1000 



Fishery Bulletin 97(4), 1999 



estuarine spawning ground during winter and spring 

 1997. 



Methods 



Laboratory observations 



Experrmental animals Winter flounder were col- 

 lected in nearshore waters between Rockaway Point 

 and Coney Island, New York, on 27 January 1997 

 with a 9.2-m otter trawl fished from RV Gloria 

 Michelle. Flounder were held in live cars with flow- 

 through seawater and transported in large ice chests 

 to the laboratory. Mature ripe males were identified 

 by the extrusion of sperm under light pressure ap- 

 plied to the body Females were identified by the pres- 

 ence of full ovaries distending the body, occupying a 

 large portion of the body cavity, and extending al- 

 most to the caudal peduncle. Ten males and ten fe- 

 males, each greater than 200 mm in TL (total length), 

 were marked with Peterson tags (13 mm diameter) 

 before release into the research aquarium. To pro- 

 vide visual identification of gender on videotape, 

 males and females were marked with orange and 

 white tags, respectively. These colors were distin- 

 guishable with low-light video cameras during both 

 day and night. At the beginning of the experiment 

 the mean total length of male flounders was 29.8 cm 

 (SD=3.2, range=24.3-34.4 cm). Mean length of fe- 

 males was 28.5 cm (SD=2.6, range=23. 5-30.8 cm). 

 Four fish (two males and two females) died near the 

 end of the 14-week observational period; these were 

 not replaced. The fish were remeasured for total 

 length at the end of the experiment to estimate 

 growth. 



Research aquarium Observations of the winter 

 flounder were made in the research aquarium at the 

 James J. Howard Marine Sciences Laboratory at 

 Sandy Hook, New Jersey. The aquarium (121 kL vol- 

 ume) is an oval 10.6 m long, 4.5 m wide, and 3 m 

 deep, with eight rectangular windows (0.7 m wide 

 and 1.2 m high), one in each end and three along 

 each side. There is a recirculating system so that 10*)^ 

 of the water is replaced each week. Seawater upwells 

 through coarse sand which covers the bottom of the 

 aquarium (46 m^) to a depth of 40 cm, then exits 

 through drains at the top of the tank. 



For this experiment, photoperiod in the aquarium 

 room, which was controlled 'by a computer-driven 

 bank of fluorescent lamps (Duro-Test, Vita-Lite: 

 model CRI91; 5500°K), was programmed to follow 

 the natural sequence at the location of the labora- 

 tory (40'='28'N, 74°00'Wi, with a 13.5-h dark period 



at the end of January, when fish were first introduced 

 into the system, and a 10-h dark period at the end of 

 the study in mid-May. During nighttime hours, light 

 intensity on the sediment surface was 0.17 lux (1.98 

 X 10"'' niEinsteins PAR [photosynthetic active radia- 

 tion]). Light intensity increased at the time of twi- 

 light to 2.0 lux (2.92 X 10-5 mEinsteins PAR) for 20 

 min, followed by a natural rise in intensity to 206 lux 

 (0.0027 mEinsteins PAR) at midday. Return to night- 

 time conditions followed an exact reverse pattern. 



Water temperature was held at 4.0°C (±0.5°C) from 

 the beginning of the observations until 27 March, 

 when a cooling system malfunction allowed tempera- 

 ture to rise to 7.5'C. To follow conditions in the 

 Navesink River, we held water temperature at 7.5°C 

 until 10 April, then raised temperature ~0.5°C per 

 day to 15°C on 14 April. This temperature was main- 

 tained until the end of observations on 13 May 1997. 



Finely chopped pieces of frozen surf clam iSpisiila 

 solidissima} were made continuously available as 

 food for the flounders. The food was renewed at least 

 once per week during the early part of the period 

 when feeding was light, and increased with inges- 

 tion rate to near-daily additions by 1 April 1997. The 

 food was placed on the bottom in an area approxi- 

 mately 2 m in diameter within camera view so that 

 all feeding bouts could be counted. 



Recording observations Two low-light-sensitive 

 video-tape cameras were set at the end windows of 

 the aquarium to make continuous recordings of court- 

 ship and spawning events over a large portion of the 

 tank from bottom to water surface. Each of these 

 cameras viewed 53'7f (24.6 m-^) of the tank bottom 

 with minimal overlap in coverage. Recordings were 

 made from about 1530 h through 0500 h between 29 

 January and 18 April 1997, after preliminary taping 

 and routine visual observation indicated that virtu- 

 ally all activity was nocturnal. The validity of this 

 approach was confirmed by cameras at the side win- 

 dows set for 24-h recording throughout the study 

 period. Twenty-four-hour recording from the end 

 cameras was initiated 19 April 1997, concurrent with 

 increasing daytime activity of the fish. 



Spawning events were divided into two classes. An 

 event was considered "definite" when characteristic 

 behavior patterns were clearly observed in the re- 

 corded image. However, because most spawning oc- 

 curred in darkness, some spawning events were not 

 observed clearly. These likely but less certain spawn- 

 ing events were scored "probable." Two forms of court- 

 ship were recorded; these are described in the "Re- 

 sults" section. Records were kept of the numbers of 

 males and females involved in each courtship and 

 spawning encounter. A feeding event was scored once, 



