Shi et al.: Growth and survival of Pleuronectes vetulus 
171 
Winberg (1956) found that individual metabolic 
requirements and food consumption increase as a 
function of W° 8 (where W=body weight), and Fonds 
(1979) showed a similar relation for young sole, Solea 
solea. The larger sizes attained by the early-settle- 
ment cohorts of English sole would probably also 
entail increased food requirements for those fish 
during May-September. The lower growth rates ob- 
served for early-settlement cohorts in comparison 
with those that settled later probably resulted from 
a combination of higher metabolic demands and re- 
duced availability of suitable prey. Growth of 0 + En- 
glish sole does not appear to be strictly linear if a 
sufficiently long period of time is examined. 
Mortality 
Estimates of mortality rates were subject to some of 
the same sources of error and bias that the growth 
estimates were. Previous work (Gunderson et al., 
1990; Shi et al., 1995) has shown that migrations of 
0 + English sole are size dependent. Typically, larger 
fish emigrate from estuaries and perhaps out of our 
nearshore survey area, whereas smaller fish immi- 
grate into our survey area. Immigration of small fish 
would cause underestimates of mortality. Previous 
analysis indicated that most immigration probably 
occurred near the settling period, i.e., May and June 
or earlier (Shi et al., 1995). Therefore, it is unlikely 
that immigration had much effect on the mortality 
estimates because only population estimates for July 
through September were used in this analysis. 
Emigration of large 0 + fish would cause overesti- 
mation of mortality rates. Although we cannot com- 
pletely ignore the possibility of emigration, previous 
analysis of length increment patterns in estuarine 
and nearshore areas has indicated that net emigra- 
tion of large fish from the study area is minor during 
July-September (Shi et al., 1995). In addition, had 
substantial emigration occurred during July through 
September, estimated mortality rates would be con- 
sistently higher during September than dui’ing July 
and August, rather than the opposite (0.0075 per day 
vs. 0.0175 per day). 
Seasonally differentiated daily mortality could be 
related to differences in temperature, individual size, 
or population density. Water temperatures, however, 
remained relatively stable in the study area during 
July through September (Fig. 3). Size-dependent 
mortality may have occurred, because 0 + English sole 
grow rapidly during the summer, with increases in 
individual size of 0 + English sole ranging from 20 to 
30 mm TL from July to September. Kramer (1991) 
estimated the mortality rates for each 5-mm size 
group of California halibut, Paralichthys californicus , 
on the basis of daily production by size group, and 
found that mortality was size specific for fish less 
than 70 days old (< 30 mm SL), smaller fish suffer- 
ing higher mortality than larger ones. For older 0 + 
California halibut (70-115 days of age or 31-70 mm 
SL), mortality varied little (0.011- 0.014 per day, with 
mean=0.0124 per day and SD=0.001 per day) and no 
trend was observed. Beverton and lies (1992) found 
a significant density-dependent mortality (/t 2 ) effect 
for North Sea plaice ranging from <15 mm to 35 mm, 
although this effect was not significant for fish larger 
than 35 mm. Both Kramer (1991) and Beverton and 
lies (1992) found that mortality of juvenile flatfish 
is highest during and immediately after settlement, 
and our results for English sole suggest the same. 
The effects of density and individual size were 
clearly confounded in our study (Fig. 6). An empiri- 
cal relation between population size in the survey 
area (P f ) and mean length (l t ) was fitted as 
In P t = A + Bl t , 
where, P t = the estimated total population size at 
time t; l t = the mean length of all 0 + English sole at 
time t (July through September); and the correla- 
tion between the two confounding factors was highly 
significant (r 2 =0.56, PcO.Ol). It should always be 
borne in mind that it is very difficult, if not impos- 
sible, to isolate these two confounding factors in field 
observations; controlled enclosure experiments would 
probably be required to disentangle them. There is 
both a theoretical (Peterson and Wroblewski, 1984) 
and an empirical (McGurk, 1986; Kramer, 1991) ba- 
sis for assuming that mortality decreases with indi- 
vidual size, and if this is the case, n 2 (the density- 
dependent mortality coefficient) would have been 
overestimated. If mortality of Q + English sole is den- 
sity dependent, it appears that this dependence is 
weak because adding a density-dependent term to 
the model (model 2) did not improve the fit to the 
data. 
Our surveys showed only a threefold difference in 
abundance of 0 + English sole during 1985-88, but 
stock synthesis analysis of commercial fisheries data 
indicates this was a period of relatively stable re- 
cruitment (Sampson 1 ). The estimated recruitment of 
age-2 + females to the commercial fishery varied by 
no more than a factor of 1.8 for the 1985-88 year 
classes, whereas it varied by a factor of 6.5 for the 
1 Sampson, D. B. 1993. An assessment of the English sole stock 
off Oregon and Washington in 1992. Appendix H in Status of 
the Pacific coast groundfish fishery through 1993 and recom- 
mended allowable biological catch for 1994. Pacific Fish. Man- 
agement Council, 2000 SW 1st Ave., Suite 420, Portland, OR, 
43 p. 
