Allman and Grimes Spawning, settlement, and growth of Lutjanus griseus from tfie West Florida shielf 



393 



was made from the anterior edge to the posterior edge 

 along the longest axis. 



We computed the hnear regression of standard length 

 on lapillus length to determine if otolith growth and so- 

 matic growth were proportional. To prepare lapilli for age 

 determination, both lateral surfaces of the otolith were 

 ground and polished on glass plates covered with a 3600- 

 and 6000-grit polishing cloth. After being polished, the 

 lapillus was mounted on a microscope slide with Pro-texx® 

 mounting medium. Thirteen sagittae were also prepared 

 to determine if sagittal section counts were consistent 

 with lapilli counts. Because of the concave-convex shape of 

 the sagittae, the technique used for examining the lapillus 

 was not possible. Therefore, the more labor intensive pro- 

 cess of embedding and sectioning the whole sagittae had 

 to be used (Secor et al., 1992). 



Growth in mm/day was estimated as the slope of the lin- 

 ear regression of standard length on the number of daily 

 increments in the lapillus. Analysis of variance (ANOVA) 

 and analysis of covariance (ANCOVA) were used to deter- 

 mine if estimated growth rates varied significantly among 

 sampling regions and years. 



Increment counts were made from sections of the la- 

 pilli and sagittae by using a compound microscope at 

 400-lOOOx magnification. Oil immersion was used at the 

 lOOOx magnification. Daily increments were distinguished 

 from subdaily increments with the method of Campana 

 (1992). Increments were counted twice by the same reader 

 and counts that were different by no greater than 5% were 

 averaged. Increment counts that differed by more than 5% 

 were counted a third time, and if the third count differed 

 by more than 57? of the previous counts, the otolith was 

 rejected. An increment correction (3 days) was added to 

 the total number counted based on the assumption that 

 the first increment was not formed until after first feeding 

 (i.e. approximately 3 days after fertilization) (Lindeman, 

 1997; Lindeman et al., 2000). This corrected increment 

 count was subtracted from capture date to determine the 

 fertilization date. 



We interpreted the settlement mark in the lapillus to 

 be where the pattern in the increment widths changed 

 markedly (Wellington and Victor, 1989). Increments were 

 counted from the primordium to the settlement mark and 

 a correction of three increments was added to estimate the 

 number of days before settlement or the planktonic larval 

 duration. Those lapilli that did not have an obvious settle- 

 ment mark were excluded from the analysis. Settlement 

 dates were calculated by subtracting the number of post- 

 settlement increments from the date of capture. ANOVA 

 was used to compare the age at settlement (i.e. planktonic 

 larval duration) with region, year, and region-year inter- 

 action. After converting lunar day of fertilization or settle- 

 ment to a circular scale (Zar, 1984), we used a chi-square 

 test to determine if fertilization and settlement dates were 

 uniformly distributed across the lunar month. 



To validate the periodicity of increment formation in la- 

 pilli we conducted an otolith marking experiment. Lapilli of 

 live fish were marked with alizarin complexone, which has 

 been shown to produce a well-defined mark in the otoliths 

 of juvenile fish without producing high mortality (Tsuka- 



moto et al., 1989; Lang and Buxton, 1993). Approximately 

 30 juveniles were captured with a benthic trawl and held 

 in an 800-gallon flow-through seawater system for several 

 days so that they might acclimate to captivity. Juveniles 

 were fed once daily ad libitum. Because of the photosensi- 

 tivity of alizarin complexone, juveniles were immersed in 

 20 L of an aerated 200 mg/L solution of alizarin complex- 

 one for 24 h in the dark. Juveniles were removed from the 

 solution and returned to the flow-through seawater tank. 

 Juveniles were sacrificed 7, 14, 21, or 28 days after marking 

 and their lapilli were removed and prepared as previously 

 described. Increments deposited after the alizarin complex- 

 one mark were counted as previously described, except that 

 transmitted ultraviolet light was used to read increments. 

 Linear regression was used to compare number of incre- 

 ments counted after the alizarin-complexone mark in the 

 lapillus to the number of days the juveniles were held after 

 marking. A /-test was used to determine if the slope of this 

 regression was significantly different from 1.0. 



Gonad histology and analysis 



To obtain an estimate of the spawning time that was inde- 

 pendent of otolith-based spawning dates, adult gray snap- 

 per gonads were examined. All adult gray snapper were 

 collected as part of an ongoing study by the National Marine 

 Fisheries Service on reef fish reproduction. Fish were col- 

 lected from commercial and recreational landings from 

 Panama City, Florida, to Ft. Myers, Florida, from January 

 to December in 1996 and from April to November in 1997. 

 Each fish was weighed whole (i.e. ungutted [g|) and total 

 length (TL) and fork length were measured (mm). All fish 

 collected were at least 252 mm TL (minimum legal size) and 

 were assumed to be sexually mature (Domeier et al., 1996). 

 Gonads were removed and stored on ice and then pro- 

 cessed by the methods of Collins et al. (1996). A small 

 sample of each gonad was examined with a dissecting mi- 

 croscope (250x) to measure the maximum oocyte diame- 

 ter (nearest 0.1 mm). The developmental stage of ovarian 

 sections was determined by using the methods of Wal- 

 lace and Selman (1981). Ovarian stages were assigned 

 on the basis of the most advanced ovarian stage or fol- 

 licle stage present: 1 — primary growth (early oocytes); 

 2 — cortical alveolar (previtellogenic); 3 — vitellogenic; 4 — 

 hydrated; and 5 — spent (i.e. presence of postovulatory fol- 

 licles). Sections of testes were staged according to the 

 methods of Moe (1969): 1 — spermatogonia; 2 — primary 

 spermatocytes; 3 — secondary spermatocytes; 4 — sperma- 

 tids; and 5 — spermatozoans. The timing and duration of 

 spawning were determined by plotting oocyte diameter 

 and histological development stage by sampling date. 



Results 



Sampling and collection 



Juvenile gray snapper were found in seagrass meadows 

 along the west Florida shelf from June to November in 

 the two most northern sampling regions, and from April 



