560 
Fishery Bulletin 99(4) 
Figure 4 
Spherical model fitted to variograms of residuals of In (sardine eggs/minute+1) for three age 
groups and total eggs, in the direction perpendicular to the coastal line during leg 2 of cruise 
9603. On the x-axis, the inner ticks are in km and the outer ticks are in nmi. 
Table 2 
Contingency table of sardine egg counts by developmental stages and two samplers: the CalVET and the CUFES, leg 1, 1996. 
Egg stages 
I 
II 
III 
IV 
V 
VI 
VII 
VIII 
IX 
X 
XI 
Total count 
CalVET 
31 
63 
116 
46 
36 
95 
23 
27 
37 
15 
13 
CUFES 
2 
74 
466 
44 
148 
104 
103 
22 
75 
21 
14 
Percent 
CalVET 
6 
13 
23 
9 
7 
9 
5 
5 
7 
3 
3 
CUFES 
0 
7 
43 
4 
14 
10 
10 
2 
7 
2 
1 
also indicated that a potential bias may exist if staged 
CUFES eggs are used to estimate daily egg production 
(P 0 ) because all stages of eggs may not be equally vulner- 
able to the CUFES. A further complication was found, in 
that egg samples taken in the CUFES were strongly cor- 
related up to distances of 22.2 km, necessitating the use of 
geostatistics if individual samples were to be used for the 
estimates. We concluded from these findings that the most 
effective way to use the CUFES in the DEPM design was 
to use it to allocate full-water-column tows with a CalVET. 
This adaptive allocation design, similar to one proposed by 
Thompson et al. (1992), takes advantage of the rapid con- 
tinuous monitoring capability of a CUFES, while preserv- 
ing the quantitative features of the CalVET. 
We decided to allocate a CalVET at 4-nmi intervals, 
whenever catch from the CUFES was greater than or 
equal to 2 eggs/min 1 When the catch from the CUFES be- 
gan to decrease to fewer than 2 eggs/min, then CalVETs 
were stopped (Fig. 6). This critical value for the allocation 
of CalVETs was equivalent to slightly more than three 
eggs per CalVET tow and was based on the conversion fac- 
tor from the pilot study of CUFES eggs/min = CalVET 0.73 
eggs/tow. The CUFES was in operation continuously along 
each transect line, and samples were taken at an interval 
