168 



Fishery Bulletin 9! |l). 1993 



Figure 3 



Appearance of weakfish Cynoscion regalis hydrated oocytes after 6-7 mo preservation in 

 2% formalin. Bar=l mm. 



Discussion 



not significantly different (F=0.0027, N=10, P=0.14) 

 from those estimated from samples preserved for both 

 3-4 and 6-7 mo (Table 1). Batch fecundities esti- 

 mated from fresh and preserved samples could be com- 

 pared because no positional effects were found between 

 counts from different areas (ANOVA, F=0.91, N=28, 

 P=0.41). 



There is need for a reliable 

 method of separating and pre- 

 serving fish oocytes. Generally, 

 separated and preserved oocytes 

 are used to estimate oocyte size 

 and develop oocyte size-frequency 

 distributions, as well as in the 

 hydrated oocyte method of de- 

 termining batch fecundity. Re- 

 searchers often want to evaluate 

 changes in egg size over the 

 spawning season, or between 

 spawning seasons (e.g., DeMar- 

 tini 1990), and oocyte size- 

 frequencies are used to assess 

 whether fish have determinate or 

 indeterminate fecundity (Hunter 

 & Macewicz 1985). The hydrated 

 oocyte method appears to be the 

 easiest and most accurate way 

 to determine batch size in serial 

 spawners (Hunter et al. 1985). 

 All of these types of analyses are 

 integral to reproductive studies 

 of multiple-spawning fishes, and yet their accuracy de- 

 pends on using either fresh oocyte samples or oocyte 

 samples separated and preserved in a reliable fashion. 

 Although still widely used, researchers are begin- 

 ning to recognize a number of problems with the use 

 of Gilson's solution as a preservative. It degenerates 

 hydrated oocytes (Hunter et al. 1985, Schaefer 1987, 

 Brown-Peterson et al. 1988), making it impossible to 

 use the hydrated oocyte method to estimate batch fe- 

 cundity (Hunter et al. 1985). It causes a high rate of 



