COWAN AND SHAW: LARVAL SCIAENIDS COLLECTED OFF WEST LOUISIANA 



previously reported. Monthly length-frequency 

 data for sand seatrout show that larvae as large 

 as 11 mm TL were first present in March samples 

 (Fig. 2A). Based on the estimated growth rate 

 determined for sand seatrout (Cowan 1985; Shaw 

 et al., in press), an 11 mm larva could be as old as 

 65 days; this further supports January spawning. 

 Sand seatrout are reported to spawn from March 

 to August, during two discrete periods — one in 

 March-May, the other August-September, with 

 little spawning between the two peak periods 

 (Hoese 1965; Daniels 1977; Shlossman and Chit- 

 tenden 1981). 



An examination of distribution and length- 

 frequency data (Figs. 1, 2A) suggests that most 

 spawning initially took place in midshelf to off- 

 shore waters at depths ranging from 15 to 80 m or 

 to about 175 km from shore. As the season pro- 

 gressed into March and April, spawning location, 

 as determined by the presence of larvae <3.0 mm 

 TL, was more inshore (5-18 m) with few small 

 larvae occurring at depths >25 m. 



Other than the indication that spawning may 

 move from offshore to inshore waters as the sea- 

 son progresses, this spatial information agrees 

 with the limited life history data available on 

 sand seatrout. Most spawning has been shown to 

 occur in the shallow waters of the Gulf of Mexico, 

 primarily between 7 and 15 m in depth (Gunter 

 1945; Moffet et al. 1979; Shlossman and Chitten- 

 den 1981). Running ripe C. arenarius have been 

 captured in deepter waters (70-90 m) in February 

 and March, but no spawning was indicated 

 (Franks et al. 1972; Perry 1979). 



In a four-way ANOVA employed to determine 

 patterns of larval sand seatrout density and dis- 

 tribution, month was a highly significant main 

 effect (P < 0.01; Table 3) reflecting spawning sea- 

 sonality and the magnitude of the density in- 

 crease in April. The test for interaction between 

 month and day-night was employed to determine 

 if daytime gear avoidance was evident as size and 

 mobility of larva increased. Most sand seatrout 

 collected, however, were small and no clear 

 monthly modal increase in larva size was evident 

 (Fig. 2A). The significant interaction (P <0.01) 

 was probably due to an increased catch in oblique 

 tows at night as the season progressed (0.0 in 

 January, 64.9/100 m"^ in April). The significant 

 interaction between month and depth group and 

 the highly significant depth group main effect 

 (P < 0.01) represents the shift in larva concentra- 

 tion from midshelf early in the study, to a more 

 coastal distribution in March and April (Fig. 1). 



Mean larva density was greatest in depth group 1 

 (23.7/100 m'^) followed by depth groups 2, 3, and 4 

 (12.7, 9.2, and 0.3/100 m^, respectively). The third 

 main effect, day vs. night tows, was highly sig- 

 nificant (P < 0.01); many more sand seatrout 

 larvae were collected at night (averge catch rates 

 in all night (74) tows combined = 21.6/100 m"^ 

 vs. day (113) tows = 7.4/100 m^). Highest night- 

 time catches occurred in oblique (49) tows (26.9/ 

 100 m'^) while the day-oblique-catch rate aver- 

 aged 7.6/100 m'^ in 76 tows. Overall, average 

 catch rate was highest in oblique (125) tows 

 (14.6 larvae/100 m-^), followed by surface (31) 

 tows (9.2/100 m^), and then bottom (1.9/100 m^; 

 31 tows). Intrepretation of the data suggests that 



Table 3. — Summary data from four-way analysis of variance done 

 on logio transformed [(larvae 100 m3) + 1] data from ichthiyoplank- 

 ton samples collected from January to April 1982. Tfie results are 

 for A. Cynoscion arenarius. B, Micropogonias undulatus, and C. 

 Leiostomus xanthurus. The four main effects tested were months 

 (Jan. -Apr.), station depth group (d.g. 1 < 10 m, 10 m < d.g. 2 < 14 

 m, 14 m < d.g. 3 < 24 m and d.g. > 24 m), day - night (2000 

 hours s night < 0500 hours) and horizontal tow type (surface vs. 

 near-bottom). 



Source 



df 



PR 



r2 = 0.75 



A. Dependent variable: 



Log 10 [(Cynoscion arenarius 1^00 m3) + 1] 



Model 21 O.OOOr* 



r^^onth 3 0.0001" 



Depth group 3 0.0001" 



Day-night 1 0.0026" 



Horizontal tow type 1 0.2574 (NS) 



IVIonth vs. Day-night 3 0.0001" 



l\/lonth vs. Depth group 9 0.0001" 



Day-night vs. Tow type 1 0.4180 (NS) 



Error 177 



Corrected Total 1 98 



B. Dependent variable: 



Logio [(Micropogonias undulatusHOO m3) + 1] 



Model 21 0.0001" r2 = o.63 



Month 3 0.0045" 



Depth group 3 0.3551 (NS) 



Day-night 1 0.0001" 



Horizontal tow type 1 0.4448 (NS) 



Month vs. Day-night 3 0.2168 (NS) 



Month vs. Depth group 9 0.0001" 



Day-night vs. Tow type 1 0.1288 (NS) 



Error 177 



Corrected Total 1 98 



C. Dependent variable: 



Logio [(Leiostomus xanthu^us/^00 m3) + 1] 



Model 21 0.0001" /-s = 0.51 



Month 3 0.0001" 



Depth group 3 0.0033" 



Day-night 1 0.1875 (NS) 



Horizontal tow type 1 0.3216 (NS) 



Month vs. Day-night 3 0.2138 (NS) 



Month vs. Depth group 9 0.0001" 



Day-night vs. Tow type 1 0.0324* 



Error 177 



Corrected Total 198 



* = Statistically significant (P < 0.05). 

 " = Highly significant (P < 0.01). 

 (NS) = Not significant. 



133 



