350 
Fishery Bulletin 11 6(3-4) 
Table t 
Data on sampling picarel (Spicara smarts ) in the Saronikos Gulf off Greece from Sep¬ 
tember 1998 through August 1999 with beach seine (SB) or otter bottom trawl (OTB). 
Sampling month 
Fishing gear 
Number of 
fish collected 1 
Number of 
fish retained 2 
Number of 
otoliths removed 
September 1998 
SB 
681 
213 
137 
October 1998 
SB 
969 
278 
152 
November 1998 
OTB 
613 
348 
144 
December 1998 
OTB 
278 
176 
129 
January 1999 
SB 
1163 
429 
181 
February 1999 
SB 
726 
372 
201 
March 1999 
SB 
272 
120 
107 
April 1999 
SB 
325 
180 
116 
May 1999 
SB 
259 
118 
93 
June 1999 
SB 
625 
345 
151 
July 1999 
SB 
293 
186 
163 
August 1999 
SB 
254 
160 
120 
Total: 
6458 
2925 
1694 
1 The number of fish collected for measurement of total length. 
2 The number of fish retained in each 5-cm length class each month. 
stead. The mesh size of the codend of the beach seine 
and trawler net between stretched knots was 8 and 
20 mm, respectively. A total of 21 hauls were made in 
an effort to obtain as many representative samples as 
possible. 
The unsorted picarel catch was placed in baskets on 
board. At the end of each haul, one basket was ran¬ 
domly selected and the total length (TL) of at least 200 
individuals was measured to the nearest 0.5 cm. Ten 
fish from each 0.5-cm length class at each haul were 
retained. These fish were measured again for TL (to 
the nearest 0.1 cm), weighed (W, to the nearest 1.0 g) 
and sex was determined. After gross examination of 
the gonads, the fish were classified to maturity stages 
according to Nikolsky (1963). The number of speci¬ 
mens, by sex and maturity stage, was calculated on a 
monthly basis. From all the fish retained each month, 
both sagittae were removed from ten fish of each length 
class. However, only one (commonly the right sagitta) 
was thereafter analyzed. These otoliths were cleaned 
and kept dry. 
The whole otoliths were examined while immersed 
in water under reflected light against a dark back¬ 
ground, thus revealing alternating hyaline and opaque 
zones. The core of the otolith was opaque. Readings 
of hyaline zones (annuli) were made on the concave 
side and on both the posterior (postrostrum) and the 
anterior (rostrum) area of the otolith (Fig. 2). Devel¬ 
opment of the 1 st annulus was detected by following 
the progression of the smaller-fish length group and 
the structure of the respective otoliths in successive 
monthly samples (Campana, 2001). 
The total otolith diameter and diameter of the annuli 
were measured along the posterior-anterior axis under 
a microscope by using a micrometer scale (Fig. 2). The 
growth increments, annuli, used for age determination, 
should be “validated” (Beamish and McFarlane, 1983). 
Among the different methods for validating absolute 
age or periodicity of growth increment formation, those 
of marginal increment (MI) and otolith edge analysis 
are suitable for validation (Campana, 2001). These in¬ 
crements, if formed on a yearly basis, are represented 
by a sinusoidal curve when plotted against time. In 
particular, MI is well suited for determining the time 
of annulus formation (Campana, 2001). In the present 
work, both the percentages of otoliths with a hyaline 
or opaque zone on the edge and the Mis were studied 
monthly. The MI was defined as the width outside the 
last completed annulus, irrespective of the type of edge 
(opaque or hyaline) (Karlou-Riga, 2000). Therefore, the 
time when the value of MI next to each annulus was 
highest corresponded with the period of the next annu¬ 
lus formation. Monthly mean values of MI were com¬ 
pared by using one-way analysis of variance. 
The time of annulus formation, as well as the date 
of capture, the otolith edge (opaque or hyaline), and 
the spawning period were used to assign the fish to 
the appropriate age group (ICSEAF, 1986; Panfili et 
al., 2002). Each otolith was then characterized either 
by the n th incomplete annulus on the edge or the n th 
complete and opaque zone on the edge. Length dis¬ 
tributions of specimens classified according to otolith 
readings were constructed. Based on similarity indices, 
these distributions were clustered for the period from 
January to June (1 st semester) and from July to De¬ 
cember (2 nd semester) separately. Whether the clusters 
derived corresponded with specimens of the same age 
groups, was then examined, according to age interpre- 
