36 



Fishery Bulletin 93(1), 1995 



but there was not enough auxiliary information to 

 quantify them. Macroscopically immature females 

 examined for histology did not show signs of recent 

 spawning, and all were microscopically staged as 

 perinucleolar. Out of 48 spent/resting females exam- 

 ined for histology, 43.8% also had perinucleolar ova- 

 ries. Seven of these had POF, direct evidence of prior 

 spawning. Of the remaining 14, atretic structures 

 were more than twice as common as in immature 

 females. Hunter et al. (1992) used atresia to distin- 

 guish immature from uncertain maturity in Dover 

 sole ovaries defined as inactive but concluded that 

 microscopic examination of oocytes in histological 

 sections may not identify all mature, postspawning 

 females. Relatively high rates of atresia may indi- 

 cate that fish have finished spawning (Wallace and 

 Selman, 1981), but this alone is insufficient to sepa- 

 rate mature from immature fish because atresia can 

 be brought on by stresses not necessarily associated 

 with spawning, such as starvation, pollution, or other 

 environmental conditions (Wallace and Selman, 

 1981; Hunter and Macewicz, 1985; Hunter et al., 

 1992). Because histological criteria could not differ- 

 entiate all mature from immature females and be- 

 cause other microscopic evidence of prior spawning 

 such as ovarian wall thickness (Burton and Idler, 

 1984) was not available, the degree of misclassi- 

 fication of mature vs. immature females staged mac- 

 roscopically could not be determined. 



Histological processing is known to cause shrink- 

 age of oocytes (West, 1990); therefore oocyte diam- 

 eters determined from processed tissue sections 

 should be considered an index rather than an abso- 

 lute measurement of oocyte size. For arrowtooth 

 flounder, Pertseva-Ostroumova (1960) reported 

 whole egg diameters are about 2.5-3.5 mm. Matarese 

 et al. (1989) lists A. stomias egg diameter as approxi- 

 mately 3 mm, three times the mean diameter of ripe 

 oocytes determined in this study. 



Seasonal bathymetric migrations are a familiar 

 pattern in flatfish. Dover sole, Microstomas pad ficus, 

 petrale sole, Eopsetta jordani, and English sole, 

 Pleuronectes vetulus, migrate seasonally across 

 depths (Alverson et al., 1964), and Dover sole tend 

 to reside at deeper depths at older ages (Hunter et 

 al., 1990). Kabata and Forrester (1974) sampled 

 arrowtooth flounder off Vancouver Island in May- 

 June 1968 and found increasing length with depth, 

 and a drop in abundance below 420 m (230 fm) con- 

 sistent with these results. Trends in arrowtooth 

 flounder catch rates by depth and season (Table 4) 

 indicate arrowtooth flounder move offshore in win- 

 ter. Arrowtooth flounder were common in shallow 

 water (<183 m) in summer, when the average size of 

 landed fish was large. In winter, smaller numbers 



and sizes of arrowtooth flounder were caught in 

 deeper water; whereas several hundred tows were 

 reported in shallow water with virtually no 

 arrowtooth flounder. Large, mature, presumably 

 spawning arrowtooth flounder may have moved out 

 of the range of the trawl fishery, possibly to deeper 

 water or north into Canadian waters. However, tar- 

 geted arrowtooth flounder trips are rare and inde- 

 pendent estimates of the amounts of arrowtooth 

 flounder discarded from trawl catches are high, from 

 nearly 76% off Oregon and Washington (Barss and 

 Demory, 1985) to over 80% in the Gulf of Alaska. 1 

 Large volumes of discards and catch unreported in 

 logbooks may obscure trends in distribution; they 

 certainly result in underestimates of arrowtooth 

 flounder CPUE. Hosie (1976) states that arrowtooth 

 flounder spawn off central Oregon from December 

 through March at about 200 fm. Hirschberger and 

 Smith (1983) reported spawning arrowtooth floun- 

 der at depths of over 350 m (191 fm) in the Gulf of 

 Alaska. The full extent of the spawning depth range 

 inhabited by arrowtooth flounder has not been de- 

 termined, but in this study in 1991 gravid females 

 were found in commercial tows out to 512 m (280 

 fm) and ripe/running females at 475 m (260 fm). In 

 1992 gravid and ripe/running females were found at 

 399 m (218 fm). Since these results also suggest that 

 in winter the bulk of the population was in water as 

 deep or deeper than 366 m, it is likely that the ma- 

 jority of arrowtooth flounder off Washington spawn 

 at depths exceeding 366 m (200 fm). 



To examine trends in length at maturity over time, 

 I fitted logistic curves to Oregon trawl survey data 

 from the 1970's 3 to compare with results from Wash- 

 ington (Table 5). The Oregon survey covered the area 

 from Newport, Oregon, south to Cape Blanco and 

 included FL and maturity for 218 male and 1,628 

 female arrowtooth flounder. Macroscopic criteria 

 used to distinguish mature from immature fish were 

 identical in both studies. Washington maturity 

 samples were collected in all months except Janu- 

 ary and February. All months except July, August, 

 and November were represented in the Oregon data. 

 Arrowtooth flounder in the present study matured 

 at a smaller size than those collected off Oregon, and 

 likelihood-ratio tests (Kimura, 1980) showed sample 

 nonlinear regressions for Oregon were significantly 

 different when compared with both Washington sur- 

 vey data (male x 2=: 74.555, P<<0.001; female 

 X 2 =137.922, P«0.001) and commercial data (male 

 X 2 =80.539,P«0.001; female x 2 =147.920,P«0.001). 

 Distribution of female maturities across lengths (Fig. 

 4) suggests that female arrowtooth flounder are 

 maturing at a smaller size than they were off Or- 

 egon in the early 1970's, or that there are latitudinal 



