210
L. Lopiano et al.
Materials and Methods
The study was carried out in the Stagnone di
Marsala (western Sicily), a shallow coastal sound
where D. puntazzo and S. salpa are transient
species. Their larger recruitment, as juveniles, is
included between last winter and early spring
(Sara et al. 1996). Samples were collected in April
1997, in the central part of the Stagnone di
Marsala. This site ranged in depth between 0.5 m
and 1.0 m and was characterised by submerged
vegetation composed of Cymodocea nodosa and
Caulerpa prolifera. Fish were caught using a
beach seine (estimated opening wide = 6 m;
length = 15 m; mesh size = 10 mm), every 4
hours over a 24 h cycle. then narcotised in chloroform and fixed in a (10%) formalin buffered
water solution. In laboratory, the standard length
(SL) and the body wet weight (BWW) of each
fish were measured to the nearest 0.1 mm and 0.1
g. respectively. In total, 86 specimens of D. puntazzo (mean SL = 31.9 ± 6.6 mm; mean BWW =
1.0 ± 0.5 g) and 100 specimens of S. salpa (mean
SL = 32.3 ± 2.9 mm; mean BWW = 0.6 ± 0.2 g)
were examined. At each sampling time, about 10
specimens of both species were used in the qualitative analysis of the diet, the remaining part
being used in the study of the gastric evacuation
pattern and the calculation of daily food ration.
For the qualitative analysis, stomachs were dissected under the stereomicroscope and the main
items of the diet classified. The contribution to
the total volume due to each item was calculated
for each stomach using the subjective method
proposed by Hellawell and Abel (1971). The principal simple indices such as the Coefficient of
Vacuity (% CV), frequency occurrence (% Fr)
(Kennedy and Fitzmaurice 1972) and the percentage in volume of prey (% V) were calculated.
Among the composite indices only the Feeding
Index (IA = % Fr * % V I 100), according to
Lauazanne (1975) was calculated. For the quantitative dietary analysis, the relative weights of the
dry eviscerated fish (BDW, g) and of the dry gastric content (DWsc, g) were measured after desiccation (60°C to constant weight). The latter was
expressed as g dry weight g-l BDW (Arrhenius
and Hansson 1994). For both species, the feeding
rhythm was derived plotting the mean corrected
DW sc at different sampling hours. The values of
gastric contents at different times, expressed as
corrected DW sc were also compared using the
Wilcoxon test (Sokal and Rohlf 1981). For both
species evacuation rates (R) were calculated from
the decreased stomach contents during nonfeeding periods (i.e., assuming that no feeding
occurred in darkness). Food consumption for
each sampling interval (C,) was calculated following Pennington (1985). Total consumption
over an entire diel cycle was obtained by summing the positive values of C, (Ruggerone 1989).
Niche breadth of both species was calculated
according to Levins (Krebs 1989). Niche overlap
between juveniles of the two species, based on %
V values, was calculated using simplified
Morisita's index, C H , (Krebs 1989).
Results
At four of the six sampling times stomachs
appeared to be full (%CV = 0), in both species; at
12 pm % CV was equal to 22 % for D. puntazzo,
and 10 % for S. salpa. All specimens caught in the
early morning (4 am) presented empty stomachs
(% CV = 100). The analysis of mean corrected
DWsc, revealed, for both species, a unimodal diel
feeding rhythm (Figure). Dipladus puntazzo
feeding phase began between 8 and 12 am (8 am
vs 12 am P < 0.01); it continued during the following hours (12 am vs 4 pm P > 0.05; 4 pm vs 8
pm P > 0.05), and ended from 8 to 12 pm (8 pm
vs 12 pm P < 0.05), without any real feeding peak.
A similar feeding rhythm was observed for S.
salpa juveniles, whose feeding phase started
between 12 am and 4 pm (12 am vs 4 pm P <
0.01) and ended from 8 to 12 pm (8 pm vs 12 pm
P < 0.01).
The values of % Fr and % V in the different
sampling times are reported in Table 1, for D.
puntazzo, and in Table 2 for S. salpa.
The analysis of Feeding Index (Table 3)
showed that, for D. puntazzo juveniles,
Amphipoda were "dominant prey" (sensu
Lauazanne 1975; 50 < IA < 100) at 8 pm and
«fundamental prey" (25 < IA < 50) at 12 pm.
Ascidiacea larvae were «fundamental prey" at the
beginning of the feeding phase (IA = 45.3 at 12
am and 42.1 at 4 pm). Copepoda, despite their
high % Fr values, were ''not negligible prey" (10 <
IA < 25) only at 8 pm, due to their low volume
contribution to the stomach content. All other
food items represented «secondary prey" (0 < IA
< 10). For S. salpa, Phycophyta were "dominant
prey" at most sampling hours, except at 4 pm,
when they were decreased to the level of "not
L. Lopiano et al.
Materials and Methods
The study was carried out in the Stagnone di
Marsala (western Sicily), a shallow coastal sound
where D. puntazzo and S. salpa are transient
species. Their larger recruitment, as juveniles, is
included between last winter and early spring
(Sara et al. 1996). Samples were collected in April
1997, in the central part of the Stagnone di
Marsala. This site ranged in depth between 0.5 m
and 1.0 m and was characterised by submerged
vegetation composed of Cymodocea nodosa and
Caulerpa prolifera. Fish were caught using a
beach seine (estimated opening wide = 6 m;
length = 15 m; mesh size = 10 mm), every 4
hours over a 24 h cycle. then narcotised in chloroform and fixed in a (10%) formalin buffered
water solution. In laboratory, the standard length
(SL) and the body wet weight (BWW) of each
fish were measured to the nearest 0.1 mm and 0.1
g. respectively. In total, 86 specimens of D. puntazzo (mean SL = 31.9 ± 6.6 mm; mean BWW =
1.0 ± 0.5 g) and 100 specimens of S. salpa (mean
SL = 32.3 ± 2.9 mm; mean BWW = 0.6 ± 0.2 g)
were examined. At each sampling time, about 10
specimens of both species were used in the qualitative analysis of the diet, the remaining part
being used in the study of the gastric evacuation
pattern and the calculation of daily food ration.
For the qualitative analysis, stomachs were dissected under the stereomicroscope and the main
items of the diet classified. The contribution to
the total volume due to each item was calculated
for each stomach using the subjective method
proposed by Hellawell and Abel (1971). The principal simple indices such as the Coefficient of
Vacuity (% CV), frequency occurrence (% Fr)
(Kennedy and Fitzmaurice 1972) and the percentage in volume of prey (% V) were calculated.
Among the composite indices only the Feeding
Index (IA = % Fr * % V I 100), according to
Lauazanne (1975) was calculated. For the quantitative dietary analysis, the relative weights of the
dry eviscerated fish (BDW, g) and of the dry gastric content (DWsc, g) were measured after desiccation (60°C to constant weight). The latter was
expressed as g dry weight g-l BDW (Arrhenius
and Hansson 1994). For both species, the feeding
rhythm was derived plotting the mean corrected
DW sc at different sampling hours. The values of
gastric contents at different times, expressed as
corrected DW sc were also compared using the
Wilcoxon test (Sokal and Rohlf 1981). For both
species evacuation rates (R) were calculated from
the decreased stomach contents during nonfeeding periods (i.e., assuming that no feeding
occurred in darkness). Food consumption for
each sampling interval (C,) was calculated following Pennington (1985). Total consumption
over an entire diel cycle was obtained by summing the positive values of C, (Ruggerone 1989).
Niche breadth of both species was calculated
according to Levins (Krebs 1989). Niche overlap
between juveniles of the two species, based on %
V values, was calculated using simplified
Morisita's index, C H , (Krebs 1989).
Results
At four of the six sampling times stomachs
appeared to be full (%CV = 0), in both species; at
12 pm % CV was equal to 22 % for D. puntazzo,
and 10 % for S. salpa. All specimens caught in the
early morning (4 am) presented empty stomachs
(% CV = 100). The analysis of mean corrected
DWsc, revealed, for both species, a unimodal diel
feeding rhythm (Figure). Dipladus puntazzo
feeding phase began between 8 and 12 am (8 am
vs 12 am P < 0.01); it continued during the following hours (12 am vs 4 pm P > 0.05; 4 pm vs 8
pm P > 0.05), and ended from 8 to 12 pm (8 pm
vs 12 pm P < 0.05), without any real feeding peak.
A similar feeding rhythm was observed for S.
salpa juveniles, whose feeding phase started
between 12 am and 4 pm (12 am vs 4 pm P <
0.01) and ended from 8 to 12 pm (8 pm vs 12 pm
P < 0.01).
The values of % Fr and % V in the different
sampling times are reported in Table 1, for D.
puntazzo, and in Table 2 for S. salpa.
The analysis of Feeding Index (Table 3)
showed that, for D. puntazzo juveniles,
Amphipoda were "dominant prey" (sensu
Lauazanne 1975; 50 < IA < 100) at 8 pm and
«fundamental prey" (25 < IA < 50) at 12 pm.
Ascidiacea larvae were «fundamental prey" at the
beginning of the feeding phase (IA = 45.3 at 12
am and 42.1 at 4 pm). Copepoda, despite their
high % Fr values, were ''not negligible prey" (10 <
IA < 25) only at 8 pm, due to their low volume
contribution to the stomach content. All other
food items represented «secondary prey" (0 < IA
< 10). For S. salpa, Phycophyta were "dominant
prey" at most sampling hours, except at 4 pm,
when they were decreased to the level of "not