Fróðskaparrit - 31.12.2000, Qupperneq 86

90
REPRODUCTION OF FAROE PLATEAU COD:
SPAWNING GROUNDS, EGG ADVECTION AND LARVAL FEEDING
Table 1. Maturity stages 3-7for Faroe Plateau cod. Talva 1. Búningarstig 3-7 hjá landgrunstoskinum.
Stage Female gonads Male gonads
3 The eggs are visible by eye and the roe fills about half the volume of the body cavity. The testes fill about half of the body cavity.
4 The eggs are clearly visible by eye as white corns. The roe fills about two-thirds of the body cavity. The testes fill about two-thirds of the body cavity.
5 The eggs are circular and many are transparent. The roe is full size. Testes are full size. Pressure on the belly causes drops of sperm to come out of the sac.
6 Light pressure on the belly causes the eggs to come out. Light pressure on the belly easily causes sperm to come out.
7 Post-spawning: the roe is nearly empty. Post-spawning: the testes are nearly empty.
The catch of cod was separated into
spawning cod (maturity stages 5 and 6, as
shown in Table 1) and non-spawning cod
(all other cod, which are not presented in
the results). Stages 1 (immature) and 2
(mature gonad, length about half of the
body cavity) were excluded since these fish
were assumed not to spawn during the cur-
rent season. First, a maturity-at-length key
was calculated from about 600 cod in 1995
and 1,000 cod in 1999, which were investi-
gated in detail (total length, round weight,
sex, maturity, and age). Second, the length
distribution of cod from each station
(scaled to total catch of cod) was split into
spawning cod and non-spawning cod using
the maturity-at-length key. By using a
length-weight key, the total weight of
spawning cod was calculated from each
station.
To investigate whether there were differ-
ences in peak spawning times between the
two years, 1995 and 1999, the average ma-
turity-stage determination was used. Since
Faroe Plateau cod migrates to its spawning
areas prior to spawning (Táning, 1940), to
get comparable results for the two years, it
was necessary to use measurements from
stations where both position and date were
as similar as possible. Only one station
(with big catches, located at the western
spawning site) could be used.
In order to investigate possible differ-
ences in maturity distribution between the
two years, the chi-square test statistic was
used. Maturity stages 3 to 5 were pooled in
order to fulfil the statistical assumption re-
quiring expected frequencies above five.
Despite a moderate violation of the as-
sumptions (one expected frequency was
4.5), a p-value well below 0.05 suggests
that the null hypothesis can be rejected
without difficulties.