Sakurai et al.: Population dynamics and stock size of Mactra chinensis 
349 
stock in April 1990-94. As Figures 3 and 4 show, 
juvenile densities in 1988, 1989, and 1991 were 
relatively high, causing the real stock sizes in 
1991, 1992, and 1994 to increase in comparison 
with each previous year. In contrast, the lower 
juvenile densities in 1987 and 1990 caused the 
real stock size in 1990 and 1993 to decrease in 
comparison with each previous year. Therefore, 
it is suggested that the population dynamics at 
Tomakomai are directly influenced by the num- 
ber of newly settled juveniles. The number of 
juveniles is determined by the settlement and 
mortality of early benthic stages (Gunther, 1992). 
We think that it is necessary to examine the fac- 
tors that cause fluctuations in the number of ju- 
veniles to gain a full understanding of the popu- 
lation dynamics of M. chinensis. 
In the present study, we define a model using 
two known population parameters, juvenile den- 
sity and age compositions, to predict the stock 
size of M. chinensis. It has been shown that the 
age groups that have relatively higher densities at 
the juvenile stage predominate for four years from 
age 2 to 5 in harvestable-size clams. Saito et al. (1982) 
reported that a large catch of M. chinensis contin- 
ued for a few years off the coast of Ishikari, Hokkaido, 
because the dominant age group was maintained. 
This was the case in our study. Therefore, we con- 
sider that our model is appropriate for predicting the 
population dynamics of M. chinensis at Tomakomai 
and other sites. Furthermore, the predicted stock size 
of our model was close to the measured value (Fig. 7), 
and consequently accurate long-term prediction of 
stock size is possible with this model, provided age 
composition and juvenile density are determined. 
On the other hand, analysis of age and growth 
based on external growth rings has been conducted 
for various bivalves including Placopecten magel- 
lanicus (e.g. Stevenson and Dickie, 1954; Claereboudt 
and Himmelman, 1996), Pecten maximus (e.g. Dillon 
and Clark, 1980), Megangulus venulosus (Goshima 
et al., 1991), Mesodesma mactroides (Defeo et al., 
1992) , Clinocardium californiense (Goshima and 
Noda, 1992), and Abra tenuis (Dekker and Beukema, 
1993) . However, it is suggested that age determina- 
tion based on external growth rings of bivalves would 
be unreliable because it is difficult to distinguish the 
growth rings from other rings owing to several 
stimuli (Dillon and Clark, 1980). Nevertheless, the 
growth rings of M. chinensis have been distinguished 
clearly from other innumerable fine rings by check- 
ing the rings with light penetrable bands (Sakurai, 
1993). These bands are regarded as nacreous layers. 
Therefore, age determination in the present study 
would be as reliable as that based on microgrowth 
Table 2 
Estimated catch of M. chinensis in the survey area and 
whole fishery ground off the coast of Tomakomai. 
Year 
Catch (metric ton) 
Survey area Whole fishery ground 7 
1992 
7.2 
36.0 
1993 
14.6 
73.0 
1994 
22.0 
108.0 
1 Based on data from Hokkaido Fisheries Statistics in 1994, 1995, 
and 1996. 
Table 3 
Comparison of the real [R) and predictive (P) stock size for 
M. chinensis. 
Year 
R 
(metric ton) 
P 
(metric ton) 
( R-P ) x 100/R 
(%) 
1992 
728.3 
651.8 
10.5 
1993 
550.3 
586.0 
-6.5 
1994 
1,127.4 
1,015.8 
9.9 
analysis with acetate peels or thin sections (Lutz and 
Rhoads, 1980), and we consider that our model would 
be able to predict the stock size without over- or un- 
derestimates. In our model, however, annual mor- 
tality is assumed to be constant; therefore, age com- 
