294 
Fishery Bulletin 117(4) 
Figure 1 
Map of locations where petrale sole (Eopsetta jordani) were 
captured during 2014—2017 off California, Oregon, and 
Washington. Northwest Fisheries Science Center (NWFSC) 
locations represent bottom-trawl survey stations; South¬ 
west Fisheries Science Center (SWFSC) and Washington 
Department of Fish and Wildlife (WDFW) locations are 
ports of landings. Map provided by R. Miller (SWFSC). 
differences among slides from the same individual, there¬ 
after, tissue was collected from the middle portion of a 
randomly selected ovarian lobe. 
Fecundity 
Determination of fecundity strategy Fecundity strategy 
was determined by examination of ovarian histological 
sections and diagrams of oocyte size—frequency distribu¬ 
tion for ovaries of macroscopic stages 2 and 3. Histological 
observations of POFs concomitant with oocytes undergo¬ 
ing vitellogenesis or maturation indicated that petrale 
sole spawned multiple times a season (batch spawning). 
Furthermore, the presence of a hiatus between primary 
and secondary growth oocytes in ovaries from fish for 
which spawning was imminent (with the leading oocyte 
cohort at late stages of vitellogenesis or further devel¬ 
oped) and in ovaries from repeat spawners (with the lead¬ 
ing oocyte cohort at late stages of vitellogenesis or further 
developed and POF present) indicated that fecundity was 
determinate (potential annual fecundity [PAF] set prior to 
the first spawning event) (Murua and Saborido-Rey, 2003). 
To confirm that fecundity was determinate and to define 
the minimum trailing oocyte stage (least developed sec¬ 
ondary growth oocyte present) threshold for fecundity 
analyses, oocyte size-frequency distributions were exam¬ 
ined from archival samples of 20 ovaries of macroscopic 
stages 2 and 3 (10 of each stage). Oocytes were teased from 
tissue and placed into welled chambers. Photomicrographs 
were recorded at 70 x magnification by using a digital 
camera (MU800 1 , AmScope, Irvine, CA) mounted to a ste¬ 
reo microscope. Measurements of oocyte diameters were 
automated from photomicrographs by using the ObjectJ 
plugin (vers. 1.03x, University of Amsterdam, available 
from website) and associated modified macros (avail¬ 
able from website) for image analysis software ImageJ 
(vers. 1.50i, National Institutes of Health, available from 
website; Schneider et al., 2012). Automated measurements 
recorded from damaged oocytes or detritus were manually 
removed. Primary growth oocytes clustered together and 
were difficult to tease apart. Because the maximum size 
for primary growth oocytes was 150 pm (on the basis of 
measurements from 10 stage-3 ovaries; Table 2), a lower 
limit of 150 pm was established for automated measure¬ 
ments. A minimum of 500 secondary growth oocytes per 
sample were measured. A gap between primary and sec¬ 
ondary oocytes was observed (Fig. 2B) when the trailing 
oocytes had reached the secondary vitellogenic stage 
(Vtg2) (meaning no secondary growth oocytes were less 
developed than Vtg2). Recruitment from the reserve of 
primary growth oocytes to the spawning stock was consid¬ 
ered complete at this time. 
Fecundity processing and analyses Only macroscopic-stage-2 
ovaries (;z=132) were considered for fecundity process¬ 
ing and analysis to determine PAF prior to any spawning 
events. Macroscopic-stage-3 ovaries were not used for fecun¬ 
dity analysis because of the observation of POFs (evidence 
of prior spawning) in all stage-3 ovaries examined histolog¬ 
ically. Histological sections from macroscopic-stage-2 ova¬ 
ries were examined for the presence of POFs (tz= 0) and to 
ensure recruitment of oocytes for the reproductive season 
was complete. Any samples with trailing oocytes earlier 
than Vtg2 (n= 56) were excluded because of the possibil¬ 
ity for incomplete recruitment of oocytes to the potential 
spawning stock. A total of 70 samples (California: n= 49; 
Pacific Northwest: n=21) were included in fecundity 
analyses. 
Homogenous oocyte density (oocytes per gram of ovar¬ 
ian weight) throughout the ovary was verified by com¬ 
paring densities from subsamples collected from the 
anterior, middle, and posterior third of each ovarian lobe 
from 6 macroscopic-stage-2 ovaries collected in the 2015- 
2016 reproductive season. The results of a nested analy¬ 
sis of variance indicate that oocyte density did not vary 
1 Mention of trade names or commercial companies is for identi¬ 
fication purposes only and does not imply endorsement by the 
National Marine Fisheries Service, NOAA. 
