Opportunities
for
hybridization
between
two
sympatric
flounder
(Platichthys
flesus)
ecotypes
in
the
Baltic
Sea
Isa
Wallin
Degree project inbiology, Master ofscience (2years), 2016 Examensarbete ibiologi 30 hp tillmasterexamen, 2016 Biology Education Centre
Supervisor: Anders Nissling
ABSTRACT
2
1. INTRODUCTION ... 3
1.1 Ecology of flounder (Platichthys flesus) ... 3
1.2 Temperature implications for fish reproduction ... 5
1.3 Prerequisites for hybridization ... 6
1.4 Aims ... 7
2. MATERIALS AND METHODS ... 7
2.1 Study of temperature impact on coastal flounder eggs and larvae ... 8
2.2 Study of hybrid viable hatch and larval size ... 9
2.3 Measurements of larvae ... 10 2.4 Statistical analyses ... 11 3. RESULTS ... 12 3.1 Temperature study ... 12 3.2 Hybridization study ... 16 4. DISCUSSION ... 17
4.1 Temperature and hybrid experiments ... 17
4.2 Temperature conditions and hybridization events – the link between reality and the lab ... 20
4.3 Hypothesis evaluation ... 23
4.4 Further research ... 24
5. ACKNOWLEDGEMENTS ... 25
3
1.
INTRODUCTION
The Baltic Sea ecosystem is unique as the salinity is significantly lower than in the rest of the world’s oceans due to irregular inflows of saline water from the Atlantic combined with run-offs from rivers and streams, abundant precipitation and low transpiration (Bernes 2005). The salinity in the Baltic Sea bottom waters is approximately 13-22 psu, practical salinity units, in the Bornholm Basin and between 10-14 psu in the Gdansk and Gotland basins (Voipio 1981), i.e. salinity decreases towards the north and the east. Thus, the species inhabiting the Baltic Sea have evolved rather differently compared to their conspecifics in the Atlantic, probably as a result of isolation, bottlenecks and a selection for specific traits (Johannesson and André 2006).
1.1 Ecology of flounder (Platichthys flesus)
The European flounder (hereafter referred to as “flounder”) is found in the Eastern Atlantic in coastal and brackish waters of western Europe, its distribution ranging from the Mediterranean and the Black Sea in the south to the White Sea in the north (FishBase 2011). It is the most abundant flatfish in the Baltic Sea (Bagge 1981), found on loose clay bottoms or occasionally on sand bottoms and distributed from the Skagerrak to the Baltic Proper; however, it is less frequently observed north of the Sea of Åland and rarely north of the N. Quark (Florin 2005). The Baltic Sea flounder abundance and distribution is a consequence of the salinity gradient, as its reproductive success is closely linked to salinity regarding spermatozoa mobility and egg buoyancy, determining egg survival (Nissling et al. 2002), even though it has been shown that the lowest salinity tolerance from hatching to the end of the yolk sac stage is merely 0-1 psu (Yin and Blaxter 1987). Further, effects of multiple ecosystem changes such as eutrophication, climate change and habitat degradation have implications for Baltic Sea flounder distribution and abundance (Jokinen et al. 2015a).
4
Fig. 1. ICES subdivisions in the Baltic Sea with sampling locations for coastal and off-shore spawning flounder in the BONUS Inspire project. The off-shore spawning ecotype is found in SD 24-26 and 28, while the coastal spawning ecotype is found in SD 25-30 and 32 (Nissling et al. 2002, Nissling et al. 2014). Map (modified) from ICES homepage, www.ices.dk.
5
has indicated that hybridization between coastal and off-shore spawning flounder should be possible. In general, potential events of hybridization may result in one of three outcomes: offspring with fitness, i.e. individual lifetime reproductive success, 1) lower, 2) higher or 3) equal to its parents´. In cases where fitness of the hybrid offspring is lower compared to the fitness of the parents there may still be positive effects of hybridization; if there, for some reason, are few potential mates belonging to the same species and thus difficult to find a suitable mate, it may be more advantageous to mate with someone from a different species than to not mate at all. Additionally, advantages and disadvantages of hybridization may change with altering conditions (Björklund 2005).
Temperature-wise, the flounder seems to prefer intermediate temperatures and the best catches are made between 8-12 °C (Florin 2005 and references therein). When surface water temperatures decrease during winter both ecotypes migrate to deeper areas, where they most likely co-occur during this time of the year (Molander 1954, Nissling et al. 2014). In spring (April-May), the coastal spawning ecotype returns to the coast to spawn while the off-shore spawning ecotype stay and spawn at greater depths (>70 m; Ustups et al. 2013, Nissling et al. 2014). When spawning is over, adult fish dwell in coastal areas where they forage during summer and autumn (Molander 1954, Bagge 1981), i.e. both ecotypes coexist during this period as well as during the winter months spent off-shore (Nissling et al. 2014). After hatching, the larvae inhabit the pelagic zone feeding on zooplankton and are transported with currents to the coast where the metamorphosed juveniles grow up in shallow waters of high temperature (<1 m; Florin 2005, Martinsson and Nissling 2011) while feeding on bottom fauna (Aarnio et al. 1996, Nissling et al. 2007, Florin and Lavados 2010). During this intense growth phase, juveniles of both ecotypes are found in these nursery areas (Nissling et al. 2014).
1.2 Temperature implications for fish reproduction
Temperature affects fish growth (Bœuf et al. 1999), rates of food consumption and metabolism (Wootton 1990), and also has a pervasive controlling effect on different aspects of reproduction and early development in fish (van der Kraak and Pankhurst 1997). Pre-reproduction, ovarian maturation, including the process of vitellogenesis, in most teleosts is controlled by a combination of photoperiod and temperature signals (Norberg et al. 1999, Bobe and Labbé 2010). In males, sperm production (Yoneda and Wright 2005), sperm motility, fertilizing ability and velocity of spermatozoa, as well as the duration of the motility period, are temperature affected (Alavi and Cosson 2005).
6
(Buckley et al. 2000, Uehara and Mitani 2009). Also during the larval stage temperature is one of the most important environmental parameters determining development (Buckley 1982, Bisbal and Bengtson 1995, reviewed by Nash and Geffen 2005). Temperature may limit both the amount of time the larvae have to establish successful feeding by controlling the rate of metabolic demands and the pace at which the yolk sac is consumed (Bisbal and Bengtson 1995 and references therein). Further, larvae and juveniles exposed to low temperatures may display both lower activity and reduced or negative growth (Malloy and Targett 1991) and survival probabilities may be affected (Buckley 1982, Bisbal and Bengtson 1995 and references therein). Accordingly, two species of flounder have been found to change their migration patterns in accordance to temperature: a flounder population off the south-west coast of England has been found to leave their estuary habitat 1-2 months earlier for migration to spawning grounds during cold years, i.e. <2 °C cooler. They were also found to arrive at the spawning ground over a shorter time period during colder years (Sims et al. 2004). Additionally, Japanese flounder (Paralichthys olivaceus) off the coast of Japan has been found to migrate off-shore in order to escape suboptimal temperatures during spawning (Yasuda et al. 2010).
1.3 Prerequisites for hybridization
The off-shore spawning flounder in the Baltic Sea appears to start spawning earlier in the spring compared to coastal spawners, and spawning also seems to last longer. At random sampling occasions all spawning flounder caught at depths of 3-10 m have been found to belong to the coastal spawning ecotype, but a certain proportion of coastal spawners in spawning condition have also been found at depths of ca 70-80 m; thus, there is a population overlap during spawning. This may be because of low temperatures in coastal areas, potentially making a share of the coastal spawning ecotype prone to stay off-shore where temperatures are more favorable, instead of migrating to coastal areas (Nissling et al. 2014). The examples of temperature-driven migration described above (Sims et al. 2004, Yasuda et
al. 2010) could indicate that also coastal spawning flounder in the Baltic Sea may be prone to
escape suboptimal temperatures during spawning. As coastal spawning flounder females produce demersal eggs, i.e. with higher egg specific gravity (Nissling et al. 2002), these eggs will, if they are released off-shore, not remain buoyant but sink to the sea floor and probably die from hypoxia (Hinrichsen et al. in prep.). However, the coastal spawning males should, in theory, be able to fertilize eggs from off-shore spawning females, as the spermatozoa of these males are active at salinities prevailing at 70-80 m depth (Nissling et al. 2002).
7
1.4 Aims
My aim with this study was to determine whether hybridization between the two sympatric Baltic Sea flounder ecotypes is possible and if temperature may act as a trigger. The questions I wished to answer were 1) Is fertilization between the two ecotypes possible?, 2) Does hybridization between the two ecotypes affect viable hatch and larval size?, and 3) Could low temperatures have a negative impact on eggs and larvae of the coastal spawning ecotype and therefore be the underlying mechanism for hybridization between Baltic Sea coastal and off-shore spawning flounder?
My overall hypothesis was that hybridization between the two sympatric flounder ecotypes in the Baltic Sea is possible and that unfavorable temperature conditions in coastal waters is the trigger. I predicted that low temperatures would adversely affect the eggs and larvae regarding development time, viable hatch and growth. I also predicted that hybridization between the two sympatric flounder ecotypes in the Baltic Sea would be possible, but with negative implications for viable hatch and larval size.
2. MATERIALS AND METHODS
In order to answer the questions above, I conducted two experiments: one aiming to determine temperature impact on egg and larval development in Baltic Sea coastal spawning flounder, and one aiming to identify potential differences in viable hatch and larval size between hybrids and non-hybrids.
Capture of spawning flounder was conducted within the BONUS Inspire project in 2014-2015. The fish were caught by gill-nets outside Herrvik, off the central eastern coast of Gotland (ICES SD 28-2; Fig. 1). In 2014, coastal spawning flounder were captured at 3-20 m depth on May 6th and 9th, while off-shore spawning flounder were captured at >65 m depth on May 8th. In 2015, coastal spawning flounder were captured at 4-20 m depth on April 21st, while off-shore spawning flounder were captured at 70 m depth on April 23rd, except for two males caught at 10 and 20 m depth, respectively. When landed, the fish were tagged with an individual number and transported in tanks filled with seawater to the Ar Research Station on northern Gotland. Upon arrival, the fish were kept in indoor tanks (1x1 m) with flowing seawater of approximately 6 °C and 6.5 psu. The tanks were provided with lids to keep the fish in darkness to minimize stress, and no food was provided.
8
neutrally buoyant; eggs buoyant in high salinities were concluded to belong to the coastal spawning ecotype (18-23 psu; Ojaveer et al. 2015), as those eggs are demersal under natural conditions, while eggs buoyant at lower salinities (approximately 11-14 psu; Ojaveer et al. 2015) were concluded to belong to the off-shore spawning ecotype, having pelagic eggs. To establish ecotype affiliation of the males, spermatozoa motility was examined at 8 °C and four different salinities: 6, 7.5, 10.5 and 15 psu, the salinities prepared from filtered Baltic Sea water (salinity approximately 6.5 psu) and Tropic Marine Sea Salt. Semen was sampled from each male with a dry pipette. For each male tested, one droplet of semen was diluted in water of known salinity and observation of spermatozoa motility began immediately under a microscope at 250× magnification. Assessments started with a high salinity followed by subsequently lower salinities until spermatozoa motility ceased. The spermatozoa motility was established through visual inspection and recorded as fast-swimming, slow-swimming or immobile. Spermatozoa motility at low salinities (i.e. 6 and 7.5 psu) indicated that the male belonged to the coastal spawning ecotype (Nissling et al. 2002).
2.1 Study of temperature impact on coastal flounder eggs and larvae
In order to test whether low temperatures could be the underlying mechanism for hybridization between Baltic Sea coastal and off-shore spawning flounder, I conducted a study of temperature impact on egg and larval development, viable hatch and larval size. The temperature experiment was performed on coastal spawning flounder, since they, according to my hypothesis, may be prone to commence spawning off-shore as a result of suboptimal temperatures in coastal areas.
Four coastal spawning females were stripped and each egg portion was fertilized with semen from coastal spawning males on May 6th 2014, and the procedure was repeated for an additional female on May 9th. The temperature study lasted for 46 days and was terminated on June 21st.
24 hours post fertilization, the fertilization levels were surveyed and found to be approximately 100 % in all cases. 750 viable eggs from each female were then placed in beakers, five for each female with 150 eggs in each, with 300 ml water of 10.5 psu. The water was prepared by mixing filtered Baltic Sea water (salinity approximately 6.5 psu) with Tropical Marine Sea Salt and adding antibiotics: ampicillin, streptomycin and nystatin, the concentrations being 0.1 g-l, 0.05 g-l and 2500 IU-l, respectively. The use of antibiotics was in line with standard procedure for incubation of eggs and larvae (see Nissling et al. 1998). One beaker for each female was then placed in 2, 4, 6, 8 and 10 °C, the eggs in 2 and 4 degrees being transferred subsequently to avoid shock from low temperatures. The event of fertilization marked the start of the temperature experiment, i.e. May 6th and 9th 2014 respectively for the different females, even though there was a delay until all the beakers had been put under different temperature conditions.
9
incubated in refrigerators and therefore light conditions varied between temperatures: eggs and larvae in 8 and 10 °C were kept in light throughout the study period, while eggs in 4 and 6 °C were kept in darkness until hatching and after that in light. The eggs and larvae in 2 °C were kept in darkness during the entire study period.
At hatching the larvae were sorted in two categories, viable and unviable. Crooked larvae and those seemingly unable to swim normally were considered unviable and thus counted as dead. Of the larvae deemed viable, 30 larvae from each beaker were collected on the day when the majority of the eggs hatched and put in new beakers. They were then kept under the same temperature regimes as before during the entire larval stage, i.e. until the yolk sac was depleted. A minimum of five larvae from each beaker were controlled every day under a stereo microscope to establish the degree of yolk sac depletion, and dead larvae were removed and counted. When the yolk sac was depleted, the mouth had opened and the eyes were pigmented, the larvae were anesthetized in a Benzocaine solution (200 ppm in water of 10.5 psu) and then euthanized in a 4 % formalin solution based on water of 10.5 psu, where they were preserved to later be measured. The larvae found dead during incubation and at the time of yolk sac depletion were discarded since mortality was not included as a response to temperature conditions in my study.
2.2 Study of hybrid viable hatch and larval size
In order to determine whether hybridization between Baltic Sea coastal and off-shore spawning flounder is possible I performed a cross-fertilization experiment in 2014 and 2015, looking at viable hatch and larval size. In 2014 only three off-shore spawning females were caught. They were all in poor condition, probably because of adverse wind conditions with up-welling of hypoxic water, killing or harming the fish while they were still in the nets. These females yielded only small amounts of eggs. To be able to collect enough data I decided to also include eggs from coastal spawners, and thus four coastal spawning females were fertilized in 2014. In 2015, I collected additional data from two off-shore and two coastal spawning females, i.e. in total 11 females of which 10 produced sufficient amounts of eggs and larvae to be used in the experiment.
10
respectively. The reason for this is unknown but may be due to poor mixing, potentially leading to lower fertilization levels. It is also possible that these egg portions consisted of eggs from two different batches, respectively. If so, part of the eggs may have been over-ripe and thus of lower quality.
300 viable eggs from each female, 150 fertilized with semen from coastal spawning males and 150 fertilized by off-shore spawning males, were placed in beakers (i.e. two for each female) with 300 ml water. The salinity was 17 psu for the eggs from off-shore spawning females, as those eggs are buoyant at such salinities, and 10.5 psu for the eggs from coastal spawning females. The water for both egg types was prepared by mixing filtered Baltic Sea water (salinity approximately 6.5 psu) with Tropical Marine Sea Salt and similar concentrations of antibiotics as in the temperature study. This arrangement was pertained throughout the experiment. The eggs were incubated at 8 °C and kept in constant light during the entire study period. Water was changed daily, approximately 200 ml for eggs from coastal spawning females and 100 ml for eggs from off-shore spawning females, because those eggs are buoyant and thus more difficult to handle. Dead eggs were counted daily and removed from the beakers.
At hatching the larvae were sorted in two categories, viable and unviable. Crooked larvae and those seemingly unable to swim normally were considered unviable and thus counted as dead. In 2014, only two (out of three) off-shore spawning females produced enough viable larvae, probably the result of poor egg quality due to gill-net sampling at inferior environmental conditions, as discussed above. Of the larvae deemed viable, 30 larvae from each beaker (i.e. 60 in total for each female) were collected on the day when the majority of the eggs hatched. These larvae were put in new beakers where they were kept during the entire larval stage. During this phase, I changed approximately 100 ml of water daily for both types of larvae. At least five larvae from each beaker were controlled every day under a stereo-microscope to establish the degree of yolk sac depletion, and dead larvae were discarded.
When the yolk sac was depleted, the mouth had opened and the eyes were pigmented, the larvae were anesthetized in a Benzocaine solution (200 ppm in water of 17 psu for larvae stemming from off-shore spawning females and 10.5 psu for larvae stemming from coastal spawning females in 2014, and 200 ppm in water of 10.5 psu for all larvae in 2015) and then euthanized in a 4 % formalin solution based on water of 10.5 psu, where they were preserved to later be measured.
2.3 Measurements of larvae
Larvae from both the temperature and the hybrid experiments were preserved in 2014 and measured a year later, along with the additional larvae from the hybrid experiment reared in 2015. A random subsample, approximately 20 larvae, of the larvae preserved for each female in each temperature and/or with both types of males, i.e. hybrid and non-hybrid larvae, were measured to determine standard length, SL (mm) (i.e. from the point between the eyes and the
11
Fig. 2. Larvae of the coastal spawning ecotype in the temperature experiment measured to determine standard length (mm).
2.4 Statistical analyses
In my test of potential differences in viable hatch at different temperatures, the number of viable larvae hatched from 150 eggs incubated from each female at the respective temperature was analyzed with a generalized linear model with Poisson distribution (JMP, version 11). The model included the viable hatch as response variable and temperature and female as fixed factors.
In order to analyze potential differences in larval size between temperatures at the time of yolk sac depletion I conducted a general linear model, univariate ANOVA (SPSS, version 22), with length as response variable and temperature and female as fixed factors.
In the analyses of hybridization, coastal and off-shore spawners were used in the same data set, i.e. only hybrids and non-hybrids were tested against each other.
12
In order to analyze potential differences in larval size between hybrids and non-hybrids at the time of yolk sac depletion I performed a general linear model univariate ANOVA (SPSS version 22) with length as response variable and female and category (hybrid/non-hybrid) as fixed factors.
3. RESULTS
3.1 Temperature study
Egg development, i.e. the time from fertilization to hatching, varied between on average 27 days at 2 °C and on average 6 days at 10 °C, showing a strong negative relationship between time to hatching and temperature (Fig. 3).
Fig. 3. Relationship between the number of days till hatching and temperature for eggs from coastal spawning flounder (Platichthys flesus). Equations of the relationship shown between temperature and time till hatching are based on mean and median values respectively.
13
Larval development up to yolk sac depletion varied between on average 19 days at 2 °C and 7 days at 10 °C, i.e. typically shorter time till yolk sac depletion at higher temperatures (Fig. 4).
Fig. 4. Relationship between the number of days till yolk sac depletion and temperature for eggs from coastal spawning flounder (Platichthys flesus). Equations of the relationship shown between temperature and time till yolk sac depletion are based on mean and median values respectively.
14
Fig. 5. Viable hatch (%) of coastal spawning flounder (Platichthys flesus) eggs incubated at different temperatures (°C).
Larval size at the time of yolk sac depletion differed significantly between temperatures (df=4, F=20.09, p<0.001; Tables 1, 2) and also between females (df=4, F=26.61, p<0.001; Table 3) with the largest larvae at 6 and 8 °C and significantly smaller at 2, 4 and 10 °C (Table 2).
Table 1. Mean size (mm) ± SE at yolk sac depletion of flounder (Platichthys flesus) larvae incubated at different temperatures.
Temperature (°C) Mean size (mm) ± SE
15
Table 2. Pairwise comparisons of larval size (mm) ± SE at yolk sac depletion after incubation at different temperatures. Temperature (°C) p 2 degrees vs. 4 degrees >0.05 6 degrees <0.001 8 degrees <0.001 10 degrees >0.05 4 degrees vs. 2 degrees >0.05 6 degrees <0.05 8 degrees <0.001 10 degrees >0.05 6 degrees vs. 2 degrees <0.001 4 degrees <0.05 8 degrees >0.05 10 degrees <0.001 8 degrees vs. 2 degrees <0.001 4 degrees <0.001 6 degrees >0.05 10 degrees <0.001 10 degrees vs. 2 degrees >0.05 4 degrees >0.05 6 degrees <0.001 8 degrees <0.001
Table 3. Mean size (mm) ± SE at yolk sac depletion for larvae from the five females used in the temperature experiment.
Female Mean size (mm) ± SE
5803 3.04±0.02
5826 3.19±0.02
5840 3.05±0.02
5845 3.02±0.02
16
3.2 Hybridization study
The viable hatch was on average 67 % for both hybrids and non-hybrids, varying between 12 and 99 % for hybrids and 11 and 97 % for non-hybrids. When analyzing the viable hatch for hybrids and non-hybrids, respectively, I found no differences between the two groups (T=21, N=10, p>0.05; Fig. 6).
Fig. 6. Viable hatch in hybrid and non-hybrid eggs from 10 different flounder (Platichthys flesus) females (6 of the coastal spawning ecotype and 4 of the off-shore spawning ecotype).
The hybrid larvae ranged in size between 3.01±0.23 and 3.53±0.09 mm (in average 3.23±0.15 mm), while the non-hybrid larvae ranged between 2.93±0.16 and 3.59±0.11 mm (in average 3.24±0.14 mm; Table 4). In the test of larval size at the time of yolk sac depletion I found no difference in size between hybrids and non-hybrids (df=1, F=0.45, p>0.05), showing that growth was similar in these two groups. However, there was a difference between females (df=9, F=57.0, p<0.001) and I also detected an interaction between the factors hybrid/non-hybrid and female (df=9, F=2.26, p<0.05). The two ecotypes display different egg sizes (eggs of the off-shore spawning ecotype being the largest; Nissling et al. 2002) and were included in the same data set, but the differences in egg size had no obvious consequences for larval size as no differences between coastal and off-shore spawning flounder larvae were detected in my material (one-way ANOVA: df=1, F=1.57, p>0.05). Instead, the interaction may result from different trends for different females and shrinkage of larvae from long-term formalin preservation (see Discussion).
0% 20% 40% 60% 80% 100% Via ble ha tch (%)
17
Table 4. Mean size (mm) ± SD at yolk sac depletion in hybrid and non-hybrid larvae from 10 different flounder (Platichthys flesus) females.
Female Hybrid Non-hybrid 5891 3.01±0.23 3.10±0.22 5898 3.19±0.14 3.14±0.16 5916 3.53±0.09 3.51±0.10 5918 3.26±0.15 3.32±0.09 5928 3.04±0.24 3.01±0.21 5929 3.05±0.19 2.93±0.16 5735 3.46±0.12 3.59±0.11 5884 3.38±0.14 3.30±0.15 5887 3.04±0.17 3.08±0.15 5982 3.36±0.07 3.46±0.09
4. DISCUSSION
4.1 Temperature and hybrid experiments
Regarding egg development for coastal spawning flounder at different temperatures, time from fertilization to hatching increased with lower temperatures. This is in line with findings from earlier studies (e.g. for yellowtail flounder, Limanda ferruginea; Benoît and Pepin 1999, Avery et al. 2004). In my analysis of time from hatching to yolk sac depletion, the results show similar trends, as higher temperatures resulted in shorter development periods. However, in both cases the curve leveled off between 8 and 10 °C. Additionally, development time for larvae appears to be less uniform at low temperatures, i.e. in 2 and 4 °C. I conclude that light regimes has little or no impact on the non-uniform development time since light regimes differed between 2 and 4 °C (larvae in 2 °C kept in darkness and larvae in 4 °C kept in light) but these larvae still displayed a similar development pattern.
In the analysis of viable hatch at different temperatures I found that the viable hatch differed significantly, showing the impact of temperature. The fish in the temperature study were caught at the end of the spawning period, meaning that the females only had limited amounts of eggs left. According to Buckley et al. (1991), winter flounder (Pseudopleuronectes
americanus) embryos produced by late-spawning fish were small and appeared to be less
viable. However, the fact that all females in my temperature experiment were caught at the end of the spawning period did not seem to have had an impact on the viable hatch. Instead, the overall high hatching success was ~90 % in temperatures >2 °C, indicating high egg quality. Additionally, as the viable hatch was high in temperatures >2 °C and the eggs in 2 °C experienced higher mortality, the theory that low temperatures could cause coastal spawning flounder to start spawning off-shore during cold years receives further support.
18
example, metabolic rates could be inferior at these temperatures and thus result in lower growth rates. The fact that higher temperatures seem to be inferior, as well as lower temperatures, is supported by Bolla and Holmefjord (1988) and Lein et al. (1997), reporting a greater proportion of deformed halibut (Hippoglossus hippoglossus) yolk sac larvae from eggs incubated at higher temperatures. As seen in Table 1 the temperatures of 6 and 8 °C yielded the largest larvae, which correlates with the highest possible temperatures the larvae would encounter under natural conditions (Andersson 2014). This may seem like a rather narrow temperature window, but the optimum temperature could in theory span between 4.5 and 9.5 °C (the incubation temperatures in my study being 2, 4, 6, 8 and 10 °C, respectively). Additionally, my results of a temperature optimum for larvae between 6 and 8 °C is in line with findings for yellowtail flounder larvae at hatching, displaying larger sizes at intermediate temperatures (7 and 9 °C; Benoît and Pepin 1999). Shrinkage from preservation in formalin should be irrelevant in this case as comparisons of larval length were performed in 2015 on specimens all preserved during 2014, i.e. shrinkage should be similar in all larvae in the temperature experiment.
19
Fig. 7. From top left larvae at the point of yolk sac depletion incubated at 2, 4, 6, 8 and 10 °C, respectively. Notice the differences in eye pigmentation, mouth width and yolk sac shape.
In conclusion, viable hatch was high (approximately 90 %) in all temperatures, except for in 2 °C (approximately 50 %). Larvae were larger in 6 and 8 °C and smaller in 2, 4 and 10 °C. My analyses suggest that both egg and larval development time, viable hatch and larval size are affected at low temperatures, resulting in larvae in poor condition at 2 °C and with somewhat reduced fitness also at 4 °C at time of yolk sac depletion, meaning that the chances of being viable at hatching or growing larger are better at >4 °C. I therefore conclude that higher temperatures (6-10 °C) have an overall positive effect, however with the exception of incubation of larvae in 10 °C as yolk sac depletion rates came to a threshold between 8 and 10 °C and the larvae also displaying somewhat smaller sizes in 10 °C. It therefore appears as if larvae are more sensitive to temperature extremes compared to eggs, and that intermediate temperatures (6-8 °C) are superior regarding both egg and larval development.
20
In the test of hybrid larval size the interaction between the factors hybrid/non-hybrid and female suggested different trends for different females. No differences in larval size were found between coastal and off-shore spawning females, despite the fact that coastal spawning females have smaller eggs. However, egg dry weight is similar for the two ecotypes (Nissling
et al. in prep.), meaning that the yolk sacs contain similar amounts of energy and thus would
allow the larvae to grow equally well. The interaction may instead result from shrinkage during preservation, as fish larvae have been shown to shrink when preserved in formalin (e.g. Hay 1981, Tucker and Chester 1984, Benoît and Pepin 1999, Buchheister and Wilson 2005). In my analysis, larvae from 4 coastal and 2 off-shore spawning females (out of 10 in total) were preserved in 2014 and measured in 2015. These larvae had probably decreased in size as a result of the long-term preservation, their expected differences in size compared to the larvae preserved in 2015 possibly contributing to the interaction. Therefore, long-term preservation of larvae should be avoided if similar studies are conducted in the future.
Further, observations indicated somewhat higher mortality in hybrid larvae compared to in non-hybrid larvae in 2014 (6-36 % in hybrid larvae and 0-23 % in non-hybrid larvae), but no such trend could be seen in 2015 (0-33 % in hybrid larvae and 3-30 % in non-hybrid larvae). It is therefore unclear if there really is a difference in mortality between hybrids and non-hybrids.
In conclusion, hybridization appears to be possible, as cross-fertilization was approximately 100 % and hybrids appear to be as viable as non-hybrids regarding viable hatch and larval size.
4.2 Temperature conditions and hybridization events – the link between reality and the lab
My overall assumption when conducting the experiments described above was that temperature conditions are suboptimal for successful reproduction in coastal waters during cold years but suitable in deeper waters, as temperatures are more stable in these areas. The likeliness for coastal spawning flounder to stay off-shore and spawn should therefore increase during cold years and possibly yield a higher number of hybrids than in years with higher temperatures. This assumption has received support by my experiments, as they indicate that low temperatures are indeed suboptimal regarding development time, viable hatch and larval size. Also, I found no proof that hybrids are less viable than non-hybrids, studying viable hatch and larval size.
21
Table 5. Mean temperatures (°C) at 3-20 and >65 m depth respectively off the east coast of Gotland ± SD, sampled during capture of spawning flounder. The temperature data was sampled in mid-April for the years 2012, 2013 and 2015, and in May (6th-9th) in 2014 (A. Nissling and A-B. Florin, pers. comm.).
2012 2013 2014 2015
Coastal waters (°C) 4.2±0.44 2.35±0.07 6.68±1.1 5.99±0.72 Off-shore (°C) 3.9±0.21 3.6 3.38±0.16 4.58±0.41
Fig. 8. Temperatures in surface waters in the Gotland Basin (station BY15). Dotted lines represent SD and black dots represent temperatures during a) 2012 (solid line represents mean temperature 1995-2004; Andersson 2012), b) 2013 (solid line represents mean temperature 1996-2010; Andersson 2013), c) 2014 (solid line represents mean temperature 1996-2010; Andersson 2014) and d) 2015 (solid line represents mean temperature 1996-2010; Andersson and Hansson 2015).
In order to investigate recruitment levels under natural conditions, I linked temperature data to results from beach seine surveys of 0-gr flounder in late July to early September in 2013, 2014 and 2015 at well-known nursery areas around Gotland. In the year 2012, with average temperatures during spawning, no beach seine surveys were conducted on Gotland later during summer, but recruitment seemed to have been successful in Finnish waters (Jokinen et
al. 2015b), indicating that recruitment may have been successful also around Gotland. Also in
2014, when temperatures were intermediate during spawning, beach seine surveys indicated that recruitment of 0-gr juveniles was high (personal observations). In 2013, when temperatures were low during spawning, virtually no 0-gr flounder were found later during summer, i.e. juveniles of both ecotypes were lacking (personal observations), indicating that
a) b)
22
recruitment failed not only for coastal spawning flounder but also for off-shore spawning flounder. Similar results were reported from Finland (Jokinen et al. 2015b). Additionally, flounder recruitment in 2015 was very low; almost no 0-gr flounder were found around Gotland during summer (personal observations) and temperatures during spawning were rather high both in coastal areas and off-shore. Therefore, it seems like years with intermediate temperatures result in successful recruitment of flounder (2012, 2014) and years with temperatures below or above average result in low flounder recruitment (2013, 2015). The poor recruitment in 2013 is in line with the results from my temperature study as the eggs and larvae of coastal spawning flounder should have encountered temperatures around 2 °C (Table 5, Fig. 8b), i.e. suboptimal temperatures. I also concluded that potential consequences of inactivity during the larval stage due to low temperatures under natural conditions could be troubles foraging, ultimately leading to starvation, and increased risk of the larvae being preyed upon. Thus, temperatures below average have a direct negative impact on flounder recruitment. An alternate explanation could be starvation due to mismatch with preferred food items (Hjort 1914), although this classic explanation is poorly studied from a Baltic Sea perspective. For example, temperature can impact on the peak abundance (Hjort 1914) and composition of zooplankton species, which may in turn affect feeding success and survival during the larval stage (Alheit et al. 2005). The mismatch theory may also apply for the off-shore spawning ecotype in 2013 as the recruitment failure was almost 100 %, i.e. for both ecotypes. Additionally, the recruitment failure in 2015 may be the result of a mismatch with preferred food items due to high temperature, i.e. an indirect consequence of extreme temperatures. On the other hand, Ustups et al. (2013) found that temperature did not explain any significant variability in flounder ichthyoplankton abundance. Instead, both the available reproductive volume, defined as the water column with dissolved oxygen of >1 ml-l and 10.6-12 psu, and spawning stock biomass significantly affected off-shore spawning flounder egg and larval abundance under natural conditions. Their results, contradicting those presented here, indicate that further studies on effects of extreme temperatures on early life stages of Baltic Sea flounder are required.
23
Instead, oxygen levels off-shore during spawning were very low, approximately 0-1 ml-l (Andersson and Hansson 2015), affecting off-shore spawning flounder recruitment in 2015. As warmer water contains less oxygen, this may in part have resulted from the rather high temperatures prevailing off-shore during spring in 2015.
In conclusion, I have shown that hybridization between the two ecotypes is possible. I base this assumption on the fact that coastal spawning males and off-shore spawning females can spawn under similar salinity conditions and because hybrids according to my study are as viable as non-hybrids. It is possible that temperature conditions control the extent of hybridization as a greater proportion of coastal males stay off-shore and spawn during cold years, even though some coastal males are found off-shore during spawning in years with average temperatures. However, my beach seine surveys indicate that recruitment for both flounder ecotypes fails during years of more extreme temperatures, which may limit the outcome of hybridization.
4.3 Hypothesis evaluation
I wished to answer the questions 1) Is fertilization between the two ecotypes possible?, 2) Does hybridization between the two ecotypes affect viable hatch and larval size?, and 3) Could low temperatures have a negative impact on eggs and larvae of the coastal spawning ecotype and therefore be the trigger for hybridization between Baltic Sea coastal and off-shore spawning flounder?
The prediction that low temperatures would adversely affect the larvae of the coastal spawning ecotype has received further support, as development time for eggs and larvae, viable hatch and larval size was inferior in <4 °C. The prediction thus has to be retained. For the prediction that hybridization between the two ecotypes in the Baltic Sea would be possible, but with negative implications for viable hatch and larval size, I found that hybridization indeed is possible for both ecotypes, i.e. both coastal and off-shore spawning females can have their eggs fertilized by males of the other ecotype. However, no negative effects on hybrid viable hatch or larval size were detected, and the prediction thus has to be rejected as hybrids seem to be as viable as non-hybrids.
24
4.4 Further research
The above study examines the event of hybridization and its possible driving forces, i.e. inferior temperature conditions. However, there are still several aspects that need to be further investigated in order to understand the possible event of hybridization between the two Baltic flounder ecotypes.
To begin with, I studied egg and larval development at different temperatures, but the eggs were all fertilized in approximately 8 °C and fertilization levels were close to 100 % in almost all cases. It has been shown that spermatozoa are adversely affected by low temperatures (Alavi and Cosson 2005). If temperature is low during fertilization under natural conditions, it may adversely affect flounder fertilization levels. The low abundance of flounder juveniles in 2013 may therefore not only be the result of low temperatures during the egg and/or larval stages but also the result of low temperatures at fertilization. Therefore, a comparison of fertilization levels at different temperatures is recommended.
Even though egg development was similar between the two ecotypes (both coastal and off-shore spawners hatched at day 7 in the hybrid experiment), larval development was prolonged by two days for larvae stemming from off-shore spawning flounder compared to larvae from coastal spawning flounder. The prolonged larval stage in larvae stemming from eggs of the off-shore spawning ecotype was evident regardless of what type of males had been used and may be the result of the greater egg-size in off-shore spawning flounder. This indicates that coastal and off-shore spawning flounder display slightly different development curves, at least during the larval stage. Therefore, I recommend a temperature study of egg and larval development and viable hatch in off-shore spawning flounder.
Additionally, as observations indicated different trends for hybrid and non-hybrid larval mortality in 2014 and 2015, respectively, I suggest further studies in order to investigate potential differences in mortality between the two groups.
My results indicate that larvae incubated in 10 °C are smaller than larvae reared in 6 and 8 °C, and it is therefore possible that temperatures in the upper range may be suboptimal for flounder growth and development also regarding other aspects than those studied here. In the light of climate change and potentially rising temperatures, this may be an important piece of the puzzle when studying flounder recruitment in the future.
25
cold years actually yield a greater proportion of hybrids or if hybrids from such years are lacking, as my beach seine surveys indicate.
5. ACKNOWLEDGEMENTS
This thesis project was conducted as a complement to BONUS (Art 185) with funding from the EU 7th Framework Programme for Research and Development and FORMAS. Sincere thanks to my supervisor Anders Nissling for guidance, support and patience. Thanks to Ann-Britt Florin, Swedish University of Agricultural Sciences, for providing data of water temperatures, studying my hybrids genetically and taking general interest in my work. Thanks to Oona Lönnstedt, Basil Lange and Sofia Lundberg for useful comments on the manuscript. Also thanks to Karl Kihlberg and Kerstin Kempe for support and encouragement.
6. REFERENCES
Aarnio K, Bonsdorff E and Rosenback N. 1996. Food and feeding habits of juvenile flounder
Platichthys flesus (L.), and turbot Scophthalmus maximus L. in the Åland archipelago,
northern Baltic Sea. J. Sea Res., 36: 311-320.
Alavi SMH and Cosson J. 2005. Sperm motility in fishes. I. Effects of temperature and pH: a review. Cell Biol. Int., 29: 101-110.
Alheit J, Mollmann C, Dutz J, Kornilovs G, Loewe P, Mohrholz V and Wasmund N. 2005. Synchronous ecological regime shifts in the central Baltic and the North Sea in the late 1980s. ICES J. Mar. Sci., 62: 1205-1215.
Andersson L. 2012. Annual report 2012. Marin Miljöövervakning. Swedish Meteorological and Hydrological Institute.
Andersson L. 2013. Annual report 2013. Marin Miljöövervakning. Swedish Meteorological and Hydrological Institute.
Andersson L. 2014. Annual report 2014. Marin Miljöövervakning. Swedish Meteorological and Hydrological Institute.
Andersson L and Hansson M. 2015. Expeditionsrapport v 17, 20-27 april. Marin Miljöövervakning. Swedish Meteorological and Hydrological Institute (in Swedish).
Avery TS, Boyce D and Brown JA. 2004. Mortality of yellowtail flounder, Limanda
ferruginea (Storer), eggs: effects of temperature and hormone-induced ovulation. Aquaculture, 230: 297-311.
Bagge O. 1981. Demersal fishes in the Baltic Sea. In: Voipio A (ed.). The Baltic Sea, pp. 311-333. Elsevier Oceanographic Series no. 30. Elsevier Scientific Company, Amsterdam. Benoît HP and Pepin P. 1999. Interaction of rearing temperature and maternal influence on
egg development rates and larval size at hatch in yellowtail flounder (Pleuronectes
ferrugineus). Can. J. Fish. Aquat. Sci., 56: 785-794.
Bernes C. 2005. Förändringar under ytan – Sveriges havsmiljö granskad på djupet. Fälth & Hässler, Värnamo (in Swedish).
Bisbal GA and Bengtson DA. 1995. Effects of delayed feeding on survival and growth of summer flounder Paralichthys dentatus larvae. Mar. Ecol. Prog. Ser., 121: 301-306. Björklund M. 2005. Evolutionsbiologi. Studentlitteratur, Lund (in Swedish).
Bobe J and Labbé C. 2010. Egg and sperm quality in fish. Gen. Comp. Endocr., 165: 535-548. Bœuf G, Boujard D and Person-Le Ruyet J. 1999. Control of the somatic growth in turbot. J.
26
Bolla S and Holmefjord I. 1988. Effect of temperature and light on development of Atlantic halibut larvae. Aquaculture, 74: 355-358.
Buchheister A and Wilson MT. 2005. Shrinkage correction and length conversion equations for Theragra chalcogramma, Mallotus villosus and Thaleichthys pacificus. J. Fish Biol.,
67: 541-548.
Buckley LJ. 1982. Effects of temperature on growth and biochemical composition of larval winter flounder Pseudopleuronectes americanus. Mar. Ecol. Prog. Ser., 8: 181-186. Buckley LJ, Smigielski AS, Halavik TA, Caldarone EM, Burns BR and Laurence CG. 1991.
Winter flounder Pseudopleuronectes americanus reproductive success. II. Effects of spawning time and female size on size, composition and viability of eggs and larvae.
Mar. Ecol. Prog. Ser., 74: 125-135.
Buckley LJ, Bradley TM and Allen-Guilmette J. 2000. Production, quality, and low temperature incubation of eggs of Atlantic cod Gadus morhua and haddock
Melanogrammus aeglefinus in captivity. J. World Aquacult. Soc., 31: 22-29.
Coombs SH. 1981. A density-gradient column for determining the specific gravity of fish eggs, with particular reference to eggs from the mackerel Scomber scombrus. Mar. Biol.,
63: 101-106.
FishBase. Froese R and Pauly D (eds.). 2011. World Wide Web electronic publication. www.fishbase.org (02/2015) http://fishbase.org/summary/Platichthys-flesus.html. Page visited February 7th, 2015.
Florin A-B. 2005. Flatfishes in the Baltic Sea – a review of biology and fisheries with a focus on Swedish conditions. Finfo: 2005:14.
Florin A-B and Höglund J. 2008. Population structure of flounder (Platichthys flesus) in the Baltic Sea: differences among demersal and pelagic spawners. Heredity, 101: 27-38.
Florin A-B and Lavados G. 2010. Feeding habits of juvenile flatfish in relation to habitat characteristics in the Baltic Sea. Estuar. Coast. Shelf Sci., 86: 607-612.
Hart PT and Purser GJ. 1995. Effects of salinity and temperature on eggs and yolk sac larvae of the greenback flounder (Rhombosolea tapirine Günther, 1862). Aquaculture, 136: 221-230.
Hay DE. 1981. Effects of capture and fixation on gut contents and body size of Pacific herring larvae. Rapp. P.-V. Reun. Cons. Int. Explor. Mer, 178: 395-400.
Hemmer-Hansen J, Nielsen EEG, Grønkjær P and Loeschcke V. 2007. Evolutionary mechanisms shaping the genetic population structure of marine fishes; lessons from the European flounder (Platichthys flesus L.). Mol. Ecol., 16: 3104-3118.
Hjort J. 1914. Fluctuations in the great fisheries of northern Europe viewed in the light of biological research. Rapp. P.-V. Reun. Cons. Int. Explor. Mer, 20: 1-228.
Johannesson K and André C. 2006. Life on the margin: genetic isolation and diversity loss in a peripheral marine ecosystem, the Baltic Sea. Mol. Ecol., 15: 2013-2029.
Jokinen H, Wennhage H, Lappalainen A, Ådjers K, Rask M and Norkko A. 2015a. Decline of flounder (Platichthys flesus (L.)) at the margin of the species' distribution range. J. Sea
Res., 105: 1-9.
Jokinen H, Wennhage H, Ollus V, Aro E and Norkko A. 2015b. Juvenile flatfish in the northern Baltic Sea – long-term decline and potential links to habitat characteristics. J. Sea
Res., 107: 67-75.
Kijewska A, Burzyński A and Wenne R. 2009. Molecular identification of European flounder (Platichthys flesus) and its hybrids with European plaice (Pleuronectes platessa). ICES J.
Mar. Sci., 66: 902-906.
27
Malloy KD and Targett TE. 1991. Feeding, growth and survival of juvenile summer flounder
Paralichthys dentatus: experimental analysis of the effects of temperature and salinity. Mar. Ecol. Prog. Ser., 72: 213-223.
Martinsson J and Nissling A. 2011. Nursery area utilization by turbot (Psetta maxima) and flounder (Platichthys flesus) at Gotland, central Baltic Sea. Boreal Env. Res., 16: 60-70. Molander AR. 1954. Skrubbskäddan, Pleuronectes flesus (L.). In: Andersson KA (ed.). Fiskar
och fiske i Norden, pp. 135-140. Natur och Kultur, Stockholm (in Swedish).
Nash RDM and Geffen AJ. 2005. Age and growth. In: Gibson RN (ed.). Flatfishes: biology
and exploitation, pp 138-163. Fish and Aquatic Resources Series No. 9. Blackwell
Science, Oxford.
Nissling A, Larsson R, Vallin L and Frohlund K. 1998. Assessment of egg and larval viability in cod, Gadus morhua: methods and results from an experimental study. Fish. Res., 38: 169-186.
Nissling A, Westin L and Hjerne O. 2002. Reproductive success in relation to salinity for three flatfish species, dab (Limanda limanda), plaice (Pleuronectes platessa), and flounder (Pleuronectes flesus), in the brackish water Baltic Sea. ICES J. Mar. Sci., 59: 93-108. Nissling A, Jacobsson M and Hallberg N. 2007. Feeding ecology of juvenile turbot
Scophthalmus maximus and flounder Pleuronectes flesus at Gotland, Central Baltic Sea. J. Fish Biol., 70: 1877-1897.
Nissling A and Dahlman G. 2010. Fecundity of flounder, Pleuronectes flesus, in the Baltic Sea – Reproductive strategies in two sympatric populations. J. Sea Res., 64: 190-198. Nissling A, Widbom B, Florin A-B and Gydemo R. 2014. Utveckling av ett hållbart
gotländskt flundrefiske – resursnyttjande och förvaltning. Fiskeområde Gotland.
Norberg B, Kjesbu OS, Tarangere GL, Andersson E and Stefansson SO. 1999. Natural environmental influences on reproduction. Proceedings of the 6th International Symposium on the Reproductive Physiology of Fish. Jon Greg AS, Bergen.
Ojaveer H, Pettai M-L, Neuenfeldt S, Casini M, Möllmann CM, Blenckner T, Horbovy J, Kallasvuo M, Kornilovs G, Lehmann A, Nilsson A, Nissling A and Polte P. 2015. Periodic progress report (incl. kick-off meeting report): Integrating spatial processes into ecosystem models for sustainable utilization of fish resources. BONUS Inspire 7.3, WP 7, UT-EMI. Sandman JA. 1906. Kurzer Bericht über in Finnland aus gefürte Untersuschungen über den
Flunder, den Steinbutt und den Kabeljau. Rapp. P.-v. Réun. Cons. Int. Explor. Mer, 5: 37-44 (in German).
Sims DW, Wearmouth VJ, Genner MJ, Southward AJ and Hawkins SJ. 2004. Low-temperature-driven early spawning migration of a temperate marine fish. J. Anim. Ecol.,
73: 333-341.
SMHI. 2016-01-28. Swedish Meteorological and Hydrological Institute. Inflöden till Östersjön. http://www.smhi.se/kunskapsbanken/oceanografi/infloden-till-ostersjon-1.4203. Page visited 2016-01-28 (in Swedish).
Tucker J and Chester AJ. 1984. Effects of salinity, formalin concentration and buffer on quality of preservation of Southern flounder (Paralichthys lethostigma) larvae. Copeia, 4: 981-988.
Uehara S and Mitani T. 2009. Effect of temperature on the development of eggs and the daily pattern of spawning of round herring Etrumeus teres. Fisheries Sci., 75: 159-165.
Ustups D, Müller-Karulis B, Bergström U, Makarchouk A and Sics I. 2013. The influence of environmental conditions on early life stages of flounder (Platichthys flesus) in the central Baltic Sea. J. Sea Res., 75: 77-84.
28
Implications for freshwater and marine fish, pp. 159-176. Cambridge University Press,
Cambridge.
Voipio A (ed.). 1981. The Baltic Sea. Elsevier Oceanographic Series no. 30. Elsevier Scientific Company, Amsterdam. 418 pp.
Wieland K, Waller U and Schnack D. 1994. Development of Baltic cod eggs at different levels of temperature and oxygen content. Dana, 10: 163-177.
Wootton RJ. 1990. Ecology of teleost fishes. Fish and fisheries Series 1. Chapman and Hall, London.
Yasuda T, Kawabe R, Takahashi T, Murata H, Kurita Y, Nakatsuka N and Arai N. 2010. Habitat shifts in relation to the reproduction of Japanese flounder Paralichthys olivaceus revealed by a depth-temperature data logger. J. Exp. Mar. Biol. Ecol., 385: 50-58.
Yin MC and Blaxter JHS. 1987. Temperature, salinity tolerance, and buoyancy during early development and starvation of Clyde and North Sea herring, cod, and flounder larvae. J.
Exp. Mar. Biol. Ecol., 107: 279-290.
Yoneda M and Wright JP. 2005. Effect of temperature and food availability on reproductive investment of first-time spawning male Atlantic cod, Gadus morhua. ICES J. Mar. Sci.,
29
Appendix 1.
Table 1. Mean temperatures ± SD during the egg stage in the temperature experiment. Temperature data from the first 48 hours, i.e. from fertilization and the two following days, are lacking.
Incubation temperature (C°) 2 4 6 8 10 Female 5803 2.17 ± 0.97 4.49 ± 1.69 6.48 ± 0.2 7.88 ± 0.1 9.9 ± 0.16 Female 5826 2.18 ± 0.95 4.56 ± 1.63 6.48 ± 0.2 7.88 ± 0.1 9.9 ± 0.16 Female 5840 2.14 ± 0.96 4.56 ± 1.63 6.48 ± 0.2 7.87 ± 0.11 9.9 ± 0.16 Female 5845 2.17 ± 0.97 4.49 ± 1.69 6.48 ± 0.2 7.88 ± 0.1 9.9 ± 0.16 Female 5929 1.83 ± 0.53 4.17 ± 0.91 6.33 ± 0.33 7.84 ± 0.05 9.98 ± 0.1
Table 2. Mean temperatures ± SD during the larval stage in the temperature experiment. No larvae stemming from female 5929 were viable in 2 °C at the point of yolk sac depletion and are thus not included.
