J. Field Ornithol., 68(3):376-381
FACTORS
PHOENIX
AFFECTING
PETREL
EGG
ON
MASS
LOSS
CHRISTMAS
IN
THE
ISLAND
G. C. WHITTOW
Departmentof Physiology
John A. Burns Schoolof Medicine
Universityof Hawaii
1960 East West Road
Honolulu, Hawaii 96822 USA
Abstract.--Field measurementsrevealed that the masslossof naturally incubated Phoenix
Petrel(Pterodroma
alba)eggswasonly76.8%of the expectedvaluefor a procellariiformegg.
The low masslosscouldnot be attributedto a low egg temperature,characteristic
of many
procellariiformeggs;egg temperature(37.3 C) washigher than expected,while the body
temperatureof incubatingadult birdswas38.2 C. The body temperature-•eggtemperature
difference (0.9 C) was the lowestreported for any bird. Laboratory measurementson the
egg shellsindicatedthat the low masslossof the eggscould not be accountedfor by a thick
eggshell;it appearedto be related to a low porosityof the eggshell.
FACTORES
QUEAFECTANLA PP•RDIDA
DE PESOEN LOSHUEVOSDE
INDIVIDUOS
DE PTERODROMA
ALBA, EN LA ISLA CHRISTMAS
Sinopsis.--Datosde camporevelaronque la p•rdida de pesode huevosincubadosnaturalmente por individuosde Pterodroma
albaresult6ser el 76.8% de losvaloresesperados
para
un huevo de procelariforme.La baja p•rdida de peso no puede ser atribuida a la baja
temperatura de los huevos,t/pica de este grupo de aves.La temperaturade incubaci6n
result6ser milsalta de lo esperado(37.3 C), mientrasque la temperau•radel aveque incubabafu• de 38.2 C. La diferencia entre la temperatura del ave incubandoy la del huevo
(0.9 C) ha resultadoser la m•s baja informada para ave alguna.Medidasdel grosorde los
huevosindican que la baja perdida de pesono puede atribuirseal grosorde estos.La misma
parece estarrelacionadaa la baja porosidadde los cascarones.
The Phoenix Petrel (Pterodromaalba) is a tropical procellariiform seabird that nests on small islands in the central Pacific Ocean (Warham
1990). It laysa singlewhite egg on the surfaceof the ground (Murphy
et al. 1954). Only fragmentaryinformation on the incubation biologyof
the egg is availablein the literature. Murphy and Pennoyer (1952) published the dimensionsof eight Phoenix Petrel eggs,and Schreiberand
Ashmole(1970) reportedthe incubationperiodto be 53 d. The present
report stemsfrom the opportunity to obtain additional data on Phoenix
Petrel eggs.The emphasisof the studywasto measurethe masslossof
naturally incubated eggsand to relate the masslossto other characteristicsof the egg.
METHODS
Observationswere made on Motu Tabu and Motu Upua isletsin the
lagoon of ChristmasIsland (1ø59'N, 157ø26'W).The dimensionsof 31
eggswere measuredwith dial micrometer calipersthat could be read to
0.025 mm. The masslossof 20 naturally incubated, unpipped eggswas
measuredby weighing the eggsat intervals of 19.5-48.1 h on an Ohaus
field balance (model 1010-10) at the nest site. The temperatureof ten
376
Vol.6s,No.•
Phoenix
Petrel
Eggs
[377
other incubatedeggswasdetermined by quicklyremovingthe egg from
the nest and insertinga fast-response,
needle-thermistorprobe (Yellow
SpringsInstrument Company,"YSI", no. 524) into the egg. This procedure required approximately20 s. All egg temperatureswere taken from
the center of the egg and they were measuredby connectingthe thermistor lead to a YSI Telethermometer (model no.43 TK). The mass of 20
adult PhoenixPetrelswasmeasuredwith an Ohausspringbalance(model
8011) graduatedin divisionsof 2 g. The stomachtemperaturesof 20 adult
birds were measuredwith a YSI thermistor probe (no. 401) connected to
the YSI Telethermometer. All measurementsof stomach temperature
were made on birds that were incubating their eggs.A record was kept
of the sequenceof eventsduring pipping of the eggs.Some measurementswere made on eggshellstransportedto the Universityof Hawaii in
Honolulu. Shell thickness was measured with micrometer calipers
equipped with a ball attachment on the spindle to fit the curvature of
the shell. The smallestmicrometer graduationswere 0.025 mm; all measurementswere made on eggshellsthat had been dried in the laboratory
at the Universityof Hawaii. The number of poresin the shellswascounted by the method describedby Tyler (1953) and Roudybushet al. (1980).
The measure of variation used in this report is one standard deviation
(SO).
RESULTS
AND
DISCUSSION
The egg measurementsare presented in Table 1, together with the
predicted valuesfor procellariiform eggsbasedeither on the massof the
freshly laid egg, or on the fresh-eggmassand the incubation period
(Rahn and Whittow 1988).
Egg dimensions,bodymass,and incubationperiod.-•The mean dimensions (length = 5.606 cm; width = 4.254 cm) of eight eggscollectedby
Murphy and Pennoyer (1952) at Christmas Island and in the Tonga
group, are very similar to the mean valuesshown in Table 1. The mean
dimensionsof the eggsmay be used to obtain an estimateof the massof
the freshlylaid eggs(Hoyt 1979). The mean value for fresh-eggmassso
obtained is 54.94 g. The mean body massof 20 adult birds measuredin
the present studyis 259.4 g _ 28.1 (SD). Hence, the estimatedfresh-egg
massis 21.2% of the adult mass--higherthan the mean value of 15.8%
for Procellariiformes (Rahn and Whittow 1988). The estimated mass of
the freshly laid egg (54.94 g) may be used to predict an expected incubation period for a procellariiform bird (Rahn and Whittow 1988). The
predictedvalue is 52.2 d, closeto the incubationperiod (53 d) measured
by Schreiberand Ashmole (1970).
Egg massloss.-•The fresh-eggmassand the incubation period may be
used to predict the daily masslossof the egg (Rahn and Whittow 1988).
The measuredvalue in unpipped eggs(Table 1) wasonly 76.8% of the
predicted value (160.62 mg/d) for a procellariiform egg weighing 54.94
g with an incubation period of 53 days.The masslossof the avian egg
during incubation is due entirely to the lossof water vapor (Rahn and
378]
G.C. Whittow
J.Field
Ornithol.
Summer
1997
TABLE1. Measured and predicted valuesfor Phoenix Petrel eggs (mean values + 1 SD).
Number of measurementsin parentheses.Predictions are from Rahn and Whittow
(1988) unless indicated otherwise.
Measured/
Predicted
Measured
Egg length (cm)
Egg width (cm)
5.771 _+ 0.198 (31)
4.168 _+ 0.114 (31)
Predicted
X 100
pd
---
---
--<0.001
Daily water lossfrom
unpippedeggs(mg/
d)
Egg temperature(C)
Shell thickness (ram)
Outer
160.62
76.8
37.28 + 0.65 (10)
35.1a
--
--
96.3
•
--
--
--
--
--
--
0.18 _+ 0.01 (142)
0.27b
shell membrane
thickness (mm)
Inner
123.34 _ 23.97 (20)
0.06
_
0.01
(56)
shell membrane
thickness(mm)
Shell pore density
(pores/cm2)
Number of pores
(pores/egg)
0.02 _+ 0.01 (7)
55.1 _ 4.5
3778.8
(8)
110.8c
49.7
<0.001
4283.4
88.2
--
Rahn (1991).
Shell and shell membranes.
Tullet and Board (1977).
Statisticalsignificanceof differencesbetweenmeasuredand predictedvalues.
Ar 1974). One of the determinantsof the water lossfrom the egg is the
egg temperature, which defines the water-vaporpressureof the contents
of the egg (Rahn et al. 1977). The higher the water-vaporpressureof the
egg, the greater the rate of water loss.
Egg and bodytemperature.--Themean egg temperature of ten Phoenix
Petrel eggs (37.3 C, Table 1) was higher than expected;the mean egg
temperature of eight speciesof Procellariiformeswas 35.1 C, according
to Rahn (1971). Although environmentaltemperatureswere not measured in this study,the climate of ChristmasIsland is equatorial (Gallagher 1960), and it seemspossiblethat the high egg temperaturewasrelated
to the hot conditions at the nesting site, as well as to the fact that the
Phoenix Petrel laysits egg on the surfaceof the ground. It is unlikely
that the relativelyhigh egg temperaturereflectsa high body temperature
of the incubatingadult bird becausethe measureddeep-bodytemperature of 20 incubating adult Phoenix Petrelswas 38.2 C ___0.56. This is
slightlylower than the mean body temperature(38.78 C) of 31 species
of Procellariiformes
(Warham1971).Noteworthyisthe smalltemperature
difference(0.9 C) betweenthe incubatingbird (38.2 C) and the egg (37.3
C)--the smallestbird-egg temperature difference reported for any bird
(Rahn 1991). This small difference may reflect the warm conditionsat
the nest site, but it also suggestsa tight coupling between the bird and
its egg, in the Phoenix Petrel. Support for the latter possibilitycomes
Vol.68,No.3
Phoenix
Petrel
Eggs
[379
from Gallagher's(1960) statement"... the parent bird sitsvery close."
The closecoupling of egg and body temperaturemay help to prevent
over-heatingof the egg.
Eggshellthickness
and porosity.
mTwo characteristicsof the eggshellmay
have a profound effect on the masslossof the egg. The thicknessof the
eggshelldefines the length of the diffusion pathwayalong which water
leavesthe egg (Ar et al. 1974). The thicker the shell,the lowerthe water
loss(and thus massloss)of the egg. The measuredthicknessof the shell
(plusthe shell membranes)is 0.26 mm (Table 1); this is very closeto the
predictedvalue (0.27 mm) for a procellariiformegg weighing54.94 g
(Rahn and Whittow 1988).
Potentiallymore important than the thicknessof the eggshellin limiting the water lossfrom the avianeggis the number of microscopicpores
in the eggshell.Water vapor diffusesout of the egg through the pores;
consequently,
with fewer pores,there is reduced egg massloss.The pore
densityof PhoenixPetrel eggswasonly 49.7% of that of birdsin general
(Table 1; Tullet and Board1977). The total numberof poresin eachegg
may be calculatedfrom the pore densityif the total surfacearea of the
eggshellis known. Shell surfaceareaswere not measuredin the present
studybut they may be calculatedfrom the estimatesof the massof the
freshlylaid eggusingthe relationshipreportedby Paganelliet al. (1974).
Multiplyingthisestimatedmeansurfacearea (68.58 cm2) bythe measured
mean pore density(Table 1), yieldsa mean value (3778.8) for the total
number of pores (Table 1). This value is considerablylessthan the predicted value (4283.4) for a procellariiformegg weighing54.94 g with an
incubation period of 53 d (Table 1). Thus, the proximate causeof the
low rate of masslossof the Phoenix Petrel egg appearsto be a reduced
porosityof the shell.
Sequence
of eventsduringpipping.--The initial event during pipping of
the Phoenix Petrel egg is a star-fractureof the shell (external pipping).
This is followed by the formation of a pip-hole in the shell and penetration of the aircell by the beak of the embryo (internal pipping). This
sequenceof eventsis similar to that in the congenericBonin Petrel (Pterodromahypoleuca),another tropical petrel (Pettit et al. 1982). The initial
star fracture of the shell, signallingthe beginning of pipping, occurred
3.75 d before hatching.Hence, the durationof the pre-pippingperiod
was49.25 (53 - 3.75) d. In three PhoenixPetrel eggs,the mean duration
of the pip-hole phasewas47.7 h. This is longer than the pip-holephase
in the Bonin
Petrel.
CONCLUSIONS
It is interestingto speculateon the reasonsfor the low egg massloss
(in unpipped eggs)and low shellporosityin the PhoenixPetrel, evenby
procellariiform standards (Rahn and Whittow 1988). Both features are
characteristicof seabirdswith prolonged incubation (Whittow 1980,
1984). The incubationperiod of the PhoenixPetrel egg, however,is no
longer than expectedfor a procellariiformegg of its size.The low egg
380]
G.C. Whittow
J.Field
Ornithol.
Summer
1997
massloss cannot be explained by the relativelyhigh egg temperature:
other things being equal, a high egg temperature implies a high watervapor pressureinside the egg, and a high rate of water lossfrom the egg
(Whittow 1984). A possibleexplanation may lie in the observationson
pipping of the eggs.Egg masslossis much greater from pipped than
from unpipped eggs (Pettit and Whittow 1983), and it is greatestof all
from eggswith a distinctpip hole. The pip-hole phase seemsto be particularly long in the Phoenix Petrel egg and, although water lossfrom
pipped eggswasnot measuredin the presentstudy,it may be safelyinferred that it is high during the pip-hole phase. It is conceivablethat
water lossfrom unpipped eggsis curtailed as a result of a low porosity,
in order to compensatefor a high masslossfrom pipped eggs.Water loss
from pipped eggsoccurslargelythrough cracksand pip holesin the shell.
The total water lossfrom eggs during incubation is closelyregulated
(Rahn and Ar 1974). Hence, unusuallyhigh masslossfrom pipped eggs
might have to be balancedby reduced masslossin unpipped eggs.Some
support for this contention comes from computation of the total mass
lossof the egg before pipping. Multiplying the duration of the pre-pipping period (49.25 d) by the daily masslossof the egg (Table 1) yields
the total masslossof the egg during the pre-pippingperiod (=6.074 g).
This represents11.0% of the massof the freshly-laidegg (54.94 g) and it
is lower than the correspondingfigure for four other speciesof Procellariiformes(Whittow et al. 1982). Thus, asa workinghypothesis,the total
masslossof the unpipped Phoenix Petrel egg is reduced to balancethe
relatively high masslossof pipped eggs.Further field measurementsof
the masslossof pipped PhoenixPetrel eggswill be needed to substantiate
this.
ACKNOWLEDGMENTS
I am very grateful to my wife, Christina,for assistance
during the field work, and to Dr.
Martin
C, Garnett and Mr. Katino Teeb'aki of the Wildlife
Conservation Unit, Christmas
Island, for considerableassistance
with transportationto Motu Tabu and Motu Upua. I am
also grateful to the Hon. David Brechtefeld,Minister for the Line and Phoenix Groups,
Kiribati,for grantinga permit for this study.The work wassupportedby a grant (No. PCM
76-12351 AO1) from the National Science Foundation.
LITERATURE
CITED
AR. A., C. V. PAGANELLI,
R. B. REEVES,
D. G. GREENE,
)•NDH. RAHN. 1974. The avianegg:
watervaporconductance,
shellthickness
and functionalporearea.Condor76: 153-158.
B)dLEY,
E. 1977. The Christmas
IslandStory.StacyInternational,London,UnitedKingdom.
88pp.
GALLAGHER,M.D.
1960. Bird notes from Christmas Island, Pacific Ocean. Ibis 102: 489-
502.
HoYr, D. F. 1979. Practicalmethodsof estimatingvolume and freshweightof bird eggs.
Auk 96: 73-77.
MURPHY,
R. C., ANDJ.M. PENNOYER.
1952. Largerpetrelsof the genusPterodroma.
Am. Mus.
Novit. No. 1580. pp. 43.
, A.M. B)dLEY,AND R. J. NIEDRACH.1954. Canton Island. Mus. Pictorial No. 10.
Denver Mus. Nat. Hist., Denver,CO. 78 pp.
Vol.68,No.•
Phoenix
Petrel
Eggs
[381
PAGANELLI,
C. V., A. OLSZOWKA,
ANDA. A•. 1974. The avian egg:surfacearea, volume, and
density.Condor 76: 319-325.
PETTIT,T. N., ANDG. C. WHITTOW. 1983. Water lossfrom pipped Wedge-tailedShearwater
eggs.Condor 85: 107-109.
, G. S. GRANT,G. C. WHITTOW,H. RAHN,ANDC. V. PAGANELLI.
1982. Respiratorygas
exchangeand growthof Bonin Petrel embryos.Physiol.Zool. 55: 162-170.
RAHN,H. 1991. Why birdslay eggs.Pp. 345-360, in D.C. Deeming,and M. W.J. Ferguson,
eds.Egg incubation:its effectson embryonicdevelopmentin birds and reptiles.Cambridge UniversityPress,Cambridge,United Kingdom.
, ANDA. AR. 1974. The avian egg: incubation time and water loss.Condor 76: 147152.
, AND G. C. WHIT•OW. 1988. Adaptations to a pelagic life: eggs of the albatross,
shearwaterand petrel. Comp. Biochem.Physiol.91A: 415-423.
, R. A. AcI•a•,
ANDC. V. P^GANELLI.1977. Humidity in the avian nest and egg
water lossduring incubation.Physiol.Zool. 50: 269-283.
ROUDYBUSH,
T., L. HOFFMAN,ANDH. RAHN. 1980. Conductance,pore geometry and water
lossof eggsof Cassin'sAuklet. Condor 82: 105-106.
SCHREIBER,
R. W., ANDN. P. ASHMOLE.1970. Sea-birdbreedingseasonson ChristmasIsland,
Pacific Ocean.
Ibis 112: 363-394.
TULLET,S. G., ANDR. G. BO•D. 1977. Determinantsof avian porosity.J. Zool., Lond. 183:
203-211.
T•R,
C. 1953. Studieson eggshellsII. a method for marking and countingpores.J. Sci.
Food. Agric. 4: 266-272.
W•a•-I•.M,J. 1971. Bodytemperaturesof petrels.Condor 73: 214-219.
1990. The petrels: their ecology/and breeding •stems. Academic Press,London.
440pp.
WHITTOW,G. C. 1980. Physiologicaland ecologicalcorrelatesof prolonged incubation in
seabirds. Am. Zool. 20: 427-436.
ß 1984. Physiologicalecology/ofincubationin tropical seabirds.Pp. 47-72, in R. W.
Schreiber, ed. Tropical seabird biology Studiesin Avian Biology/No. 8. The Cooper
Ornithological Society,Allen Press,Lawrence,Kansas.
, R. A. AcI•a•,
C. V. P^GAN•I, ANDT. N. P•T•IT. 1982. Pre-pipping water loss
from the eggsof the Wedge-tailedShearwater.Comp. Biochem.Physiol.72A: 29-34.
Received17Jun. 1996;accepted11 Sep. 1996.