156 



Fisheiy Bulletin 88(1), 1990 



Few studies have assessed the effect of short-term 

 starvation on the accuracy of age estimates and growth 

 histories of the early life stages of fish derived from 

 the analysis of otolith microstructure. This paper ex- 

 amines the reliability of sagittal microstructure of lar- 

 val Atlantic menhaden to estimate age, detect stressful 

 events, and infer the growth chronology of individual 

 larvae. Two age groups of laboratory-reared larvae 

 were subjected to short periods of starvation and 

 optimal feeding conditions to determine the relation- 

 ship between age, growth history, and microstructural 

 growth patterns in sagittae. 



The Atlantic menhaden is a commercially and eco- 

 logically important species along the Atlantic east coast 

 (Reintjes 1969, Reish et al. 1985). Recruitment of this 

 species has undergone marked fluctuations since moni- 

 toring of the fishery began in the 1940's (Ahrenholz 

 et al. 1987). Investigations of the early life stages of 

 Atlantic menhaden indicate that physical and biological 

 factors operating during larval drift may influence the 

 growth and survival and hence recruitment (Nelson 

 et al. 1977, Checkley et al. 1988). Information derived 

 from these laboratory experiments will serve as a basis 

 for interpreting the microstructure of sagittal otoliths 

 of Atlantic menhaden larvae collected in nature (Maillet 

 1988). 



Methods 



Spawning and rearing conditions 



Fertilized eggs (henceforth called the "stock popula- 

 tion") were obtained from an induced spawn of a cap- 

 tive stock (Hettler 1981, 1983) and placed into three 

 circular tanks containing 60-L of filtered (20 pim) 

 seawater. Eggs and larvae were incubated at 19°C 

 (18.9 ± 0.1°C; X ± SE) in lightly aerated, static water 

 with overhead fluorescent lighting on a 12 L: 12 D 

 photoperiod. Salinity ranged from 29 to 33 g/kg. Furan 

 II (7 mg/L, Aquarium Pharmaceuticals) was added to 

 retard the growth of bacteria and fungi. During early 

 development (first feeding to 12 days postfertilization), 

 larvae were offered cultured algae Nanochhris spp., 

 rotifers 5rr(c/(/o*i'M.s plicatilis, and wild microzooplank- 

 ton (70-250 ^m size range) ad libitum. During later 

 development, larvae were offered only wild microzoo- 

 plankton. Dead larvae and settled plankton were re- 

 moved every 1-3 days. Water level was maintained by 

 removal of seawater each time food was added. 



Reference to trade names does not imply endorsement by tl 

 National Marine Fisheries Service. NOAA. 



Experimental procedures 



Eggs were sampled daily during development and pre- 

 served in 95% ethanol to investigate otolith formation 

 in embryos. A total of 15 eggs were used in the anal- 

 yses. To estimate the time to first increment forma- 

 tion and to test for subsequent daily increment forma- 

 tion, 10 to 15 larvae were sampled from the stock 

 population at 3, 4, 5, 6, 7, 8, 11, 14, 25, 34, and 35 days 

 postfertilization. The effect of starvation on the period- 

 icity of increment formation was investigated by ex- 

 posing two different age groups of larvae (13-20 days, 

 and 28-36 days postfertilization) to short periods wit h 

 no food. For each group, larvae of various sizes were 

 randomly selected from the stock population and trans- 

 ferred to eight 10-L experimental tanks filled with 

 20-Mm filtered seawater. Initial densities of larvae in 

 experimental tanks were 5 larvae/L (13-20 days) and 

 4 larvae/L (28-36 days). Environmental conditions, in- 

 cluding illumination, water temperature, salinity, and 

 the use of Furan II, were identical to those of the stock 

 tanks. Four experimental groups, consisting of (a) con- 

 tinuous feeding (controls), (b) 1-day starved, (c) 2-day 

 starved, and (d) 3-day starved treatments, were ran- 

 domly assigned to duplicate experimental tanks. Con- 

 trol larvae were fed wild microzooplankton ad libitum 

 immediately after transfer; treatment larvae received 

 food at the end of the respective starvation interval and 

 were allowed to continue feeding for several days after 

 this period. 



All larvae sampled from the stock and experimental 

 containers were first anesthetized with tricaine me- 

 thanesulfonate (Cresent Research Chemical), measured 

 for standard length (tip of upper jaw to end of noto- 

 chord) to the nearest 0.1 mm, and then preserved in 

 95 % ethanol. Larvae were stored individually in 10-mL 

 vials. The preservative was changed once after 48 

 hours to maintain sagittae in optimal condition. 



Otolith preparation and analysis 



Sagittae were examined within two months after pres- 

 ervation. The right sagitta was teased from the inner 

 ear with minuten needles, cleaned of excess tissue, and 

 mounted medial-side-up in Flo-Texx (Lerner Labora- 

 tories). Specimens were examined with transmitted 

 light under a compound microscope fitted with a lOOx 

 objective and a video camera and monitor, thereby in- 

 creasing the total magnification (monitor image/actual 

 size) to 3600 X. An electronic caliper was used to 

 measure growth increments on the video monitor. This 

 system allowed electronic enhancement of otolith 

 images (e.g., contrast between the incremental and 

 discontinuous zones). 



Increment counts were made in triplicate on masked 

 (i.e., of unknown origin) samples, to minimize bias, and 



