Fey et al.: Daily deposition of growth increments in sagittae and lapilli of larval Esox lucius 
307 
Figure S 
Images of otolith microstructure of a laboratory-reared larval northern pike f Esox lucius) 28 
days after hatching: (A) sagitta and (B) lapillus. The white dot indicates the hatching check 
(Hch) and arrows indicate stress checks that correspond with days of handling (i.e., sampling 
of larvae from aquaria). 
age estimates, it can also remove increments and lead 
to the underestimation of ages. The narrowest incre¬ 
ments at the otolith center, in particular, can be affect¬ 
ed during preparation (Secor et al. 3 ; Fey et al., 2005). 
The most important source of inaccuracy is interpre¬ 
tation error among investigators (Campana, 1990). In¬ 
terpretation depends on the experience of the reader 
and clarity of the otoliths, the latter of which may be 
affected by increment width. The increment width in 
turn is not only species specific, but it can also be af¬ 
fected by factors such as temperature, feeding condi¬ 
tions, and somatic growth (Campana, 1983; Folkvord 
et al., 2000; Fukuda et al., 2009). Although Wang and 
Eckmann (1992) reported that increment deposition 
started 2-3 dah, they referred to fish held in waters 
at a temperature of 10°C. For fish held at a temper¬ 
ature of 16°C, the delay in first increment formation 
described by Wang and Eckmann (1992) was only 1-2 
days. It is possible that the clarity of the microstruc¬ 
ture in the otoliths from the present study conducted 
with fish held at a temperature of 18°C was greater 
and therefore first increment formation was found to 
occur at hatching. Thus, it can be suggested that incre¬ 
ment formation starts at hatching but, depending on 
the temperature at hatching, some age underestima¬ 
tion may occur because of problems with recognition 
of increments that are very narrow. Considering the 
3 Secor, D. H., J. M. Dean, and E. H. Laban. 1991. Manual 
for otolith removal and preparation for microstructural ex¬ 
amination, 85 p. Electric Power Res. Inst., Palo Alto, CA, 
and Belle W. Baruch Inst. Mar. Biol. Coast. Res., Clemson, 
SC. 
above conditions, the importance of repeated evalua¬ 
tions of the timing of increment deposition and first 
increment formation is emphasized. 
The relationship of otolith size to fish size was linear 
for lapilli and exponential for sagittae. Rapid differen¬ 
tiation in otolith size between sagittae and lapilli, i.e., 
when daily increments in sagittae become significantly 
wider than those in lapilli, occurred when fish growth 
in weight increased. Therefore, if larvae larger than 17 
mm SL are used for age estimates, wider increments in 
sagittal otoliths would be more suitable for measure¬ 
ments of increment width otolith back-calculations of 
growth. Fey et al. (2005) made the opposite recommen¬ 
dation (i.e., lapilli are better for back-calculations of 
growth) for larval spot. Although sagittae had wider 
increments than lapilli, their irregular pattern made 
width measurements along one radius impossible. 
When aging itself was concerned, in the present 
study sagittae provided slightly better visibility and 
clarity of increments than lapilli in fish of up to 25 mm 
SL. In larger specimens (25-33 mm SL), the useful¬ 
ness of sagittae decreased, and that of lapilli increased 
slightly. In sagittae, difficulties resulted particularly 
from wide and relatively irregular increment formation 
with the result that many subdaily structures that can 
sometimes be difficult to distinguish from “real” incre¬ 
ments. In lapilli, when optimal focus of the microscope 
is used, increment patterns are seen as more regular 
and are seen as daily increments, but when the focus 
is changed from that optimum, the daily increments 
have a tendency to merge and result in an underes¬ 
timation of real age. Notice, however, that the qual¬ 
ity of otolith microstructures in laboratory-reared fish 
