Conrath: Reproductive potential of Sebastes variabilis and 5. polyspinis 
143 
Table 3 
Classifications used to determine maturity of northern rockfish (Sebastes polyspinis) and light dusky rockfish (S. variabilis) col¬ 
lected in May and December 2014 in the Gulf of Alaska. POF=postovulatory follicles. For stages of oocyte and embryo development, 
see Table 2. 
Oocyte or embryo 
Class 
development 
Description 
Immature 
Primary growth 
No oocyte development beyond stage 2. 
Abortive maturity 
Oocyte development is initiated, then oocytes are resorbed. No evidence of prior 
spawning. Identified histologically by presence of widespread alpha atresia, but 
no POFs or beta or delta atresia. 
Mature 
Development 
Oocyte or embryo development at stage 3 or at a more advanced stage of 
development. 
Partial development 
Some oocytes within the ovary are at stage 3 or at a more advanced stage of devel- 
(incomplete 
opment. Some oocytes are not developing, because of incomplete fertilization. 
fertilization) 
No development 
(skipped spawning) 
Identified histologically by the presence of oocytes at the one-cell blastodisc stage. 
Resting 
No initiation of oocyte development. Identified histologically by development in 
stage 1 or 2 with evidence of a prior spawning (beta or delta atresia or POFs). 
Resorbing 
Oocyte development is initiated, but oocytes are later resorbed. Identified histologi¬ 
cally by widespread alpha atresia and evidence of a prior spawning (beta or delta 
atresia or POFs). 
Fertilization failure 
Oocytes develop but are not fertilized; therefore, no embryogenesis occurs. Identi¬ 
fied histologically by the presence of oocytes at the one-cell blastodisc stage. In 
skipped spawners, all developing oocytes are at this oocyte stage of development. 
The ovary was patted dry at the laboratory, and a 
wet weight was recorded. A gonadosomatic index (GSI) 
was calculated for all mature fish by using the follow¬ 
ing equation: GS/=gonad weight/total body weight. The 
mean and standard error of these values for each spe¬ 
cies during each season were determined. Ovary weights 
were unavailable for some fish whose ovary contained 
advanced embryos because the ovary structure was 
difficult to maintain. This delicate ovary structure was 
particularly prevalent in samples collected in May that 
contained eyed embryos with a thin and fragile ovarian 
wall that made it difficult to weigh the ovary without 
losing some of the embryos. 
Fecundity was estimated at each site by using all 
individuals caught in December that had both lobes of their 
ovary and all specimens caught in May that had at least 
one complete lobe of their ovary. The difference between 
months sampled in which individuals were included was 
necessary because of the small number of individuals with 
a complete ovary that were collected in May. If only one 
lobe of the ovary was available, the total ovary weight was 
estimated by multiplying the weight of the complete lobe 
by 2. Individual potential annual fecundity was estimated 
by using a gravimetric approach in which the weight of 
a subsample of oocytes or embryos present was extrapo¬ 
lated to estimate the total ovary weight. Fecundity was 
estimated by using only fish that had an ovary that was 
composed predominately of oocytes or embryos at a devel¬ 
opment stage of 3 or higher and that had no evidence 
of partial or total reproductive failure. In December, all 
ovaries examined contained oocytes that were in either 
the migratory nucleus or ovulation stages, whereas in 
May all ovaries contained eyed embryos. Therefore, val¬ 
ues of fecundity for specimens captured in December are 
given in number of oocytes or number of oocytes per gram, 
and values of fecundity for specimens captured in May are 
given in either number of embryos or number of embryos 
per gram. Two subsamples were collected from randomly 
chosen quadrants of one lobe of the ovary. If the coefficient 
of variation from the 2 initial samples was greater than 
5%, 2 additional subsamples were collected. The individ¬ 
ual potential annual fecundity was determined by aver¬ 
aging the fecundity estimates taken for each sample and 
multiplying by the total ovary weight. In addition, relative 
fecundity was determined by dividing potential annual 
fecundity by the total weight of the fish. 
Reproductive failure in adult light dusky and northern 
rockfish was defined as partial or total lack of oocyte or 
embryo development in mature fish. Partial reproduc¬ 
tive failure was identified by the presence of developing 
oocytes and non-developing oocytes within the ovary of a 
fish and was caused by incomplete fertilization (Fig. 2). 
Mature fish with no oocyte development were considered 
to be experiencing total reproductive failure and were 
classified as skipped spawners. Three types of skipped 
spawning similar to those defined by Rideout et al. (2005) 
were identified: resting skipped spawning (no initia¬ 
tion of oocyte development), resorbing skipped spawning 
