus, 

 Sicyonia, and Solenocera). Using discrete depth 

 sampling nets, they showed some increase in 

 migratory ability with the development of the 

 larval stages, as well as the variable patterns 

 seen on different sampling occasions. They did 

 not measure light penetration and attributed the 

 variations of larval distribution to differing con- 

 ditions of water column structure and stability, 

 as characterized by the presence or absence of a 

 thermocline. The same degree of variation in 

 vertical migratory patterns was seen in this 

 study under isothermal conditions, in calm to 

 slight seas. Therefore, turbulence seemed to be of 

 minimal importance in explaining the differ- 

 ences in distribution patterns. However, even 

 under near uniform wind and sea conditions, 

 light penetration in the shallow coastal stations 



was quite variable and this variation was re- 

 flected in larval behavior. It therefore appears 

 that light penetration is the dominant environ- 

 mental variable affecting larval behavior. There 

 is little doubt, under conditions of strong vertical 

 mixing in shallow water, that both turbulence 

 and the concomitant increase in turbidity would 

 result in a mixed vertical distribution. Temple 

 and Fischer (1965) also believed that there was 

 evidence for a reversal of vertical distribution 

 with growth from zoeal to postlarval stage. Their 

 summary figure (fig. 3, p. 62) is not convincing 

 and again may be biased by the change in be- 

 havior of postlarvae to a tidally dominated one 

 used to enter coastal estuaries (Temple and 

 Fischer 1965). In the Gulf of Carpentaria sam- 

 pling, there was no evidence of a reversal with 

 development, but these samples were taken well 

 offshore so a tidal-coastal behavior pattern cued 

 by salinity or pressure differences was probably 

 not in evidence. 



Only one other study of larval penaeid vertical 

 migration has been undertaken; Jones et al. 

 (1970) sampled P. duorarum larvae off southern 

 Florida. Summary figures of pooled data also 

 show the ontogenetic change in vertical migra- 

 tory ability of the larvae, as well as variability in 

 the patterns between sampling periods. No envi- 

 ronmental data are provided to help explain the 

 difference. 



All these studies (Racek 1959; Eldred et al. 

 1965; Temple and Fischer 1965; Jones et al. 1970) 

 were undertaken to help explain how the vertical 

 distribution of penaeid larvae affects dispersal 

 from offshore spawning grounds to nearshore or 

 estuarine nursery grounds. None of the studies, 

 however, was able to explain the variable results 

 encountered, and furthermore, the currents that 

 would be responsible for this onshore movement 

 were not measured. Penn (1975) attempted to 

 overcome these deficiencies by using an ideal- 

 ized larval behavior, one which required all lar- 

 vae to move the full height of the water column 

 from the surface to the bottom on a strict night- 

 day cycle (12:12 larva in Figs. 7, 8, 9). This, how- 

 ever, did not allow for differential larval abilities 

 and/or environmentally induced variations in 

 behavior. Furthermore, because of the lack of in 

 situ current recordings, he used differences in 

 predicted tide heights to calculate residual 

 flows. Penn found, even with these constraints, a 

 general mechanism by which larvae could be 

 transported farther inshore at night than off- 

 shore during the day, against the prevailing cur- 



551 



