FISHERY BULLETIN: VOL. 87, NO. 1 



METHODS 



Experimental Design 



White perch adults were collected by otter trawl 

 from the Potomac and Patuxent Rivers, MD dur- 

 ing April and May 1982. Eggs were collected from 

 at least four females and milt from four males from 

 each river system. Gametes were stripped into 8 L 

 polycarbonate containers. Fertilized eggs were then 

 transported to the laboratory and placed in well- 

 aerated, 38 L tanks divided into three temperature 

 groups: 13°, 17°, and 21 °C. Salinity was maintained 

 at I'Voo. After hatching, yolk-sac larvae were trans- 

 ferred to 38 L culture tanks. Feeding studies were 

 initiated with larvae that had some yolk remaining 

 and which had pigmented eyes, indicating readiness 

 to initiate feeding (Blaxter 1969). 



All feeding experiments were conducted for a 

 period of 8 days. Partial water changes of 25% were 

 made in each feeding tank every other day to mini- 

 mize buildup of metabolites. Fluorescent lighting 

 provided constant 200-300 lux, with photoperiod 

 maintained on a 13 h light: 11 h dark cycle. Tem- 

 perature was controlled to the nearest 0.5°C by 

 maintaining aquaria in water baths of ambient 

 Patuxent River water and regulating individual 

 tanks by aquarium heaters. 



Food for larvae consisted of rotifers {Brachionus 

 plicatilis) cultured in the laboratory on the green 

 alga Chlorella sp. Field studies demonstrated that 

 Brachionus constitutes the bulk of the diet of first- 

 feeding white perch larvae (Martin and Setzler- 

 Hamilton 1983). Based on a size analysis of zoo- 

 plankton prey consumed by Potomac River larvae, 

 rotifers were graded as follows: Day 1 to day 3: 

 100-150 imi in breadth provided; and day 4 to day 

 8: all sizes (100-180 nm) provided. Food levels were 

 measured in the feeding tanks by calculating the 

 mean values of three 100 mL aliquots taken four 

 times daily. Food concentrations subsequently were 

 adjusted to nominal levels. 



Four food groups were established, representative 

 of high, low, and delayed-high food conditions. 

 Group 1 was a well-fed group maintained at 800 

 rotifers/L; group 2 was maintained at 50 rotifers/L 

 concentrations for 2 days and then fed at 800 

 rotifers/L levels for 6 days; group 3 was fed at 50 

 rotifers/L levels for 4 days and then fed at 800 

 rotifers/L concentrations for 4 days; group 4 was 

 maintained at low levels of 50 rotifers/L for the en- 

 tire study period. The food levels of 800 and 50 

 rotifers/L were representative of high and low 

 microzooplankton levels that typically occur in tidal 



freshwaters of the Chesapeake (Lippson et al. 1980). 



Each food group was tested at three tempera- 

 tures: 13°, 17°, and 21°C. At each temperature, 10 

 eggs and 10 newly hatched larvae were sampled 

 from the rearing tanks and fixed in 4% formalin to 

 test for possible incubation temperature or paren- 

 tal stock effects on egg and newly hatched larva 

 sizes. Just prior to feeding, 10 larvae were removed 

 from each temperature stock tank and preserved for 

 initial length and dry weight measurements. 



At each temperature, 150 larvae were assigned 

 to each of two replicates for each food group (with 

 four food groups per temperature). Once feeding 

 was initiated, at 2 d intervals, subsamples of 3 or 

 4 larvae were removed from each tank and pre- 

 served in 4% formalin for growth analyses. 



Sample Analyses 



Mean egg diameter, larval hatching length, and 

 length at first-feeding were measured and compared 

 among temperatures. Yolk and oil globule dimen- 

 sions of eggs and larvae were measured by ocular 

 micrometer and converted to yolk and oil volumes 

 (mm^); the stage-specific volumes were then com- 

 pared among temperatures. Regressions also were 

 developed predicting the duration of the egg and 

 yolk-sac stages in relation to temperature. 



Expected mean survival after 8 days of feeding 

 was calculated based on the relationship: N, = 

 N^e ~^', where A^, = number of survivors at t days 

 after first-feeding (8 days), A^o = initial number of 

 larvae (150), t = number of days of feeding (8), and 

 Z = instantaneous total mortality rate. Also, Z = 

 F + M, where F = sampling mortality and M = 

 natural mortality rate. The number of larvae pre- 

 served for analyses was considered sampling mor- 

 tality (F), and all other mortality was M. Thus, when 

 Nf), N,, t, Z, and F were known, it was possible to 

 solve for M, from which expected number of sur- 

 vivors was calculated as N, (Expected) = A^o^ ~^' 

 (Ricker 1975). 



Growth rates were calculated from the subsam- 

 ples of preserved larvae. Lengths were measured 

 after three weeks of preservation using a Wild^ 

 dissecting microscope fitted with an ocular microm- 

 eter. Lengths were recorded to the nearest 0.1 mm. 

 Dry weight was obtained by drying larvae at 60°C 

 for 48 hours, dessicating, and weighing to the near- 

 est 0.1 ng on a Cahn electrobalance. Growth in 

 length was estimated by linear regression: L; = a 



^Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



64 



