I B I S 132: 68-77

Weight loss of Antarctic Fulmars Fulmarus glacialoides during incubation and chick brooding

H . W E I M E R S K I R C H Centre National de la Recherche Scientifique, Centre &Etudes Biologiques des

Animaux Sauvages, 79360 Beauvoir sur Niort, France

Submitted 10 October 1988; accepted 24 April 1989

Weight losses incurred by incubating birds of both sexes, and weight changes at the beginning and end of shifts between successive incubation and brooding shifts were studied in the Antarctic Fulmar Fulmarus glacialoides in Adklie Land, Antarctica. The overall rate of weight loss during incubation fasts was 4.46% per day in both sexes but 3.91 % per day after the first 2 days of fast once the stomach contents had been digested. Antarctic Fulmars appear to have rates of body-weight loss significantly greater than the rates of weight loss observed in any other species of albatross or petrel. The validity of methods using the rate of weight loss to estimate incubation metabolic rates is discussed, and it appears that for birds incubating for periods of less than 2-3 days the digestion of the stomach contents can bias the estimation. The absence of change in weight at the beginning of shifts during incubation and brooding, together with the short time spent on foraging trips, suggests that food availability is high and/or food is readily accessible to Antarctic Fulmars. The availability of food allows the Antarctic Fulmar to compensate for the high energy cost of breeding that could be due to the cold Antarctic environment and perhaps also to a higher intrinsic metabolism.

All the fulmarine petrels but one, the Fulmar Fulmarus glacialis, breed in the southern hemisphere, concentrating along the Antarctic continent. They have a contracted breeding cycle, with short incubation and fledging periods, short incubation shifts, a high frequency of feeds to the chick and more rapid chick growth rates compared to other similar-sized petrels or albatrosses (Mougin 1975, Croxall 1984). This contracted cycle coincides with the rich availability of food during the short summer in the vicinity of the Antarctic continent. Croxall(l984) has pointed out that it is not clear what facets of their biology enable fulmarine petrels, rather than other groups, to take full advantage of the feeding opportunities over the summer period. In particular, the accelerated breeding cycle with the high frequency of foraging trips between the nest and feeding zones might be associated with an increased energy cost of breeding. While the number of studies on the cost of breeding has increased dramatically in recent years, concentrating noticeably on Procellariiformes (Croxalll982, Rickleffs 1983, Adams & Brown 1984, Grant 1984), no such study has yet been made on fulmarine petrels.

This paper examines the weight changes of Antarctic Fulmars Fulmarus glacialoides over the incubation and brooding shifts, in order to determine whether there are any differences in the rate of weight loss during incubation and brooding between a fulmarine petrel and the other Procellariiformes.

Methods The study was carried out on the Pointe Geologie Archipelago (66.40’S, 140.01’E), Adtlie Land (Antarctica) in January and February 1988. The single Antarctic Fulmar colony of Adelie Land varies

I990 BREEDING W E I G H T LOSS I N A N T A R C T I C FULMARS 69

between 30 and 50 pairs (Jouventin & Weimerskirch in press). Since 1963 all the birds present on the colony have been ringed and thereafter recaptured every year. The sex of birds was determined from the incubation routine and from a comparison of the culmen-length between the two birds of each pair: males have a significantly longer culmen than females, with less than a 5% overlap in the range of lengths (Mougin 1967). To limit stress, birds were not weighed in a sack but were gently removed from their egg or chick, and placed in a box on an electronic balance accurate to 0.1 5% of capacity. Incubating birds were weighed every 8-14 h. Brooding birds and their chick were weighed every 6-12 h. The weight of adult birds at the beginning of brooding shifts was calculated by adding to the weight of the adult the estimated weight gain of the chick. The total weight gain of the chick was estimated from the net weight gain (difference between the two weighings at 6-12 h apart) corrected for mean rates of defaecation and assimilation of feeds (calculated from data of Ricketts & Prince 1984). To determine the frequency of visits by adults to the nest after the chick had been left alone, the nest was inspected every 3.5 h between 07.00 and 21 .OO h local time. The presence or absence of adults was noted and the chick was weighed at each inspection. If no adult was present, an increase in the chick's weight was considered to indicate that a parent had visited during the interval.

Weight loss during fasts is best described as an exponential function of time (see Croxall 1982). The weight loss by adult fulmars during each shift was described by:

.

W,= Woe-''

where W , is the weight t days from the start of the shift, WO is the weight at the start of the shift and k is the proportion of the bird's weight lost each day. Only incubation shifts lasting more than 2 days were analysed to estimate the overall rate of weight loss. For each complete shift and for the two periods subsequently determined within a shift, the pooled rate ( k ) was estimated by regression of the logarithm of weight against time. Instantaneous rates of body weight loss were estimated from rates over periods of 8- 14 h, indexed to a time midway between the two weighings.

Results

Duration of incubation and brooding shifts in males and females Incubation shifts tended to decrease in length up to hatching in both sexes (Table 1). The female was generally present at hatching, as found by Mougin (1967) at the same colony in 1964. The length of shifts after hatching was fairly constant throughout the brooding period for males and females, and both sexes stayed with the chick for similar periods. When the chick was left alone on the nest, it was fed every 1-32 0.21 days (range 0.93-1.86 for 20 chicks weighed four times a day over 7 days).

Table 1. Duration (days) of stays on the nest by male and female Antavctic Fulrnars before, a t , and after hatching

Females Males Shift number Mean2s.d. n range Mean&s.d. n range

Before hatching ~ 3.7+ 1.0 7 2.5-5.5

- 2 3.3 f 1.1 7 2-5 3.4k0.6 8 2.5-4.5 -1 2.7k0.7 7 2-4 2.0T0.9 8 0-5-3

At hatching 0 1.5f1.0 8 0.5-2.5 ___

Brooding period 0.96 f 0.3 66 0.5-2 0.94 2 0.4 67 0.5-2

70 H . WEIMERSKIRCH IBIS 132

Table 2. Weight at beginning of incubation shifts and weight loss by incubating male and female Antarctic Fulmars (mean? s.d.; vange in parentheses)

Initial Daily Overall rate of No.of No. of weight weight loss weight loss" shifts weighings (9) (g/day) (%/day)

Females 21 126 931.7k53.3 44.4k7.8 4.46k0.14 (845-1020) (24.9-60) (3.14-7'07)

Males 24 147 1048.9 73.6 49.1 & 12.6 4.46 & 0.1 (935-1180) (33.3-85) (2.63-7.35)

*Calculated by regressing log. weight against time

Weight loss during incubation shifts Antarctic Fulmars lost 2.6-7.3y0 of their initial weight each day during an incubation fast. Males were significantly heavier than females at the beginning of incubation shifts and lost 5 g per day more than females (Table 2). The rate of weight loss was the same, 4.46% per day in both sexes (Table 2).

The instantaneous rate of daily weight loss was calculated for periods of 8-14 h to evaluate the length of the period of digestion of the stomach contents. It declined significantly during the first 2 days of the incubation fast, but showed no significant changes thereafter (Fig. 1). The pooled rate of daily weight loss was 5.47 k O.l8yO per day during the first 2 days of fast, significantly greater than that estimated after the first 2 days (3.91 + 0.15% per day).

Changes in weight during the incubation and brooding periods

Weights at the beginning of incubation shifts during the second half of the incubation period remained approximately constant in birds of both sexes (Fig. 2; y = 22.5 + 0*99x, P < 0.001 for the regression line of the weights at the beginning of two successive shifts for 11 females and 13 males). Over the same period the weight gain while at sea was 35.8 k 13.0 g per day (range 13.3-60.1 g) for 20 females and .

' 0 6.6 -1.42~

r = 0.45, P< 0.001

0

- . 0

0 . 0 0 0

0 0

.O . .

0 . . 0

0

1 I I I I I 2 3 4

Length of the fast (days)

Figure 1. Changes in daily rate of weight loss of Antarctic Fulmars during the incubation fasts: males (0); females (0).

I990 BREEDING W E I G H T LOSS I N A N T A R C T I C FULMARS 71

Figure 2. Changes in the weights of (a) male and (b) female Antarctic Fulmars at the beginning and end of incubation and brooding shifts during a SO-day period including hatching. The values are expressed as a percentage of the weight at the end of the shift including or following hatching.

H. W E I M E R S K I R C H I B I S 132 72 y = 22.5 + 0.99x, P < 0.001 for the regression line of the weights at the beginning of two successive shifts for 11 females and 13 males). Over the same period the weight gain while at sea was 35.8 rf: 13.0 g per day (range 13.3-60.1 g) for 20 females and 51.9 -t 18-2 g per day (range 26-2-90.0 g) for 19 males. The length of stay at sea was significantly correlated with the relative weight gain only for the shifts just before hatching ( r = 0.67, n = 18, P < 0.01) with eight females gaining 46.7 f 9-3 g per day and nine males gaining 54 26-1 g per day. Over the previous incubation shifts birds spent more time at sea than during the last shift before hatching (shift - 1 in Table l), but gained similar absolute weight. The weight at the end of incubation shifts increased progressively in males during the 20 days before hatching, but showed no significant changes among females (Fig. 2). After hatching, the weight at the end of shifts decreased significantly in both sexes, while the weight at the beginning of shifts did not change significantly in females and decreased only slightly in males (Fig. 2).

Discussion

The rate of weight loss by fasting birds during incubation has been recognized to be a reliable index of energy cost (Croxall 1982, Adams & Brown 1984) although it is highly dependent on the composition of the material lost (Groscolas 1988). However, this last restriction may be of less importance in comparative studies, especially those involving closely related species in which the fuel used is likely to be the same. Croxall (1982) has shown that even though there is extensive individual variation in the rate of proportional weight loss in most species, for petrels and albatrosses there is a significant relationship between the proportional weight loss and body weight. Despin & Mougin (1 988) used data from 20 species and came to the same conclusion. For 24 species of petrels and albatrosses (see Table 3 for references) the relationship between the logarithms of proportionate weight loss and body weight is described by the regression equation (a): y = - 0.767 - 0 . 3 2 4 ~ ( P < 0.01). The value obtained for male and female Antarctic Fulmars lies well outside the confidence limits (at P < 0-01) of the regression line described by this equation (Fig. 3) and is the only one to have a significantly large residual (at least twice that of any other species, P < 0.05). Antarctic Fulmars appear to have a much higher proportio- nate weight loss than any other procellariiform species.

Some caution should be exercised in interpreting this high rate of weight loss as a high incubation metabolic rate. Indeed, Groscolas (1 988), using Croxall’s (1982) data, argued that the incubation metabolic rates calculated from rates of body weight loss are overestimated in species fasting less than 7 days. He stated that in short-term fasting species, the fasting shift partly coincides with the digestion of the last meal and of the oil, derived from seafood (Warham 1977), that is stored in the stomach of most Procellariiformes. However, most of the species incubating for periods less than 7 days (see Table 3) are evenly distributed along the line, and within the confidence limits (at P < 0.05), of the regression equation (b) calculated for species incubating for periods of more than 7 days: y = - 0.882 - 0 . 3 ~ ( r = 0.89, P < 0.01). This result suggests that stomach contents are probably rapidly digested and do not greatly bias the values of the overall rate of weight loss used to estimate incubation metabolic rates. However, for species incubating for periods shorter than 3 days (see Table 3) the rates of weight loss lie outside the confidence limits of equation (b) ( P < 0-05) indicating that the digestion of stomach contents could bias the estimation of metabolic rates using measurements of body weight loss in these species. In the Antarctic Fulmar, the digestion of the stomach contents probably ends after 1.5-2 days of fast, when the daily weight loss becomes stable (Fig. 1). The pooled daily

Tab

le 3

. W

eigh

ts a

nd r

ates

of w

eigh

t lo

ss in

24

spec

ies of p

etre

ls a

nd a

lbat

ross

es

DE

X W

ande

ring

Alb

atro

ss D

iom

edea

exu

lans

D

ME

Bla

ck-b

row

ed A

lbat

ross

D. m

elan

ophr

is

DC

R G

rey-

head

ed A

lbat

ross

D. c

hrys

osto

ma

DIM

Lay

san

Alb

atro

ss D

. im

mut

abili

s F

UG

Ant

arct

ic F

ulm

ar F

ulm

arus

gla

cial

oide

s C

DB

Cor

y’s

She

arw

ater

Cal

onec

tris

dio

med

ea b

orea

lis

PM

A G

reat

-win

ged

Pet

rel

Pter

odro

ma

mac

ropt

era

CD

D C

ory’

s S

hear

wat

er C

alon

ectr

is d

iom

edea

dio

med

ea

PP

S D

ark-

rum

ped

Pet

rel P

tero

drom

a ph

aeop

ygia

sand

wic

hens

is

PU

P M

anx

She

arw

ater

Puf

inus

puf

inus

P

PP

Dar

k-ru

mpe

d P

etre

l Pte

rodr

oma

phae

opyg

ia p

haeo

pygi

a P

IN M

ottl

ed P

etre

l P. i

nexp

ecta

ta

PH

Y B

onin

’s P

etre

l P. h

ypol

euca

H

AC

Blu

e P

etre

l Hal

obae

na c

aeru

lea

PP

Y P

ycro

ft’s

Pet

rel P

tero

drom

a py

crof

ti

PU

L A

udub

on’s

She

arw

ater

Puf

inus

Ihe

rmin

ieri

P

AT

Fai

ry P

rion

Pac

hypt

ila

turt

ur

PE

G S

outh

Geo

rgia

n D

ivin

g P

etre

l Pe

leca

noid

es g

eorg

icus

B

UB

Bul

wer

’s P

etre

l Bul

wer

ia b

ulw

erii

FR

T B

lack

-bel

lied

Sto

rm P

etre

l F

rege

tta t

ropi

ca

OC

F F

ork-

tail

ed S

torm

Pet

rel

Oce

anod

rom

a fur

cata

O

CL

Lea

ch’s

Sto

rm P

etre

l 0

. leu

corh

oa

OC

C M

adei

ran

Sto

rm P

etre

l 0. c

astr

o 001 W

ilso

n’s

Sto

rm P

etre

l O

cean

ites

ocea

nicu

s

Wei

ght

(g)

Rat

e of

w

eigh

t los

s ( O

oida

y)

Len

gth

of

incu

bati

on

shif

ts (d

ays)

9800

38

10

3690

30

00

990

985

670

620

460

450

430

370

202

188

178

168

134

125

102

62

59

53

42

42

0.8

1.2

1.2

1.1

4.5

1.9

1.3

2.3

1.6

2.5

2.9

2.1

3.2

2.9

2.0

3.6

5.9

3.8

2.9

4.0

6.3

4.5

3.8

6.5

21.0

13

.0

12.0

15

.0

2.8

8.4

10.0

8.

5 12

.4

5.9

10.0

13

.0

8.0

6.0

6.0

2.5

1.3

3.0

2.3

3.1

6.0

2.5

Ref

eren

ce

Cro

xall

& R

icke

tts

1983

P

rinc

e, R

icke

tts

&T

ho

mas

1981

P

rinc

e, R

icke

tts

& T

hom

as 1

981

Fis

her

1967

T

his

stud

y D

espi

n &

Mou

gin

1988

Im

ber

1976

R

isto

w &

Win

k 19

80

Sim

ons

1985

H

arri

s 19

66

Har

ris

1970

W

arha

m, K

eele

y &

Wils

on 1

977

Gra

nt &

W’it

tow

198

3 M

orin

& G

rosc

olas

unp

ubl.

D

unne

t 19

85

Har

ris

1969

a H

arpe

r 19

76

Rob

y &

Ric

klef

s 19

83

Jouv

enti

n &

Mou

gin

1981

B

eck

& B

row

n 19

71

Sim

ons

1981

R

ickl

efs,

Rob

y &

Wil

liam

s H

arri

s 19

69b

Bec

k &

Bro

wn

1972

e z 3 z 4

P

3 z

74 H . W E I M E R S K I R C H IBIS 132

\ \ \ '.

I I I I I I I I I 1.5 2 2.5 3 3.5 4 4.5

Log body weight ( g )

Figure 3. Log-log relationship between rate of weight loss and body weight in 24 species of Procellariiformes (see Table 3). The dotted lines indicate the limits of the 99% confidence zone of the regression line.

weight loss after the first 2 days of fast (3.91% per day) remains well outside the confidence limit at P < 0.01 of equations (a) and (b). Consequently, it appears that Antarctic Fulmars have a particularly high rate of weight loss during incubation.

The higher rate of weight loss during incubation for Antarctic Fulmars compared to other Procellariiformes could result, at least in part, from the cold Antarctic climate. At the Ad6lie Land colony, Antarctic Fulmars incubated their eggs under mean ambient temperatures of - 0.7"C (range - 9.2-5.2"C). This temperature is lower than the temperatures encountered by all the 24 procellariiform species used in the calculation of the regression line. Among the 24 species, six incubate in a tropical climate, nine in a temperate climate and ten in a sub-Antarctic or sub-Arctic climate. Most of the species fit well within the confidence interval of the regression, indicating that despite the wide range of temperatures encountered, most have similar relative energy expenditure. Antarctic Fulmars could be incubating under temperatures below their lower critical temperature (the tempera- ture below which body temperature cannot be maintained without an increase in heat production; Kendeigh et al. 1977) and consequently could suffer an additional weight loss compared to other species of similar size but incubating under milder conditions. However, because of the extent of the shift in values obtained for Antarctic Fulmars compared to the other Procellariiformes, additional or alternative factors could have contributed to their high rate of weight loss. In particular, members of the fulmarine petrel group may have a higher intrinsic metabolism than the other Procellariiformes. A higher metabolism would explain the short incubation stints by fulmarine petrels compared to other similar-sized petrels (see Table 3 and Croxall 1984). Indeed, if fulmarine petrels had a relative incubation metabolism similar to those of other petrels, it would be difficult to understand why they do not incubate for longer periods in order to limit the number of foraging trips between the colony and the feeding zones. Increasing the length of incubation shifts would have the additional advantage of reducing the number of nest reliefs which are the major cause of nesting failure (unpubl. data).

I990 BREEDING W E I G H T LOSS I N ANTARCTIC FULMARS 75

Over the second part of incubation and during brooding the weights of male and female Antarctic Fulmars at the beginning of the shifts after foraging trips did not vary significantly (Fig. 2). This stability in weight during incubation, which has also been observed in other Procellariiformes (Fisher 1967, Brooke 1978, Prince et al. 1981, Croxall & Ricketts 1983), implies that the birds can regain the weight loss during the incubation fast while they are at sea. However, during the brooding period, their weights at the end of the brooding shifts decrease abruptly (Fig. 2), corresponding to the progressive increase in weight of the chick. The observation that during brooding the weight at the end of shifts can fall below 75% of the weight at the end of the shift overlapping hatching indicates that during the incubation period the birds could have spent longer shifts on the nest. Indeed, their weights at the end of incubation shifts rarely go below 80% of their reference weights. Thus, as suggested by Ricklefs (1983) for Procellariiformes, the period when adults brood small chicks appears to be the most demanding energetically because parents can feed only half of the time yet must support both themselves and the growing chick. During brooding the time spent at sea by males and females remains constant, indicating that birds are probably unable to complete the foraging trips in less than 1 day. However, the higher energetic cost of brooding seems to be easily absorbed by Antarctic Fulmars as their weights do not change extensively during this period. When the chick is large enough to be left alone on the nest each parent visits the chick at a minimum interval of 1.3 days when the food needs of the chick are the highest. Antarctic Fulmars feed mostly on krill Euphausia superba offshore from Adelie Land (Ridoux & Offredo 1989). The short interval between feeds to the chick and their high growth rate (Mougin 1975) compared to any other medium-sized or large procellariiform (Croxall & Prince 1980, Jouventin & Mougin 1981, Croxall 1984) indicate that krill must be abundant and/or easy to catch in the Adelie Land zone. With an estimated flying speed of 7 m/s (calculated from Jouventin & Mougin 1981 and Pennycuick 1982), and using our measurements on the length of stays at sea, birds of AdPlie Land have a maximum foraging radius during the incubation, brooding and fledging periods of 900-1 100 km, 290 km and 390 km, respectively. This range permits them to travel to oceanic waters, where krill occurs (Baker 1965), some 110 km from the colony (Arnaud 1974).

The nesting of the Antarctic Fulmars in the cold antarctic environment of AdPlie Land, which should dramatically increase the energetic cost of breeding, might thus largely be compensated for by the high availability of an abundant food resource nearby. Further comparative studies should indicate whether high rates of weight loss incurred during incubation by Antarctic Fulmars result only from cold temperatures or if fulmarine petrels also have a particularly high daily energy expenditure associated with their contracted breeding cycle.

This study was supported by the Administration of Terres Australes et Antarctiques FranGaises and RCP 764 and GDR 001 of CNRS, and logistic support was provided by the Expeditions Polaires FranGaises. It is part of the programme on the Ecology of Antarctic Birds and Mammals directed by Dr P. Jouventin. Many thanks are due to N. Sadoul for his help in Adelie Land, Dr C. P. Doncaster for improving the English and to Drs Y. Cherel, R. Groscolas, P. Jouventin, P. A. Prince and the editor Dr P. J. Jones for comments and suggestions on the manuscript.

References

ADAMS, K. J. & BROWN, C.R. 1984. Metabolic rates of subantarctic Procellariiformes: a comparative

ARNAUD, P.M. 1974. Contribution a la bionomie benthique des regions antarctiques et subantarctiques. study. Comp. Biochem. Physiol. 77A: 169-173.

Tethys 6: 465-656.

76 H . WEIMERSKIRCH IBIS 132

BAKER, A. DE C. 1965. The latitudinal distribution of Euphausia species in the surface waters of the Indian Ocean. Discovery Rep. 33: 309-334.

BECK, J.R. & BROWN, D.W. 1971. The breeding biology of the Black-bellied Storm Petrel Fregetta tropica. Ibis 113: 73-90.

BECK, J.R. & BROWN, D.W. 1972. The biology of Wilson’s storm petrel, Oceunites oceanicus (Kuhl) at Signy Island, South Orkney islands. Br. Antarct. Surv. Sci. Rep. 69: 1-54.

BROOKE, M. DE L. 1978. Some factors affecting the laying dates, incubation and breeding success of the Manx shearwater, Pufinus pufinus. J. Anim. Ecol. 47: 477-495.

CROXALL, J.P. 1982. Energy costs of incubation and moult in petrels and penguins. J. Anim. Ecol. 51: 177- 194.

CROXALL, J.P. 1984. Seabirds. In Laws, R.M. (ed.), Antarctic Ecology. Vol. 2: 533-618. London: Academic Press.

CROXALL, J.P. &PRINCE, P.A. 1980. Food, feeding ecology and ecological segregation of seabirds at South Georgia. Biol. J. Linn. SOC. 14: 103-131.

CROXALL, J.P. & RICKETTS, C. 1983. Energy costs of incubation in the Wandering Albatross Diomedea exulans. Ibis 125: 33-39.

DESPIN, B. & MOUGIN, J.L. 1988. Evaluation de ladtpense tnergttique et de la consommation alimentaire du Puffin Cendrt Calonectris diomedea borealis d’apris l’ttude de la dtcroissance ponderale au cours du jeQne. Oiseau 58: 28-43.

DUNNET, G.M. 1985. Pycroft’s petrel in the breeding season at Hen and Chickens islands. Notornis 32: 5- 21.

FISHER, H.I. 1967. Body weights in Laysan Albatrosses Diomedea immutabilis. Ibis 109: 373-382. GRANT, G.S. 1984. Energy cost of incubation to the parent seabird. I n Whittow, G.C. & Rahn, H. (eds),

Seabird Energetics: 59-71. New York: Plenum. GRANT, G.S. & WHITTOW, G.C. 1983. Metabolic cost of incubation in the Laysan Albatross and Bonin

Petrel. Comp. Biochem. Physiol. 74A: 77. GROSCOLAS, R. 1988. The use of body mass loss to estimate metabolic rate in fasting seabirds: a critical

examination based on emperor penguins (Aptenodytesforsterz~. Comp. Biochem. Physiol. 90A: 361- 366.

HARPER, P.C. 1976. Breeding biology of the fairy prion (Pachyptila turtur) at the Poor Knights islands, New Zealand. N. Z. J. 2001. 3: 351-371.

HARRIS, M.P. 1966. Breeding biology of the Manx Shearwater Pu&uspu&u. Ibis 108: 17-33. HARRIS, M.P. 1969a. Food as a factor controlling the beeding of Pufinus Iherrninieri. Ibis 111: 139-156. HARRIS, M.P. 1969b. The biology of storm petrels in the Galapagos Islands. Proc. Calif. Acad. Sci. 33:

HARRIS, M.P. 1970. The biology of an endangered species, the dark-rumped petrel (Pterodroma phaeopygia) in the Galapagos islands. Condor 72: 76-84.

IMBER, M.J. 1976. Breeding biology of the Grey-faced Petrel Pterodroma macropteragouldi. Ibis 1 1 3: 51- 64.

JOUVENTIN, P. & MOUGIN, J.L. 1981. Les strategies adaptatives des oiseaux de mer. Terre Vie 35: 217- 272.

JOUVENTIN, P. & WEIMERSKIRCH, H. 1990. Population monitoring of seabirds and seals in the French Austral territories, with an analysis of the decreasing population of Emperor Penguin in Adtlie Land. In Kerry, K. (ed.), Ecological Changes and the Conservation of Antarctic Ecosystems. Berlin: Springer-Verlag. In press.

KENDEIGH, S.C., DOL’NIK, V.R. & GAVRILOV, V.M. 1977. Avianenergetics. I n Pinowski, J. & Kendeigh, S.C. (eds), Granivorous Birds in Ecosystems: 127-204. Cambridge: Cambridge University Press.

MOUGIN, J.L. 1967. Etude tcologique de deux esptces de fulmars, le fulmar atlantique (Fulmarus glacialis) et le fulmar antarctique (Fulmarus glacialoides). Oiseau 37: 57-103.

MOUGIN, J.L. 1975. Ecologie comparee des Procellariidae antarctiques et subantarctiques. Com. Nat. Franq. Rech. Antarct. 36: 1-195.

PENNYCUICK, C. J. 1982. The flight of petrels and albatrosses (Procellariiformes) observed in South Georgia and its vicinity. Phil. Trans. R. SOC. Lond. B 300: 75-106.

PRINCE, P.A., RICKETTS, C. &THOMAS, G. 1981. Weight loss in incubatingalbatrosses and its implications for their energy and food requirements. Condor 83: 238-242.

RICKETTS, C. & PRINCE, P.A. 1984. Estimation by use of field weighings of metabolic rate and food conversion efficiency in albatross chicks. Auk 101: 790-795.

RICKLEFS, R.E. 1983. Some considerations on the reproductive energetics of pelagic seabirds. Stud. Avian Biol. 8: 84-94.

95-165.

I990 B R E E D I N G W E I G H T L O S S I N A N T A R C T I C F U L M A R S 77 RICKLEFS, R.E., ROBY, D.D. & WILLIAMS, J.B. 1986. Daily energy expenditure by adult Leach's Storm

Petrel during the nesting cycle. Physiol. Zool. 59: 649-660. RIDOWX, V. & OFFREDO, C. 1989. The diet of five summer breeding seabirds in AdClie Land, Antarctica.

Polar Biol. 9: 137-146. RISTOW, D. & WINK, M. 1980. Sexual dimorphism of Cory's Shearwater. 11-Merill 21: 9-12. ROBY, D.D. & RICKLEFS, R.E. 1983. Some aspects of the breeding biology of the diving petrels

Pelecanoi'des georgicus and P. urinatrix exsul at Bird island, South Georgia. Br. Antarct. Surv. Bull.

SIMONS, T.R. 1981. Behavior and attendance patterns of the fork-tailed storm petrel. Auk 98: 145-158. SIMONS, T.R. 1985. Biology and behavior of the endangered Hawalian dark-rumped petrel. Condor 87:

WARHAM, J. 1977. The incidence, functions and ecological significance of petrel stomach oils. Proc. N. Z.

WARHAM, J., KEELEY, B.R. & WILSON, G.J. 1977. Breeding of the mottled petrel. Auk 94: 1-17.

59: 29-34.

229-245.

Ecol. Soc. 24: 84-93.