610 
Fishery Bulletin 95(3), 1997 
the opposite for arrowtooth flounder; her highest L 50 
values were derived from fish taken prior to spawn- 
ing and her lowest value was derived from fish taken 
during the spawning season. The Oregon study oc- 
curred from September through June (Hosie and 
Barss 1 ), the Washington study occurred nearly year- 
round (Rickey, 1995), and the British Columbia study 
occurred only in June (Fargo et al., 1981). 
Histological examination revealed that most ma- 
ture males in this study had only a small portion of 
their testes filled with spermatozoa; and thus they 
were not yet ready to spawn. It is likely that, as these 
large males continued to develop sexually during the 
season, other smaller males would have become sexu- 
ally mature, thus lowering the male L 50 . The male 
L 50 value of 42.2 cm determined in this study should 
be viewed as a high estimate. Male GSI values 
started increasing at around 30 cm in length, and 
CF values began declining at 34.5 cm; both trends 
indicate a transition from somatic growth to gonad 
maturation at a much smaller size than that for the 
L 50 reported here. 
In general, the largest females were the most ma- 
ture in this study, indicating that they might spawn 
the earliest. The high values of CF, GSI, and HSI for 
these largest, most mature females also show that 
these fish are best able to support the burden of 
spawning. The noticeable but insignificant drop in 
CF for females in the cortical alveoli and early vitel- 
logenesis stages (Fig. 2), at around 50-60 cm in 
length (Fig. 3A), if real, can be explained by two pos- 
sibilities. Either these mid-size fish are affected more 
by vitellogenesis than are the larger fish, or all fe- 
males suffer losses in CF in the early stages of vitel- 
logenesis and recover during later maturity stages. 
Gonadosomatic index was also highest in the largest 
males; thus they appear to mature earlier in the sea- 
son than smaller males. The largest, most mature 
males had decreasing CF values that indicated an 
impact of maturing testes on body composition. 
The spawning habits of arrowtooth flounder are 
not well known. Shuntov (1970) was unable to de- 
termine accurate spawning times for arrowtooth 
flounder in the eastern Bering Sea but nonetheless 
stated that they were close to those of Kamchatka 
flounder (Atheresthes evermanni), which were found 
in spawning condition in January and March. Fargo 
et al. (1981), using macroscopic observations of go- 
nads collected in June from Hecate Strait, concluded 
that spawning takes place prior to June, probably in 
spring months. Rickey (1995) showed that spawning 
occurred off the Washington coast from September 
through December, and possibly as late as February. 
Hosie and Barss 1 reported a December-March 
spawning period for arrowtooth flounder off the Or- 
egon coast. Pertseva-Ostroumova (1961) reported 
arrowtooth flounder spawning in the Bering Sea from 
January through March. The results presented here, 
that a spawning season begins after September, are 
supported by all of the studies mentioned above. 
Total fecundity 
Both macroscopic and microscopic observations 
showed that this study was made prior to the spawn- 
ing season: none of the females were “ripe and run- 
ning,” had hydrated oocytes, or had postovulatory 
follicles, and none of the males were ready to spawn. 
Thus no bias due to loss of oocytes was expected. 
Total fecundity estimates for arrowtooth flounder 
had not been previously reported in the literature. 
The only other member of the genus, Kamchatka 
flounder, has an estimated fecundity range of 
130,000-500,000 oocytes (Pertseva-Ostroumova, 
1961), which is much lower than what is reported 
here for arrowtooth flounder. As with many other flat- 
fish species, arrowtooth flounder total fecundity in- 
creases linearly with fish weight and in a curvilin- 
ear fashion with length (Hempel, 1979). The largest 
arrowtooth flounder in this study had about 10 times 
as many oocytes as the smallest fish for which total 
fecundity was estimated. The unimodal frequency 
distribution of maturing oocyte diameters is sup- 
ported by Rickey’s (1995) determination that 
arrowtooth flounder is a group-synchronous spawner. 
The significant difference in weight between the 
eyed-side and blind-side lobes has not been previ- 
ously reported for arrowtooth flounder but has been 
reported for sole ( Solea solea, Witthames and Walker, 
1995). Nichol 2 found that blind-side lobes were sig- 
nificantly larger than eyed-side lobes in yellowfin sole 
( Pleuronectes asper). This finding suggests that in 
flatfish species both ovarian lobes should be consid- 
ered when calculating GSI or total fecundity. Because 
there were no significant differences in mean oocyte 
diameter or mean oocyte density within or between 
ovarian lobes, total fecundity samples may be taken 
from any portion of the ovary. In his review paper, 
West ( 1990) mentioned that typically there is no dif- 
ference in oocyte size or diameter frequency distri- 
bution between ovarian lobes, but differences along 
the length of the ovary and in cross sections do occur 
in some species. Hunter et al. (1992) found no differ- 
ences in oocyte density between ovarian lobes, along 
the length of a lobe, or by cross section of a lobe in 
Dover sole. 
2 Nichol, D. G. 1995. Resource Assessment and Conservation 
Engineering Div., Alaska Fish. Sci. Center, 7600 Sand Point 
Way NE, Seattle, WA 98115. Personal commun. 
