Fey et at: Daily deposition of growth increments in sagittae and lapilli of larval Esox lucius 
305 
Table 1 
Number of increments deposited on sagittae and lapilli of known-age larval north¬ 
ern pike (Esox lucius) reared in a laboratory in 2015 at the Center of Aquaculture 
and Ecological Engineering, University of Warmia and Mazury in Olsztyn, Poland. 
n=the number otoliths analyzed (left or right otolith was used depending on which 
provided a clearer microstructure). 
Number of days 
after hatching 
Sagittae 
Lapilli 
n 
Mean number 
of increments 
SD 
n 
Mean number 
of increments 
SD 
7 
17 
7.6 
0.88 
16 
7.4 
0.97 
14 
21 
14.4 
0.65 
18 
13.9 
1.47 
22 
22 
22.0 
0.97 
20 
21.6 
1.51 
28 
18 
27.8 
1.41 
20 
28.1 
1.26 
slopes of these two regressions were not statistically 
different from 1 (ANCOVA, P> 0.05). The intercept of 
these 2 regression lines was not statistically different 
from 0 (ANCOVA, P>0.05), indicating that the first in¬ 
crement was formed at hatching (when the tempera¬ 
ture is 18°C). The mean number of increments counted 
in relation to real age is presented in Table 1. The SD 
for lapilli was higher than that of sagittae for all age 
classes (7, 14, and 22 days) except for the age of 28 
days. 
A stress check (i.e., a distinct dark band formed be¬ 
tween regular increments) was formed on most (79%) 
of the sagittae on the days of handling, especially on 
day 7, and less frequently on days 14 and 22. Stress 
checks were less commonly observed on lapilli (18%). 
The mean number of increments between the sagittal 
edge and stress check formed on day 7 was 6.7 (SD 
0.49, n=14; for 14-day-old fish), 14.8 (SD 0.67, n=17; for 
22-day-old fish), and 21.4 (SD 0.51, n= 19; for 28-day- 
old fish). Therefore, the number of increments between 
stress check and otolith edge corresponded to the num¬ 
ber of days elapsed after the formation of the stress 
check. The lower SD for the counts from the stress 
check to the otolith edge described above compared 
with the SD for the counts from the otolith center to 
the edge (Table 1) indicate that the source of error in 
increment counts is related largely to the analysis of 
the area close to the otolith center. 
The relationship of otolith size to fish size was lin¬ 
ear for lapilli (tz= 115) and exponential for sagittae 
(n=115) (Fig. 3). The rapid change in otolith size be¬ 
tween sagittae and lapilli, i.e., when daily increments 
on sagittae become significantly wider than those on 
lapilli, occurred around an SL of 17 mm and an age 
of 9-10 days (Fig. 3). This SL and age corresponded 
to the starting point of increase in fish weight (Fig. 4). 
The clarity of the otolith microstructures in both 
sagittae (Fig. 5A) and lapilli (Fig. 5B) varied among 
all individuals from good, with easily distinguishable 
increments, to unclear and difficult to read. Generally, 
the lowest confidence in increment recognition was for 
the smallest fish (< 17 mm SL), for both sagittae and la¬ 
pilli. Differences between these two otolith types were 
evident when fish of different sizes were compared. 
Sagittae are preferable for northern pike larvae of up 
to approximately 25 mm, but lapilli are more suitable 
for larger specimens. 
Discussion 
The number of increments on otoliths, both sagittae 
and lapilli, corresponded in this study to the known 
age of the larvae. These results can be compared with 
those of Wang and Eckmann (1992) who also confirmed 
the daily deposition of increments on the lapilli of lar¬ 
val northern pike. No published data on increment for¬ 
mation exist for sagittae that could be used for com¬ 
parison with data from the present results. The results 
obtained in the current study at a relatively high tem¬ 
perature of 18°C should be supplemented in the future 
with data obtained at lower temperatures that could 
potentially cause non-daily increment formation or the 
formation of increments that are too narrow to be iden¬ 
tified with a transmitted light microscope. Wang and 
Eckmann (1992) analyzed increment development in 
lapilli at 16°C and 10°C, but it would still be valuable 
to analyze samples at temperatures below 10°C. Such 
temperatures that can result in age underestimation 
because the increments are too narrow to record have 
been previously reported for Atlantic herring (Clupea 
harengus ) and turbot (Scophthalmus maximus ) (Geffen, 
1982) and Arctic charr (Salvelinus alpinus) (Radtke 
and Fey, 1996). 
Depending on the species, the first increment oc¬ 
curs either at hatching, a day after hatching, or af¬ 
ter the first feeding and yolk-sac absorption (Jones, 
1986; Thorrold and Hare, 2002). In the present study 
on northern pike, conducted at a temperature of 18°C, 
increment deposition started at hatching in both the 
