Theriogenology 67 (2007) 786–794

www.theriojournal.com

Characterization of rainbow trout egg quality: A case study

using four different breeding protocols, with emphasis
on the incidence of embryonic malformations
Emilie Bonnet, Alexis Fostier, Julien Bobe *
Institut National de la Recherche Agronomique, INRA SCRIBE, IFR 140, Campus Beaulieu, 35000 Rennes Cedex, France
Received 29 May 2006; received in revised form 7 September 2006; accepted 13 October 2006

Abstract
The aim of this study was to set up a methodology to accurately evaluate the effects of various husbandry practices on a fish
broodstock based on the developmental potential of the egg. For that purpose, long–short photoperiod manipulations (tested twice,
PM1 and PM2 groups), spawning induction by injection of a GnRH analog (SI group), and a 16-day post-ovulatory ageing of eggs
(POA group) were used in rainbow trout (Oncorhynchus mykiss). Females without any treatment were used as a control group.
Survival at eying (E) and yolk-sac resorption (YSR) were recorded and malformations at YSR were monitored according to a
detailed typology that included cyclopia, torsion, incomplete YSR, prognathia, and others. Egg weight was also monitored. A
deleterious effect of photoperiod manipulation was observed on egg quality in both PM1 and PM2 groups. Incomplete YSR
appeared as the predominant malformation while cyclopia type was nearly absent. In the SI group, a limited effect on egg quality
was observed in comparison to the other experimental groups, although the percentage of normal alevins at YSR was significantly
lower than in the control group. Finally, the most important effects on egg quality were observed in the POA group. The percentage
of normal alevins was only 14  6% (mean  95% confidence interval) while the percentage of malformed embryos reached
49  11%. The proportion of cyclopia was significantly higher than in the control group. In conclusion, the type of egg quality
alteration is extremely dependent on the applied breeding protocols, and the proposed methodology is able to discriminate those
experimental conditions even when the impact on egg quality is limited.
# 2006 Elsevier Inc. All rights reserved.

Keywords: Fish; Egg quality; Photoperiod; Spawning induction; Post-ovulatory ageing

1. Introduction embryo. It is well known that rainbow trout egg quality

Fish egg quality is an important issue in the fish
farming industry, especially for intensively cultured
species such as rainbow trout (Oncorhynchus mykiss).
Fish egg quality can be defined as the ability of the egg
to be fertilized and subsequently develop into a normal
* Corresponding author. Tel.: +33 2 23 48 57 24;
fax: +33 2 23 48 50 20.
E-mail address: Julien.Bobe@rennes.inra.fr (J. Bobe).
can be highly variable [1,2]. Understanding those
variations would be of great economic value to
hatcheries [3]. Variations of fish egg quality have been
hypothesized to be correlated with significant effects of
external factors and/or breeding conditions (see [4,5]
for review). Until now, the most studied external factors
have been food composition and nutrient availability
during vitellogenesis [6]. They are believed to influence
egg quality through the building of yolk components
and thus involve the nature and quantity of trophic
stores used during early embryonic development. Many

0093-691X/$ – see front matter # 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.theriogenology.2006.10.008
 E. Bonnet et al. / Theriogenology 67 (2007) 786–794
other extrinsic factors such as photoperiod regimes [7],
stress [8], water temperature [9,10], induction or
synchronization of ovulation [11,12], over-ripening of
the eggs [13,14] have also been studied in rainbow trout.
However, the compilation and analysis of most of the data
devoted to fish egg quality is complex due to the use of
heterogeneous criteria mainly based on fertilization
success and embryonic survival at specific stages. A
standardized methodology that would take into account
classic criteria (e.g., fertilization success and embryonic
survival) but also survival at later stages and the
occurrence of embryonic malformations would be of
great interest. For that purpose, the effects of various
breeding protocols were analyzed using an autumn-
spawning rainbow trout brood stock strain. Four specific
experimental protocols were applied to different groups
of females: (1) no manipulation resulting in natural and
spontaneous spawning in November (control group), (2)
a widely used long–short photoperiod regime designed to
obtain spawning in July, (3) a classic protocol for
spawning induction, and (4) a long delay between
ovulation and egg collection for fertilization known to
have a major negative impact on egg quality. For each
group, egg quality was analyzed by recording survival at
two different developmental stages as well as the
occurrence of embryonic malformations. Finally, data
originating from all experimental groups were analyzed
together to test whether the effects of specific breeding
protocols could be characterized and discriminated using
this method.
2. Materials and methods
2.1. Animals
Investigations were conducted according to the
guiding principles for the use and care of laboratory
animals and in compliance with French and European
regulations on animal welfare. Male and female
rainbow trout (1:1 sex ratio) from an autumn-spawning
strain were held under natural photoperiod until their
first reproduction in an INRA experimental farm (Sizun,
France). Three or 4 weeks before expected ovulation,
fish were transferred into a recirculating water system at
12 8C in INRA facilities (Rennes, France). Before any
manipulation, fish were anesthetized in 0.05% 2-
phenoxy-ethanol. Each fish was individually weighed
and tagged. During the reproductive season, all females
were checked for ovulation three times a week by
applying a manual pressure onto the abdomen.
In the control group, eggs were collected 5 days after
ovulation from 25 females (1813 g  260) in their first
787
reproductive season (Fig. 1A). In the post-ovulatory
ageing (POA) group, eggs were collected a second time,
16 days after detected ovulation from the same 25
females (Fig. 1A).
In the spawning induction (SI) group, eggs were
collected from 35 females (2030 g  360) in their first
reproductive season that had been given an intraper-
itoneal injection of [Des-Gly10,DArg6,Pro-NHEt9]-
GnRH analog (Bachem, Allemagne) at 60 mg kg 1
b.w. (Fig. 1B). Unfertilized eggs were collected 5 days
after detected ovulation.
In the photoperiod-manipulated group 1 (PM1), eggs
were collected from 17 females (2618 g  300) that had
been subjected to a photoperiodic control of their
spawning date. After their first reproduction, fish were
isolated in light-tight tanks and exposed to an artificial
photoperiod (Fig. 1C). Beginning on 15 January, all fish
were held under constant light (24L:0D) for 490 8C day.
Then, beginning on 27 March, they were held under
short photoperiod (8L:16D) until ovulation
(1230 8C day). Light was supplied by four neon tubes
(58 W). Unfertilized eggs were collected 5 days after
ovulation was detected.
In the photoperiod-manipulated group 2 (PM2), eggs
were collected from 20 females (2640 g  460) that had
been subjected to the same photoperiod protocol as
PM1 fish (Fig. 1D). In addition, three sub-groups of
females were made based on the timing of ovulation
during the first reproductive season: early (PM2-1,
n = 10), intermediate (PM2-2, n = 5) and late (PM2-3,
n = 5) spawning date. Unfertilized eggs were collected
5 days after ovulation.
2.2. Gamete collection
Five days (C, SI, PM1 and PM2 groups) or 16 days
(POA group) after ovulation was detected, unfertilized
eggs were collected in 50 mL plastic tubes by manual
stripping. Individual egg weight was calculated for each
egg sample. Two batches of 5 mL of eggs (approxi-
mately 100–200 eggs per batch) were used for
fertilization. On each day of egg collection, sperm
samples were collected from 10 mature males
originating from the same experimental group and
were used to fertilize eggs with a pool of sperm. Sperm
samples were obtained by manual pressure on the
abdomen and kept at 4 8C for a short time before use.
2.3. Fertilization and early development
Fertilization was performed under previously
described standardized conditions [15]. Immediately
788 E. Bonnet et al. / Theriogenology 67 (2007) 786–794

Fig. 1. Experimental manipulations: (A) control and post-ovulatory ageing (POA) group (n = 25). (B) Spawning induction using a 60 mg kg 1
intraperitoneal injection of [Des-Gly10,DArg6,Pro-NHEt9]-GnRH analog (SI, n = 33). (C) Long–short photoperiod regime group 1 (PM1, n = 17).
(D) Long–short photoperiod regime group 2 (PM2, n = 20). Photoperiod (continuous line), temperature (dotted line) and ovulation period (OV) are
shown for each experimental protocol.

prior to fertilization, coelomic fluid was discarded and
eggs were washed using 5 mL of 10 8C Ovafish1 rinsing
solution (IMV, L’Aigle, France). After removing rinsing
solution by pouring the eggs onto a plastic strainer, 5 mL
of pooled semen were pipetted onto the eggs, and 5 mL of
10 8C Actifish1 solution (sperm motility activating
saline solution, IMV, L’Aigle, France) were immediately
added. After 5 min, the Actifish1 was drained and
fertilized eggs were transferred into compartmentalized
incubation trays supplied by recirculated water (10 8C).
Water temperature and chemistry were routinely
monitored and maintained constant over the entire
incubation period. Dead eggs and embryos were
periodically removed and survival rates were estimated
as percentages of the initial number of eggs used for
fertilization. Survival at eyeing (150 8C day) and
completion of yolk-sac resorption (YSR, 550 8C day)
were monitored. The occurrence of noticeable morpho-
logical malformations (spinal cord torsion, head or
caudal fin malformations, etc.) at YSR was expressed as a
percentage of alevins present at YSR. The number of
normal alevins at YSR was expressed as a percentage of
the initial number of eggs. In addition, the occurrence of
specific types of malformations was recorded at YSR.
 E. Bonnet et al. / Theriogenology 67 (2007) 786–794 789

and were removed from the experiment. In the other

groups, all females completed ovulation.

3.2. Egg weight

A significantly lower egg weight was observed in the

PM2 group in comparison to C, POA and SI groups
(Fig. 2). In contrast egg weight was not significantly
different in PM1 and PM2 groups (Fig. 2). No
significant egg weight differences were observed
among C, POA, SI and PM1 groups.

Fig. 2. Egg weight (mean  S.D.) monitored in egg samples originat-
ing from females of the control group (C), the post-ovulatory ageing
group (POA), the spawning induction (SI) group and photoperiod-
manipulated groups (PM1 and PM2).
2.4. Statistical analysis
Egg quality data were statistically analyzed using
nonparametric Mann–Whitney U tests. Paired statistical
t-tests were performed to compare survival and
malformations rates between C and POA groups. Egg
weight differences among groups were analyzed by an
ANOVA followed by a Tukey’s honest significance
difference (HSD) test. All statistical analyses were
carried out using Statistica 7.0 software (StatSoft, Tulsa,
OK, USA).
3. Results
3.1. Timing of ovulation
Females from the photoperiod-manipulated groups
(PM1, n = 17 and PM2, n = 20) ovulated during the
summer, 3 months earlier than the control group (C,
n = 25), i.e. ovulation peak was observed in July instead
of November (Fig. 1). Females of the spawning
induction (SI) group (n = 33) ovulated within a week
of injection. Two females of the SI group did not ovulate
3.3. Survival
A significantly lower survival at eyeing was observed
in three experimental groups in comparison to the
control ( p < 0.001, Table 1). Embryonic survival was
only 49  18% (mean  95% CI) in the PM1 group,
38  17% in the PM2 group, and 37  11% in the POA
group, while it was up to 93  3% in controls. In
addition, inter-female variability of survival rates was
much higher in the experimental groups than in controls
(Fig. 3A). In contrast, no significant difference of
survival rates could be detected between the SI and C
groups (Table 1).
In all experimental groups, the percentage of alevins
without noticeable malformations (normal alevins) at
yolk-sac resorption (YSR) was significantly lower than
in controls ( p < 0.001), the lowest values being
observed in the PM1, PM2, and POA groups (Table 1).
In addition, differences between PM2 subgroups
were detected (Fig. 4A). The intermediary group PM2-2
exhibited significantly higher survival rate at eying and
YSR in comparison to the early PM2-1 group (not
significant) and the late PM2-3 group ( p < 0.05).
3.4. Malformations
The observed frequency of malformed alevins
(Table 1) was significantly higher ( p < 0.001) for

Table 1
Embryonic survival (%) at eyeing stage and normal alevins at YSR expressed as percentages of the initial number of eggs used for fertilization
(mean  95% confidence interval)
C POA SI PM1 PM2
*** ***
Survival at eyeing 93  3 37  11 84  8 49  18 38  17***
Normal alevins at YSR 84  5 14  6 *** 65  9*** 37  16 *** 24  13***
Malformations at YSR 52 49  11 *** 84 15  9 *** 19  10***
n 25 25 33 17 20
Malformation occurrence at YSR is expressed as a percentage of the number of alevins present at YSR. C: control, POA: 16-day post-ovulatory
ageing of the eggs, SI: induction of ovulation by GnRHa, PM1 and PM2: long–short photoperiod manipulation. n: sample size. Significantly different
from control group at ***p < 0.001.
790 E. Bonnet et al. / Theriogenology 67 (2007) 786–794

Fig. 3. Egg quality parameters recorded in control group (C, n = 25), egg post-ovulatory ageing group (POA, n = 25), spawning induction group (SI,
n = 33) and photoperiod-manipulated groups (PM1, n = 17 and PM2, n = 20). (A) Boxplot representations of survival at eyeing (190 8C day) and
normal alevins at yolk-sac resorption (YSR, 550 8C day) expressed as a percentage of the initial number of eggs used for fertilization. The inter-
quartile range is represented by the box and the whiskers indicate maximum and minimum values. (B) Boxplot representation of morphological
malformations at YSR expressed as a percentage of alevins present at YSR.

Fig. 4. Egg quality parameters recorded in the 3 subgroups of the PM2 photoperiod-manipulated group. The three sub-groups were designed based
on the timing of ovulation during the first reproductive season, PM2-1 = early, PM2-2 = intermediate and PM2-3 = late spawning date. (A) Survival
at eyeing (190 8C day) and percentage of normal alevins at yolk-sac resorption (YSR, 550 8C day) expressed as a percentage of the initial number of
eggs used for fertilization. (B) Morphological malformations expressed as a percentage of the number of alevins present at YSR. Different letters
indicate significant differences between groups.
 E. Bonnet et al. / Theriogenology 67 (2007) 786–794 791

POA (49  11%), PM1 (15  9%) and PM2
(19  10%) groups than for the control group
(5  2%). Intra-group heterogeneity was higher in
the POA group. Thus, half of the POA females
produced malformed alevins at a frequency ranging
between 20% and 70%. In the SI group, some females
produced a high frequency of malformed alevins,
although it was not significantly different from the
control (Fig. 3B).
Due to small sample size (PM2-1, n = 3 and PM2-3,
n = 1), statistical differences between PM2 sub-groups
could not be evaluated. However, the percentage of
malformation observed in PM2-2 was lower than in the
two other PM2 subgroups (Fig. 4B).
The types of malformed alevins observed in this
study are presented in Fig. 5B. The frequency
distribution of malformation types (Fig. 5A) was
clearly different between the groups. The control group
was characterized by a low occurrence of each type of
malformation. In the POA group, cyclopia was the most
frequent malformation (Fig. 5) and occurred signifi-
cantly more than in the control group (Table 2). In the SI
group, the occurrence of other types of malformations
was significantly higher than in the control group
(Table 2). In both PM1 and PM2 groups, yolk-sac
resorption and torsion defects were the most frequently
recorded malformations in comparison to cyclopia and
prognathia (Fig. 5). In addition, the occurrence of
cyclopia was significantly lower in PM1 ( p < 0.05) and
PM2 ( p = 0.06) groups than in the control group
(Table 2) while the occurrence of YSR defects was
higher than in the control group (Table 2, p = 0.06).

Fig. 5. Distribution of specific types of malformations recorded at yolk-sac resorption in each experimental group (550 8C day). (A) Malformations
types at YSR expressed as a percentage of the number of alevins present at YSR (mean  S.D.). For each graph, different letters indicate significant
differences between groups. (B) Types of identified morphological malformations.
792 E. Bonnet et al. / Theriogenology 67 (2007) 786–794
Table 2
Frequencies of malformation types (mean  95% confidence interval) at YSR expressed as a percentage of the number of malformed embryos in
each experimental group
Malfo. types C POA
Cyclopia 98 34  11 **
Torsion 24  11 10  4
YSR defect 30  12 25  10
Prognathia 11  7 23
Others 26  15 29  11
n 23 18
C: control, POA: 16-day egg post-ovulatory ageing, SI: induction of ovulation by GnRHa, PM1 and PM2: long–short photoperiod manipulation; n:
sample size. Significantly different from control group: *p < 0.05, **p < 0.01, ap = 0.06.
4. Discussion
While the effect of breeding conditions on embryo-
nic survival at early stages is usually well documented,
their impact on survival at later stages (e.g. yolk-sac
resorption) or on the occurrence of embryonic
malformations has received far less attention. In the
present study, we aimed at developing a discriminating
and sensitive method to characterize egg quality under
different experimental conditions. Indeed, we tested our
methodology using four specific breeding conditions.
Females from an autumn-spawning strain with eggs
collected 5 days after ovulation were considered as a
control group. The other groups were compared to this
control group and showed specific patterns of egg
quality defects.
4.1. Post-ovulatory ageing
In the present study, a 16-day egg post-ovulatory
ageing (POA group) induced the most negative effects on
egg quality in comparison to the other groups including
the control group. Thus, a 93% embryonic survival rate at
eyeing stage was recorded for control eggs while a very
low embryonic survival (37%) was observed after
retention of the eggs in the body cavity for 16 days.
This observation is in total agreement with existing
studies that have looked at post-ovulatory ageing in
salmonid eggs [13–15]. At the yolk-sac resorption stage
(YSR), the percentage of normal alevins was low and
exhibited very little inter-female variability (14  6%;
mean  95% CI). In addition, POA resulted in an
increased occurrence of malformed alevins at YSR
(49  11%). In rainbow trout, high percentages of
malformations were observed after post-ovulatory
ageing of eggs for 14 or 21 days [13,15]. A relation
between loss of egg quality and increased malformations
was previously noted in catfish (Silurus glanis) after post-
ovulatory ageing [16]. Moreover, these previous studies
SI PM1 PM2
55 0  0* 1  1a
14  7 30  15 23  10
a
30  9 50  18 41  13a
11  8 1  2* 33
40  9 * 19  15 32  12
29 16 10
in rainbow trout reported that one specific type of
malformation seemed highly represented in the post-
ovulatory ageing-induced embryonic malformations. In
the present study, we showed that a significantly higher
occurrence of the cyclopia malformation was observed in
the POA group in comparison to controls. Together, this
demonstrates that POA preferentially increases the
occurrence of cyclopia. In zebrafish (Danio rerio), a
mutation called cyclops lead to the cyclops phenotype
[17]. Authors showed that cyclops locus encodes the
nodal-related protein Ndr2, a member of the transform-
ing growth factor beta superfamily. In interaction with
other molecules, this protein is supposed to be required
for ventral midline patterning of the embryonic central
nervous system. Thus, egg post-ovulatory ageing process
could be associated with mechanisms affecting the
proper expression of Ndr2 gene in the developing
embryo.
4.2. Long–short photoperiod regime
A long–short photoperiod regime is a common
practice used to obtain an advanced spawning period
[9]. In the present study, we chose a widely used
photoperiod design to advance the spawning date by 4
months. After a first reproductive season in the fall, the
artificial photoperiod regime was applied in January and
led to a reproductive period during early summer. This
long–short photoperiod regime significantly decreased
embryonic survival at eyeing and yolk-sac resorption.
This was observed in both experiments conducted on
two different groups of animals (PM groups 1 and 2).
Most of the existing studies on photoperiod manipula-
tion and egg quality have focused on a combination of
photoperiod and temperature variations termed photo-
thermal changes [9,18]. However, the effect of
photoperiod by itself on egg quality has received far
less attention. Furthermore, such studies were based on
the measurements of egg size and female fecundity to
 E. Bonnet et al. / Theriogenology 67 (2007) 786–794
estimate reproductive success [9,19]. In the present
study, egg quality was monitored separately for each
female. Our data clearly show lower survival at two
developmental stages in comparison to the control
group. An increase of inter-individual variation was also
recorded within the two photoperiod-manipulated (PM)
groups. At YSR, some females exhibited a percentage
of normal alevins close to 0% while others exhibited a
percentage of normal alevins close to 85%. In
conjunction with the decrease in survival, we observed
an increase in the occurrence of malformations at YSR.
After a detailed analysis, we observed a similar
distribution of malformation types for the two PM
groups. A lower occurrence of specific types of
malformations such as cyclopia (0 and 1 cyclop alevin
observed in PM1 and PM2, respectively) was con-
comitant with a higher occurrence of other malforma-
tions, such as incomplete yolk-sac resorption. To our
knowledge, no prior analysis of embryonic malforma-
tions was ever done after a photoperiod manipulation of
female rainbow trout. Clearly, our observations on both
PM groups strongly suggest that a photoperiodic
manipulation of spawning date can induce specific
embryonic malformations linked to the resorption of the
yolk-sac. In the present study, a photoperiod manipula-
tion of spawning date resulted in a reduced length of the
oogenetic process, including the vitellogenic phase.
Based on the types of malformation observed, we can
speculate that photoperiod manipulation affects egg
quality by impacting mechanisms linked to yolk
accumulation or processing. These observations are
consistent with the lower egg weight observed in PM2
group in comparison to the other groups. However, the
observed differences were rather limited and not
significant in the PM1 group.
In conclusion, we performed a full evaluation of the
impact of the photoperiodic control of spawning date
on egg quality. This methodology allowed us to
demonstrate that photoperiod manipulation had a
negative impact on egg quality characterized by
embryonic mortalities and specific morphological
abnormalities. Together, the high inter-female varia-
bility in PM1 and PM2 groups and the analysis of egg
quality in PM2 sub-groups strongly suggest that the
negative impact of photoperiod on egg quality could be
modulated by the time-period between the first
ovulation and the beginning of photoperiod control.
Thus, females, with early or late ovulation during the
first reproductive season seemed to be more affected by
a modified photoperiod regime than females ovulating
in the middle of the reproductive season. Our
observations stress the importance of the life history
793
of each fish in the success of reproduction when
artificial photoperiod is used to control the spawning
date. This observation is original and opens new
prospects for understanding the impact of photoperiod
on vitellogenesis. Indeed, the appropriate moment to
start photoperiod manipulation should be well defined
when it comes to egg quality.
4.3. Induction of ovulation
The stimulation of post-vitellogenic females with
GnRH agonist ([Des-Gly10,DArg6,Pro-NHEt9]-GnRHa
at 60 mg kg 1 b.w.) resulted in the successful induction
of ovulation. GnRH-treated females ovulated within 7
days of hormonal stimulation. This result is in total
agreement with past results obtained in rainbow trout
after treatment with [DArg6Pro9Net] sGnRH
20 mg kg 1 b.w. which induced 90% ovulation within
7 days after injection [20]. In rainbow trout, it was
previously demonstrated that GnRHa-induced ovula-
tion had no effect on fertilization rates [20]. Another
study showed that fertility, survival at eyeing and
hatching success of brown trout (Salmo trutta) eggs
collected from females injected with 10 mg kg 1
GnRHa were significantly lower than in the control
[21]. However, this observed decrease in egg quality
was very limited and a hatching success around 95%
was reported for the treated group. Similarly, a recent
study in rainbow trout showed GnRHa when adminis-
tered in Freund’s Incomplete Adjuvant did not affect
egg quality based on fertilization, eyeing and hatching
percentages [11]. In the present study, embryonic
survival at eyeing was not affected by the spawning
induction treatment. This observation is in total
agreement with existing studies. In contrast, survival
at YSR was significantly lower than in the control
group. Together, these observations strongly suggest
that a spawning induction treatment can induce limited
but significant egg quality defects. These defects are
characterized by late embryonic mortalities occurring
after eyeing and possibly after hatching. Furthermore, it
is noteworthy that, in contrast to the other two
experimental conditions studied (egg post-ovulatory
ageing and photoperiod manipulation), no specific
malformation type seems to be associated with
spawning induction-induced egg quality loss.
5. Conclusions
In the present study, we used a new methodology to
evaluate egg quality under four different breeding
conditions. We showed that post-ovulatory ageing of
794 E. Bonnet et al. / Theriogenology 67 (2007) 786–794
eggs, photoperiod manipulation and spawning induc-
tion had significant impacts on egg developmental
potential after fertilization. In addition, we showed that
each experimental manipulation had a specific impact
on egg quality characterized by specific malformation
types, specific timing of mortalities and different inter-
female variability. We can therefore propose that each
experimental manipulation has its own signature
characterized by specific embryonic mortalities and
malformation patterns. Furthermore, our observations
strongly suggest that, for some manipulations, specific
estimators of egg quality have to be used to avoid the
underestimation of the egg quality loss. Indeed, our
methodology showed that two breeding protocols (i.e.
photoperiod control and spawning induction) widely
used in the fish farming industry, have a negative impact
on egg quality even though those two processes were
previously believed to have little or no impact on egg
quality. In addition, the use of this discriminating
methodology can provide new hints for the identifica-
tion of some mechanisms triggering egg quality losses.
Acknowledgement
Funding for this work was supported through an
INRA-AGENAE-IFOP grant to JB.
References
[1] Bromage N, Jones J, Randall C, Thrush M, Davies B, Springate
J, et al. Broodstock management, fecundity, egg quality and the
timing of egg production in the rainbow trout (Oncorhynchus
mykiss). Aquaculture 1992;100(1–3):141–66.
[2] Su G, Liljedahl L, Gall GAE. Genetic and environmental
variation of female reproductive traits in rainbow trout (Oncor-
hynchus mykiss). Aquaculture 1997;154(2).
[3] Lee CS. Biotechnological advances in finfish hatchery produc-
tion: a review. Aquaculture 2003;227(1–4):439–58.
[4] Brooks S, Tyler CR, Sumpter JP. Egg quality in fish: what makes
a good egg? Rev Fish Biol Fish 1997;7:387–416.
[5] Kjorsvik E, Mangor-Jensen A, Homefjord I. Egg quality in
fishes. Adv Mar Biol 1990;26:71–113.
[6] Izquierdo MS, Fernandez-Palacios H, Tacon AGJ. Effect of
broodstock nutrition on reproductive performance of fish. Aqua-
culture 2001;197(1–4):25–42.
[7] Taranger GL, Haux C, Stefansson SO, Bjornsson BT, Walther
BT, Hansen T. Abrupt changes in photoperiod affect age at
maturity, timing of ovulation and plasma testosterone and
oestradiol-17 beta profiles in Atlantic salmon, Salmo salar.
Aquaculture 1998;162(1/2):85–98.
[8] Schreck C, Contreras-Sanchez W, Fitzpatrick M. Effects of
stress on fish reproduction, gamete quality, and progeny. Aqua-
culture 2001;197(1–4):3–24.
[9] Davies B, Bromage N. The effects of fluctuating seasonal and
constant water temperatures on the photoperiodic advancement
of reproduction in female rainbow trout, Oncorhynchus mykiss.
Aquaculture 2002;205(1/2):183–200.
[10] Yamamoto T, Kobayashi W, Kuramoto T. Twin malformation
induced in chum salmon eggs by elevated water temperature,
with a suggestion as to its mechanism. Can J Zool 1996;74(3).
[11] Arabaci M, Diler I, Sari M. Induction and synchronisation of
ovulation in rainbow trout, Oncorhynchus mykiss, by adminis-
tration of emulsified buserelin (GnRHa) and its effects on egg
quality. Aquaculture 2004;237(1–4):475–84.
[12] Gillet C, Breton B, Mikolajczyk T. Effects of GnRHa and
pimozide treatments on the timing of ovulation and on egg
quality in Arctic charr (Salvelinus alpinus) at 5 and 10 8C.
Aquat Living Resour 1996;9(3):257–63.
[13] Aegerter S, Jalabert B. Effects of post-ovulatory oocyte ageing
and temperature on egg quality and on the occurrence of triploid
fry in rainbow trout, Oncorhynchus mykiss. Aquaculture
2004;231(1–4):59–71.
[14] Springate JRC, Bromage N, Elliot JAK, Hudson DL. The timing
of ovulation and stripping and their effects on the rates of
fertilization and survival to eyeing, hatch and swim-up in the
rainbow trout (Salmo gairdneri R.). Aquaculture 1984;43:313–
22.
[15] Aegerter S, Jalabert B, Bobe J. Large scale real-time PCR
analysis of mRNA abundance in rainbow trout eggs in relation-
ship with egg quality and post-ovulatory ageing. Mol Reprod
Dev 2005;72(3):377–85.
[16] Varkonyi E, Bercsenyi M, Ozouf-Costaz C, Billard R. Chromo-
somal morphological abnormalities caused by oocyte ageing in
Silurus glanis. J Fish Biol 1998;52(5).
[17] Rebagliati MR, Toyama R, Haffter P, Dawid IB. Cyclops
encodes a nodal-related factor involved in midline signalling.
Proc Natl Acad Sci USA 1998;95(17):9932–7.
[18] Breton B, Maisse G, Lemenn E. Contrôle photopériodique de la
saison de reproduction en salmoniculture: une expérience pilote
en Bretagne. Bull Fran Piscicult 1983;288:35–45.
[19] Bon E, Breton B, Govoroun M, Menn F. Effects of accelerated
photoperiod regimes on the reproductive cycle of the female
rainbow trout. II. Seasonal variations of plasma gonadotropins
(GTH I and GTH II) levels correlated with ovarian follicle
growth and egg size. Fish Physiol Biochem 1999;20(2):143–54.
[20] Breton B, Weil C, Sambroni E, Zohar Y. Effects of acute versus
sustained administration of GnRHa on GtH release and ovulation
in the rainbow trout, Oncorhynchus mykiss. Aquaculture
1990;91(3/4).
[21] Mylonas C, Hinshaw J, Sullivan CV. GnRHa-induced ovulation
of brown trout (Salmo trutta) and its effects on egg quality.
Aquaculture 1992;106(3/4):379–92.