Hunter et al.: Fecundity, spawning, and maturity of Microstomus paaficus 



I 19 



Figure 8 



Relative frequency distribution of the fraction of the random 

 sample of advanced yolked oocytes measured that were atretic {a 

 stage) from the Dover sole Microstomus pacifieus females used in 

 estimates of total fecundity shown by state: Oregon females. A' 189; 

 California females, N 361. 



and Pikitch (1989) for females caught along the Oregon 

 coast. Those authors used an exponential model and 

 expressed annual fecundity as a function of length. The 

 distribution of points in our data broadly overlapped 

 the data of Yoklavich and Pikitch (1989) (Fig. 7). To 

 compare our Oregon data with theirs, we truncated 

 ours so that the length ranges of the two sets coincided 

 and applied an analysis of covariance to log-trans- 

 formed data. Analysis of covariance indicated that no 

 significant difference existed between the two equa- 

 tions (Fi 81 2.03, P 0.158). In Table 10, the exponen- 

 tial equation for fecundity as a function of length is 

 given for our data (not truncated) and for that of 

 Yoklavich and Pikitch (1989). In summary, we found 

 no statistical difference between California and 

 Oregon, nor between our Oregon data and that of 

 Yoklavich and Pikitch (1989). 



Atretic losses 



In fishes with determinate fecundity, a 

 key question is whether atretic losses dur- 

 ing a season constitute an important frac- 

 tion of the potential annual fecundity. We 

 identified whole atretic oocytes under a 

 microscope while doing our fecundity 

 work. To measure atretic losses, we 

 counted the number of atretic oocjrtes (a 

 advanced yolked oocytes) occurring in a 

 random sample of 30 advanced yolked 

 oocytes for each of the females used to 

 estimate total fecundity (N 550). 



In the fish used to estimate fecundity, 

 the average fraction of advanced yolked 

 oocytes that were atretic was low with 

 the mean 0.015 (SD 0.032, N 361) in Cali- 

 fornia, and even lower in Oregon females 

 (mean 0.0033, SD 0.014, A^ 189). Atretic 

 oocytes were observed in only 26% of 

 California females and in only 6% of 

 Oregon females (Fig. 8). 



The total fecundity of California fe- 

 males was negatively correlated with the 

 fraction of oocytes in the ovary that were 

 atretic. A stepwise multiple regression of 

 female weight, elapsed time, and fraction 

 atretic on total fecundity (Table 11) indi- 

 cated that the coefficient for the fraction 



Table 1 1 



Analysis of the relation between total fecundity (Yp) of Dover sole Micro- 

 stomus pacificus ovary-free body weight, elapsed days since 1 November, and 

 fraction of atretic oocytes using stepwise regression. Specimens from California. 



Source 



DF 



Analysis of variance 



SS MS 



Regression 



Error 



Total 



3 

 357 

 360 



7.58x10'" 

 1.13x10" 

 1.89x10" 



2.53 xlC" 

 3.16x10' 



79.99 < 0.001 



Source 



DF Sequential SS 



Weight 

 Elapsed days 

 Fraction atretic 



6.28x10'" 

 1.15x10'" 

 1.49x10' 



*ForP = 0.05, t 1.97. 



