580 
Fishery Bulletin 95(3), 1997 
the spawning season, especially for coverage of the 
central stratum. To counteract this, the stations in 
the central stratum were divided randomly between 
two time strata to reduce the time between occupation 
of this high-density stratum between subsurveys. 
In the DFRM analysis, the plankton samples used 
for the estimation of egg production were from two 
consecutive AEPM subsurveys that were occupied 
near the peak of the spawning season and that were 
not subject to “downtime” from ship and equipment 
failures. These two subsurveys had no time break 
between them and were treated as a single survey, 
in which the central stratum was occupied four times 
and each of the surrounding strata was occupied 
twice. The plankton sampling in the entire survey 
was done from 14 June to 7 July, and the two 
subsurveys used in the DFRM analysis were done 
from 28 June to 6 July (subsurveys 3 and 4). 
Egg sampling and staging, count standardization, 
and production estimation The plankton net used 
in sampling had a cylinder-cone design with 900-pm 
mesh, a mouth area of 2 m 2 , and was fitted to a 125- 
kg flat-steel ring. It was designed to be efficient, with 
the ratio of the open area in the mesh to mouth area 
being >5:1 (Tranter and Smith, 1968). The net was 
deployed from a starboard crane while the ship was 
stationary (i.e. not under power) and its starboard 
side faced the wind. The winch had dynamic 
tensioning, to minimize surging of the net as a re- 
sult of the rolling motion of the ship. Tow depths were 
within 30 m of the bottom to the surface if the bot- 
tom was less than 950 m and from 850 m if the bot- 
tom was deeper than 950 m. A conductivity-tempera- 
ture-depth (CTD) probe or a net sonde within the 
mouth of the net was used to measure net depth. 
Warp payout was measured with winch instrumen- 
tation. Warp payout and recovery rates were 1 m / 
sec, also measured with winch instrumentation. 
Eggs were staged (Grimes et al. 3 ) on board, prior to 
preservation, and generally within 0.5 h of landing the 
net, except for two samples with many eggs. For these, 
staging was done partly on board and partly in the labo- 
ratory on 4% formaldehyde-preserved eggs. All eggs < 
stage 7 (32-cell) were grouped, because it was not pos- 
sible to identify with confidence the stages from germi- 
nal disk through 32 cells within the plankton samples 
because most of these younger eggs were damaged (76% 
of the 8,293 eggs < stage 7), which caused the cell walls 
of the embryos to rupture, the cells to fuse, and the 
perivitelline space to collapse. Justification for assum- 
ing that the damaged eggs were all < stage 7 are given 
in Zeldis et al. (1995) and Grimes et al. 3 
The standardization from egg count to egg density 
(eggs/m 2 ) was based on the formula density = count 
x correction factor, where the correction factor takes 
into account the mouth area of the net and the vol- 
ume of water filtered by it. With a vertical haul, the 
correction factor is 0.5 (because the net mouth area=2 
m 2 ). However, because the vessel almost always 
drifted (owing to wind and current) during shooting 
and hauling, hauls were not vertical and therefore 
the correction factors were almost always <0.5. Dis- 
tance towed was not estimated by using flowmeters 
because flowmeters were found to record spurious 
revolutions during the deployment (descent) phases 
of tows in subsequent tests (Grimes 9 ). Instead, to 
calculate the volume of water filtered by the net, it 
was necessary to use global positioning satellite 
(GPS) vessel positions, warp length, depth, and cur- 
rent velocities to infer the path of the net (which, 
because of the ship’s drift, would be curved in the 
vertical dimension during hauling; Appendix 1). 
With a curved net trajectory, there was a different 
correction factor for each combination of plankton 
tow and egg stage because the different egg stages 
occupied different depths as they ascended the wa- 
ter column and because the net sampled more water 
in layers of equal thickness in the upper water col- 
umn than in the lower column. To calculate egg age 
and depth range (Table 1), data on egg development 
rate as a function of temperature, buoyancy by egg 
stage, and temperature as a function of depth were 
9 Grimes P. 1996. NIWA, P.O. Box 14-901, Kilbirnie, Welling- 
ton, New Zealand. Unpubl. data. 
Table 1 
Egg age (h) and depth (m) ranges used in calculating cor- 
rection factors for count standardization for Ritchie Bank 
plankton samples. Egg stages are described in Grimes et 
al. (Footnote 3 in the text). 
Stage 
Min. age 
Max. age 
Max. depth 
Min. depth 
0 
0.0 
0.0 
850 
725 
1 
0.0 
5.1 
850 
784 
2 
5.1 
8.2 
784 
743 
3 
8.2 
11.2 
743 
704 
4 
11.2 
14.0 
704 
667 
5 
14.0 
16.7 
667 
631 
6 
16.7 
19.3 
631 
596 
7 
19.3 
21.8 
596 
563 
8 
21.8 
24.1 
563 
531 
9 
24.1 
26.3 
531 
500 
10 
26.3 
28.4 
500 
470 
11 
28.4 
33.4 
470 
400 
12 
33.4 
40.0 
400 
301 
13 
40.0 
45.5 
301 
216 
14 
45.5 
50.2 
216 
143 
15 
50.2 
54.3 
143 
78 
