110 
Fishery Bulletin 99(1 ) 
was examined for each female. Grid counts were repeated 
4 to 12 times at different locations within each ovary sec- 
tion, until counts totaled approximately 180 oocytes. 
Testing for homogeneity within the ovary 
Prior to subsampling for total fecundity, we tested (two- 
way AN OVA; t-tests) for differences in oocyte density 
(number of oocytes per gram of ovary tissue) and mean 
oocyte diameter between eyed-side and blind-side ovarian 
lobes and among three ovary positions (anterior, middle, 
posterior). Twenty ovaries histologically identified with AY 
oocytes (maturity-code 2) and no evidence of prior batch 
spawnings (i.e. no POFs) were selected from fish over a 
broad length range. AY oocytes were defined as those with 
yolk filling more than half the volume of the oocyte. Tissue 
samples averaging 10.5 mg (SD=2.9559) and 267 oocytes 
(SD=90.0) were taken from anterior, middle, and poste- 
rior positions along the long axis of each ovarian lobe (six 
subsamples per fish). All AY through MN stage oocytes 
from each tissue sample were counted manually with the 
aid of a dissecting microscope. Fifty of these oocytes per 
tissue sample were randomly selected for oocyte area mea- 
surements. Oocyte areas were measured with a micro- 
scopic image analysis system with Optimas 5.0 software 
(BioScan, Inc., 1992). Black and white images were gener- 
ated with a video camera attached to a dissecting micro- 
scope (transmitted light) and were viewed on a 13-inch 
(diagonal) color monitor. The resolution was set to 640 x 480 
pixels, corresponding to 4.149 pin/pixel at 25x magnifica- 
tion. Oocyte diameter was calculated from oocyte area by 
number of oocytes per gram of ovary tissue and unyolked 
oocytes in terms of percent frequency of the 50 measured. 
Total fecundity and hatch fecundity estimation 
Because of findings from the testing of homogeneity with- 
in ovaries (see “Results” section), two tissue subsamples 
were taken from the ovaries of each fish: one from the 
posterior third of either ovary, and one from either the 
middle or anterior third of either ovary. To estimate total 
fecundity, defined as the standing stock of AY through 
HY oocytes, only ovaries with no evidence of prior batch 
spawnings (no ova, no POFs, nor residual chorion mate- 
rial) were used. To eliminate less developed ovaries that 
contained many PY oocytes, we also limited fecundity 
samples to those with maximum oocyte diameters >0.35 
mm as measured from histological slides. Area measure- 
ments of AY through MN stage oocytes with image analy- 
sis and counts were conducted in the same manner as the 
above test for homogeneity. 
To determine the proportional mass of each ovary sec- 
tion ( WF p , WF m , and WF a ), one of the paired ovaries 
from each of the 20 fish tested for oocyte homogeneity 
was cut into thirds, mid-way between anterior-middle and 
mid-way between middle-posterior tissue positions. Each 
section was weighed to the nearest 0.001 g. The propor- 
tional mass (section wt/sum of section wts) of anterior, 
middle, and posterior sections averaged respectively 0.536 
(SE=0.0095), 0.311 (SE=0.0086) and 0.153 (SE=0.0059) of 
the total ovary mass. 
Total fecundity was computed as 
Diameter = 2 ■ 
( 1 ) 
Total fecundity 
^(WF,) + ^ 
wt,, ' wt . 
( WF 
POW , 
Oocyte size distributions 
Oocyte areas were measured to determine whether there 
was a hiatus between distributions of AY oocytes and less 
advanced oocytes. Areas of oocytes in partially yolked (PY ), 
unyolked, and AY stages were measured by using preserved 
tissue (as above), and calculated diameters were plotted 
for 75 of the 324 females examined for fecundity. These 
were the same females used to examine ooctye atresia, 
thus representing prespawning through partially spawned 
females, as well as those with residual chorion material 
present. PY-stage oocytes were considered as those with 
yolk that filled less than half the volume of the oocyte. Sub- 
samples were taken from the anterior or middle portion 
of one ovary and weighed to the nearest 0.001 g. All PY 
and AY oocytes were counted separately. For the purpose of 
oocyte counts and measurements, MN oocytes were not dis- 
tinguished from AY oocytes. No attempt was made to count 
unyolked oocytes or measure unyolked oocytes less than 
approximately 0.05 mm. Random oocyte area measure- 
ments by image analysis (at 25x) included 50 AY oocytes, 
50 PY oocytes (if they existed), and 50 unyolked oocytes. 
The three oocyte diameter distributions were then plotted 
for each fish as PY and AY oocyte distributions in terms of 
where no „ = 
wt„ = 
wt = 
WF. = 
WF.. 
POW = 
number of AY-HY oocytes in tissue sample 
from posterior position of either ovary; 
number of AY-HY oocytes in tissue sample 
from either anterior or middle positions of 
either ovary; 
weight of tissue sample from posterior 
position of either ovary; 
weight of tissue sample (g) from either 
middle or anterior position of either 
ovary; 
weighting factor computed as the average 
proportional mass of the posterior third of 
either ovary; 0.153 g O?=20); 
weighting factor computed as the average 
proportional mass of anterior and middle 
ovary sections combined; 0.536 g + 0.311 g 
= 0.837 g (n=20)] and 
paired ovary weight (g). 
Batch fecundity was estimated from those ovaries con- 
taining HY oocytes (maturity-code 3). Subsample weights 
averaged 52 mg. Both HY oocytes (batch oocytes) and AY oo- 
cytes (remaining oocytes ) were counted. Batch fecundity was 
computed as above (Eq. 2), substituting no p and no am with 
