Watanabe and Nakamura: Growth of Sardinops melanostictus 
901 
5 . 34.0 °N 
137.0 °E ’ 137.5 °E 
Figure 1 
Location of shirasu fishing grounds in the coastal 
waters along Atsumi Peninsula. Minor fishing 
grounds are found in Ise Bay at the tip of Chita 
peninsula. Dots indicate locations of monthly SST 
observations. A solid line along the Pacific coast of 
western and central Japan denotes an example of 
the meandering Kuroshio current axis. 
range is a function of growth rate, slower growth 
results in longer duration time and an increase in 
the available number of larvae. Watanabe and Kuroki 
(1997) backcalculated the growth history of larval 
sardine Sardinops melanostictus in the coastal wa- 
ters off Miyazaki in western Japan. They found that 
growth rates reached a maximum (0.80-0.85 mm/d) 
at around 10-12 mm in total length (TL) but slowed 
thereafter, exhibiting asymptotic growth trajectories. 
According to this growth pattern, duration time from 
15 to 30 mm TL was calculated to be 35 days. If, 
however, growth rates of 0.80-0.85 mm/d were main- 
tained in the late larval stage, duration time of the 
same size range could be as short as 18 days. 
Growth rate in early life stages is thus important 
as a potential determinant of year-class strength as 
well as of available stock sizes of shirasu larvae of 
the Japanese sardine. The daily growth rate of fish 
can be estimated by regressing size-at-age data to a 
growth model (Campana and Jones, 1992). When we 
apply this method, we need data points throughout 
an age range, from first feeding up to the fish size of 
concern (Watanabe et al., 1997). Because the size 
range of sardine larvae fished by the boat seine fish- 
ery is usually from 15 to 30 mm, we do not have a 
complete range of data points for early larvae, and 
therefore the regression method is not applicable to 
describe the growth history from first feeding to size 
at capture. Instead, backcalculation of size at age 
from the relationship between otolith radius and fish 
length makes it possible to draw a growth trajectory 
for individual fish (Campana, 1990; Campana and 
Jones, 1992). We backcalculated growth trajectories of 
individual larval sardines caught in coastal waters off 
Atsumi Peninsula, 1990 and 1991, and compared them 
with growth trajectories of larvae from other waters. 
Materials and methods 
Larval sampling 
Two shirasu fishing boats, towing a seine net, were 
used to sample waters off Atsumi Peninsula and in 
Ise Bay (Fig. 1). They usually departed before dawn 
from the fishing port, together with a catch-loading 
boat, set the net several times in the morning, and 
returned to the port for offloading. Catches were 
stored with ice before landing. In our study, we ran- 
domly sampled shirasu larvae from catches in the 
ports ofMorozaki and Toyohama (tip of Chita Penin- 
sula) four times in 1990 (11, 16, 27 April and 7 May) 
and four times in 1991 (15, 23 April and 7, 14 May). 
