422 



Fishery Bulletin 92|2), 1994 



water fished by the net ranged from 41 to 805 m 3 but 

 most sets filtered 150-350 m 3 . Ichthyoplankton 

 samples were preserved with 95% ethanol and di- 

 luted so that the final alcohol concentration was 

 >70%. Samples not sorted within 24 hours were 

 rinsed and represerved in 70% ethanol. As for bongo 

 nets, these catches were standardized as the num- 

 ber of Atlantic menhaden/100 m 3 of water fished. The 

 mean of the density data from all sets on a given 

 night (bongo nets in 1985—86 and neuston nets in other 

 years) was used as the estimate of density of Atlantic 

 menhaden larvae recruited during the flood tide. 



Simultaneous larval collections were made with 

 bongo and neuston nets on 17 December 1986 (3 sets) 

 and 18 March 1987 (4 sets) to test for differences in 

 menhaden catch density between gear type. There 

 were no significant differences in density of menha- 

 den caught by gear type (AN OVA, P>0.54) or among 

 sets (ANOVA, P>0.24). Hence, the catch data from 

 bongo and neuston nets are comparable. The mean 

 age or standard length (SL) also did not differ be- 

 tween the two types of nets. 



Larval ageing 



Larvae for ageing were subsampled from individual 

 weekly night net sets in proportion to their contri- 

 bution to the total nightly catch. In catches of up to 

 20 fish, all larvae were used. In catches of >20 fish, 

 subsample sizes were proportional to catch but gen- 

 erally no more than 50 fish were aged per week. The 

 ages of 3,864 larvae were determined. 



The standard length of each larva to be aged was 

 measured to the nearest 0.1 mm with an ocular mi- 

 crometer. Sagittal otoliths were removed from their 

 surrounding soft tissue, cleaned in distilled water, 

 and placed on a glass microscope slide under a thin 

 layer of Flo-Texx mounting medium. Otoliths were 

 observed with transmitted light on a compound mi- 

 croscope fitted with a television camera. Growth in- 

 crements were counted from images on a video moni- 

 tor at microscope magnifications of 400x or l,000x. 

 One person made dual readings of otoliths from each 

 fish. Readings were averaged if they differed by a 

 count of four or less; if they differed by >5 increments, 

 the fish was excluded from further analyses. Incre- 

 ment counts of about 2% of the aged fish differed by 

 five or more. Estimated age was the number of in- 

 crements counted plus an empirically derived value 

 for the number of days (5) from spawning to first 

 increment formation ( Warlen, 1992). The otolith age- 

 ing technique for Atlantic menhaden larvae has been 

 validated by Maillet and Checkley (1990), who es- 

 tablished that larvae form one otolith growth incre- 

 ment per day. I assumed that the age at initial incre- 

 ment deposition did not vary and that the otolith 



increment deposition rate was constant within and 

 between sampling seasons. A spawning date 

 (birthdate) was assigned to each ageable larva by 

 using the estimated age of the fish in days to back- 

 calculate from the date of capture. Larvae spawned 

 in a given calendar week were considered in the same 

 calendar birthweek cohort. 



Estuarine recruitment of birthweek cohorts 



The percentage contribution of Atlantic menhaden 

 larvae of all birthweek cohorts to the total recruit- 

 ment was measured for each of the seven years from 

 1985-86 to 1991-92. For each weekly collection, the 

 percentage of larvae spawned in each back- calcu- 

 lated birthweek was determined. Each percentage 

 was then multiplied by the total larval density (num- 

 ber/100 m 3 ) for the week. Based on these results, den- 

 sity estimates were made for larvae from each 

 birthweek cohort recruiting to the estuary on a given 

 collection night. 



Densities for each birthweek cohort were summed 

 over all collections within a sampling year. The pro- 

 portion that each birthweek cohort contributed was 

 determined by dividing the individual birthweek 

 sums by the total density of all birthweek cohorts 

 for the recruitment year. These computations pro- 

 duced estimates of the relative contribution that each 

 birthweek cohort made to the total recruitment of 

 menhaden larvae into the estuary near Beaufort. 



Results 



Larval recruitment abundances 



Larval Atlantic menhaden recruited to the estuary 

 near Beaufort over a 5 to 5V2 month period from mid- 

 November to the end of April (Fig. 2). Larvae were 

 generally recruited in highest densities during Feb- 

 ruary-April and contained, on average, about 86% 

 of the total estuarine recruitment during those 

 months (Table 1). Recruitment was low in Novem- 

 ber and seldom extended into May. There was a dis- 

 tinct density mode in each of the seven years (Fig. 

 2). Except for 1988-89, modal density always oc- 

 curred in March or April (Fig. 2, Table 2). Highest 

 densities in a 3-week period (modal week plus ad- 

 joining weeks) in any year contributed 32-89% of the 

 total density of Atlantic menhaden larvae for the year 

 (Table 2). 



During the recruitment period, most Atlantic men- 

 haden larvae recruited to the estuary in pulses. Mean 

 density varied from zero to about 570 larvae/100 m 3 

 and most samples contained <20 larvae/100 m 3 (Figs. 

 2 and 3). Catch densities varied among sets on any 

 given night and the standard deviation in catch den- 

 sities generally increased with the mean catch den- 



