McQuinn: Year-class twinning in sympatric spawning populations of Clupea harengus 
129 
(Postuma, 1974; Messieh 9 ), whereas newly hatched 
autumn spawners are still larvae (Fig. 1). Thus to 
assign correctly the age of a herring hatched in the 
autumn, one must add one year to the number of 
winter rings read from the otolith. Because spring- 
hatched herring metamorphose before their first win- 
ter and thus produce a winter ring in their first year, 
their proper age is equal to the number of winter 
rings. However, for any spring-hatched herring that 
subsequently reproduces in the autumn, its spawn- 
ing season would be considered autumn with the 
maturity-stage method, and one year would be added 
to the number of rings read from the otolith. Follow- 
ing the same logic in reverse, if an autumn-hatched 
individual subsequently spawned in the spring, a 
year would not be added to the number of rings read 
from the otoliths and it would be assigned an age 
that was one year younger than its actual age. There- 
fore, the number of rings will be used when compar- 
9 Messieh, S. N. 1974. Problems of ageing Atlantic herring 
(Clupea harengus harengus L.) in the ICNAF area. Inter- 
national Commision for the Northwest Atlantic Fisheries 
(ICNAF) Res. Doc. 74/59, Ser. 3274, 6 p. ICNAF, P.O. Box 638, 
Dartmouth, Nova Scotia, Canada B2Y 3Y9. 
ing the biological characteristics for a given age be- 
tween an autumn-spawning year class with the 
spring-spawning year class of the following year. 
Results 
The proportion of spring- and autumn-hatched her- 
ring was estimated from the mature members of the 
1979-80, 1981-82 and 1986-87 autumn- and spring- 
spawning year classes, respectively (Table 1 ). Accord- 
ing to their otolith characteristics, the vast majority 
(>90%) of the 1980 and 1982 spring-spawning her- 
ring were also spring hatched. This pattern is con- 
sistent from age 3 through age 5 (age=no. of rings). 
However, a large percentage of the 1979 (40%) and 
1981 (77%) autumn-spawning herring were also 
spring hatched, as judged from their otolith charac- 
teristics. Therefore, in both situations, there was a 
significant crossover from the strong spring-hatched 
cohort to the autumn-spawning population, in com- 
parison with the number of autumn-hatched indi- 
viduals. However, because the resulting 1979 au- 
tumn-spawning year class was also large, whereas 
the 1981 autumn-spawning year-class was not, this 
Table 1 
Percentage of autumn- and spring-hatched herring that became mature autumn and spring spawners within the 1979, 1981, and 
1986 autumn-spawning and within the 1980, 1982, and 1987 spring-spawning year classes off western Newfoundland (age is 
expressed as the number of otolith rings — see text). 
1979 autumn- 
■spawning year class 
1980 spring-spawning year class 
No. of 
Autumn- 
Spring- 
Spring- 
Autumn- 
Year 
rings 
hatched (%) 
hatched (%) 
n 
hatched (%) 
hatched (%) 
n 
1983 
3 
34.5 
65.5 
712 
90.6 
9.4 
402 
1984 
4 
55.6 
44.4 
1923 
92.3 
7.7 
1438 
1985 
5 
60.4 
39.6 
1760 
92.4 
7.6 
2590 
1981 autumn-spawning year class 
1982 spring-spawning year class 
No. of 
Autumn- 
Spring- 
Spring- 
Autumn- 
Year 
rings 
hatched (%) 
hatched (%) 
n 
hatched (%) 
hatched (%) 
n 
1985 
3 
54.4 
45.6 
68 
92.2 
7.8 
192 
1986 
4 
30.7 
69.3 
140 
95.2 
4.8 
481 
1987 
5 
22.6 
77.4 
186 
97.2 
2.8 
966 
1986 autumn- 
•spawning year class 
1987 spring-spawning year class 
No. of 
Autumn- 
Spring- 
Spring- 
Autumn- 
Year 
rings 
hatched (%) 
hatched (%) 
n 
hatched (%) 
hatched (%) 
n 
1990 
3 
86.0 
14.0 
57 
68.8 
31.2 
32 
