346 
investigated during September and November in 
1987-94 as tabulated in Table 1. Thirty sampling 
stations used for the collection of juveniles were ar- 
ranged as shown in Figure 1. Each of the stations 
was set up at six transects including depths of 3, 5, 
7, 9, and 11 m for evaluating the average density of 
juveniles in the fishing ground. The distribution of 
juveniles had been previously observed at 9 m, par- 
allel to the shoreline in Tomakomai (Sakurai, 1994). 
Two samples were collected at each station with a 
Smith-Mclntyre grab (sampling area: 0.05 m 2 ), sieved 
through a 1-mm mesh, and preserved immediately 
in 5% buffered formalin in the field. Sorting, identi- 
fication based on morphological characteristic of the 
shell (Habe, 1977), and counting of juveniles were 
done in the laboratory. 
Harvestable-size clams with shells >60 mm long 
were collected at 18 sampling stations in April, 1987- 
94 (Fig. 1; Table 1). Samples were collected with a 
hydraulic jet dredge (width=1.2 m, mesh size of net 
bag=70 mm, spacing of teeth=36 mm, and angle of 
tooth=60°) normally used for collecting P sachali- 
nensis. The dredge was towed parallel to shore for 
50-100 m at each station. Distance of each tow was 
estimated by measuring the length of the dredge- 
connected rope wrapped by a winch. Sorting was con- 
ducted in the field. Shell lengths were measured to 
the nearest 0.1 mm with a sliding caliper in the labo- 
ratory. Clams with shells >60 mm long were grouped 
into 5-mm intervals and counted at each station. 
These counts were subsequently adjusted on the ba- 
sis of catch efficiencies of the dredge for the different 
5-mm-long intervals (60-65 mm to 0.75, 65-70 mm 
to 0.82, 70-75 mm to 0.88, 75-80 mm to 0.95, and 
>80 mm to 1.00), which were based on the selectiv- 
ity curve of M. chinensis (Nashimoto, 1984), before 
density (ind./sample area) was estimated. The den- 
sity was then converted to biomass (g/sample area) 
Table 1 
Sampling dates of newly settled juveniles and harvestable- 
size clams of M. chinensis. 
Newly settled 
Harvest-size 
Year 
juveniles 
clams 
1987 
20 Nov 
2-5 Apr 
1988 
15 Nov 
1-4 Apr 
1989 
21 Nov 
2-5 Apr 
1990 
4 Sep 
11, 12 Apr 
1991 
26 Nov 
10, 11 Apr 
1992 
15 Oct 
16, 18 Apr 
1993 
20 Sep 
12, 13 Apr 
1994 
29 Nov 
20, 21 Apr 
Fishery Bulletin 96(2), 1 998 
by using the weight (g) equivalent for the median 
shell length of each size interval. The length-weight 
equivalent was determined from 107 clams with 
shells 57.8-83.4 mm long sampled at station N (Fig. 1) 
in April 1991 as follows: 
log W = 3. 16 log L - 4.071, (r = 0.9728, P < 0.01) (1) 
where W = the weight (g); and 
L = the shell length (mm). 
Next, the stock size (N g ; unit: t [age in years]) in the 
survey area was estimated with the following equation: 
18 
N e = 18- 1 A^bjr'w- 1 x 1(T 6 
i= 1 
where A = survey area (72 x 105 m 2 ); 
b t = biomass (g/sample area) at each station; 
l. = tow distance (m) at each station; and 
w = width of the dredge (1.2 m). 
In M. chinensis, an external growth ring with a 
light penetrable band, as shown in Figure 2, is formed 
annually on the shell during declining water tem- 
perature, November-January, and is able to be dis- 
tinguished from other rings (Sakurai, 1993). For the 
stock size prediction, therefore, the age of clams was 
determined by counting these rings. Clams with 
shells >60 mm long sampled at station N (Fig. 1) in 
April 1991 were used for age determination because 
the distribution of harvestable-size M. chinensis was 
observed at 9 m parallel to the shoreline in Toma- 
komai (Sakurai, 1996) and because specimens were 
considered typical of age composition in the survey 
area. Age determination was conducted by first wash- 
ing out the shell periostracum of the specimens with 
a chlorine bleaching agent and by drying it in the 
shade. Next, the specimens were put on a light box, 
and external growth rings with light penetrable 
bands were counted. 
Age-determined clams were grouped into 5-mm 
intervals at each age and counted. These counts (ind./ 
sample area) were adjusted by using the above effi- 
ciencies, and then density (ind./m 2 ) was estimated 
at each age by multiplying the counts by tow dis- 
tance and width of the dredge. 
Stock size prediction 
The predicted stock size (N pi ) [=1992,1993, . . . , 1997; 
unit: /), based on the age composition in April 1991, 
was estimated for each year class as follows: 
( 2 ) 
