Kurita et ai.: The effect of temperature on the duration of spawning markers in Pciralichthys olivaceus 
83 
Figure t 
Histological and whole mount images of oocytes and postovulatory follicles (POFs) of Japanese 
flounder ( Paralichthys olivaceus). (A, B) Oocyte at yolk granule stage; diameter ca. 500 pm. 
(C, D) Early migratory nucleus stage; diameter ca. 600 pm. (E, F) Late migratory nucleus stage; 
diameter ca. 650 pm. (G, H) Early hydrated stage; diameter ca. 720 pm. (I, J) Late hydrated stage; 
diameter ca. 900 pm. (K) New POF. Scale bars = 100 pm. See Table 1 for detailed definitions. o = oil 
droplet; n=nucleus. 
set at 4.5% which was the same proportion as that for 
the HD-stage oocytes. 
The lower size limit of hydrated oocytes at a given 
temperature was considered equal to the maximum 
average size of MN(L)-stage oocytes among the suc- 
cessive 48-h samplings (Table 2) at that temperature 
because the growth rate of MN(L)-stage oocytes was 
very slow and the maximum average size of MN(L)- 
stage oocytes can be considered as the smallest diam- 
eter for the hydrated oocytes (see Fig. 3). The average 
egg diameter was considered as the maximum diam- 
eter for the hydrated oocytes. The lower and upper 
limits of hydrated oocyte diameters were obtained 
from six (9.2-19.7°C) and seven (9.2-22.6°C) time se- 
ries samples, respectively. Changes in those diameters 
were analyzed in relation to temperature. 
Average diameters of oocytes at the MN(E) stage 
and the leading cohort of oocytes at the yolk granule 
stage (YG[LC]> were also calculated for each sam- 
ple. The proportions of the number of oocytes at the 
MN(E) and the YG(LC) stages to the developing oo- 
cytes were determined in the same manner as that 
for MN(L) stage (Fig. 2). The leading cohort of YG- 
stage oocytes is usually defined as the largest 10% of 
YG-stage oocytes (Kjesbu, 1994) to specify the matu- 
rity phase. In this study, however, the proportion of 
YG(LC)-stage oocytes to the developing oocytes was 
set at the same proportion of HD-stage oocytes (e.g., 
4.5 % in Fig. 2) to correspond to the expected batch 
of oocytes. 
Body size and individual effects on the growth rate 
of hydrated oocytes 
To evaluate size effect on the growth rate of hydrated 
oocytes, multiple regression analysis was conducted 
for 20 data sets of log-transformed growth rates of 
hydrated oocytes (dependent variable), temperature, 
and total length (independent variables). In addition, 
to evaluate the individual effect on the growth rate 
of hydrated oocytes, the relationship between the log- 
transformed growth rate of hydrated oocytes and tem- 
perature (T) was compared for two females (fish no. 
2; TL 710 mm, 11.3<T<16.7°C, n = 6: fish no. 3; TL 707 
