Brouwer and Griffiths: Reproductive biology of Argyrozona argyrozona 



261 



ing fraction (Hunter and Macewicz, 1985). POFs were 

 aged by comparing them with known age POFs from 

 spawning females under captive conditions. Female 

 carpenter were held in a flow-through system at ambi- 

 ent sea temperature (mean 16C, range 9.5-20°C) in 

 5000-liter circular tanks, were stimulated to ovulate 

 with a commercially available GnRH-analogue (Davis, 

 1996). Three fish were sacrificed immediately after 

 ovulation and then three fish every six hours over the 

 following 48-hour period. Histological analysis of ova- 

 ries revealed three clearly defined POF stages (Fig. 2). 

 The proportions of wild-caught fish with stage-1 POFs 

 (the spawning fraction) were inverted to produce an 

 estimate of spawning frequency (Wilson and Nieland, 

 1994). 



Batch fecundity was estimated from counts of hy- 

 drated oocytes from ovaries without POFs or atretic 

 oocytes (Hunter and Macewicz, 1985). A ±1.00-g section 

 was removed from the middle of the right ovary. This 

 was weighed to the nearest 0.01 g and the hydrated oo- 

 cytes were separated according to the method described 

 by Lowerre-Barbieri and Barbieri (1993). Hydrated 

 oocytes were suspended in water and counted at 8-10 

 times magnification in a Bokkeroff tray and measured 

 to 0.1 mm with an ocular micrometer along the longest 

 diameter. 



Annual fecundity was calculated as follows: 



Af t 





xfb t , 



where Af t = the annual fecundity for fish t; 



Is = the length of the spawning season (days) 



for fish of size class j; 

 sf = the spawning frequency (days) for fish of 

 size class j (all months combined); and 

 fb t = the batch fecundity of fish t. 



Spawning season was established by calculating the 

 monthly proportions of macroscopic gonad stages and 

 mean monthly gonodosomatic index (GSI) for fish larger 

 than L 50 : 



GSI- 



xlOO, 



where m = the gonad mass (g); and 



m s = the somatic mass (g) (minus gonad and 

 stomach mass). 



In order to investigate the relationship between 

 spawning and temperature, temperature data were 

 collected at the sampling site with a Seamon Mini (Hu- 

 grun, Iceland) recorder stationed at at a depth of 35 m 

 on the reef from which the biological samples were col- 

 lected. A thermistor array consisting of four underwater 

 temperature recorders (UTRs) at depths of 12 m, 19 m, 

 27 m, and 35 m recorded the temperature every minute 



Figure 2 



Postovulatory follicle (POF) stages deter- 

 mined from carpenter [Argyrozona argyro- 

 zona) chemically induced to spawn in an open 

 circulating system housed at the Tsitsikamma 

 National Park. (A) = stage 1 (0-6 hours), 

 ( B ) = stage 2(7-24) hours and ( C I = stage 3 

 125-48) hours. 



