Macchi et al.: Effects of skipped spawning on the reproductive potential of Merluccius hubbsi 
399 
Therefore, to estimate length at first maturity with 
data from the August survey, only the traditional in- 
terpretation of maturity assignment (resting stage as 
mature female) was employed. 
Estimation of size and age distributions of female 
Argentine hake 
The number of females was estimated from data col- 
lected during each survey between 2005 and 2013. 
These cruises covered a wide geographic range from 
43°S to 47°S, but to analyze the abundance of females 
that would skip spawning or would spawn, we used 
only data from sampling stations located in the area 
of active reproduction, which covers approximately 
65% of the study zone (Fig. 1). The same area, with an 
extension of about 50,000 km 2 , is assessed every year 
during a fixed-station trawl survey to estimate the 
potential egg production of the Patagonian stock of 
Argentine hake (Macchi et al. 3 ). Information obtained 
from sampling the trawl catch was expanded to ob- 
tain estimates of the number of individuals per length 
class, by using the method described by Macchi et al. 
(2004). The number of mature females for each survey 
was estimated by multiplying the number of Argen- 
tine hake within each length class by the proportion 
of mature females estimated from the L 50 ogive ob- 
tained for each survey. To obtain the incidence of SS 
by length class, we compared the length distributions 
obtained with the estimated maturity ogive with the 
traditional interpretation and the revised interpreta- 
tion for each year. 
From the age-length keys obtained with data collect- 
ed in each survey, we estimated the age distributions 
of female Argentine hake in the sampling area during 
each January between 2005 and 2013. The number of 
mature females by age class for each year was esti- 
mated by multiplying the number of females in each 
age class by the proportion of mature females obtained 
from the A 50 ogive corresponding to that year. We com- 
pared the age distributions of mature females by using 
the maturity ogive constructed with the traditional and 
the revised interpretation. In this case, to determine 
the effect of spawning omission, we estimated the aver- 
age age distributions obtained after grouping all years 
sampled during the period 2005-2013. 
Estimation of reproductive potential 
Because the Argentine hake is a batch spawner with in- 
determinate annual fecundity, we estimated the batch 
fecundity (number of oocytes released per spawning) 
by the hydrated oocyte method of Hunter et al. (1985). 
3 Macchi, G. J., M. Estrada, H. Brachetta, and V. Abachian. 
2013. Estructura y production potencial de huevos del efec- 
tivo desovante de merluza ( Merluccius hubbsi) al sur de 41°S 
durante enero de 2013. Inf. Tec. 88, 12 p. [Available from 
Instituto Nacional de Investigation y Desarrollo Pesquero, 
Paseo Victoria Ocampo Nro. 1, B7602HSA Mar del Plata, Ar- 
gentina]. 
After histological diagnosis, we selected only ovaries 
with hydrated oocytes and without evidence of recent 
spawning (no postovulatory follicles), and removed 
3 pieces of tissue (0. 1-0.2 g each) from the anterior, 
middle and posterior parts of a gonad. These samples 
were weighed (to the nearest 0.1 mg), and the number 
of hydrated oocytes was counted to estimate the mean 
of hydrated oocytes per unit of weight. Batch fecundity 
value for each female was obtained by multiplying the 
mean number of hydrated oocytes and the total weight 
of the ovaries. The relationships of batch fecundity to 
TL and age were described by using standard regres- 
sion analysis (Draper and Smith, 1981). 
We could not estimate a relationship between spawn- 
ing frequency and TL or age for each year sampled. For 
this reason, the index of reproductive potential of the 
Patagonian stock of Argentine hake in January from 
2005 to 2013 was estimated by multiplying the number 
of mature females in each length class or each age class 
by the batch fecundity corresponding to that length 
class or age class. Therefore, this index represents the 
egg production of the stock during one spawning event. 
Estimates of egg production, by TL and age, by us- 
ing the maturity ogive created with the traditional 
interpretation and the ogive created with the revised 
interpretation were compared to obtain the percentage 
of reduction in egg production attributed to nonrepro- 
ductive adult females during the spawning season by 
year. 
Results 
Length and age at maturity 
When we applied the traditional interpretation for 
maturity diagnosis, the L 50 estimates for females of 
Argentine hake during January 2005-2013 ranged 
between 32.22 and 34.53 cm TL (Table 2). However, 
when we used the revised interpretation and consid- 
ered that females in a resting stage were individu- 
als that had skipped spawning, the values of L 50 in- 
creased to range of 34.57-38.01 cm TL. In all cases, 
the standard error of L 50 was lower when SS was 
considered for estimation of the maturity ogives — a 
result that may be associated with the increase in 
the number of immature individuals in length classes 
close to L 50 . In addition, when SS was included in the 
L 50 estimate, the slope of the logistic model decreased 
(Table 2), suggesting that all females reached sexual 
maturity at larger sizes than those estimated with the 
traditional interpretation. Comparison of the maturity 
ogives estimated by the different interpretations of 
maturity showed highly significant differences for all 
years sampled (P<0.001). 
The A 50 curves estimated with samples collected be- 
tween 2005 and 2013 showed a similar pattern after 
inclusion of the proportion of females that SS in rela- 
tionship both to the A 50 and the slope of the models. 
The A 50 values estimated with the traditional inter- 
