85
Candling and field atlas of early egg development in Common 
Eiders Somateria mollissima in the central Baltic
Ovoskopija in terenski atlas zgodnjega razvoja pri gagah Somateria mollissima 
v Osrednjem Baltiku
Svend-Erik Garbus1,2, Peter Lyngs3, Anders Popp Thyme2, Jens Peter Christensen2, Christian 
Sonne1* 
1 Aarhus University, Department of Bioscience, Arctic Research Centre (ARC), Frederiksborgvej 399, PO Box 
358, DK-4000 Roskilde, Denmark; e–mail: cs@bios.au.dk
2 University of Copenhagen, Faculty of Health and Medical Sciences, Department of Veterinary and Animal 
Sciences, DK-1870 Frederiksberg C, Denmark
3 Christiansø Scientific Field Station, Christiansø 97, DK-3760 Gudhjem, Denmark
Here we present the results of candling 258 eggs from 50 nests of 
Common Eiders Somateria mollissima in a colony in the central Baltic. 
Of these, 223 (86%) had a developing foetus. Among the 35 (14%) 
failed eggs, 15 were unfertilized and 20 contained a dead embryo. The 
prevalence of failed eggs is similar to the average proportion of failed 
eggs reported previously by the Christiansø Scientific Field Station 
during 1998–2014. The reason for the high percentage of failed eggs is 
unknown; however, low pre-incubation body mass and energetic stress 
is likely to be the main factor. In addition, we incubated 8 eggs in the 
laboratory from day 0 to hatch in order to follow the development and 
foetal morphology. This resulted in a field atlas from which it is possible 
to estimate date of incubation start using candling in early incubation 
(days 1–12). The atlas is a new possibility for field biologists to estimate 
the first day of incubation of breeding eiders and the prevalence of 
unfertilized and rotten eggs, which is important for studying their 
biology and population dynamics.
Key words: Common Eider, Baltic, candling, fertilization, starvation, 
stress
Ključne besede: gaga, Baltik, ovoskopija, oploditev, stradanje, stres
Acrocephalus 39 (178/179): 85–90, 2018
10.1515/acro-2018-0007
1. Introduction
The Christiansø archipelago in the southern part of 
the Baltic Proper holds the second largest Danish 
colony of breeding Common Eiders Somateria 
molissima (Christensen & Bregnballe 2011). 
It consists of 1,500–1,700 pairs and has been 
monitored since 1950 (Lyngs 1992, 2009, 2014). 
This long-term monitoring revealed that during 
1970–1990, the population increased from approxi-
mately 1,000 to approximately 3,000 breeding pairs, 
while from 1990 until today the population has 
declined by approximately 50% (Lyngs 2009, 2014). 
The reasons for this is unknown, however, limited 
access to food in wintering areas and the breeding 
location, infectious diseases, blooming toxic algae 
86
and failed eggs have been suggested as some of the 
main reasons (Buchman 2010, Camphuysen 
et al. 2002, Christensen et al. 2008, Larsson 
et al. 2014, Laursen & Møller 2014). 
In order to understand the fluctuations in 
the number of breeding eiders and clutch size, 
we established a research programme running in 
2015–18. Briefly, the study showed that incubating 
females underwent extreme physiological stress 
over the approximately 26-day incubation period. 
The weight loss for a certain proportion of the 
colony in some years is borderline to cachexia and 
breeding failure (Garbus 2016, Garbus et al. 
2018). As part of this, the prevalence of failed eggs 
has been studied, since earlier indications suggest 
that this could be a potential problem for the colony 
and population dynamics (Lyngs 2009, 2014). 
Here, we present the results of the controlled 
study of candling incubating eider eggs and 
provide the first field atlas of developmental 
stages in Common Eiders. Eggs were monitored 
and candled, and controlled incubation in the 
laboratory was conducted. The present publication 
is of value to field studies that measure fertility 
ratios and estimations of date of incubation start in 
breeding eiders.
2. Materials and methods
During April-May 2015, 258 eggs and egg 
membranes from 50 nests of incubating habituated 
eiders were monitored at Christiansø located 
northeast of Bornholm in the Baltic Proper 
(55°19’N 15°11’E, Figure 1). All 258 eggs were 
candled to investigate the frequency of failed eggs 
and the relation to the health of female eiders. An 
oviscope was constructed from a flashlight (Cree 
XPE High Power LED 3 W 150 Lumen) modified 
with a rubber cup on the top (Figure 2). The 
flashlight had a sufficient brightness to examine 
the eider eggs. The entire candling procedure was 
performed in darkness using a mackintosh covered 
cardboard box. Using nitrile examination gloves, 
the base of the egg was held between thumb and 
forefinger, placed directly against the light and 
tilted slightly to one side while rotating the egg. 
The candling was conducted from day 8 to 12 of the 
incubation stage as recommended for ducks having 
long incubation periods (Ernst et al. 2004). The 
10-second candling is considered to pose no risk to 
the further egg development. 
Birds were kept in a cotton bag during all 
procedures to reduce stress. In cases where all eggs 
were removed from the nest due to complete failure, 
artificial replacement eggs were used to keep birds at 
their nests and thereby reduce stress levels. All nests 
were checked daily and visually inspected at distance. 
Eggs were divided into two groups: active eggs 
and failed eggs (Hemmings et al. 2011). Failed eggs 
were further divided into infertile eggs or dead 
embryos. Information on the methods and their 
interpretations is described in Ernst et al. (2004). 
In addition to this, newly laid eggs in 2016 (n=4) 
and 2017 (n=4) were brought for incubation in the 
laboratory. The eggs were incubated on days 1–28 and 
the hatched ducklings were brought back to the nest 
and taken over by either the mother or a nanny. The 
nests were visually observed, and all ducklings were 
successfully accepted by their mothers or nannies. 
An America motor incubator (America A/S, Thisted, 
Denmark) was used for incubation. A thermometer 
and hygrometer were placed inside the incubator 
in order to monitor and adjust temperature and 
humidity. The settings used were 38.5°C on the first 
21 days with a humidity of 55%. In the last 4–5 days 
of incubation, a temperature of 38.9°C was applied. 
From day 8, the eggs were humidified daily with a spray 
containing water at 38.5 °C. The eggs were turned four 
times daily (8.00, 12.00, 18.00 and 22.00). All eggs 
were photographed daily at 13.00 and simultaneously 
cooled during transport (1–2  minutes). Around the 
time of hatching, humidity was adjusted to 70%. For 
photodocumentation of egg development, a Cree 
XPE High Power LED 3 W 150 lumen was used for 
early development and an EC4S NITECORE® LED 
Flashlight 2150 Lumen for late development. Lumen 
settings varied from 1000–2150. Both flashlights 
were modified with rubber cubs placed in a holder 
fitting the morphology of the eggs. Pictures were 
taken with a Nikon D3S using a 60 mm macro lens 
(f/9, 1250 ISO, 1 sec exposure) in a partly dark room.
For basic training, the candling procedure 
was tested at the Department of Disease Biology, 
University of Copenhagen in March 2015 on day 
10 of incubation. Candling of 200 chicken eggs 
showed fourteen unfertilized and seven dead 
embryos adding up to 10% failed eggs. No false 
negatives were observed.
S-E. Garbus et al.: Candling and field atlas of early egg development in Common Eiders Somateria mollissima in the central Baltic
87
Christiansø
Figure 1. Map of the archipelago Christiansø study area
Slika 1: Zemljevid preučevanega območja v arhipelagu Christiansø
Figure 2. Egg candling using a modified Cree XPE High 
Power LED® as oviscope inside a cardboard box 
Slika 2: Ovoskopija z uporabo modificiranega orodja Cree 
XPE High Power LED® kot oviskopa v kartonski škatli
Acrocephalus 39 (178/179): 85–90, 2018
3. Results and discussion
3.1. Candling and hatching
The candling and egg membrane counting showed 
that 223 (86%) eggs had a developing foetus. 
Among the 35 (14%) failed eggs, fifteen were un-
fertilized (6%) and twenty contained dead embryos 
(8%). This percentage of failed eggs (14%) is the 
same as the average proportion of failed eggs found 
by the Christiansø Scientific Field Station during 
1998-2014 (14%; range 8–23%) (Lyngs 2014). 
To the best of our best knowledge, no other 
large-scale investigations on failed eider eggs have 
been conducted, making it difficult to compare 
with the proportion of failed eggs in other colonies. 
However, the loss of incubated eggs from modern, 
high-hatching chicken strains, stored under 
optimal conditions, should be no more than 10%. 
88
Losses in waterfowl may be slightly higher (Ernst 
et al. 2004). The reasons for the relatively large 
proportion of failed eggs in the Christiansø colony 
is unknown, however, low initial body weight 
and energetic stress is likely to be the main reason 
(Garbus 2016, Garbus et al. 2018, Hemmings 
et al. 2011). Nutritional deficiencies, early infections 
and contaminant exposure constitute the common 
cause of dead-in-shell embryos in chickens, which is 
a supporting weight of evidence for similar cause and 
effects in the present investigation of the Christiansø 
breeding colony (Alcorn 2008, Hoffman 1990).
3.2. Candling atlas
Figure 3 shows the atlas of the day 1–12 stages 
of incubating Christiansø eider foetus. It is seen 
that the centrally located embryo takes form as a 
small-condensed area forming into a red / dark 
area. From the embryo, blood vessels radiate to the 
extra-embryonic membrane. The simple circulatory 
system evolves into a more advanced system and the 
embryo takes gradually more space of the egg. After 
day 12, the examination of the egg is complicated or 
no longer suitable as a measure of foetal age as the 
foetus gets darker and takes up more space. During 
the last period of the egg development, the foetus is 
only visible as a dark shadow with a prominent air 
sac. In addition, Figure 4 shows examples of dead and 
infertile eggs, respectively, at Christiansø. A blood 
ring is seen in the dead egg, and the infertile egg 
appears clear and transparent with no development.
3.3. Future considerations
The atlas and candling method provide field 
biologists with a tool to estimate first day of 
incubation of breeding eiders and prevalence of 
unfertilized and rotten eggs, which is important 
for studying their biology and population dynamics 
(Robertson & Cooke 1993) including what 
is known about males during the winter season 
(Garbus et al. In press). 
Acknowledgements
The study was approved by the Danish Nature 
Agency (SVANA). For Funding of the study, we 
acknowledge The Danish Environmental Protection 
Agency, 15. Juni Fonden, Jægernes Naturfond, 
SKOV A/S (Glyngøre 7870 Roslev, Denmark) and 
BONUS BALTHEALTH that has received funding 
from BONUS (Art. 185), funded jointly by the EU, 
Innovation Fund Denmark (grants 6180-00001B 
and 6180-00002B), Forschungszentrum Jülich 
GmbH, German Federal Ministry of Education 
and Research (grant FKZ 03F0767A), Academy of 
Finland (grant 311966) and Swedish Foundation for 
Strategic Environmental Research (MISTRA). In 
addition, we are grateful to the people of Christiansø 
for their help with reporting and collecting dead 
eiders in 2007 and 2015.
Povzetek
Avtorji članka predstavljajo rezultate ovoskopije 
258 jajc iz 50 gnezd gage Somateria mollissima v 
koloniji na Osrednjem Baltiku. Od teh jih je 223 
(86  %) vsebovalo razvijajoče se zarodke. Med 35 
(14  %) propadlimi jajci je bilo 14 neoplojenih, 20 
pa jih je vsebovalo mrtve zarodke. Prevladujoče 
število propadlih jajc je podobno povprečnem deležu 
propadlih jajc, o katerem so v obdobju 1998–2014 
poročali z Znanstvene terenske postaje v arhipelagu 
Christiansø. Razlog za ta visoki odstotek propadlih 
jajc sicer ni znan, vendar avtorji domnevajo, da gre 
glavni faktor za takšno stanje po vsej verjetnosti 
pripisati telesni masi pred valjenjem in energetskemu 
stresu. Da bi sledili razvoju in zarodkovni morfologiji, 
so se avtorji odločili za inkubacijo osmih jajc v labo-
ratoriju od dneva 0. Rezultat je bil terenski atlas, iz 
katerega je mogoče oceniti datum začetka inkubacije 
z uporabo ovoskopije med zgodnjo inkubacijo (dnevi 
1–12). Atlas ponuja terenskim biologom novo 
možnost ocenjevanja prvega dne valjenja gnezdečih 
gag in prevladujoče število neoplojenih in gnilih jajc, 
kar je pomembno za preučevanje njihove biologije in 
populacijske dinamike.
4. References
Alcorn MJ (2008): How to carry out a field 
investigation. pp. 14-38. In: Jordan F. T. W, 
Pattison M., Alexander D., Faragher T. (eds.): 
Poultry Diseases. – W.B. Saunders, London.
Buchmann K. (2010): Døde edderfuglene af et 
parasitangreb? (Did the eiders die from parasitic 
infections?)  – Natur på Bornholm 8: 18–21. (in 
Danish)
S-E. Garbus et al.: Candling and field atlas of early egg development in Common Eiders Somateria mollissima in the central Baltic
89
Day 1 Day 2 Day 3
Day 4 Day 5 Day 6
Day 7 Day 8 Day 9
Day 10 Day 11 Day 12
Common Eider eggs
Figure 3: Stages of egg incubation on days 1–12 at Christiansø
Slika 3: Faze valjenja v dnevih 1–12 v arhipelagu Christiansø
Acrocephalus 39 (178/179): 85–90, 2018
90
Christensen JP, Bojesen MA, Bisgaard M. (2008): 
Fowl cholera. pp. 149–154. In: Jordan F. T. W, 
Pattison M., Alexander D., Faragher T. (eds.): 
Poultry Diseases. – W.B. Saunders, London.
Christensen TK, Bregnballe T. (2011): Status of 
the Danish breeding population of Eiders Somateria 
mollissima 2010.  – Dan Ornol Foren Tidsskr 105: 
195–205. 
Ernst RA, Bradley FA, Abbott UK, Craig RM 
(2004): Egg candling and breakout analysis. – ANR 
Publication 8134.
Garbus SE (2016): Health, behavior, egg failure and 
starvation-mortality of incubating common eiders 
(Somateria mollissima) at Christiansø, Central Baltic 
Sea. MSc thesis, University of Copenhagen, Denmark.
Garbus SE, Lyngs P, Christensen JP, Buchmann K, 
Eulaers I, Mosbech A, Dietz R, Gilchrist HG, 
Sonne C. (2018): Common eider (Somateria mollissima) 
body condition and parasitic load during a mortality 
event in the Baltic proper. – Avian Biol Res 11:167–172.
Garbus SE, Krogh AKH, Jacobsen ML, Sonne C. 
(Accepted for publication – In press): Pathology and 
plasma biochemistry of common eider (Somateria 
mollissima) males wintering in the Danish part of 
Western Baltic. – J Avian Med Surg.
Hemmings N, West M, Birkhead TR (2012): Causes 
of hatching failure in endangered birds. – Biol Lett 
8: 964–967. 
Hoffman DJ (1990): Embryotoxicity and Teratogenicity 
of Environmental Contaminants to Bird Eggs. 
In: Ware G.W. (ed.): Reviews of Environmental 
Contamination and Toxicology. Reviews of 
Environmental Contamination and Toxicology, vol 
115. – Springer, New York.
Larsson K., Hajdu M., Kilpi R., Larsson R., Leito A., 
Lyngs P. (2014): Effects of an extensive Prymnesium 
polylepis bloom on breeding eiders in the Baltic Sea. – 
J Sea Res 88: 21–28.
Laursen K., Møller AP (2014): Long-Term Changes 
in Nutrients and Mussel Stocks Are Related to 
Numbers of Breeding Eiders Somateria mollissima at 
a Large Baltic Colony. – PLoS ONE 9:e95851.
Lyngs P. (1992): Ynglefuglene på Græsholm 1925–90. – 
Dan Orn Foren Tidsskr 94: 12–18.
Lyngs P. (2009): Christiansø Fieldstation  – [http://
www.chnf.dk/aktuelt/edf09/edfugl09_2sp.php], 
01/01/2016. 
Lyngs P. (2014): Christiansø Fieldstation.  – [http://
w w w.ch n f .d k /a kt uelt/ed f 14/ed f ug l14 .php], 
01/01/2016.
Robertson GJ, Cooke F. (1993): Intraclutch egg-
size variation and hatching success in the common 
eider. – Can J Zool 71: 544–549.
Prispelo / Arrived: 17. 9. 2018
Sprejeto / Accepted: 12. 12. 2018
Figure 4: From top to bottom: active egg (day 7), 
unfertilized egg, egg containing dead embryo at 
Christiansø. Note the circulatory system in the active egg 
and that the infertile egg appears clear and transparent 
with no development, while the beginning of a blood 
ring is seen in the dead egg (arrow). 
Slika 4: Od zgoraj navzdol: aktivno jajce (7. dan), 
neoplojeno jajce, jajce z mrtvim zarodkom. Glej obtočni 
sistem v aktivnem jajcu in kako je neplodno jajce videti 
čisto in prozorno brez znakov razvoja, medtem ko je 
v mrtvem jajcu opaziti začetek oblikovanja krvnega 
prstana (puščica).
Camphuysen CJ, Berrevoets CM, Cremers HJWM, 
Dekinga A., Dekker R., Ens BJ, Van der Have TM, 
Kats RKH, Kuiken T., Leopold MF, van der Meer 
J., Piersma T. (2002): Mass mortality of common 
eiders (Somateria mollissima) in the Dutch Wadden 
Sea, winter 1999/2000: Starvation in a commercially 
exploited wetland of international importance.  – 
Biol Conserv 106: 303–317. 
The active egg (day 7) Note the circulatory system
The unfertilized egg Note the clear yolk
The dead embryo The blood rings observed indicates 
death of embryo
S-E. Garbus et al.: Candling and field atlas of early egg development in Common Eiders Somateria mollissima in the central Baltic