484 
Fishery Bulletin 95(3), 1997 
CD 
O 
c 
03 
"D 
c 
-Q 
CO 
"CO 
£ 
_co 
c 
CO 
0 
E 
■O 
0) 
co 
E 
50 
25 
0 - 
225 
150 
75 
0 - 
12 1 
6 - 
O) 
LU 
B. tyrannus 
L. xanthurus 
P. dentatus 
P. albigutta 
—i 1 
100 
50 
0 
130 
65 
0 
12 
10 
M. undulatus 
I 5 
i i 
i i 
L. rhomboides 
P. lethostigma 
M. cephalus 
-i r ~ 
14 30 
— l 1 1 1 1 1 r 
2 3 4 5 7 14 30 
Sampling interval (days) 
Figure 2 
Estimated effect, based on daily sampling data, of sampling interval on estimated larval fish 
abundance. 
B. tyrannus data obtained with a 1 x 2 m net fished 
near the surface (Warlen, 1994). Sampling with a 
1 x 2 m surface net occurred at night, at approximately 
the same hour, from the same sampling platform, and 
during the same immigration season (1991-92) as 
our basic study. The seasonal weekly mean of the 
weekly means calculated from daily means obtained 
with the 1-rm net was significantly different (P<0.Q5) 
from the seasonal weekly mean density obtained with 
the 1 x 2 m surface net. Sampling daily with the 1-m 
oblique net resulted in an average of 10.0 B. tyrannus 
larvae per 100 m 3 compared with 7.2 larvae in weekly 
sampling during the same season with the 1 x 2 m sur- 
face net (Warlen, 1994). The abundance of B. tyrannus 
in catches of the 1-m oblique net tows made only on the 
nights that the 1 x 2 m surface net was fished appeared 
similar to the weekly average of the daily catches with 
the 1-m oblique net tows (10.2 vs. 10.0 larvae per 100 
m 3 ), but this similarity could not be statistically tested 
because the samples were not independent. 
