JOURNAL
OF FIELD
ORNITHOLOGY
Published by
Association of Field Ornithologists
Vo•.. 69, No. 2
Spring 1998
P^GES149-352
J. Field Ornithol., 69(2):149-159
IMPLANTED
TRANSPONDERS
IN PENGUINS: IMPLANTATION,
RELIABILITY, AND LONG-TERM EFFECTS
JuI)¾C•m•
Australian
•r)
Kxo•q•Es I•P•¾
An taretic Division
Channel Highway
Kingston,Tasmania7050 Australia
Abstract.--Subcutaneously
implanted transpondershave proved to be a reliable meansof
identifying individual penguins.We found survivalof Ad61iePenguins(t:5,goscelis
adeliae)
carryingtransponders
overfiveseasons
to be equalto or betterthan that of birdswith flipper
bands,although not statisticallysignificanton an annual basis.There were, however,occasional problems associatedwith the use of implanted transponders.The transponderremoved from one bird had developed a slimy biofilm harboring potentially pathogenic organismsincorporated at the time of implantation. If such contaminationis common, it is
possiblethat the long-term survivalof groupsof birds carrying implanted transpondersmay
be lower than that of unmarked populations.Migration of transpondersawayfrom the injection site has alsobeen demonstratedand may compromisesurvivalin someindividuals.
These risk factorscould limit the use of implanted identificationdevicesin long-livedor
endangeredspecies.Introduction of bacteria can be minimized by careful injection techniquesand cleansingof instrumentsand skinwith iodineor alcohol.The choiceof a suitable
implantation site, such as midwaydown the back, from which transpondersmay migrate
withoutimpingingupon vital structuresis alsoimportant.%rebelievethat transponders,
when
usedwith care, providea usefulalternativeto flipper bandsin demographicstudiesof penguins.
TRANSPONDORESIMPLANTADOSEN PINGUINOS: IMPLANTACI(SN,CONFIABILIDAD, Y EFECTOS
A LARGO
PLAZO.
Sinopsis.--Transpondores
implantadossubcutaneamentehah demostradoset una forma
confiablepara identificarpinguinosindMduales.Hallamosla supervivencia
de Pygoscelis
adeliae conteniendotranspondoresmils de cinco temporadasserigualo roejot que la de aves
con bandasen las aletas,aunque no estadisticamente
significativoen t6rminosanuales.Sin
embargo, hallamos problemasocasionalesasociadosal uso de transpondoresimplantados.
E1 transpondorremovidode un ave habia desarrolladouna biopeliculababosaque tenia
organismospotencialmentepat6genosincorporadosal momentode la implantaci6n.De esta
contaminaci6nset cornfin, es posibleque la supervivenciaa largo plazo de gruposde aves
cargandotransportdotesimplantadospuede set inferior a la de poblacionessin marcar.La
migraci6nde transpondores
lejosdel lugar de inyecci6ntambi6n seha demostradoy puede
comprometerla supervivencia
de algunosindividuos.Estosfactoresde desgopodrianlimitar
el uso de arteiactosde identificaci6nimplantadosen especiesde larga vida o en peligro de
exfinci6n. La introducci6n de bacteriasse puede minimizar con tficnicasde inyecci6ncuidadosay de la limpiezade losinstrumentos
y de la piel conyodoo alcohol.La selecci6nde
149
150]
J. Clarhe
andK. Kerry
j. Field
Ornithol.
Spring 1998
un lugar apropiadopara la implantaci6n,tal como a mitad de la espalda,del cual los transpondores
puedenmigrarsin afectarestructuras
vitalesesigualde importante.Creemosque
los transpondores
sonuna alternativaconverientealas bandasde aletasen estudiosdemograffcosde pinguinossi son utilizadoscon el cuidadoapropiado.
Flipper bands have traditionallybeen used to mark penguinsfor longterm study,and manual observationmethodsemployedto detect returning individuals.The detrimental effectsof bands are well documented
and include mortality during molt (Ainley et al. 1983), physicalinjury
(Sallaberryand Valencia1985), and increasedenergyexpenditureduring
swimming (Culik and Wilson 1991). Band lossis also known to occur
(Ainley and DeMaster 1980), and returns of someproportion of individuals can be expectedto be missedannually (Ainley et al. 1983).
The disadvantages
of flipper-bandinghave led to the searchfor alternative marking techniquesfor penguinsand other seabirds.The use of
implanted identification transpondershas become popular in the livestockindustry and in zoosover recent years (Behlert and Willms 1992;
D. Spielman, pers. comm.). We have developedan automatedweighing
and identificationsystemfor usein our studyof Ad61iePenguins(Pygoscelis adeliae)that incorporatesimplanted transpondersand a data-logging
deviceto identify and record individualbirds as they travelin and out of
their colony (Kerry et al. 1993). Other researchershave also started to
use such identification systemsin colonies of Little (Eudyptula minor),
King (Aptenodytes
patagonicus)
, Royal (Eudyptes
schlegeli)
, Gentoo (P papua), and Ad61ie Penguins (A. Chiaradia, pers. comm.; Gendner et al.
1992; Hindell et al. 1996; McCormick et al. 1993; P. Wilson, pers. comm.).
We describein this paper an assessment
of the practicaluseof implanted transpondersin Ad6lie Penguinsand presentdata on mortality rates
and transponder durability on the basisof six years of records from a
combinedtaggingand banding study.We drawattention to the potentially
seriouseffectsof biofilmson implantsthat mayharbor and releasepathogenic bacteriainto the bloodstream.We also documentthe problemsof
transpondermigration and loss,and discusspossiblelong-term effectsof
implantson individualsand populations.We believe it is important that
thoseintending to use implanted transpondersas identificationmarkers
be alerted to potential problemswith these devices.
METHODS
Our studycommencedin the australsummerof 1991-1992 when 132
breeding adult Ad61iePenguinsat BfchervaiseIsland (67ø35S,62ø49E)
near Mawson base, eastern Antarctica were implanted with glass-encapsulatedelectronictranspondersproducedby TIRIS• (TexasInstruments,
USA). Sixty of these implanted birds were also fitted with stainlesssteel
flipper bands. A further 85 and 116 penguins received transpondersin
the 1992-1993 and 1993-1994 seasonsrespectively(Table 1); 40 and 49
of thesewere alsogivenbands.(Not includedin Table 1 are an additional
165 birdsthat receivedtranspondersin a seasondifferent to that in which
they were first flipper-banded.)
Vol.69,No.2
Transponders
in Penguins
[] 5]
TABLIt1. Multi-year return rates of Adelie Penguinscarrying flipper bands and implanted
transponderscomparedto thosecarrying transpondersonly.All birdswere firstbanded/
implanted as adults and are grouped accordingto the year in which the band and/or
transponderwere first applied. Annual differenceswere not statisticallysignificant.(X2
> 3.84, df = 1, P < 0.05).
Birds carrying bandsand transponders
N
Bandedand tagged91-92
Birds carrying transpondersonly
%
60•
N
Tagged91-92
%
Difference
72*
return 92-93
45
75%
return 92-93
56
78%
3%
return 93-94
36
80%
return 93-94
49
88%
8%
return 94-95
return 95-96
return 96-97
31
21
19
86%
68%
90%
return 94-95
return 95-96
return 96-97
41
32
29
84%
78%
91%
-2%
10%
0%
Tagged92-93
45
87%
90%
89%
87%
4%
8%
26%
5%
91%
80%
90%
7%
14%
1%
Banded and tagged92-93
return
return
return
return
93-94
94-95
95-96
96-97
Banded and tagged93-94
return 94-95
return 95-96
return 96-97
40
33
27
17
14
83%
82%
63%
82%
49
41
27
24
84%
66%
89%
return
return
return
return
93-94
94-95
95-96
96-97
39
35
31
27
Tagged93-94
67
return 94-95
return 95-96
return 96-97
61
49
44
• These numbersdo not include birds tagged in a seasondifferent to that in which they
were banded, and thusdiffer from thosepublishedin Clarke and Kerry (1994).
The transpondersimplanted in the first two seasonswere 30-mm long
and 3-mm diameter; after 1992-1993 a shorter version 23-mm long became available (Fig. la). Use of transponderswithout banding has continued from 1994-1995 onwardsat B6chervaiseIsland aswell asat nearby
Verner Island and at Edmonson Point (74ø20S,165ø09E) in the RossSea.
Chicks have received transponders(but not bands) in all seasonsand
locationsat a rate of up to 300/yr (except in 1994-1995 at Mawsonwhen
starvationcausedthe whole cohort for that year to die; Kerry et al. 1995).
Overall, 1142 adults and 1929 chicksreceived transpondersby the end of
the 1996-1997 breedingseason.
TIRIS• transponders
were chosenbecausethey can be detectedfrom
a distanceof 0.7 m comparedto those of other manufacturerswhich,
although smaller,have a shorter range. Unique identification numbers
are transmitted passivelyfrom each transponder in responseto electromagneticinterrogationfrom a reader.The TIRIS• systemenablesautomatic detectionof birds passingin and out of a fenced colonythrough a
1-m wide passageusing an antenna connectedto a dam-loggingsystem
(Kerry et al. 1993).
Returns of implanted birds were detected annually by our Automated
Penguin Monitoring System(APMS), which ran continuouslythroughout
eachbreedingseason.In addition,the colonywasmanuallycheckedusing
binoculars during the incubation period each seasonfor banded birds
152]
J. Clarke
and/C Kerry
J.Field
Ornithol.
Spring 1998
10%
,500
Vol.69,No.e
Transponders
in Penguins
[ 153
that had moved to sub-coloniesother than thoseservicedby the APMS.
The majority of birds carrying both transpondersand bandscame from
marked nests;thesewere checkedmanuallywith a hand-heldtransponder
detectorto pick up any bandedbirdswhosetranspondershad failed.
Transponderimplantation.--Transponderswere implanted subcutaneously at the back of the neck or between the shoulder blades. These
locationswere chosenfor easeof implantation asthey have the mostloose
skin. Transponderswere injected using an applicator provided by Texas
Instruments. This device takes cartridges of ten sterile transpondersindividuallysurroundedby iodine gel and allowsinjection via a large bore
needle without direct handling of the implant. The needle was cleaned
with iodine solution(Betadine•) betweeneach applicationto minimize
the opportunitiesfor infection.
Transponderswere injected with the needle pointing awayfrom the
head and the neck extended. Transponderswere massagedwell under
the skin to prevent them being lost from the injection wound before it
healed. After injection the head was allowed to move freely and the
wound rechecked to ensure that the transponder had remained properly
under the skin. A cyanoacrylateskin glue (Vetbond•) was used as a
wound sealanton approximately90 penguinsduring 1995 and 1996.
The site of transponderimplantationwas changedin 1996-1997 to
midway down the back subsequentto radiographic investigationof the
extent of transpondermigration in 20 individualsimplanted in previous
seasons.Dorso-ventraland lateral radiographsof the neck and thorax of
thesebirds were taken using a portable X-ray machine, and the distance
and direction of transpondermigration measured.
RESULTS
AND
DISCUSSION
Typesof transponders.--Arange cf identification transpondersis now
availablefrom different manufacturers.These vary in size,surfacecoating, and distance over which they may be read. Our experience as reported here is restrictedto the TIRISc• transpondersmanufacturedby
TexasInstruments,USA. These transpondersare larger than most,however their size enablesthem to be read over correspondinglygreater distances.Identical transpondersare being usedin studiesof Little Penguins
(the smallestof all penguinspecies)and canbe readfrom outsideshallow
FIGURE1. a) Electronicidentificationtranspondersproducedby TIRIS@ (TexasInstruments,USA). These transponders(23 mm long and 3 mm in diameter) were implanted
subcutaneously
at the back of the neck or betweenthe shoulderblades.Longer (30
mm) transponderswere usedduring the first twoyearsof our study.b) Injury due to a
flipper band openingand penetratingthe radio-carpaljoint of an adult Ad6lie penguin
(after 1 year of wear). The band was subsequentlyremoved and, although die joint
remained swollenfor severalmonths, the bird returned to the colony to breed in the
following two seasons.c) Electron micrographshowingbacteria in the slimy biofilm
coatingthe glasssurfaceof an encapsulatedtransponderremovedfrom a fledgling.
154]
J. Clarhe
andK. Kerry
J.Field
Ornithol.
Spring 1998
burrowsin sand and through the wood of artificial nest boxes (A. Chiaradia, pers. comm.).
We consideredthe possibilitythat the glassencapsulatingthe electronics of the transponder could implode under the pressureof diving and
harm the bird. Ad61iePenguinscan dive to 175 m (Whitehead1989) and
Emperor Penguins (A. forsteri) have been recorded to depths of 534 m
(Kooymanand Kooyman1995). A number of TIRIS• transponders
from
different batcheswere tested to the equivalent of 1000 m depth and
showedno damage.We considerthesetransponderssuitablefor all penguin species.Other brandsof transpondersshould be pressuretestedto
such depths before use on deep-divingspecies.
Transponderlossand long-termfunction.--Transponder lossin the first
3 wk after injection proved to be a major problem. During the first 2 yr
of our studywe lost up to 30% of all implants.At that stagewe were
injecting transponderswith the needle pointing towardsthe head as it
was easier to part the feathers in that direction. We discoveredthat transponderswere falling out, assistedby gravity,before the wound healed
properly. The problem was exacerbatedin thin dehydratedbirds during
the incubation period (when most of our taggingwascarried out). Regular nest checksenabledus to detecttransponderlossesand we re-tagged
all such individualsbefore the end of each breeding season.We found
that if a transponderwasretained for 3 wk after taggingthen in almost
all cases it continued
to function
into
the future.
We now inject transponderswith the needle pointing awayfrom the
head and take care to massagethe transponderwell under the skin.We
re-check
the hole
after
the bird
has retracted
its neck
to make
sure the
transponderis not being expelled. Use of 23-mm long transpondersinsteadof the 30 mm onesusedin the first twoyearshasalsoimprovedthe
retention rate. However, the receiving antenna on the APMS needed to
be more finely tuned to detect reliably these smallertransponders.We
now experiencea transponderlossrate of only 3-5% and are finding that
the use of skin glues lowersthis rate further. Only one transponderwas
lostfrom the 90 Adfilie Penguinson whichVetbond• skin glue wasused
during 1995 and 1996, and the sameglue hasbeen usedsuccessfully
as
a sealant for injection wounds in Little Penguins (A. Chiaradia, pers.
comm.).
Once a transponderhas been successfully
implanted and retained for
3 wk, it is generally highly reliable. However, we have found some evidence of long-term transponderfailure over the years that birds have
been carrying these implants to date. Three hundred and fourteen in-
dividualshave now been carrying bandsand transpondersfor 3-5 yr.
From these individuals,five transpondershave failed (after 1-4 winters)
and six bands have been lost (after 2-5 yr of wear). Two of the birds
whosetranspondershad failed were radiographedand in neither wasthe
implant visible in the neck or thoracic region. We assumethat in these
casesthe transpondershave migrated right out of the bird, especially
Vol.
69,No.S
Transponders
in Pen•ffuins
[ 155
since our radiographic survey showed that tag migration over distances
up to 5 cm was a common occurrence.
Numerous flipper bands have been observed to be partially open after
one or more seasonsof wear and we believe it likely that the penguins
are opening bands with their bills, thus contributing to band loss and
increasingthe danger of injuries suchas that depicted in Figure lb where
the end of the band had penetrated 5 mm into the tissue.We have removed severalbands becausethey had opened to such a degree that
injury was present or imminent. Four of the six missingbands were lost
during the fifth year of wear, suggestingthat the rate of band losswill
increaseover time as partially opened bandsfall off.
Return ratesof taggedversusbandedadults.-•The return rates of tagged
and banded adults over the first two seasonsof our studywere reported
in Clarke and Kerry (1994). At that stagewe could not demonstrateany
differences in survival between the two groups over the short period.
Hindell
et al. (1996) also found no difference in the survival rates of
banded and taggedRoyalPenguinsover a singlewinter.We now havesix
seasonsof data and have analysed the annual survival rates over each
winter of birds banded and taggedin the first three seasonsof our study.
The return ratesof birds giventranspondersalonewere greater (7891%) than thoseof birds given bandsin addition to transponders(6390%) for almostall seasons
followingthe marking (Table 1). The differencesare, however,statisticallyinsignificantdue to small samplesizes.We
assumethat all birds that fail to return are dead becausebreeders rarely
emigrate to other coloniesor missa season,our detection rate of tagged
birds using the APMS is extremely high, we manually check all unfenced
sub-colonies
rigorously,and our tag lossrate is lessthan 3%.
The apparent mortality rate of birds with bandswasparticularlyhigh
over the 1995 winter for all three gr>ups (Table 1). This winter followed
a summer of severefood shortage (Kerry et al. 1995), and our results
suggestthat, when prey is scarce,the extra energy required to swimwith
a flipper band (quantified by Culik and Wilson, 1991) may actuallycompromise the survival of such birds. There is so far no evidence that implanted transpondersincrease mortality compared to bands; in fact the
long-term survivalof tagged birds may overall be better.
Chick survival and return rates.--Five hundred and thirty-nine chicks
were implanted with transpondersbetween 1992 and 1994. Forty-twoper-
cent of thesehavesubsequentlyreturned to the colony,the majorityfirst
appearing as 3-yr-olds.The percentagesof each cohort known to have
survivedtheir firsttwowintersrangedfrom 38-45%, and 32-40% of each
cohort are known to have reached at least 3 yr of age (Table 2a). One
hundredand seventy-two
3-5-yr-oldsreturnedto the colonyin 1996-1997,
and APMS recordsindicated that some of thesewere attempting to breed.
The survivorshipof the three cohorts of chicksto 2, 3 and 4 yr of age is
greater than that determined by Ainley et al. (1983) using banded populationsof fledglings.In fact it is alsogreater than the rate theseauthors
predicted for unbanded fledglings.
156]
J. ClarkeandK. Kerry
j. Field Ornitho1.
Spring 1998
TABLE2. Survival and age-specificreturn rates of chickscarrying implanted transponders.
a) Percentagesof each cohort known to surviveto minimum agesof 2-5 yrs. b) Return
rates of pooled cohorts of 2-, 3- and 4-yr-olds.
a) Chick survival
1992
Number of chickstagged:
34
% chickssurvivingto age:
2 yr
3 yr
4 yr
5 yr
44%
32%
21%
12%
1993
244
38%
37%
25%
1994
261
45%
40%
b) Chick return rates
N
%
2-yr-olds
Return as 3-yr-olds
3-yr-olds
Return as4-yr-olds
4-yr-olds
Return as 5-yr-olds
48
30
76
63%
(3 cohortspooled)
44
58%
(2 cohortspooled)
5
2
40%
(1 cohort)
The age-specificannual return rates of 2-4-yr-oldshave ranged from
40-63% (Table 2b), averaging59%. These figuresare lower than those
of banded and/or tagged adults (Table 1), and it will be of interest to
see if return rates improve once the birds are fully mature. The rate of
transponder loss in chicks is unknown; however, there is no reason to
expect it to be different to that in adults.
Overall, our resultsindicate that the use of transpondersis a good
alternativeto bandingfor shortto medium term demographicstudiesat
least,providingthat visualobservationis not necessary
for the detection
of tagged individuals.
Transponder
migration.-•A major problem with implanted transponders
is the potential for migration awayfrom the original injection site. Implant migration mayor may not causeproblemsdependingon where the
deviceendsup. Larger transpondersand thosecoatedwith syntheticplastics have been shownto be more likely to migrate than small glassimplants (Behlert and Willms 1992).
The transpondersthat we use are fairly large (in order to be detected
from a distance)but are encapsulatedin glass.In contrastto Behlert and
Willms (1992) suggestion
that glassimplantsare unlikelyto move,we have
noticedthat someproportion of transpondersretrievedfrom chickskilled
by skuashave migrated awayfrom the original site of injection. In most
casesthe transpondershave only migratedslightly,usuallyto one sideor
the other of the neck. However,in two chickstransponderswere found
alongsidethe tracheaand esophagus.
Transponder migration over distancesof 1-5 cm wasdemonstratedin
13 of 20 previouslyimplantedadultsexaminedradiographicallyin 19961997. Fivebirdsshowedno transpondermovement,and in twoindividuals
the transponderno longer existed.Most transpondershad movedeither
laterallyor caudoventrally,
severalinto the thoraciccavity.While it is pos-
Vol.
69,No.9
Transponders
in Penguins
[ 157
sible that such transpondersmay never cause damage to surrounding
organs,there is somedangerthat they mayobstructblood flow or damage
nerves in the region.
Transpondermigration is a causefor concern, despitethe fact that the
return rates of implanted birds compare favorablyto those of banded
individuals,and that the birds in which migration has occurred appear
healthy.Implantation of transpondersfarther down the back may reduce
the chancesof migration into the throat and thoracic cavity.The transponderswe implanted into adults and chicks during 1996-1997 were
injected midwaydown the back. Subsequentradiographyof four individuals showedonly slight lateral migration after two months,and no tendency of transpondersto move into body cavities.
Woundinfectionand long-termeffectsof implants.--Localizedwound infectionwasnot a problemin the daysfollowingtransponderimplantation
in either adults or chicks. However, it is possible that minor discharge
contributed to the early lossof transpondersin some birds during the
first fortnight post-implantation.
The use of Betadine• to clean the injector needle and the presenceof iodine gel surroundingeach transponder help minimize bacterial contamination.
Significantbacterial growth can occur,however,within the biofilmssurrounding implantsin humansand animalsover the long term (Costerton
and Lappin-Scott1989, Deighton and Balkau 1990, Vaudauxet al. 1994).
Most implant-relatedinfectionsdescribedin the medical field are associated with devicesmade from metal and plastic, the surfacesof which
tend to erode and facilitatebacterialgrowth (Deighton and Balkau 1990).
Glassis immunologicallyneutral and lesssubjectto pitting, but maychemicallyinteract with surroundingtissuesto form a stabilisingcapsule(Behlert and Willms 1992).
Bacteria associatedwith implants tend to remain localized, either attached to the implant or encapsulatedwith it. Occasionally,however,they
become detached and disseminatevia the bloodstream. The speciesof
bacteriafound in biofilmstend to be thosenormally presenton the skin
(commonly Staphylococcus
epidermidisand less frequently S. aureus)
which, although non-pathogenicin their normal environments,can become virulent when attached to surfacesof implanted materials (Deighton and Borland 1993, Vaudaux et al. 1994).
We have not seen any evidenceof chronic localized or systemicinfection attributabledirectlyto transponderimplants.However,suchaffected
birds may be more likely to die at sea than in the colony.Our annual
survivorshiprates also give us no cause for concern. Although we have
not killed any adult birds to check the statusof their implants,we have
dissecteda number of taggedfledglingskilled by skuasto investigatethe
tissuereactionaround their transponders.In mostcaseswe haveobserved
very little tissuereaction ranging from no macroscopicchangeto encapsulationin thin fibrous collagenlayers.These changescorrelatewell with
thosereported by, for example, Behlert and Willms (1992) for zoo animals carrying implanted glasstransponders.
]58]
J. ClarkeandK. Kerry
J.Field
Ornithol.
Spring 1998
A notable exceptionwasone implanted transponderremovedfrom a
nearly fledged chickwhich died of causesunrelated to implantation.The
transponderwas found to be encapsulatedand surrounded by a thick
layer of purulent material. A heavybiofilm had developedon the glass
surfaceof the transponder (Fig. lc). This material contained both cocci
and rods, the latter of which appeared to be activelydividing. Culture of
the purulent contents of the lesion under both aerobic and anaerobic
conditionsshowedthe presenceof Clostridium
perfringens,C. sordelliiand
some unidentified Bacillusspecies.No coccoidorganismscould be cultured; presumablythe long frozen-storagetime betweendeath of the bird
and culture of the lesion had allowed only spore-forming speciesto survive.
We do not knowyet what long-termeffectsmay resultfrom transponder
implantation. Our population of Addle Penguinscould normally be expected to live up to 20 yr (Ainley et al. 1983). If biofilm formation and
persistentbacterial infectionsare common then we may find that we are
actuallyshorteningthe life expectancyof our birds to somedegree.The
level of suchan effect may be difficult to determine, especiallyaswe have
only banded birds availablefor comparisonand these are also likely to
be adverselyaffectedby their identificationmarkers.
Further researchis required to quantify and reduce the incidenceand
effectsof both biofilm developmentand transpondermigration,especially as such risk factors may limit the use of implanted transpondersin
endangered or long-lived species.Although transponder migration may
be difficult to prevent, introduction of bacteria into the implantation site
can be minimised.Cleansingof the skin at the injectionsite with 70%
alcohol will help prevent incorporation of micro-organismsinto the capsule around the implant, as will disinfectionof the implantation device
between individuals. In the case of the fledgling described earlier, the
transponder had been unintentionally implanted within muscle tissue,a
factor which may have exacerbatedthe tissuereaction. Glass-coatedimplants are likely to be safer than those with plastic surfaces,and care
should be taker to ensure that each transponder is positioned subcutaneouslyin a location from which it is unlikely to migrate towardsvital
organs.
ACKNOWLEDGMENTS
We thank Margaret Deighton for adviceon biofilms,SusanRenaldsonand SilviaKirov for
undertaking the microbiologicalcultures,Geraldine Nash and Anna McEldowneyfor producingthe electronmicrographs,
Andre Chiaradiafor sharinghis experiencein the useof
transpondersin Little Penguinsand Heather Gardner,Ruth Lawless,Ruth TrEmont,Enrica
Franchi, Simonetta Corsolini, Stefan Eberhard and Sarah Brown for assistancein tagging
and carryingout autopsieson Addle adultsand chicks.We are alsogratetiffto A. Chiaradia
for his helpful commentson the draft manuscript.
LITERATURE
CITED
AINLEY,D. G., •ND D. R DEMASTER.
1980. Survivaland mortality in a population of Ad•lie
penguins.Ecology61:522-530.
Vol.69,No.')
Transponders
in Penguins
[ 159
--,
R. E. LERESCHE,
ANDW.J.L. SLADEN.1983. Breeding biology of the Ad•lie penguin.
Universityof California Press,Berkeley,California, 240 pp.
BEHLERT,O., ANDN. WILLMS.1992. Gewebsreaktionenauf implantierte Transponder eines
elektronischenMarkierungssystems.
Kleintierpraxis37:51-54.
CLARKE,
J. R., ANDK. R. KERRY.1994. The effectsof monitoring procedureson Ad•51iepenguin colonies. CCAMLR ScienceJournal 1:155-164.
COSTERTON,
J. w., AND H. M. LAPPIN-SCOTT.
1989. Behavior of bacteria in biofilms. ASM
News 55:650-654.
CULIK,B. M., AND R. P. WILSON.1991. Swimmingenergeticsand performance of instrumented Ad•lie penguins,Pygoscelis
adeliae.J. exp. Biol. 158:355-368.
DEIGHTON,M., AND R. BORLAND.1993. Regulation of slime production in Staphylococcus
epidermidis
by iron limitation. Infection and Immunity 61:4473-4479.
--,
ANDB. BALKAU.1990. Adherence measuredby microtitre assayas a virulence marker
for Staphylococcus
epidermidis
infections.J. Clin. Microbiol. 28:2442-2447.
GENDNER,J.p., J. GILLES,E. CHALLET,C. VERDON,C. PLUMERE,g. RABOUD,Y. HANDRICH,
ANDY. LEMAYO.1992. Automatic weighing and identification of breeding king penguins.
Pp. 29-30, in I. G. Priede and S. M. Swift, eds. Wildlife telemetry: remote monitoring
and tracking of animals. Ellis Horwood Ltd., England.
HINDELL,M. A., M. LEE,ANDC. L. HULL. 1996. The effectsof flipper bandson adult survival
rate and reproductionin the RoyalPenguin,Eudyptes
schlegeli.
Ibis 138:557-560.
KERRY,
K., J. CLARKE,
ANDG. ELSE.1993. The use of an automatedweighingand recording
systemfor the study of the biology of Ad6lie penguins,Pygoscelis
adeliae.Proc. NIPR
Symp.Polar Biol. 6:62-75.
, --,
H. GARDNER,R. MURPHY,F. HUME, AND P. HODUM. 1995. Ad61ie penguin
chick deathsinvestigated.ANARE News 75:18-20.
KOOYMAN,
G. L., ANDT. G. KOOYMAN.
1995. Diving behaviourof Emperor Penguinsnurturing
chicks at Coulman Island, Antarctica. Condor 97:536-549.
MCCORMICK,D. F., S. D. EMSLIE,N.J. KARNOVSKY,
L. W. KEITH, S. G. TRIVELPIECE,
AND W.
Z. TRI;qrLPIECE.
1993. Continuing studiesof the demographyand foraging behaviourof
the Pygoscelis
penguins.AntarcticJ. Review 1993:151-152.
SALLA•EP,•Y,
M., ANDJ. VALENCIA.1985. Wounds due to flipper bands on penguins.J. Field
Ornith.
56:275-277.
VAUDAUX,P. E., D. P. LEW,ANDF. A. WALDVOGEL.
1994. Host factorspredisposingto and
influencingtherapyof foreign body infections.Pp. 1-7, in A. L. Bisnoand F. A. Waldvogel, eds. Infections associatedwith indwelling medical devices,2nd ed. American Society for MicrobiologyWashington,D.C.
WHITEHEAD,M.D. 1989. Maximum diving depth of the Ad•lie penguin, Pygoscelis
adeliae,
during the chick-rearingperiod in Prydz Bay,Antarctica. Polar Biol. 9:329-332.
Received I Aug. 1996; accepted 11 Dec. 1996.