Criales and Lee: Larval distribution and transport of penaeoid shrimps 



473 



The cruise was divided into four legs (fifty-six sta- 

 tions) from west of the Dry Tortugas to Looe Key in 

 the middle Florida Keys (Fig. 1). The sampling data 

 are summarized in Table 1. Stations of leg 1 were 

 repeated in the other three legs. During leg 2, one 

 transect was added at Marquesas and another at 

 Halfmoon. Leg 3 repeated stations of legs 1 and 2 in 

 addition to an upstream transect (Western transect). 

 Leg 4 included a downstream transect (Looe Key) in 

 addition to the Tortugas, Rebecca, NW Patch, and 

 Marquesas transects (Fig. 1). 



Sampling was carried out continuously depending 

 on the time the ship arrived at the station (Table 1). 

 Samples were taken during the day at 35 stations 

 and at night at the remaining 21 stations. Three 

 moon phases were covered during this cruise. 

 Twenty-six stations were sampled during the full 

 moon, eight during the last quarter, and eight dur- 

 ing the new moon. 



Three ARGOS satellite-tracked surface drifters 

 were deployed on the northern side of the gyre as 

 part of the physical oceanographic survey. Two sta- 

 tions were sampled on the track of drifter Halley af- 

 ter a 15-day interval (day 1=30 May, day 15=13 June). 



0630 h; n=2l, 1=91.4, SD=115.5) stations was not sig- 

 nificantly different tt-test, < 54 =0.82, P>0.05); therefore 

 day and night data were pooled (Sokal and Rohlf, 1981). 

 To characterize the vertical distribution of the dif- 

 ferent larval stages per species, the depth of the cen- 

 ter of mass (Z m ) was calculated (Ropke et al., 1993): 



Z m =^(P i xZ i ) 



where P i is the standardized (no./lOO m 3 ) number of 

 larvae in the i th depth stratum; Z ( is the mean sam- 

 pling depth of the i th depth stratum; C- is the con- 

 centration of larvae in 100 m 3 ; andH i is the width of 

 z th depth interval. Day and night stations were sepa- 

 rated to determine any daily vertical larval migra- 

 tion. Differences in Z m between day and night were 

 not significant for any of the three species (P. 

 duorarum t-test, t 24 =0.63, P>0.05; Sicyonia sp. t-test, 

 £ 36 =0.95; Solenocera sp. t-test, £ 50 =0.28, P>0.05), sug- 

 gesting no daily vertical migration of larvae; there- 

 fore data were pooled. However, this may be biased by 

 the greater number of day stations (35 day, 21 night). 



Larval identification and standardization 



Plankton samples were preserved in 4% formalde- 

 hyde-seawater solution and later transferred to 70% 

 ethanol. Aliquot sampling was performed after the 

 similarity of duplicate samples was statistically 

 evaluated U-test, t 9 =0.04, P<0.05). Each sample was 

 adjusted to a volume of 500 mL, and the penaeoid 

 larvae sorted from five 25-mL aliquots. The mean of 

 the five counts was standardized to the concentra- 

 tion of larvae (no./lOO m 3 ) or to abundance (no./lO 

 m 2 of sea surface). No nauplii were caught because 

 the mesh size of the net was too large to retain them. 

 Penaeoid shrimp larvae were identified to genus 

 or species by using existing keys and larval descrip- 

 tions (Heldt, 1938; Pearson, 1939; Gurney, 1943; 

 Dobkin, 1961; Cook, 1966; Heegaard, 1966; Sub- 

 rahmanyam, 1971a). Postlarvae of Penaeus sp. were 

 identified from the keys of Williams (1965) and the 

 morphometric study of Chuensri (1968). Individual 

 stages were separated only for P. duorarum because 

 it was the only species of interest for which larval 

 stages have been described (Dobkin, 1961). The re- 

 maining penaeoid identifications were made to a 

 generic level. 



Statistical analysis 



Mean abundance of total larvae caught at day (0631- 

 2000 h; rc=35, 3c =98.3, SD=111.6) versus night (2001- 



Results 



Horizontal distribution and abundance 



Penaeus duorarum These larvae showed an on- 

 shore distribution concentrated near the Dry 

 Tortugas (Fig. 2). The density ranged from 1.5 to 57.1 

 larvae/10 m 2 represented by 60.3% zoeae, 15% myses, 

 and 24.7% postlarvae. Distribution of zoeae was 

 nearshore; 90% were caught no farther than 25 km 

 offshore from the Dry Tortugas. Myses were distrib- 

 uted a little farther offshore and eastward than zoeae 

 but no more than 45 km from the coast. The trend of 

 distribution of postlarvae was also close to shore; 83% 

 of postlarvae I— III were located inshore at the 

 Tortugas, Halfmoon, NW Patch, and Rebecca 

 transects. The other 17% were found at the offshore 

 stations of Marquesas associated with the Florida 

 Current front, and no postlarvae were captured at 

 the Looe Key transect (Fig. 2). 



Larval stages in the spawning area at the NW 

 Patch and Tortugas transects showed a progression 

 of ages during the four legs of sampling (Fig. 3). In 

 late May zoeae II— III showed a high concentration 

 (33.8-45.4 larvae/10 m 2 ) at the spawning area, few 

 myses were found at these stations, and no postlarvae 

 were caught. In early June, 15 days after the first 

 sampling, the abundance of myses and postlarvae in- 

 creased. Postlarvae II— III reached their peak of con- 



