Favero et al: Variation in egg size of Engraulis anchoita in the southeastern Brazilian Bight 
525 
Table 4 
Mean and standard deviation values of temperature (Temp. °C) and salinity (Sal.) and results from analysis of variance 
(ANOVA) and a posteriori paired Tukey’s honestly significant difference (HSD) test that compared data from the 3 areas 
sampled (areas 1, 2, and 3) in the southeastern Brazilian Bight during the 5 years studied (December 1975, January 1988, 
December 1990, December 1991, and January 1993). N is the number of sampling stations considered. The 3 areas were 
from Cape Frio to Sao Sebastiao Island (area 1), from Sao Sebastiao Island to Paranagua Bay (area 2), and from Paranagua 
Bay to Cape Santa Marta Grande (area 3). 
Area 
ANOVA 
Tukey’s HSD 
1 
2 
3 
F 
P 
2 vs. 1 
3 vs. 1 
3 vs.; 
Dec 75 
N 
53 
44 
43 
Temp. 
22.06 (2.17) 
23.76 (0.97) 
23.47 (0.87) 
17.72 
<0.01 
0 
0 
0.65 
Sal. 
35.99 (0.67) 
35.30 (0.76) 
35.49 (0.94) 
9.76 
0 
0 
0 
0.51 
Jan 88 
N 
24 
35 
19 
Temp. 
24.16 (1.37) 
26.40 (1.34) 
25.40 (2.93) 
10.41 
0 
0 
0.08 
0.15 
Sal. 
35.41 (0.33) 
35.22 (0.60) 
35.14 (0.73) 
1.27 
0.29 
Dec 90 
N 
25 
40 
24 
Temp. 
22.71 (1.92) 
24.81 (0.56) 
24.12 (2.78) 
10.44 
<0.01 
0 
0.02 
0.31 
Sal. 
35.95 (0.63) 
34.84 (0.63) 
34.74 (0.86) 
24.37 
<0.01 
0 
0 
0.86 
Dec 91 
N 
33 
48 
29 
Temp. 
20.34 (2.94) 
24.47 (1.91) 
24.15 (1.86) 
36.11 
<0.01 
0 
0 
0.81 
Sal. 
36.25 (0.70) 
35.49 (0.68) 
35.54 (0.52) 
24.11 
<0.01 
0 
0 
0.06 
Jan 93 
N 
32 
47 
29 
Temp. 
21.33 (3.33) 
25.13 (1.79) 
24.55 (2.82) 
21.11 
<0.01 
0 
0 
0.62 
Sal. 
36.01 (0.55) 
35.12 (0.59) 
34.91 (0.47) 
36 
<0.01 
0 
0 
0.25 
growth, seasonal changes in the age structure of the 
spawning populations, changes during oogenesis that 
are correlated with some environmental conditions 
(e.g., temperature, photoperiod, and food supply) (Blax- 
ter and Hunter, 1982; Chambers, 1997) or a co-occur¬ 
rence of these factors, as suggested by Llanos-Rivera 
and Castro (2004) for eggs of anchoveta ( Engraulis 
ringens) off the Chilean coast. Ciechomski (1973), who 
reported that the volume of eggs of Argentine anchoita 
off the Argentinian and Uruguayan coast decreased 
23% throughout the spawning season, probably ow¬ 
ing to multiple spawning or length composition of the 
spawners (or both) correlated with the differences in 
growth rate in relation to temperature. 
Egg size increases with latitude, as shown by Lla¬ 
nos-Rivera and Castro (2004) who studied anchoveta 
from Chile and by Ciechomski (1973), who showed 
that egg size of Argentine anchoita increased from the 
north (35-36°S) to the south (40-41°S) and that this 
increase was not correlated with temperature or salin¬ 
ity because both parameters were similar throughout 
the area sampled. Egg size typically varies inversely 
with temperature and salinity, as has been shown for 
European anchovy (Engraulis encrasicolus ) by Demir 
(1974) and for Argentine anchoita by Phonlor (1984). In 
our results, area 1, which usually exhibited the lowest 
temperature and highest salinity values, was the area 
that had the smallest eggs in 4 out of 5 years sampled. 
The exception occurred in January 1988, when the 
largest eggs were collected in that area. January 1988 
was also a period of a weak SACW intrusion at Cape 
Frio (Fig. 6), which is reflected by greater temperature 
variability (SD, Table 4). Temperatures and salinities 
were similar throughout the 3 areas during this period 
in 1988. Conversely, in area 3 the highest tempera¬ 
ture and lowest salinity values were found, and this 
area usually had the largest eggs. It is important to 
highlight that area 3 encompasses most of the bonae- 
rense stock, whereas area 1 contains only the Brazil¬ 
ian Southeastern Bight stock. According to Carvalho 
and Gastello (2013), who analyzed fewer samples than 
we did, the boundary between the Brazilian South¬ 
eastern Bight stock and the bonaerense stock (~27°S) 
is not well defined. Because the geographic limits of 
each stock are not clear, and the eggs from area 2 ex¬ 
hibited no consistent pattern, with the largest eggs in 
some samples, or with egg sizes exhibiting no signifi¬ 
cant differences between area 1 or area 3, area 2 was 
considered a transition zone, and will not be considered 
further in this discussion. 
Individual Argentine anchoita from southeastern 
Brazil (22-27°S) were shorter at comparable ages than 
