120 



Fishery Bulletin 91(1), 1993 



NC in the summers of 1983 and 1985. Menhaden were 

 held at the laboratory in Beaufort at ambient tem- 

 peratures until water temperatures fell below 20° C. 

 Thereafter, temperatures were maintained at ca. 20°C. 

 Ten menhaden were induced to spawn at each spawn- 

 ing (methods followed Hettler 1981, 1983). Generally, 

 spawning occurred in 20° C water at night, and eggs 

 were collected in the morning. Eggs were held either 

 in 10 L black-sided tanks with static water immersed 

 in a temperature-controlled water bath or in 100 L 

 black-sided rearing tanks with static water. Tempera- 

 tures in the 100 L tanks varied with slight changes in 

 room temperature ( 19-20° C). Salinities ranged from 

 approximately 30 to 35^. Two 40W fluorescent lamps 

 were placed 90 cm above each large rearing tank and 

 40 cm above each small tank. The tanks were il- 

 luminated for a 12 h dark:12h light cycle. Rotifers 

 Brachionus plicatilis were the sole source of food. Ex- 

 perimental temperatures were 16°, 20°, and 24° C. 

 These temperatures were chosen because 20° C approxi- 

 mates the average temperature of most intensive 

 spawning for both species (Shaw et al. 1985, Checkley 

 et al. 1988). The 16° and 24° C temperatures were ar- 

 bitrarily chosen to bracket the 20° C temperature and 

 were in accord with a previous study (Powell & Phonlor 

 1986). 



Eggs 



Egg, yolk, and oil globule diameters were measured on 

 approximately 50 live eggs per spawning for each spe- 

 cies. Yolk and oil diameters were used to calculate 

 volumes. Because the spheroidal oil globule is embed- 

 ded in the prolate spheroidal yolk mass, oil volumes 

 were subtracted from initial yolk volume estimates to 

 give a more accurate measurement of yolk volume. 



Relationship between egg size and latitude was in- 

 vestigated by using collections of preserved (59c for- 

 malin) Atlantic menhaden eggs from eight cruises dur- 

 ing 1979-81 by the Sandy Hook Laboratory, NMFS 

 Northeast Fisheries Science Center, Highlands, NJ. 



Larvae 



Starvation experiments were conducted at 16°, 20°, 

 and 24° C. Approximately 20-25 larvae were removed 

 from the rearing tanks and randomly placed in each 

 1L fingerbowl at each experimental temperature on the 

 day prior to initiation of feeding by the larvae. Food 

 was withheld from all larvae. Mortalities were recorded 

 at 24 h intervals. Results from repeated independent 

 experiments (n=6 and 3 for Atlantic and gulf menha- 

 den, respectively) were averaged. 



Mean standard lengths (SLi and mean dry weights 

 of menhaden larvae at hatching and at first-feeding 



were determined from larvae reared at 16°, 20°, and 

 24° C in 10 L tanks. Mean lengths and mean dry 

 weights were obtained from approximately 10 preserved 

 larvae (5% buffered formalin) per experiment. To de- 

 termine when first-feeding occurred, 10 larvae at each 

 experimental temperature were removed daily, pre- 

 served, and inspected for the presence or absence of 

 rotifers. The presence of rotifers in the guts of any of 

 the 10 larvae was considered to be first-feeding. Ob- 

 servations on development were also made (e.g., fully- 

 pigmented eyes, functional mouth, and differentiated 

 foregut and midgut). All observations were made at 

 50 x magnification. 



Mean rate of yolk absorption was determined from 

 approximately 10 larvae sampled daily at 16°, 20°, 

 and 24°C, from time of hatching until the time when 

 yolk and oil were completely used. Because preserva- 

 tion had a significant effect (P<0.01, n-50) on size and 

 condition of the yolksac (preserved yolk diameters were 

 smaller than those measured live), measurements were 

 made only on live material. Yolk volumes were calcu- 

 lated using the formula for a prolate spheroid, 



V = (7t/6)lh 2 , 



where 1 is length and h is the height of the yolk mass 

 (Blaxter & Hempel 1963). As with the measurement of 

 eggs, oil volumes were calculated and subtracted from 

 the calculated yolk volumes to give a better estimate 

 of yolk volume. 



Two series of growth experiments were conducted 

 on each species. One examined the relationship be- 

 tween growth and survival of menhaden larvae in re- 

 lation to temperature and food density. Groups of 50 

 larvae, approximately 6d old, were removed from 100 L 

 rearing tanks and transferred to each 10 L experimen- 

 tal tank. The larvae were slowly acclimated (~4h) to 

 16°, 20°, or 24° C. Food levels were designed to reflect 

 high, moderate, and low food availabilities. Food lev- 

 els the first year ( 1984) were 50, 25, and 5 rotifers/mL, 

 and two replicates were performed at each food level 

 for each species. These food levels did not limit growth. 

 Food levels of 10.0, 1.0, 0.1, and 0.025 rotifers/mL were 

 employed the following years ( 1985 and 1986) and three 

 replicates were performed. Over the course of a 7d 

 experiment, food densities were monitored daily and 

 adjusted on the basis of the average of three samples 

 from each rearing tank. Growth was determined as 

 the difference between mean size of the preserved sur- 

 vivors at the end of the experiment and mean size of 

 50 preserved larvae removed from the 100 L tank at 

 the beginning of the experiment. Larvae for each tank 

 were combined, dry weights obtained, and values ex- 

 pressed as mean dry weight. The gain in biomass over 

 a 7d period was calculated to weight the number of 



