6 
Fishery Bulletin 95(1 ), 1997 
0.20 
0.15 
=§ 0.10 - 
0.05 
40 
0.20 r 
0.18 - 
E 
g 
® 0.16 - 
0.14 - 
0.12 - 
45 50 55 
Age at metamorphosis (d) 
60 
B 
0.10 
5.5 6.0 6,5 7.0 7.5 
Length at 30 d (mm) 
8.0 
Figure 4 
Larval and juvenile growth for fish reared individu- 
ally in both periods. (A) Growth rate from week 0 to 
week 3 of the juvenile period versus age at meta- 
morphosis. (B) Growth rate from week 0 to week 3 
of the juvenile period versus length at 30 d, an in- 
dex of larval growth rate. 
0.20 r 
f 0.10 - 
<D 
0.05 1 1 1 
30 40 50 60 
Age at metamorphosis (d) 
0.22 r 
0.20 - 
E 0.18 - 
E 
CD 
5 0.16 - 
sz 
O 0.14 - 
CT> 
a) 
| 0.12 - 
z> 
0.10 - 
0.08 - 
0.05 0.06 0.07 0.08 0.09 0.10 0.11 0.12 0.13 
Average larval growth rate (mm/d) 
Figure 5 
Larval and juvenile growth for fish reared individu- 
ally (n= 52) as juveniles. (A) Growth rate during 
weeks 1—4 of the juvenile period versus age at meta- 
morphosis. (B) Growth rate during weeks 1-4 of the 
juvenile period versus average larval growth rate. 
Discussion 
Our data on growth dynamics of larval fishes show 
that size at age is highly variable during the larval 
period. Patterns in CV’s for size at age demonstrate 
that most of the variation upon which selection can 
act is found during the early to mid phase of the lar- 
val period. Chambers et al. (1988), who analyzed 
average growth rates of larvae reared in groups, also 
found that CV’s for size at age increased from hatch- 
ing to a peak of 0.135 at 28 d and subsequently de- 
clined as metamorphosis approached. In this study, 
most larval growth occurred during the first 30 d. 
Individual larvae that grew most rapidly and reached 
the largest size at about day 30, midway through the 
larval period, metamorphosed at the youngest ages. 
Travis (1981), who reared anuran larvae individu- 
ally, also reported that age at metamorphosis was 
inversely related to size midway through the larval 
period. Despite the nonlinear growth observed, the 
length at 30 d and the average larval growth rate 
were positively correlated. Thus, the general conclu- 
sion (derived from those studies where average lar- 
val growth was used and larval growth was assumed 
to be linear) that rapid growth reduces the duration 
of the larval period is supported (Chambers and 
Leggett, 1987; Chambers et al., 1988; Bertram et al., 
1993). 
