410 
Trace metals in four structures of fish and 
their use for estimates of stock structure 
Bronwyn M. Gillanders 
School of Biological Sciences A08 
University of Sydney 
New South Wales 2006, Australia 
Present address: Department of Environmental Biology 
University of Adelaide, South Australia 5005, Australia 
E-mail:bronwyn. gillanders@adelaide.edu.au 
Abstract— Trace elements in calcified 
tissues have been suggested as one of 
the most powerful means for stock dis- 
crimination yet developed. The struc- 
ture of choice for determining elemental 
composition or fingerprints is the oto- 
lith, although other structures also 
incorporate trace elements into their 
matrix. The aim of this study was to 
compare the elemental fingerprints of 
four structures (otoliths, scales, eye 
lenses, and spines) of a territorial reef 
fish to determine whether there were 
correlations between otoliths and each 
of the other structures. Elemental fin- 
gerprints of juvenile (<3 years of age) 
and adult fish (which may reach a 
maximum age of 37 years) were also 
compared for each structure to deter- 
mine whether there may be differences 
between size classes of fish. All struc- 
tures were analyzed by solution-based 
inductively coupled plasma-mass spec- 
trometry (ICP-MS). Otoliths, scales, 
spines, and eye lenses differed in com- 
position. Calcium dominated otoliths, 
scales, and spines but was not detected 
in eye lenses. Some elements, for exam- 
ple barium, showed significant correla- 
tions between the otolith data and that 
of scales and spines of both juvenile and 
adult fish. A multivariate test of matrix 
correspondence (Mantel's test) detected 
significant relationships between the 
otolith data and the data matrices for 
each of scales and spines of both juve- 
nile and adult fish. Significant rela- 
tionships between the otolith data and 
the eye lens data were detected only 
for juvenile fish. Significant differ- 
ences were also found between juvenile 
and adult fish for all structures. The 
BIOENV multivariate analyses showed 
that the highest rank correlation was 
found between the otolith data and the 
scale or spine data for both juvenile 
and adult fish. These data suggest that 
the use of scales and spines may pro- 
vide a nonlethal alternative to the use 
of otoliths for future stock discrimina- 
tion studies. 
Manuscript accepted 16 February 2001. 
Fish. Bull. 99:410-419 (2001). 
Studies of the population dynamics of 
marine fishes depend on the ability to 
distinguish separate “stocks” that may 
make up the total population of a single 
fish species. Although much of fisheries 
management assumes that a single pop- 
ulation is being monitored, measures of 
growth, natural mortality, recruitment, 
and reproduction are often made for the 
total population because data on move- 
ment or stock separation or fidelity (or 
the combination of all three) are inad- 
equate to allow separation of stocks. 
A variety of methods exist for iden- 
tifying stocks (e.g. population param- 
eters, capture-mark-recapture studies, 
physiological and behavioral characters, 
morphometric and meristic characters, 
calcareous characters, cytogenic charac- 
ters, and biochemical characters; Ihssen 
et al., 1981), but few methods can pro- 
vide a reliable measure of stock identity. 
Genetic techniques are frequently used 
but fail to differentiate stocks because a 
small amount of larval or adult mixing 
among populations makes differences 
undetectable (Hartl and Clark, 1989). 
Recently, many studies have focused 
on the use of “elemental fingerprints” 
( sensu Campana et al., 1994), or the ele- 
mental composition of the otoliths, as a 
measure of stock identity. 
Initial microchemistry studies tend- 
ed to focus on diadromous species and 
distinguished between freshwater and 
marine life history phases of fish (e.g. 
Bagenal et al., 1973; Belanger et al., 
1987; Kalish, 1990; Coutant and Chen, 
1993; Rieman et al., 1994). The elec- 
tron microprobe was widely used for 
such studies but because of the “typi- 
cal” concentrations of Ca, Na, Sr, K, S, 
and Cl found in otoliths, electron probes 
were effectively limited to detecting 
these six elements. Recently, there have 
been dramatic advances in analytical 
techniques and studies now indicate 
that other elements also exist in otoliths 
(e.g. Edmonds et al., 1989, 1991, 1992; 
Sie and Thresher, 1992), many of which 
are thought to be reflective of the envi- 
ronment (Fowler et al., 1995). Stock dis- 
crimination capabilities have improved 
with the sensitivity of recent instrumen- 
tation and the inductively coupled plas- 
ma-mass spectrometer (ICP-MS) has 
now become the instrument of choice 
for simultaneously quantifying the con- 
centration of multiple elements and iso- 
topes (Houk, 1986; Date, 1991). 
ICP-MS has recently been used for 
assays of various fish tissues including 
otoliths (e.g. Edmonds et al., 1991; Cam- 
pana and Gagne, 1995; Campana et al., 
1994, 1995; Dove et al., 1996; Gillanders 
and Kingsford, 1996), scales (e.g. Cout- 
ant and Chen, 1993, Wells et al., 2000), 
eye lenses (e.g. Dove and Kingsford, 
1998) and soft tissues (e.g. Beauchemin 
et al., 1988; Ishii et al., 1991; Hellou et 
al., 1992). 
Sample-specific differences in ele- 
mental concentrations in tissues other 
than otoliths have also been reported 
(e.g. vertebrae — Mulligan et al., 1983; 
Behrens Yamada et al., 1987; scales — 
Bagenal et al., 1973; Lapi and Mulli- 
gan, 1981; eye lenses — Dove and Kings- 
ford, 1998; soft tissue — Hellou et al., 
1992). Such sample-specific differences 
in trace elements provide evidence for 
the nonmixing and segregation of pop- 
ulations. Otoliths are an ideal natural 
marker of elemental composition be- 
cause they grow throughout the life of 
the fish and once the elements are de- 
