Begg et al.: Use of otolith morphometries for identification of stocks of Melanogrommus aeglefinus 
3 
Table 1 
Details of haddock samples used in analysis of internal otolith morphometries for stock identification. 
Region 
Date captured 
Age group 
(years) 
Length range 
(cm) 
n 
Eastern Georges Bank (EGB) 
10-11 Apr 1995 
1 
21-28 
7 
2 
31-37 
10 
21-23 Apr 1996 
1 
21-32 
6 
3 
41-53 
9 
4 
45-54 
9 
6-8 Apr 1997 
2 
38-45 
18 
3 
40-47 
10 
4 
41-58 
16 
5 
49-62 
10 
Western Georges Bank (WGB) 
13 Apr 1995 
1 
24-31 
11 
16-20 Apr 1996 
2 
31-47 
9 
3 
40-55 
9 
25 Mar-23 Apr 1997 
2 
36-44 
17 
3 
39-60 
44 
4 
47-67 
37 
5 
50-73 
10 
tions. Although, linear morphometries and shape analysis 
of “external” or whole otoliths have been used for fish 
stock identification, there has been little use of “internal” 
otolith morphometries, probably because of the time, ex- 
pense, and potential difficulties in obtaining consistent 
otolith sections. However, most fisheries agencies now use 
a standardized procedure for sectioning otoliths. These 
standardized procedures enable consistent and compara- 
ble otolith sections to be obtained, providing a consistent, 
rapid, and readily accessible structure that can be exam- 
ined for stock identification. 
We investigated the feasibility of using internal otolith 
morphometries for fish stock identification, by considering 
Georges Bank haddock as a representative case study. For 
the purposes of our study, we considered a stock, or spawn- 
ing component, as a semidiscrete, self-reproducing group 
of fish with definable morphometric characteristics that 
are assumed to be homogeneous for management purpos- 
es (Begg and Waldman, 1999). Variation in these char- 
acteristics is assumed to be evidence that distinct geo- 
graphic regions are partially occupied throughout the life 
history of the fish, thereby providing a phenotypic basis 
for stock identification (Ihssen et al., 1981). 
Materials and methods 
Sample collection 
Haddock samples were collected in 1995, 1996, and 
1997 during spring Northeast Fisheries Science Center 
(NEFSC) stratified random bottom trawl surveys, when 
the fish were assumed to be on or near their spawning 
grounds. Most of the adults (ages 2+) sampled were in 
spawning condition. Samples were collected from survey 
stations throughout eastern (EGB) and western Georges 
Bank (WGB) (Fig. 1; Table 1). At sea, haddock samples 
were measured (fork length [FL], to the nearest cm), sex 
and maturity were determined by macroscopic examina- 
tion of the gonads, and sagittal otolith pairs were removed 
from each sample. In the laboratory, one otolith from each 
pair was sectioned and assigned an age by following stan- 
dard methods for northwest Atlantic species (Pentilla and 
Dery, 1988). Standardized sectioning procedures ensured 
consistent and comparable otolith sections for morphomet- 
ric measurements and subsequent statistical analyses. 
Internal otolith morphometries 
Internal otolith morphometries were obtained from each 
otolith section by using the OPTIMAS™ (version 6.2) 
image analysis system (Media Cybernetics, 1996). All mea- 
surements were taken at a magnification of 15x. The 
perimeter of each otolith section was traced in a counter- 
clockwise direction to allow morphometries to be calcu- 
lated for each structure (shape). Six linear morphometries 
(length, width, Al, HI, A2, and H 2) were measured for 
each sample, where A 1 is the growth increment to the first 
annulus, HI is the width of the hyaline band of the first 
annulus, A2 is the growth increment to the second annu- 
lus, and H2 is the width of the hyaline band of the second 
annulus (Fig. 2). In addition, four shape variables (area, 
perimeter, circularity, and rectangularity) were measured 
for each sample. Circularity was defined as the perimeter 
of the otolith section squared divided by its area, and rect- 
angularity was defined as the otolith section area divided 
