Karlou-Riga et al.: Sex change and oscillating growth pattern of Spicara smaris in the Saronikos Gulf (Greece) 
355 
with the findings of Vidalis and Tsimenidis (1998), 
who also validated annulus completion by MI analy¬ 
sis. Denaxa et al. (2014), who determined picarel ages 
in Greek waters by means of otolith microstructure, 
reported that the first annulus is completed between 
January and April and formed between 180 (6 months) 
and 372 days (12.4 months) of life, which correspond to 
8.63 and 14.3 cm TL respectively. The data of Denaxa 
et al. (2014) seem to be slightly lower than those of 
the present study, where the first annulus appears in 
otoliths from fish between 9.7 and 15.0 cm TL. Clus¬ 
ter analysis based on the similarity indices, and after 
we compared the length frequency distributions of fish 
classified according to their otolith readings, showed 
both the validity of the age interpretation criteria and 
an alternative way to assign the fish to age groups. It 
can be concluded that fish with similar lengths belong 
to the same age group, a relationship that many times 
in the past has been verified (Francis and Campana, 
2004). 
The application of the VBGF as a nonlinear ap¬ 
proach to fitting iength-at-age data is commonly used to 
describe fish growth, given that it is still the most use¬ 
ful tool for growth analysis despite the criticism of its 
use. As stated by Kimura (1980), rejection of the VBGF 
curve must ultimately be based on superior alternative 
curves or methods of analysis. In our work, we noticed 
good fitting (see Fig. 5), and the growth parameters 
were found to be very reasonable (L„ close enough to 
the L mas values observed, t 0 close enough to zero). The 
good fit of data in the present work is probably due to 
both a careful reading of annuli and the large num¬ 
ber of specimens caught monthly. The mean observed 
lengths at ages 1, 2, 3, and 4 for sexes combined as 
reported by Tsangridis and Filippousis (1992) are in 
general agreement with those of the present study. The 
smaller lengths at age estimated by the VBGF as re¬ 
ported by Ismen (1995), Vidalis and Tsimenidis (1996), 
and Duldic et al. (2003), when compared to those of our 
study, are probably due to the interpretation of pre-an¬ 
nual (false) zone as the 1 st annulus. Slight differences 
in lengths at age were also found between those in the 
present work and those in the work by Rizkalla (1997). 
It is noted, however, that lengths in Rizkalla (1997) 
were estimated with the back calculation method. It is 
worth noting that in both works (Vidalis and Tsimeni¬ 
dis, 1996; Rizkalla, 1997), differences in the growth 
rate between sexes (males become longer than females) 
were not observed. 
An oscillating pattern in the growth of picarel was 
mentioned by Tsangridis and Filippousis (1992). In par¬ 
ticular, they attribute the slowing down of growth rate 
to either maturity or sex change. In more detail, they 
observed only two oscillations related to two growth 
periods: one linked to a very fast female growth period 
(up to 15.0 cm TL) and the second linked to a male 
slow period (up to 19.0 cm TL). In the present study, 
the decrease in the observed average length for each 
sex is noticed from summer to autumn during the 2 nd , 
3 rd , and 4 th year of life and is most likely related to the 
