Rocha-Olivares: Age, growth, and mortality of Lutjanus peru 
571 
the first 10-12 years (Fig. 5). The very few males 
sampled beyond the ascending limb of the curve may 
have prevented a more constrained curve fit in this 
region. 
The results of this and other studies made in 
Mexico and Costa Rica reveal the following patterns: 
1) scale-based studies, using traditional methods for 
growth parameter estimation, yield higher growth 
rates resulting from age underestimation if large 
fishes are involved (Table 5, reference A) but not so 
if only small fishes are included (Table 5, reference 
B); 2) small asymptotic lengths may result from a 
restricted length range (Fig. 7B); 3) length-frequency 
based methods for growth parameter estimation yield 
fewer age classes and unrealistic growth rates, which 
have been observed elsewhere (Grimes, 1987; see also 
Rocha Olivares, 1991, for examples of unpublished 
data). It is possible that part of the observed differ- 
ences in the growth parameters of L. peru between 
peninsular and continental populations are “true”; 
however the diversity of fishing gears and selectivities 
used in different states (Aguilar Salazar, 1986) and the 
poor effort invested in systematic biological data col- 
lection make interpopulation comparisons difficult. 
The presence of an individual of ca. 100-cm TL rep- 
resents a length record for the species. The sectioned- 
otolith age of this fish (31 years) also consti- 
tutes a record for the maximum age of an 
eastern Pacific lutjanid, and adds L. peru to 
the list of long-lived snappers (i.e. exceeding 
30 years of age): L. bohar (38 years ), L. adetti 
(37 years), L. sebae (35 years) (Loubens, 
1980), L. malabaricus (46 years) (Mathews 
and Samuels, 1985), and L. quinquelineatus 
(32 years) (Newman et al., 1996). When com- 
pared with the growth trends observed in the 
family, growth performance of L. peru (0=1.74) 
falls above the mean value for lutjanids ( 1.65) 
but within the expected range (1.08-2.15; 
SB=0.35) (Munro, 1983). Furthermore, the 
combination of L x = 97.32 cm and K = 0.1111/ 
yr of the Baja California population of Pacific 
red snapper is very close to the functional rela- 
tionship of log 10 (/O versus log 10 (LJ reported by 
Manooch ( 1987) for snappers (Lutjanidae) and 
groupers (Serranidae). 
Mortality 
In this study, estimates of M were restricted 
to the use of empirical relationships between 
M and K because no effort data were avail- 
able and no unfished areas could be sampled. 
It has been repeatedly observed that these 
two parameters are inversely related across 
a large number of fish taxa (Pauly, 1980). 
Hoenig’s ( 1983) empirical formula yielded an 
unrealistically low estimate of mortality, as 
it probably would in other long-lived snap- 
pers. When compared to total mortality rates 
(Z), estimates of M are consistent and yield 
much lower exploitation rates and F/M ra- 
tios than those reported for other lutjanids 
(cf. Table 8.2 in Ralston, 1987), probably as 
the result of an overestimation of M by the 
empirical formulae. 
Male predominance in population sex ra- 
tios is not uncommon among lutjanids (cf. 
females 
Z = 0.282 ± 0.025 
n = 587 r 2 = 0.900 
1000 
100 
10 
1 
0 
o' 
males 
Z = 0.366 ±0.030 
o 
o 
n = 691 
r2 = 
0.923 
all individuals 
Z = 0.345 ± 0.032 
n = 5,323 r 2 = 0.878 
10 
15 20 
Age (years) 
25 
30 
Figure 8 
Catch curves and estimates of total mortality of Lutjanus peru fe- 
males, males and sexes combined, caught by hook and line off the 
southeast coast of Baja California Sur. The age composition was 
computed with an age-length key and the length composition of the 
massive and biological samples. Total mortality rates Z (per year ± 
standard error) are presented for females, males, and the entire 
population. Black dots represent data points included in the calcu- 
lation of the slope (solid line). 
