Yoneda et ai.: Reproductive cycle, fecundity, and seasonal distributions of Lophius litulon 
359 
Table 2 
Summary of results of oocyte density (the number of secondary yolk stage oocytes per unit sam 
female Lophius litulon from various locations and results of two-way analysis of variance, int. 
pie weight [g] ) 
= interior; ext. 
in the ovaries of five 
= exterior. 
TL 
(mm) 
Position of sample in ovary* 
Right ovary 
Left ovary 
Anterior 
(int. or ext.) 
Middle 
(center) 
Posterior 
(int. or ext.) 
Anterior 
(int. or ext. ) 
Middle 
(center) 
Posterior 
(int. or ext.) 
546 
28.32 ext 
2711 
2802 ext 
2768' nt 
2960 
3000 int 
647 
3008 ext 
3113 
3067 ,nt 
3059 int 
3160 
2980"“ 
662 
2917 ext 
3009 
2759 int 
2913 lnt 
2635 
3216"“ 
796 
2540 int 
2771 
2691 ext 
2662 ext 
2990 
2780"“ 
847 
2607 int 
2572 
2821 ,nt 
2680 ext 
2800 
2800 exl 
Mean 
2781 
2835 
2828 
2816 
2909 
2955 
Two-way analysis of variance 
Source of variation 
df 
SS 
MS 
F 
Right vs. left ovary 
1 
23130 
23130 
0.73 
Position within ovary 
2 
105500 
52730 
1.66 
Interaction 
2 
48140 
24070 
0.76 
Error 
24 
764700 
31860 
imens collected in the spawning season, 410 specimens 
(187 males and 223 females) were used to estimate age at 
sexual maturity. The vertebral centra of the other 72 spec- 
imens were either damaged in preparation or lost. To esti- 
mate the total length (L 50 ) and age at which 50% of males 
and females are sexually mature, the fraction of mature 
fish in each interval ( 10-mm length or year of age) was fit- 
ted with a logistic function with the Marquardt method 
(Draper and Smith, 1966). 
All specimens TL > L 50 were used to determine the 
monthly changes in gonadosomatic index (GSI) and hepa- 
tosomatic index (HIS) for adult males and females. The 
GSI and HSI were calculated in the following manner: 
GSI = (GW / (BW - VW)) x 100. 
HSI = {LW KBW- VW))x 100. 
The Kruskal-Wallis test (one-way analysis of variance, 
ANOVA) followed by Dunn’s multiple comparison test 
were used to test for significant differences between the 
GSI and HSI values of gonadal stage groups of fish. 
Estimation of batch fecundity followed Yoneda et al. 
(1998a). Batch fecundity was estimated only from spec- 
imens with ovaries containing oocytes in the secondary 
yolk stage and no gelatinous material. Only 15 females 
contained such ovaries during our study; these fish were 
thought to be ready to spawn for the first time during 
that spawning season because their ovaries contained nei- 
ther postovulatory follicles nor atretic oocytes. To deter- 
mine whether secondary yolk stage oocytes were randomly 
distributed throughout the ovary densities (no. oocytes/g 
ovary wt.) of secondary yolk stage oocytes from the six 
locations within the ovaries of five fish were compared (Ta- 
ble 2): two samples from the center of the right and left 
ovarian lobes, two samples from the posterior section (ei- 
ther from the interior or exterior) and two samples from 
the anterior section (either from the interior or exterior). 
A two-way ANOVA was performed to test for the effect of 
sample location on oocyte density within each ovary. There 
was no significant change in oocyte density by location 
of oocytes within the ovaries (Table 2). Advanced yolked 
(secondary yolk stage) oocytes were randomly distributed 
within the ovary and samples could be taken from any lo- 
cation without bias. Fecundity samples were collected from 
six different parts of the ovary in the anterior, middle, and 
posterior portion of each ovarian lobe. Ovarian tissue sam- 
ples (30-120 mg), each containing approximately 100-350 
oocytes, were placed on a slide in water and covered with a 
cover slip. The most advanced oocytes were counted with a 
profile projector (50-100x). Batch fecundity for each female 
was calculated as the product of the number of secondary 
yolk stage oocytes per unit of weight (of each of the six sam- 
pling sites) multiplied by the total ovarian weight. Linear 
regression analysis was used to examine the relationship 
between batch fecundity and the total length of the fish. 
To examine the seasonal distribution of fish, the num- 
bers of specimens collected at each sampling station dur- 
ing each of the three study periods (September, Novem- 
ber-January, and February— May) were compared. The 
September samples were collected in the 1993 SNFRI 
trawl survey. Samples for the other two periods came from 
