8 
Fishery Bulletin 99(1 ) 
0.4- 
-0.4- 
- 1 . 2 - 
- 2 . 0 - 
0.6 
■ ^"oo o 
I °o 
1 gpO a 
o° ° 0 
°°io °c 
oo 0?) 
o 
1.4 
41 
2.2 
3.0 
1 .6 1 
1.5- 
1.4- 
1.3 
1.2 
0.6 
■ 
oft) O 
o o o 
o o 
o° 0 o 
o ° °o 
° °°o 
1.4 
41 
2.2 
3.0 
-2.4-1 
-3.0- 
-3.6- 
-4.2- 
■o f " 
o ° o 
° o o 
jpo ® 
° O _ 
0.6 
1.4 2.2 
41 
3.0 
0.74 1 
■=0.71 - 
(13 
CD 
C 
ro 
q) 0.68 
0.65 
m 9 ° 
"" <F° < 
0.6 
1.4 
2.2 
41 
3.0 
Figure 4 
Scatter plots of internal otolith morphometric variables for 3-year-old haddock ( 1994 year class) show- 
ing grouping patterns of eastern Georges Bank (squares) and western Georges Bank (circles). 
Table 6 
Age- and year-class-specific jack-knifed cross-validation 
classification matrices of the frequency of assigned had- 
dock samples from eastern (EGB) and western Georges 
Bank ( WGB). 
Correct classification (%) of 
haddock samples by age group (year class) 
Age 1 
(1994) 
Age 2 
(1995) 
Age 3 
(1993) 
Age 3 
(1994) 
Age 4 
(1993) 
Age 5 
(1992) 
n 
18 
35 
18 
54 
53 
20 
EGB 
71 
72 
67 
80 
75 
80 
WGB 
64 
88 
56 
84 
59 
70 
Total 
67 
80 
61 
83 
64 
75 
the development of regional differences in otolith morpho- 
metries (Smith, 1992; Campana and Casselman, 1993). 
Growth (i.e. metabolic activity) alters the rate of otolith 
crystalline growth, which determines the size and shape 
of individual crystals, and ultimately the size and shape 
of individual otoliths (Gauldie, 1990). Differences in oto- 
lith shape may be mediated through differences in orien- 
tation and packing of otolith crystals that are influenced 
by rate of crystalline growth (Smith, 1992). Fast growth 
tends to form long thin crystals, in contrast to slow growth 
which forms short compacted crystals (Gauldie and Nel- 
son, 1990). 
Differences in growth rates (and hence, otolith struc- 
ture) of eastern and western Georges Bank haddock ap- 
pear to derive mainly from differences in water tempera- 
ture and diet. Eastern Georges Bank haddock are affected 
more by colder Scotian Shelf waters than western Georges 
Bank haddock, which are affected more by warmer Gulf 
of Maine waters (Drinkwater and Mountain, 1997). Fur- 
thermore, eastern Georges Bank haddock appear to have 
less available food and have a diet that is less rich in 
protein (Garrison 3 ). Hence, the colder waters and poorer 
diets experienced by eastern Georges Bank haddock cor- 
respond to slower growth rates. Moreover, this correspon- 
dence suggests that environmental factors are major de- 
terminants of otolith growth as has been inferred in other 
3 Garrison, L. P. 2000. Unpubl. data. Food Web Dynamics 
Program, Population Biology Branch, Northeast Fisheries Sci- 
ence Center, 166 Water St., Woods Hole, MA 02543. 
