Be
ae

I N Te

le Fan a Fu |

N N
BE OBPRETEEBLLEN
a I u Re L

et Fe tn ln
Re En N
Nee,

re ne
een a ne ae ee ee
ee rrgieng
N N ee ee

a FE

Veen N Mete

Denen
ee N het m Tu

men

BER SRH IE PESENSER TEE

DENT

Se VENEDIG
rn ne De FT ir re ID
LM SROE EEE 25 “

DIDaNEn
er ZEN

TEN en te at
ne Dre nee! na te
an ee a .

I

ee

wi in
wur nen

Re

en ge
NEN
De Ne ee

PEReRT:

Bao DLR ES We TE STR EZ
nd

AT Ne

MÄR RENT NGREL SEEN
u TEN Me

Kr Der 57 ce
Br Te m an Er Nie

en
RT Er
a

mern dene”

De
ne ae
Wi et A

ee 222
Rat en

:
ARTEN Pi
Be en

de ß N
Men y sera PERLE En
wer 5

w
EIN

it
Runde NEULIENE
BUS EEE ENEOLI EEE DEE

Dee m
F De ERNEOTITNT
BAR ie u Bene TE
RETTEN $

FIRST Bew en 215 >
I STR Te 7 ES ae wu ep
ee Te T ET IEN NVRENN e N

m AND en EREN ED
ea EN RT ng FETT REN
RE ATI RAR UNBTSER IN ES EEE

TUT DE on
a WEN TEEN DES

PREEOTEREESTEEE NV EFT EEE ENDBIEE

IM So £ . ERLANGEN TEN EHEN

Er ER > SAH ATETEA MAN

Pe 7 22 \ “ eTEERT PERL ZE DEReS Eur Du

Ss Dessen ERDE NN Ä
EN

Aa
PEROEUR T
eternaaten sen BUEIRIELTEN EZ
N an M.

re
een

un NN
ir

AN

NDR 5 4
f TEN

Mn EN en,

TUN

5
Pe We

et ee
ELZWIRSe SERZEZ

Eu N

re a AUREN IE 7 7 2.5720

Zu
Br Se Ne En

ENT EN

A A ET
nn N

TE u te

2”

ER SER PLN
%

ve 15
ToRERUH
Stat

BER SE UELI
ERREES SET DE KO NEE
BP BRTENR TORE REES EN ESS
BETEN ORT
nen S NIS SE Zr
EROBERTE RE EEE RIRLILENE

NER NE!
ws

t
Se

PAPLLET ZN

vr ee
BEWERTE EP

ern et eN
TEN PU WERTNETE TA S
MRRBRNES IE DER TEL LT Te
ENDE IEUNIE LG N N

A TET
ON ET NETT DEN
ee N

Bar r IE
KRceTr:

ELSPETITLN
ee
EIRTIOHEFLTUTIEL DNA
AT baten

rn REN AM
EEE A ER a ea
et
that

BER SP 7705
ARSTER ENT -
N

DET EUD ET ETN EUIDE

NA = — = er = Ne u =

‚SaIUVugıT_LIBRARI ES SMITHSONIAN _ INSTITUTION NOILNLILSNI_NVINOSHLINS S3 |

J = Lu) \ 2 u = =

= 2, 4#,D# 2 Rn = =

E = <Y A = < = <

| : 28, 5 : = }

E SZ = 5 -

f = = i zZ ur ee

„INSTITUTION NOLLNLILSNI _NVINOSHLINS _S31UVU 8 I] _LIBRARI Es_ SMITHSONIANTINS

S ze 9 = 5;

E S N E > = 2

= = N\' E > > 1>

” > = N N, > — re 3 / m
LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31

ae = U: zu. <

1 z per > = = 3

= oO I 5 z Fi 5 =

2 E: 2 8 ‘2 2

J = = = e = 2,4

: z ea = 3

ii _NVINOSHLINS, S31UVYAIT_LIBRARIES SMITHSONIAN INS

: MM) = [MM TITG z # -

x = = = pe =

{ < =. = 4 = er

x 3 5 S B K

; [60) 5 & = mr =

> =, 2 2 zZ u‘ =

s314Vyg11 el BRARIES ZOMITHSONIDN LINSTELIE ON ENDELOEES NVINOSHLINS S3

; ei = = I: > 5

- 2 5. = 5 > =

> %

: Pe] = Z ee) = = r

m 2 m 2 m ®

z 7, = 77) = 177) =

NOILNLILSNI S31IUV4UHIT_LIBRARIES _SMITHSONIAN INS

2 z = Are 2

= E = =. 35 =

° Do: 7 ; 2 0RS 5 2

S E 3 EN 2 =

= > = > = NO =

) 4 (92) r (92) ge 77

LIBRARIES SMITHSONIAN INSTITUTION NOILNLI

=

SN

LIBRARIES, SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31

j: Mi SHLINS
z “ .5 wu 3 N wu
- = n. = x = 2 ©
[ © <<; © < = <
- = & S a2 = a
° 2 = = ° 2 la
J zZ ar $ z — zZ x
„INSTITUTION _ NOILNLILSNI_NVINOSHLIWS _S313VY811_ LIBRARIES SMITHSONIAN_ INS
- ö EEE. ö g ö ei u
J — X N _ & — Z &
| E s N = = E @2m2
B ZN E > u GP: >
= NN Fa 2) — PF KG»

- _ N N = =. \ _
1 2 m a I 2 m BL m
) z (47) EN z 7) zZ [77]

12]

z

=

LE

nm

oO

=

>

2

IE IE IIWI IT EG
“

G
NVINOSHLIWNS

NVINOSHLINS

SMITHSONIAN

IE
N
B\

INSTITUTION NOILNLILSNI _ NVWINOSHLINS

7a No,

S314UVYg17 LIBRARIES SMITHSONIAN INS?

>
WI Zs

&

iS
3.
r
GL;

Str
> uns

5 %
2

9

A ARE

TARIES SMITHSONIAN
/
TARIES SMITHSONIAN

NLILSNI

NLILSNI
ARIES

IN
wsTtITyn

%
N
,

(€)
Wwasn\S
(HSON
5
ä

(®)
IN DC.T
NLILSNI

IN.DC

OO 7GEDR z ce) OO WEI ER NN on eo) Me u,
2 GI 5 a 2 Tu NN 2 WU E °
> £ H ON.oc.> > ; 0, Y/ ı= =
> = re > = n > WTS> > =
zZ (42) 4 (77) * > = =

|] _LIBRARIES SMITHSONIAN INSTITUTION NOILNMNLILSNI NVINOSHLIWS S3IUVygIT

ee. 7) == (49) = [dp] =
_ m) = u zZ SÜ\ m =
z = “2 = NE z
| — <<“ je. X a II < A
< = < cc = RN e =
e) 21.04 ° = ö I e
Pr an = u > Ne zZ
N _ NOILNLILSNI _NVINOSHLINS_ S31IUVY911 LIBRARIES SMITHSONIAN INSTITUTION N
= ul: E A a: E
_ N = o = ee) _
= 2 NN = & = =
| = > NN = > > > E
= >) EIERN — = — Pe) de
> 2 NN 0b = m, 2 =
z o Ih = 7, zZ 7) zZ
1_LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS SIINYVYHIT L
zu; u zu 7) z u z
< = < s MT; < = AR: =
&ö en S 2 fr 5 ns DA ©
" g z WG 3 .DE 2
E = E RE 2,4 E
= IN > = > = > =
(77) = = 72) z 177) 2 7)
N NOILNLILSNI _ NVINOSHLINS LIBRARIES SMITHSONIAN INSTITUTION N
en [42] ul 7) ul 7 Ww
X — & = & er 04
| <z = » X = L- @ = &£ |
= Bun “ = 2; ö =
Bar = _ 2 ur — |
1 LIBRARIES SMITHSONIAN INSTITUTION NOILNALILSNI NVINOSHLINS S3luv4gI? L
= ö = 5 = 5 2
(es) _ “&0 Fu [so] — &
Ey = Ey = o = E
Ps] m? 2 = >) rn >)
= 72) un Fo 7) ja:
o = on zn o 2 oo
N NOILNLILSNI S31UWYAII_LIBRARIES .SMITHSONIAN INSTITUTION N
o > m — I u > a Pr
= z = = N = EZ: =
I r- - RIES | & _
EL =) 2; 2 IIND 5 2‘ 5
2° re 2 = re = >
zZ aa =
> 7 > 5 A is
7 SMITHSONIAN _ INSTITUTION NOILNLILSNI NVINOSHLINS Sa1yVyaıı L
a any NER N a
= <W/ ; am < = NN < pen
= eG 5 e£ 5 e =
oO ken G fe} == UTE> ö = 7 oO
z — == 4 2 SE : =
N _ NOLLNLILSNI _NVINOSHLINS „3 Iavyg 1% LIBRARIES SMITHSONIAN_ INSTITUTION = N
= e Ä oO | = = HG = 5 =)
5 2. = = re TG = >
= = > G. =)
= = = 52 re. Aal: 2 Er
= m \ o = n°® m n
= (77) Ye = 7) z 7) 2
I_LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NYVINOSHLINS. SI3IUVYAIT_L
= R 2} un zZ ® [42] zZ 142) ; 2 |
E =: 2. = Een S S
a 7, 7) SL T,) 7) [07] [2]
Pa je) E N ae FF; Mn (@) I
= | = En 2, = = =s
= Ben > = \ >" = > >
02] “ 2 (77) van z 7) = on |
I „NOILNLLLSNI_NVINOSHLINS S3 IUVYUEAIT_ LIBRARIES SMITHSONIAN INSTITUTION u
= 2 on , = — |
5 N a | = 7 ED 5 (ET a = DER 2 U = |
FINE NE IE N ZEN

Er:

ZEITSCHRIFT FÜR
SAÄUGETIERKUNDE

INTERNATIONAL JOURNAL
OF MAMMALIAN BIOLOGY

Organ der Deutschen Gesellschaft für Säugetierkunde
Volume 59, 1994 ISSN 0044-3468

Herausgeber / Editors

P. J.H. van Bree, Amsterdam — W. Fiedler, Wien — H. Frick, München - G. B. Hartl,
Wien -W. Herre, Kiel — R. Hutterer, Bonn — H.-G. Klös, Berlin — H.-J. Kuhn, Göttingen —
E. Kulzer, Tübingen -— W. Maier, Tübingen — J. Niethammer, Bonn - O. Anne E. Rasa,
Bonn — H. Reichstein, Kiel -— M. Röhrs, Hannover — H. Schliemann, Hamburg -
D. Starck, Frankfurt a.M. — E. Thenius, Wien — P. Vogel, Lausanne — H. Winking,
Lübeck

Schriftleitung/Editorial Office

D. Kruska, Kiel — P. Langer, Gießen

Mit 140 Abbildungen

Verlag Paul Parey Hamburg und Berlin V

Wissenschaftliche Originalarbeiten

AGUILERA, M.; CorTI, M.: Craniometric differentiation and chromosomal speciation of the
genus Proechimys (Rodentia: Echimyidae). — Kraniometrische Differenzierung und chro-
mosomale Artbildung in der Gattung Proechimys (Rodentia: Echimyidae) ..........

ARLETTAZ, R.; BECK, A.; GÜTTINGER, R.; LUTZ, MIRIAM; RuUEDI, M.; ZınGs, P.: Oü se situe la
limite nord de r&partition geographique de Myotis blythii (Chiroptera: Vespertilionidae) en
Europe centrale? - Where ıs the northern border of the distribution range of Myotis blythü

BorGH1, C. E.; GIANNONI, STELLA M.; MARTINEZ-RıcA, J. P.: Habitat segregation of three
sympatric fossorial rodents in the Spanish Pyrenees. — Habitatwahl bei drei Arten von
Kurzohrmausenknldenspanıschemibyreniene re ee Rn # Eee.

Brooks, ]J. E.; AHMAD, E.; Hussarn, I.: Reproductive biology and population structure of
Rattus rattus in Rawalpindi, Pakistan. - Reproduktionsbiologie und Populationsstruktur
VOonrRattusrattus myRawalpındı Pakıstane en er

BRUNET-LECOMTE, P.; VOLOBOUEV, V.: Comparative morphometry and cytogenetics of Micro-
tus (Terricola) multiplex (Arvicolidae, Rodentia) of the western French Alps. - Morphome-
trie und Cytogenetik von Microtus (Terricola) multiplex (Arvicolidae, Rodentia) der west-
lichenktranzosischentAlpeniimMereleichnn 22

CAnova, L.; MAISTRELLO, LARA; EMILIANI, D.: Comparative ecology of the Wood mouse
Apodemus sylvaticus in two differing habitats. — Vergleichende ökologische Untersuchungen
an Waldmäusen, Apodemus sylvaticus, in zwei unterschiedlichen Lebensräumen .......

CHRISTIAN, S. F.: Dispersal and other inter-group movements in badgers, Meles meles. —
Ausbreitung und andere Ortsbewegungen zwischen Gruppen von Dachsen, Meles meles.. . .

DAnNELID, E.: Chromosome polymorphism in Sorex alpinus (Mammalıa, Soricidae) in the
western Alps and the Swiss Jura. - Chromosomenpolymorphismus von Sorex alpinus
(Mammalıa, Soricidae) in den Westalpen und im Schweizer Jura ........ 2.222220.

FEER, F.: Observations ethologiques sur Bubalus (Anoa) quarlesi Ouwens, 1910 (Ruminantıa,
Bovidae) en captivite. - Fthological observations on Bubalus (Anoa) quarlesi Ouwens, 1910
(Ruminantia, Bovidae) in captivity. — Ethologische Beobachtungen an Bubalus (Anoa)
Guarlesı Ouwwens, 1910 (RumınantıayBoyidae) m Zoo

FıLıppuccı, MARIA G.; BURDA, H.; NEvo, E.; Kocka, J.: Allozyme divergence and systematics
of Common mole-rats (Cryptomys, Bathyergidae, Rodentia) from Zambia. — Allozymati-
sche Divergenz und Systematik der Graumulle (Cryptomys, Bathyergidae, Rodentia) aus
Sambia, =... Dame en an ee ee

GALLARDO, M. H.; KÖHLER, NELiDAa: Demographic changes and genetic losses in populations
of a subterranean rodent (Ctenomys maulınus brunneus) affected by a natural catastrophe. —
Demographische Veränderungen und Verluste von genetischer Variabilität bei Populationen
des grabenden Nagers (Ctenomys maulinus brunneus) als Folge einer Naturkatastrophe ...

GIACOMETTI, M.; RaTTı, P.: Zur Reproduktionsleistung des Alpensteinbockes (Cypra 1. ibex L.)
in der Freilandkolonie Albris (Graubünden, Schweiz). - On the reproductive performance
of the free-ranging alpine ibex population (Capra 1. ıbex L.) at Albris (Grisons, Switzerland) .

HAFERKORN, J.; STUBBE, M.: Zur Altersstruktur von Apodemus flavicollis in einem Auwald an
der mittleren Elbe. - Age structure of Apodemus flavicollis in a floodplain forest in the
middle partottheriverPlbe sy... 2.0. 2.0 2

HERNANDEZ, Lucına; DELıBes, M.: Seasonal food habits of coyotes, Canıs latrans, ın the
Bolsön de Mapimi, Southern Chihuahuan Desert, Mexico. — Jahreszeitliche Ernährungs-
gewohnheiten von Kojoten, Canis latrans, im Bolsön-de-Mapimi-Reservat, südliche Chi-
hüahua-Wüste, Mexiko... S.8..2.8.0 00 Wa el. 2 ER er a

HUTTERER, R.: Island rodents: a new species of Octodon from Isla Mocha, Chile (Mammalıa:
Octodontidae). — Nagetiere auf Inseln: eine neue Octodon-Art von der Isla Mocha, Chile
(Mammalia: Octodentidae). .... 20 a. en en en A SE

Juste B., J.; IBANEZ, C.: Contribution to the knowledge of the bat fauna of Bioko island,
Equatorial Guinea (Central Africa). - Beitrag zur Kenntnis der Chiropterenfauna der Insel
Bioko, Aquatorial-Gumea ... un... 2 a ee ee

KaıLko, ELISABETH K. V.; HAanDL£y, C. O., Jr.: Evolution, biogeography, and description of a
new species of Fruit-eating bat, genus Artibeus Leach (1821), from Panamä. — Evolution,
Biogeographie und Beschreibung einer neuen, fruchtfressenden Fledermausart der Gattung
Antibeıs Leach (11821),aus Panamagı 4. 2020022 per

KASTELEIN, R. A.; MULLER, M.; TERLOUW, A.: Oral suction of a Pacific walrus (Odobenus
rosmarus divergens) in air and under water. — Orales Saugvermögen eines Pazifischen
Walrosses (Odobenus rosmarus divergens) ın Luftund unter Wasser... 2... 2.2220...

Kry$turer, B.; FıLıppuccı, MARIA GRAZIA; MACHOLÄN, M.; ZIMA, J.; VUJosEvIc, M.;

366

181

52

209

116

193

218

161

139

42

358

174

21

82

27

274

257

105

Sımson, S.: Does Microtus majorı occur in Europe? - Kommt Microtus majorıın Europa vor? ..
LINDSTRÖM, E. R.: Placental scar counts in the Red fox (Vulpes vulpes L.) revisited. —- Zählungen
von Implantationsnarben beim Rotfuchs (Vx/pes vulpes L.), eineRevision. ..........
Lorinı, MARIA Lucia; OLIVEIRA, J. A. DE; PERSSON, VAnESsA G.: Annual age structure and
reproductive patterns in Marmosa incana (Lund, 1841) (Didelphidae, Marsupialia). —
Jahreszeitliche Altersstruktur und Fortpflanzung bei Marmosa incana (Lund, 1841) (Didel-
DiiidaewMlarsupralia)bwen.n. >... 0. 0 ee nee si
LÖöPEZ-LUNA, PILAR; AREvALO, F.; Burcos, M. J.; Der Hoyo, N.: Lipid deposits in pregnant
and non-pregnant bats (Pıpistrellus pipistrellus). — Fettspeicher ın trächtigen und nicht
krächtigenäbledermausen (Wıpzstrellussprpistnellusyp 200 0
LUCHERINI, M.; CREMA, GIULIA: Seasonal variation in diet and trophic niche of the Red fox in
an Alpine habitat. — Saisonale Variabilität von Nahrung und trophischer Nische beim Fuchs
mealpmenelabıatee SEE m a ee aan de
MATSsoNn, J. O.; BLooD, B. R.: A report on the distribution of small mammals from Namibia. —
Merbreitungsmusterszonlemsausernun Namibia. 2.0 nenn...
PETERS, G.; Hast, M. H.: Hyoid structure, laryngeal anatomy, and vocalızation in felids
(Mammalia: Carnivora: Felidae). - Hyoidbau, Kehlkopfmorphologie und Lautgebung bei
Behden(Nammalas@arnivora-iEelidaoye 0 wen...
Perıt, P.; De Boıs, H.; DE Meurichy, W.: Chromosomal reduction in an Okapı pedigree
(Okapia johnstoni).- Chromosomenreduktion in einer Okapı-Familie (Okapia johnstoni) . .
Pıuc, Ina: Springbok, Antidorcas marsupialis (Zimmerman, 1780) from the past. — Spring-
böcke, Antidorcas marsupialıs (Zimmerman, 1780), aus vergangenen Zeiten . ... 2.2...
Purvgs, M. G.; Kruuk, H.; Net, J. A. J.: Crabs Potamonautes perlatus in the diet of Otter
Aonyx capensis and Water mongoose Atılax paludinosus ın a freshwater habitat in South
Afrıca. -— Krabben, Potamonautes perlatus, in der Nahrung von Kapfıngerotter, Aonyx
capensis, und Wassermanguste, Atılax paludinosus, in einem Süfswasserhabitat ın Südafrika. .
RucksTUHL, K.; InGoLp, P.: On the suckling behaviour of Alpine chamoıs Rupicapra rupıicapra
rupicapra. — Zum Saugverhalten bei der Alpengemse Rupicapra rupicapra rupicapra . ......
Saucy, F.: Fates of fossorial Water voles, Arvicola terrestris, as revealed by radıotelemetry. —
Erfassung von Einzelschicksalen bei Ostschermäusen (Arvicola terrestris) mit Hilfe der
Radıorelemerziegg AN I lee a
SPOTORNO, A. E.; SUFAN-CATALAN, J.; WALKER, LAURA ].: Cytogenetic diversity and evolution
of Andean species of Ehgmodontia (Rodentia, Muridae). — Cytogenetische Vielfalt und
Evolution von Eligmodontia-Arten in den Anden (Rodentia, Muridae) ............
STUBBE, ANNEGRET; WIEGAND, SABINE: Ontogenesis,of pelage and the course of moulting in
Microtus brandti (Radde, 1861). - Ontogenese des Haarkleides und Fellwechselverlauf von
nero brand Badde 130er. re ee
STUBBE, ANNEGRET; WIEGAND, SABINE: Influence of photoperiod and temperature on moulting
processes ın Microtus brandtı (Radde, 1861). — Der Einfluß von Photoperiode und Tempera-
turaußkEellwechselprozesse bei Mzerotus brandtı (Radde, 1861) ................
TURNI, H.; SCHÖNHERR, R.: Neue Nachweise der Schabrackenspitzmaus (Sorex coronatus) ın
Baden-Württemberg durch Polyacrylamidgel-Elektrophorese. - New records of the Jersey
shrew (Sorex coronatus) ın Baden-Württemberg by polyacrylamide gel electrophoresis . . . .
TORRE, 1.; TELLA, J. L.: Distribution of the Cabrera water shrew (Neomys anomalus) ın
Northeastern Spain. — Verbreitung der Sumpfspitzmaus (Neomys anomalus) ım Nordosten
SPACE REOE LUDER TER HR ERBE ar BED EA 2%
VAsSART, M.; GRANJON, L.; GRETH, A.; CATZEFLIS, F. M.: Genetic relationships of some
Gazella species: an allozyme survey. — Genetische Verwandtschaft einiger Gazella-Arten:
ene Allozyme Ungsicmag Lern
Vos, A. C.: Reproductive performance of the Red fox, Vulpes vulpes, in Garmisch-Partenkir-
chen, Germany, 1987-1992. — Die Reproduktion des Rotfuches (Vulpes vulpes) im Land-
kreis@armisch- Partenktechen, Deutschland, 198721992 rm:
WEBER, J.-M.; MEIA, J.-S.; AuBRY, $.: Activity of foxes, Vulpes vulpes, in the Swiss Jura
mountains. — Aktivität des Rotfuchses, Vulpes vulpes, im Schweizer Jura. ...........

WHITAKER, J. O., JR.; SHALMON, B.; Kunz, T. H.: Food and feeding habits of insectivorous bats
from Israel. - Nahrung und Ernährungsverhalten einiger insektivorer Fledermäuse aus Israel.
ZAMBELLI, A.; VıDaL-RıoJA, LiDIA; WAINBERG, R.: Cytogenetic analysis of autosomal poly-
morphism in Graomys griseoflavus (Rodentia, Cricetidae). - Cytogenetische Analysen von
Autosomen-Polymorphismen bei Graomys griseoflavns (Rodentia, Cricetidae) ........

349

169

65

289

87

153

246

932

230

342

232

199

309

321

282

Wissenschaftliche Kurzmitteilungen

REUMER, J. W. F.: Ehomys (Hypnomys) onicensis nomen novum, to replace the homonym
Hypnomys intermedius Reumer, 1981 (Rodentia: Gliridae) from Majorca. — Eliomys Hypno-
mys onicensis nomen novum als Ersatz für das Homonym Hypnomys intermedius Reumer,
IS Rodentıas.Glırıdae) von Mallorca re 380
SCHREIBER, A.; DmocH, R.: Chromosomes of two rare species of neotropical mammals:
Southern pudu (Pudu puda) and Bush dog (Speothos venaticus). -— Karyotypen von zwei
seltenen südamerikanischen Säugerarten: Südpudu (Pudu puda) und Waldhund (Speothos
VENALIEUS)) ze. une deals Dre Sutenp wenns net eher ee ken 317
SEILER, A.; KRÜGER, H. H.; FEsTETIcs, A.: Reaction of a male Stone marten (Martes foina
Erxleben, 1777) to foreign faeces within its territory: a field experiment. — Reaktion eines
Steinmarderrüden (Martes foına Erxleben, 1777) auf Fremdlosung innerhalb seines Territo-

rıums; ein Freilandexperiment 0... 0. a. u Se 58
LoDe, TH.: Feeding habits of the Stone marten and environmental factors in western France. —
Ernährungsgewohnheiten des Steinmarders und Umweltfaktoren in Westfrankreich ....... . 189

MEIJAARD, E.; BREE, P. J. H. van: Cases of dental malocclusion in populations of Red foxes
(Vulpes vulpes) in the state of Victoria, Australia. — Fälle von Gebißanomalie in Populatio-
nen des Rotfuchses (Vulpes vulpes) aus dem Staat Victoria, Australien ............. 378
MusTRanGI, MEIKA A.: Marmosops scapulatus Burmeister, 1856, and the brown mutation in
didelphids (Marsupialia). - Marmosops scapulatus Burmeister, 1856, und die braune Muta-

tiombei Didelphiden (Marsupialia)n 2200 252
VırA, BısranA L.: Use of dung piles by neighbouring vicunas. — Benutzung von Kotplätzen
dureh benachbarte Vieunas ..) . „unctena 0 ec en Se 126

SEHEN. nei u new er Are et RE u ee 61521925982

SCHEN a. en a a a 64, 255

This journal is covered by Biosciences Information Service of Biological Abstracts, and by Current Con-
tents (Series Agriculture, Biology, and Environmental Sciences) of Institute for Scientific Information

Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch
begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme
von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der Mikroverfilmung oder der
Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugsweiser Verwertung, vorbehalten.
Das Vervielfältigen dieser Zeitschrift ist auch im Einzelfall grundsätzlich verboten. Die Herstellung
einer Kopie eines einzelnen Beitrages oder von Teilen eines Beitrages ist auch im Einzelfall nur in den
Grenzen der gesetzlichen Bestimmungen des Urheberrechtsgesetzes der Bundesrepublik Deutschland
vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungs-
pflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtsgesetzes. Ge-
setzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Vervielfältiger
zu kennzeichnen.

© 1994 Paul Parey. Verlag: Paul Parey GmbH & Co. KG, Hamburg. Anschrift: Spitalerstraße 12,

D-20095 Hamburg, Bundesrepublik Deutschland. Printed in Germany by Westholsteinische Verlags-
druckerei Boyens & Co., Heide/Holstein

| ISSN 0044-3468 59 (1-6) 1-384 (1994) |

Vol. 59 “ ) 1-64, Februar 1994 ISSN 0044-3468 C 21274 F
Ä 2
2487

= ITSCHRIFT FÜR
SKUGETIERKUNDE

INTERNATIONAL JOURNAL

OF MAMMALIAN BIOLOGY

Pe der Deutschen Gesellschaft für Säugetierkunde

| Lucherini, M.; Crema, Giulia: Seasonal variation in diet and trophic niche of the Red fox in an Alpine habitat. —
' Saisonale Variabilität von Nahrung und trophischer Nische beim Fuchs in alpinem Habitat 1

| Weber, J.-M.; Meia, J.-S.; Aubry, S.: Activity of foxes, Vulpes vulpes, in the Swiss Jura mountains. — Aktivität des

Rotfuchses, Vulpes vulpes, im Schweizer Jura 9
Zambelli, A.; Vidal-Rioja, Lidia; Wainberg, R.: Cytogenetic analysis of autosomal polymorphism in Graomys
griseoflavus (Rodentia, Cricetidae). — Cytogenetische Analysen von Autosomen-Polymorphismen bei Graomys
| griseoflavus (Rodentia, Cricetidae) 14
Haferkorn, J.; Stubbe, M.: Zur Altersstruktur von Apodemus flavicollis in einem Auwald an der mittleren Elbe. — Age
ı structure of Apodemus flavicollis in a floodplain forest in the middle part of the river Elbe 21
Hutterer, R.: Island rodents: a new species of Octodon from Isla Mocha, Chile (Mammalia: Octodontidae). —
Nagetiere auf Inseln: eine neue Octodon-Art von der Isla Mocha, Chile (Mammalia: Octodontidae) 27
Filippucci, Maria G.; Burda, H.; Nevo, E.; Kocka, J.: Allozyme divergence and systematics of Common mole-rats
(Cryptomys, Bathyergidae, Rodentia) from Zambia. — Allozymatische Divergenz und Systematik der Graumulle
(Cryptomys, Bathyergidae, Rodentia) aus Sambia 42
Borghi, C. E.; Giannoni, Stella M.; Martinez-Rica, J. P.: Habitat segregation of three sympatric fossorial rodents in
| the Spanish Pyrenees. — Habitatwahl bei drei Arten von Kurzohrmäusen in den spanischen Pyrenäen 52
|Wissenschaftliche Kurzmitteilung
'Seiler, A.; Krüger, H. H.; Festetics, A.: Reaction of a male Stone marten (Martes foina Erxleben, 1777) to foreign
faeces within its territory: a field experiment. — Reaktion eines Steinmarderrüden (Martes foina Erxleben, 1777)
ı auf Fremdlosung innerhalb seines Territoriums: ein Freilandexperiment____ 58
Mitteilungen der Gesellschaft Ä 2
|Buchbesprechungen 64

|
|
& Paul Parey Hamburg und Berlin
|
|

HERAUSGEBERY/EDITORS

P. J. H. van BreE, Amsterdam — W. FIEDLER, Wien -— H. Frick, München - G. B.

HarTL, Wien -—- W. HERRE, Kiel - R. HUTTERER, Bonn - H.-G. Krös, Berlin - H.-].

Kunn, Göttingen — E. Kurzer, Tübingen -— W. MAIER, Tübingen — J. NIETHAMMER,

Bonn - ©. Anne E. Rasa, Bonn - H. ReıcHsteEin, Kıel -— M. Rönrs, Hannover —

H. ScHLIEMAnn, Hamburg — D. STARcK, Frankfurt a. M. — E. THentus, Wien - P. Vo-
GEL, Lausanne - H. Wınkıng, Lübeck

SCHIRTETBELBUNGYEDTEORTIZOEEN@E

D. Kruska, Kiel - P. LAnGEr, Gießen

This journal is covered by Biosciences Information Service of Biological Abstracts, and by Current Con-
tents (Series Agriculture, Biology, and Environmental Sciences) of Institute for Scientific Information

Die Zeitschrift für Säugetierkunde veröffentlicht Originalarbeiten und wissenschaftliche Kurzmittei-
lungen aus dem Gesamtgebiet der Säugetierkunde, Besprechungen der wichtigsten internationalen
Literatur sowie die Bekanntmachungen der Deutschen Gesellschaft für Säugetierkunde. Verantwort-
licher Schriftleiter im Sinne des Hamburgischen Pressegesetzes ist Prof. Dr. Dieter Kruska.

Zusätzlich erscheint einmal ım Jahr ein Heft mit den Abstracts der Vorträge, die auf der jeweiligen
Hauptversammlung der Deutschen Gesellschaft für Säugetierkunde gehalten werden. Sie werden als
Supplement dem betreffenden Jahrgang der Zeitschrift zugeordnet. Verantwortlich für ihren Inhalt
sind ausschließlich die Autoren der Abstracts.

Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z. Hd. Prof. Dr. Dieter
Kruska, Institut für Haustierkunde, Biologiezentrum der Christian-Albrechts-Universität, Am
Botanischen Garten 9, D-24118 Kiel, Bundesrepublik Deutschland. Für die Publikation vorgese-
hene Manuskripte sollen gemäß den „Redaktionellen Richtlinien‘ abgefaßt werden. In ihnen
finden sich weitere Hinweise zur Annahme von Manuskripten, Bedingungen für die Veröffent-
lichung und die Drucklegung, ferner Richtlinien für die Abfassung eines Abstracts und eine

Korrekturzeichentabelle. Die Richtlinien sind auf Anfrage bei der Schriftleitung und dem Verlag
erhältlich.

Sonderdrucke: Anstelle einer Unkostenvergütung erhalten die Verfasser von Originalbeiträgen und
Wissenschaftlichen Kurzmitteilungen 50 unberechnete Sonderdrucke. Mehrbedarf steht gegen
Berechnung zur Verfügung, jedoch muß die Bestellung spätestens mit der Rücksendung der
Korrekturfahnen erfolgen.

Vorbehalt aller Rechte: Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich
geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks,
des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der
Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei-
ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch ım Einzelfall grund-
sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines
Beitrages ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber-
rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom
24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen
den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind
mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen.

Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of
the first page of an article in this journal indicates the copyright owner’s consent that copies of the article may
be made for personal or internal use, or for the personal or internaluse of specific clients. This consent is given
on the condition, however, that the copier pay the stated percopy fee through the Copyright Clearance
Center, Inc., 21 Congress Street, Salem, MA 01970, USA, for copying beyond that permitted by Sections 107

or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying
for general distribution, for advertising or promotional purposes, for creating new collective, or for resale.
For copying from back volumes of this journal see “Permissions to Photo-Copy: Publisher’s Fee List” of the
GEC.

Fortsetzung 3. Umschlagseite

© 1994 Paul Parey. Verlag: Paul Parey GmbH & Co. KG, Hamburg und Berlin. Anschriften: Spitalerstr. 12,
D-20095 Hamburg; Seelbuschring 9-17, D-12105 Berlin, Bundesrepublik Deutschland. — Printed in Germany by
Westholsteinische Verlagsdruckerei Boyens & Co., Heide/Holst.

Z. Säugetierkunde 59 (1994) 1-8
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Seasonal variation in diet and trophic niche of the Red fox
in an Alpine habitat

By M. LucHERINI and GIULIA CREMA

Department of Evolutionary Biology, Ethology and Behavioural Ecology Group,
University of Siena, Siena, Italy

Receipt of Ms. 30. 10. 1992
Acceptance of Ms. 23. 4. 1993

Abstract

The diet and trophic niche breadth of the fox (Vulpes vulpes) were studied in a high-elevation Alpine
ecosystem to determine their monthly variations. The analysis of 270 faeces showed that the fox used
all the potential resources present in its habitat: of total diet, small mammals were the most frequent
food category, but Orthoptera and marmots were also very important. Correlations between diet and
weather variables suggested a seasonal shift in diet. From June to November foxes preyed mainly on
insects and marmots, whereas ungulates, Lagomorpha, garbage and earthworms were eaten more
often from December to May. These changes appeared related both to seasonal differences in food
availability and to the presence of alternative preferred foods. Trophic niche was wide during the
entire year, with a moderate increase in winter-spring when the absence of some main prey types
forced the fox to exploit a larger range of items.

Introduction

The adaptable behaviour of the red fox has enabled it to colonize a variety of habıtats,
including those typically montane (MAacponarpd 1987). Although the fox ıs probably the
most common carnıvore ın the Alps, only two recent studies (CAnTını 1991; STORCH and
KLEıne 1991) have dealt with the diet of the fox ın the Alpine region, and both were carried
out at elevations between 300 and 1700 m. The diet of the fox in the highest part of its
altitudinal distribution ın Europe has been locally investigated only by Leınarı et al.
(1960).

Montane ecosystems are strongly seasonal in terms of climate and productivity. Given
the opportunistic feeding habits of the fox (e.g. ENGLUND 1965; GOSZCZYNsKI 1986;
Carıstı et al. 1990), it can be expected that its diet in the Alps would largely reflect the
seasonal changes ın food availabiılıty.

Our study area, in the Italian Western Alps, is frequented by at least 370 Alpine
chamois (Rupicapra rupicapra) (minimum density: 10 individuals/km?), about 20 Alpine
ibex (Capra ibex), at least 25 roe deer (Capreolus capreolus) and 5-15 red deer (Cervns
elaphus). Furthermore, although no estimates are available, the study area supports wild
boars (Sus scrofa) and dense populations of Alpine marmot (Marmota marmota) and
insects (mainly Orthoptera). The results of the research conducted in the Alps by Leınarı
et al. (1960) showed that ungulate carrion, especially in late winter, and marmots, ın
summer, are well represented ın the fox diet in the Alps, but did not include Orthoptera
among the most important food items. Nevertheless, since strong predation on Orthoptera
has been reported in other habitats (Cazıstı et al. 1990 for a review), we expected that the
local abundance of marmots and Orthoptera would largely influence the diet of the fox in
the months when they are available. The densities of small mammals in the study area were
not known, but they can also be a main component of the fox diet, as found in northern
highly seasonal habitats (LINDSTRÖM 1989).

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5901-0001 $ 02.50/0

2 M. Lucherini and Giulia Crema

The aım of this paper is to present a picture of the seasonal variation of the fox trophic
nıche in a typıcal Alpine ecosystem. We were particularly interested in determining the
extent of fox predation on marmot and Orthoptera and in documenting the use of ungulate
carcasses. |

Study area

The study area in the Western Italian Alps, about 80 km west of Torino, comprises the Val Troncea
Natural Park (3280 ha) and the small part of the homonymous valley at lower altitude outside the park
boundaries. Elevation ranges from 1560 m to 3280 m a. s. 1. Topography, vegetation and climate of the
valley are typically Alpine. Precipitation peaks in spring and autumn, while winter is characterised by
4-6 months of permanent snow cover and low temperatures. Mean annual temperature is about 8°C
(Fig. 1). Larch (Larix decidua) forests and pastures cover the slopes to about 1900-2100 m, while
Alpıne meadows are prevalent at higher altitudes. At the end of the valley lies a small tourist village,
inhabited mainly during summer and Christmas holidays. Many tourists visit the park from May to
August.

Material and methods

From May 1990 to June 1991, 3-4 monthly collections of faecal samples were made at elevations
between 1560 and 2850 m. In a total of 46 excursions on foot or skis 270 faeces were collected (sample
size: May: 4; Jun: 25; Jul: 16; Aug: 18; Sep: 44; Oct: 23; Nov: 31; Dec: 7; Jan-Feb: 16; Mar: 26; Apr:
15; May 1991: 22; Jun 1991: 23). On the two trails followed each month, all scats were collected;
outside these trails, only obviously fresh samples were collected. Scats were stored in polythene bags
and frozen at -20°C, to be thawed later and analysed as described in Kruuk and ParısH (1981),
CIAMPALINI and Lovarı (1985) and Carıstı et al. (1990). To allow comparisons with other studies,
results are reported as percentage of occurrence (number of occurrence of each food/total number of
occurrences X 100) (CAavaLLını and Lovarı 1991), percent frequency of occurrence (number of
occurrence of each food/number of faeces x 100), and percentage of volume (estimated volume of each
food/total estimated volume x 100) (Kruuk and PArısH 1981). In February we only found two scats,
therefore we pooled data for January and February. In winter, frequent snowfalls often prevented
faecal collections, both by limiting the movements of the foxes and rapidly covering tracks and scats
(PATALAno and Lovarı 1993). As a result, more faecal samples were found during summer-autumn.
In the evaluation of the total yearly diet, scats collected in May 1990 and June 1991 were excluded so
that each month would be represented only once. The total percentages of occurrence and volume
were computed as means of monthly values.

A standardized index of trophic niche breadth (B,..) (COLWELL and Furuyma 1971) was calculated
from both percentages of occurrence and percentages of volume. The index has the formula B,.. =
B-1/Bax-1, where B is the Levıns’ index of niche breadth (Levıns 1968) and B,..x is the total
number of food categories recognized. B,,., values can range between O0 (minimum niche breadth) and 1
(maximum niche breadth).

The availability of Orthoptera was estimated by direct counts of the number of individuals seen
along two fixed transects (100 m each) (CavarLını and Lovarı 1991). One transect was located ın
pastures at 1900 m, the other in Alpine meadows at 2200 m.

To identify seasonal variations in consumption of the main food categories, Spearman rank
correlations (SIEGEL 1956) were performed between percentages of occurrence and volume and
climatic varıables (Carıstı et al. 1990; Cavarıını and Lovarı 1991). Differences between sum-
mer-autumn and winter-spring diets were tested with a G-test of independence (SoKAL and ROHLF
1981) on the occurrences (expressed as frequencies of scats where the category was found over the
total number of scats) and with Mann-Whitney U-test (SoRAL and ROoHLF 1981) on volumes. Values
of p less than 0.05 were considered significant.

Results
Diet

In the total diet, the most frequent and abundant prey remains were of small mammals
(mainly Arvicolidae, followed by Muridae) (Tab. 1), with little seasonal varıatıon (Fig. 2).
Orthoptera, which were only available from June to November (Fig. 1) and consumed
mainly from August to November (Fig. 2), were nevertheless the second most important

_ Orthoptera availability (N)

mean temperature (°C)

Diet and trophic niche of the Red fox in an Alpine habitat 3

16
20 SNOW ® )

| precipitations (I)

% occurrence

BEE GARBAGE

b) 0474 KA \ @ PLANT MATTER
80 = VERTEBRATES
S INVERTEBRATES

60
40
20

[6) fa

2 £53 22.9 22 L. 3 5 >55 > £ce5 22.922 L 5 5 >55
=s3>253024°23<33 ESF RSSHE FREE

Fig. 1. Monthly changes of: a) Mean temperature and precipitation in the study area; b) Orthoptera
availability in the study area; c) Percent occurrence and volume of the main groups of food categories
in the diet of the red fox in Val Troncea

category in the annual diet (Tab. 1). Alpine marmots (often young individuals) and wild
ungulates (mostly carcasses of Alpıne chamoıs and roe deer, but also some young of roe
deer and wild boar) were the only other items accounting for more than 10% of total
volume, but their occurrences were remarkably lower (Tab. 1). The presence of marmot ın
the dıet showed a peak in June-July, whereas that of ungulates was higher ın January-Feb-
ruary than in the rest of the year (Fig. 2). Coleoptera remains were found in all months
except December (Fig. 2). They were eaten very often but usually ın small amounts (Tab.
1; Fig. 2). A similar discrepancy between occurrences and volumes was shown by
Lepidoptera larvae and earthworms (Tab. 1). As a whole, fruits were found ın almost 15 %
of the scats and composed 4.5 % of the annual volume, but the main kind of fruits eaten by
the fox (Rosa sp. fruits) dıd not reach 3% of the volume (Tab. 1). More important was the
role of garbage, which was present in 21 % of all the scats with a total volume of 6.6 %
(Tab. 1). Birds (mainly Passeritormes) and Lagomorpha (mainly mountain hare, Lepus
timidus) were seldomly consumed by foxes in the Troncea valley (Tab. 1).

Overall, vertebrates composed the bulk of the diet (38.2 % occurrence and 56.2 %
volume; see also Fig. 1). The seasonal pattern of use of this food source was rather
constant, with a moderate decrease during summer (Fig. 1). Invertebrates (33.3 % occur-
rence and 30% volume) showed the opposite trend, decreasing markedly during winter
(Fig. 1). Plant matter and garbage were taken mainly in winter and early spring,
respectively (Fig. 1).

Correlations

In most cases, correlations between diet and main clımatic varıables confirmed the
seasonality described above for some food categories (Tab. 2). Coleoptera consumption
was inversely correlated to snow cover (percentage of days with snow-covered ground)
and directly to mean temperature. The presence of marmot in diet showed the same kind of

4 M. Lucherini and Giulia Crema

Table 1. Annual diet composition of the red fox in the Val Troncea Natural Park
(n = 243 faecal samples)

Food items % freq. occ.

Coleoptera
Orthoptera
Dermaptera
Coleoptera larvae
Lepidoptera larvae
Diptera larvae
Earthworms

Other invertebrates

Small mammals
Marmots
Lagomorpha

Ungulates

Other mammals

Birds

Reptiles

Amelanchier ovalıs fruits
Rubus sp. fruits

Rosa sp. fruits

Other fruits

Other plant matter

oo vr WOoOo owr-u
DD Hamm m oNN

Garbage
Bea

S
un
oO

B;:.: standardized trophic niche breadth index (see text).

Table 2. Correlations, on a monthly basis, of food categories in diet with mean temperature (°C),
. [7 . y 5 . pP
precipitations (ml) and snow cover (% days with snow-covered ground)

(n = 13 months)

Food items Mean temp. Precipitations Snow cover
IE P Ys P

Coleoptera freq. occ. 0.835 Se 0.271 0.35
Occ. 0.896 ne 0.202 0.48
vol. 0.732 5 0.396 AS

Marmots freq. occ. Er AS
Occ. = 4 AS
vol. = 5

Ungulates freq. occ. 0.42
Occ. 0.36
vol. 0.58

Fruits freq. 2 , Ä L 028
Occ. $ s 0.32
vol. ß = i 0.26

AS = almost significant (0.10>p> 0.05); * = p<0.05; ** = p<0.01.

correlations as Coleoptera and a tendency to increase with precipitation. Fruits were eaten
more frequently during cold and snowy months than during the rest of the year. This
seemed also to be the case for ungulates, but significance was not reached. No trend was
shown by Orthoptera, small mammals and garbage. Use of Orthoptera increased together
with their estimated availability (occurrence: r, = 0.570, p <0.05; frequency of occurrence:

Diet and trophic niche of the Red fox in an Alpine habitat 5

COLEOPTERA

40

50
20 25
0 0

ORTHOPTERA

40
1
0

MALL MAMMALS

50

oO

MARMOTS

®)

>

2100 40,>
501 < 20
Sh S

S S

< ®)
=) S
0 0
&

(Um

S

©
oO

100 40 UNGULATES
50 20
0 Sp ee: 0 | F Er
MJJASONDJFMAMJ MJ J ASONDFIFMAMJ
Months Months Months

Fig. 2. Monthly varıation of the main food categories in the red fox diet

r, = 0.587, p = 0.05; volume: r, = 0.747, p = 0.01; n = 13 months). No seasonal pattern was
found for the varıation of trophic niche breadth (B,.. values; in occurrences, May: 0.358;
1097455 [ul2 04162: 2 Aue: 20419 7Sep:202522 O2 0.4335, N 042 0.2345 7Dee:7 0.267;
Jan-Feb: 0.315; Mar: 0.369; Apr: 0.306; May 1991: 0.361; Jun 1991: 0.129; ın volumes,
IMEyz20A12.1- Sfun2 0415354]ul:2041457Au2:20,122525ep270.1535 Oct: 0,323 7Nov: 0.1157 Dee:
0.228; Jan-Feb: 0.109; Mar: 0.256; Apr: 0.133; May 1991: 0.298: Jun 1991: 0.196) and no
significant correlation between it and the monthly percentages of the main food categories
or the number of scats collected each month.

Seasonality

Two main periods were distinguished in the diet of the fox in the study area on the basıs of
its monthly varıation and the correlations shown by the main food categories (Fig. 3).
From June to November foxes ate more insects (Orthoptera: occurrence G = 91, p <
0.001; volume Z = -2.37, p < 0.05; Coleoptera: occurrence G = 19.8, p < 0.001; volume Z

6 M. Lucherini and Ginlia Crema

% Occurrence

SUMMER-AUTUMN
WINTER-SPRING

% volume

COLEOPTERA
ORTHOPTERA
EARTHWORMS
MARMOTS
LAGOMORPHA
UNGULATES
GARBAGE

[7]
,
<
=
=
<
=
_
I
<
=
(72)

Fig. 3. Percent occurrence and volume of the main red fox food categories in summer-autumn and
winter-spring diets

= -2.01,p < 0.05) and marmots (occurrence G = 23.4, p < 0.001; volume Z = -1.73, p <
0.08) than from December to May. Conversely, ungulates (occurrence G = 24.4, p <
0.001; volume Z = -2.56, p < 0.05), Lagomorpha (occurrence G = 6.6, p = 0.001; volume
Z = -1.1, p = 0.27) garbage (occurrence G = 26.9, p = 0.001; volume Z = -1.46, p = 0.14)
and earthworms (occurrence G = 19.4, p <0.001; volume Z = -1.55, p = 0.12) appeared to
be more frequent in winter-spring than in summer-autumn. Small mammals did not show
significant seasonal differences (occurrence G = 2.4, p = 0.06; volume Z = -1.64, p = 0.10).

The winter-spring diet was dominated by mammals (62.8% volume; see also Fig. 3).
Nevertheless, the trophic niche appeared to be larger in this period (occurrence B,., =
0.489; volume B,.. = 0.316) than in summer-autumn (occurrence B,.. = 0.372; volume B,..
= 0.234), although the difference did not reach significance (T-test: t=5.7,n =2,p =
0.055).

Discussion

Our data show that the fox can use all the potential sources of food present in its habitat.
Small mammals, Orthoptera, marmots and ungulates supplied the fox with protein-rich
food items. In particular, our results suggest that, as could be expected on the basıs of their
qualitative analysıs (Cavanı 1991), Orthoptera can be an important food resource not only
in warm Mediterranean habitats (Carıstı et al. 1990), but also ın elevated Alpine regions
(see also PATALAno and LovaRrI 1993, for the Italian Apennines). In a sımilar area, LEINATI
et al. (1960) found a greater presence of vegetables and ungulate carrıons and a lower
existence of invertebrates in the fox diet. These differences are likely to be related to the
local varıiation of food availability and also to the inclusion of lower-altitude ranges in the
study area of LEINATT et al.

In our study area, the diet of the fox reflected the alternation of two main seasons
typical of the montane ecosystems (PATALANo and Lovarı 1993). As shown for Orthop-
tera (see the correlations between availability and consumption), the temporal variation in
feeding habits was likely to be mainly determined by the availability of different food

Diet and trophic niche of the Red fox in an Alpine habitat L

resources (DONCASTER et al. 1990; Cavarıını and Lovarı 1991). For example, insects and
marmots are unavaılable in winter, and were present almost exclusively in the sum-
mer-autumn diet. As could be expected, the winter peak in mountain ungulate mortality
(e.g. GEIST 1971; FEsTA-BIANCHET 1989) was reflected by an increase of this item in winter
diet. The higher percentage of earthworms in the cold season may appear surprising.
Actually, the use of this food was concentrated in the months between March and May, the
only months when soil temperature and humidity were probably high enough to permit
frequent earthworm activity on the ground (Kruuk and ParısH 1985; LAMBERT 1990).
Therefore, the variation in the consumption of this invertebrate may also be explained by
that of its availability. On the other hand, the increased use of certain food categories in
winter-spring would confirm the influence of the availability of alternative food resources
on the feeding behaviour of an opportunistic carnıvore, as suggested by WECKWERTH and
HawLey (1962). In our study area, garbage is mostly available during the summer months,
when tourists are numerous. Nevertheless, rubbish remains were found in the scats mainly
in winter. During this season important food items (1. e. insects and marmots) were absent
and the fox exploited other resources, less preferred (i.e. garbage) or difficult to obtain
(1.e. Lagomorpha).

The seasonality of the dıet was not clearly reflected by the trophic nıche sıze, as found
in other areas (CaLıstı et al. 1990). If the number of scats collected each month is not large
enough to be representative of the diet, the monthly value of niche breadth index might be
influenced by sample size. No evidence of such a relation was found (correlations between
B;:. values and monthly number of scats collected were far from significant) (IRIARTE et al.
1990). PrıGıonı (1991) reviewed studies of fox diet in Italy and calculated for them an
index of trophic nıche breadth on a reduced number of main categories. The application of
the same method shows that, ın our study area, the fox has a very wide trophic niche.
These findings could suggest that the lack of correlation between trophic niche breadth and
climatic varıables in our study might have been more correctly related to ıts large breadth
(ef. Carıstı et al. 1990) than to uncorrect sampling. In the Troncea Valley, the fox had a
varıed diet in each period of the year, with a moderate widening of the trophic niche during
winter-spring. This varıation was likely due to the increased presence in the diet of
“secondary” items (e.g. garbage, fruits, Lagomorpha) in cold months, when some main
food categories are absent and the fox ıs forced to exploit a larger array of items to meet its
nutritional requirements.

Acknowledgements

We wish to thank M. FEsTA-BIANCHET, $. LOvARI and two anonymous referees, whose suggestions
greatly improved our first draft. We are grateful to the staff of the Val Troncea Natural Park, which
helped us ın many ways. The abstract was kindly translated by U. BREITENMOSER. The research was
partially supported by the Val Troncea Natural Park, Italy, and by a grant to $. Lovarı from Italian
Ministry of Education.

Zusammenfassung

Saisonale Variabilität von Nahrung und trophischer Nische beim Fuchs in alpinem Habitat

Mit dem Ziel, die monatliche Variabilität zu bestimmen, untersuchten wir die Nahrungswahl und die
trophische Nische des Fuchses (Vulpes vulpes) ın höheren alpinen Stufen. Die Analyse von 270
Exkrementen zeigte, daß sämtliche in diesem Habitat vorhandenen Nahrungsquellen genutzt wurden.
In der gesamten Nahrung waren Kleinsäuger die am häufigsten gefundene Nahrungskategorie, gefolgt
von Orthopteren und Murmeltieren. Korrelationen zwischen Nahrung und Klimavariablen wiesen
auf einen saisonbedingten Wechsel in der Nahrungswahl hin. Von Juni bis November erbeuteten die
Füchse hauptsächlich Insekten und Murmeltiere, während von Dezember bis Mai häufiger Ungulaten,
Lagomorphen, Abfälle und Regenwürmer gefressen wurden. Dieser Wechsel scheint sowohl mit
saisonalen Unterschieden in der Verfügbarkeit von Futter als auch mit dem Angebot an bevorzugter
Alternativnahrung zusammenzuhängen. Die trophische Nische war während des ganzen Jahres breit
und zeigte eine leichte Ausweitung im Winter und Frühjahr, wenn das Fehlen einiger Hauptbeuten die
Füchse veranlaßte, ein breiteres Nahrungsangebot zu nutzen.

8 M. Lucherini and Ginlia Crema

References

CALISTI, M.; CIAMPALINI, B.; LOvarı, $.; LuUcCHERINI, M. (1990): Food habits and trophic niche
variation of the red fox Vulpes vulpes (L., 1758) in a Mediterranean coastal area. Rev. Ecol. (Terre
vie) 45, 309-320.

Canrinı, M. (1991): Alimentazione della volpe in aree boscate delle Alpı Orobie. Hystrix 3, 83-89.

CaAvauLint, P.; Lovarr, $. (1991): Environmental factors influencing the use of habitat in the red fox
Vulpes vulpes. J. Zool. (Lond.) 223, 323-339.

Cavanı, C. (1990): Qualitä della dieta della volpe (Vulpes vulpes) in un’area costiera mediterranea
(Italia centrale). Hystrix 3, 63.

CIAMPALINI, B.; Lovarı, $. (1985): Food habits and trophic niche overlap of the badger (Meles meles
L.) and the red fox (Vulpes vulpes L.) in a Mediterranean coastal area. Z. Säugetierkunde 50,
226-234.

CoLwELL, R. R.; Futuyma, D. ]J. (1971): On the measurement of niche breadth and overlap. Ecology
52, 567-572.

DONCcASTER, C. P.; DicKkMAn, C. R.; MAcDonaLD, D. W. (1990): The feeding ecology of red foxes
(Vulpes vulpes) in the city of Oxford, England. J. Mammalogy 71, 188-194.

EnGLUND, J. (1965): Study on the food ecology of the red fox (Vulpes vulpes) in Sweden. Viltrevy 3,
375485.

FEstAa-BIAncHET, M. (1989): Survival of male bighorn sheep in southwestern Alberta. J. Wildl.
Manage. 53, 259-263.

GEIST, V. (1971): Mountain sheep. Chicago: University of Chicago Press.

GOoszczynsk1, J. (1986): Diet of foxes and martens in Central Poland. Acta Theriol. 31, 491-506.

Harrıs, $. (1986): Urban foxes. London: Whittet Editions.

IRIARTE, J. A.; FRANKLIN, W. L.; JoHnson, W. E.; REDFORD, K. H. (1990): Biogeographic varıation
of food habits and body size of the America puma. Oecol. 85, 185-190.

Kruuk, H.; ParısH, T. (1981): Feeding specialization of the European badger Meles meles ın
Scotland. J. Anım. Ecol. 50, 773-788.

— - (1985): Food, food availability and weight of badgers (Meles meles) in relation to agricultural
changes. J. Appl. Ecol. 22, 705-715.

LAMBERT, A. (1990): Alimentation du blaireau euroasien (Meles meles) dans un Ecosysteme forestier:
variations spatiales du regime et comportament de predation. Gib. Faune Sauv. 7, 21-37.

LEINATI, L.; MANDELLI, G.; VIDESOTT, R.; GRIMALDI, E. (1960): Indagini sulle abitudini alımentari
della volpe (Vulpes vulpes L.) del Parco Nazionale del Gran Paradiso. La Clinica Veterinarıa 83,
305-328.

Leviıns, R. (1968): Evolution in changing environments. New Haven: Princeton University Press.

LINDSTRÖM, E. (1989): Food limitation and social regulation ın a red fox population. Hol. Ecol. 12,
70-79.

MacponaDd, D. W. (1987): Running with the fox. London, Sidney: Unwin-Hyman.

PATALANO, M.; Lovarı, S. (1993): Food habits and trophic niche overlap of the wolf (Canis lupus, L.
1758) and the red fox Vulpes vulpes (L. 1758) ın a Mediterranean mountain area. Rev. Ecol. (Terre
vie) 48, 279-294.

Prıcıont, C. (1991): Lo studio della dieta della volpe Vulpes vulpes. Hystrix 3, 51-62.

SIEGEL, $. (1956): Nonparamerric statistic for the behavioral sciences. New York: McGraw-Hill.

SoKAL, R. R.; ROHLEF, F. J. (1981): Biometry. 2nd ed. New York: W. H. Freeman and Co.

STORCH, ]., KLEINE, C. (1991): Zur Nahrungswahl des Fuchses in den Voralpen. Z. Jagdwiss. 37,
267-270.

WECKWERTH, R. P.; HawLEy, V. D. (1962): Marten food habits and population fluctuations in
Montana. J. Wildl. Manage. 26, 55-74.

Authors’ addresses: MAURO LUCHERINI, Dipartimento dı Biologia Evolutiva, Universita dı Siena,
V. Mattioli 4, I-53100, Siena, Italy, and GIULIA CREMA, Estacıön Biolögica de
Donana, C. S. I. C., Apartado 1056, E-41080 Sevilla, Spain

Z. Säugetierkunde 59 (1994) 9-13
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Activity offoxes, Vulpes vulpes, in the Swiss Jura mountains

By J.-M. WEBER, J.-S. MEIA, and $. Außry

Institut de Zoologie, Universite de Neuchatel, Neuchatel, Switzerland

Receipt of Ms. 26. 1. 1993
Acceptance of Ms. 21.9. 1993

Abstract

Studied the activity pattern of seven radio collared foxes in a mountainous habitat in Switzerland.
Foxes were mainly nocturnal and their nocturnal activity pattern was acyclıc. Activity periods varied
according to seasons, being shortest during summer. On the other hand effective activity did not vary
significantly throughout the year. Fox active period was generally interrupted by several inactive
phases, most of which were short (< 15 min.).

Introduction

Red fox activity has been studied in different lowland habitats (AgLes 1969; ArToıs 1985;
WOOLLARD and HARRIS 1990 amongst others), but little is known of red fox activity in
mountainous habitats. In their study of fox habitat use in the Swiss Alps, Carr and
STALDER (1988) emphasızed the nocturnal pattern of fox activity. Most authors usually
described fox activity pattern and its variations ın general terms (e.g., ARTOIS 1989). There
are few quantitative data about activity level and time budget. Recently, WOOLLARD and
Harrıs (1990) recorded the duration of inactive and active bouts occurring during the
active period of urban foxes.

Therefore, the objective of this study is to describe the activity pattern and to quantify
the activity level of foxes ın a mountainous habıtat.

Material and methods

This study was carried out in a rural area of the Swiss Jura mountains (47°09’ N, 6°56’ E; altitude: 995
to 1288 m). This area is mainly composed of pastures and wooded pastures, as described by WEBER
and Ausry (1993).

Seven vixens (F2, F3, F8, F10, F11, F12, and F19) were tracked between September 1989 and May
1992. The foxes were snared, and then tranquilized with an injection of ketamin hydrochloride. Each
fox was tagged with 1 or 2 colored eartags (Dalton Supplies, Ltd., Henley-on-Thames, UK). Adult-
sızed individuals were fitted with activity monitoring transmitters (Wildlife Materials, Inc., Carbon-
dale, Ill., USA).

Radio tracking data were collected in two different ways. Radio tagged animals were located daily
(1 fix/fox/day) to determine their diurnal resting sites. The collared anımals were also tracked for one
24-h period (12.00 to 12.00; 1 fix/15 min.) one day per week during the first month following their
capture. They were then tracked for 24 hours (1 fix/15 min.) once every two weeks. Radio tracking
was made either by car or on foot using portable telemetry equipment.

Variations ın radıo sıgnal pulse rate indicated whether foxes were moving (active) or not (inactive);
hence fıxes were recorded as either active or inactive. Two periods were distinguished within a day:
the daytime/daylight (dawn to dusk) and the nighttime (dusk to dawn). The total active fixes (diurnal
and nocturnal) constituted fox effective activity (EA). The activity period (AP) was determined by
nocturnal active and inactive fixes as well as diurnal activity fiıxes. A prolonged inactive phase
(>1 min.) during the nocturnal activity period was considered as “rest during the active period”

(RAP).

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5901-0009 $ 02.50/0

10 J.-M. Weber, J.-S. Meia, and S. Aubry

Results and discussion
Activity period and seasonal variations

The activity period (AP) of all vixens was mainly nocturnal, but the individuals were also
active to different degrees in daylight (Tab.1). Two foxes (F2 and F3) were seldom or
never active ın the daytime (1.7% (9546) and 0% (0/107) of diurnal active fixes
respectively). Other vixens were significantly more diurnal (x? test, p < 0.05). For F8, F10
and F19, the total proportion of diurnal active fixes did not exceed 9%, whereas it
represented 15.4% and 11.7% for F1l and F12, respectively. There was no significant
trend ın seasonal variations, but some individuals were more diurnal in one season than in

another: F1l was more active during daytime in winter than in spring and summer (y? =
108.89, d.f. = 2, p< 0.001).

Table 1. Seasonal frequency (%) of diurnal active fixes recorded in foxes

n: number of diurnal active fixes

Summer Autumn

AsgLes (1969) noticed that such variations could be influenced by climatological factors,
for instance shorter diurnal activity in summer due to higher temperature and insolation.
EsucHi and NAKAZoNO (1980) emphasized human disturbance as a possible cause of fox
diurnal activity, at least for individuals which were resting above ground during daytime.

Although we did not quantity the climatological factors, it seems that their influence
was neglıgible, since there was no seasonal trend in diurnal activity. However, extreme
weather conditions during the nıght could contribute to an increase of diurnal activity.
This was observed for F11 during winter, when some nights were cold (ca. -25 °C.) On the
other hand, we consider the human factor as very important. The high hunting pressure
occurring in the region probably induced a nocturnal pattern of activity. In areas where
cover was scarce, some foxes (F2 and F3) spent the day in dens and were strietly nocturnal
most probably for safety reasons. For other foxes, diurnal activity was usually limited to
movements between resting sites (MEIA and WEBER 1993). Foraging activity was uncom-
mon in daylıght, and even F8 dıd not show any increase of diurnal activity when rearıng
cubs during spring as described for another breeding vixen (PHıLLıps and CArtLing 1991).

80 tl
40 | || IIIKIJEINTNU N
a | III
0 nr — Aal |
24 6 12

12 18

8

ACTIVE FIXES (%)

TIME

Example of fox activity pattern. F10: autumn and winter

Activity of foxes, Vulpes vulpes, in the Swiss Jura mountains 11

The nocturnal activity pattern
of the seven vixens was acyclıc
(Fig. 1). Most authors found a
cyclicity ın fox nocturnal activity
pattern, being either bimodal
(Ares 1969; Arrtoıs 1985) or
trımodal (EGucHI and
NARAZoNO 1980). According to
ABLeEs (1969), fox actıvity peaks
were synchronized with those of
the prey. In our study area, the
main prey of foxes was the water
vole, Arvicola terrestris scherman
(WEBER and Ausry 1993). Little
is known about its activity pat-
tern, but ın Britain, no varıatıon
in activity was found between
day and nıght (Boyce 1991). Our
field observations indicated simi-
lar behaviour of water voles. No
difference in trappability was
found between day and night.
Predators such as farm cats, Felıs
catus, and some raptors, regular-
ly preyed on them in the day-
time. This suggests a great varıia-
bility in the actıvity pattern of
water voles and could explain fox
acyclic activity pattern in our
study area to some extent.

There were seasonal varıations
in actıvity level of foxes (Tab. 2).
Their actıvity period was shortest
during summer and generally
longest during winter (Mann-
Whitney U test, p<0.05). Be-
sıdes, there were few varıations
between individuals. F12’s au-
tumn APs were longer than those
of other foxes (Mann-Whitney U
test, p<0.05), and a similar
trend was also observed in spring
and summer.

Since fox activity period was
mostly nocturnal in our area, the
seasonal character of ıts duration
was likely related to nighttime
length (EsucHı and NAKAZONO
1980; CAPtT and STALDER 1988).
Some intrinsic factors could ex-
plain the differences in the AP
length observed between F12 and

Table 2. Seasonal duration (min.) of fox activity periods, rest during the activity period and effective activity
X: mean SD: standard deviation

Effective activity
Rest during AP

Activity period
spring

spring
autumn
winter
summer
autumn
winter
summer
autumn
winter

summer

112 J.--M. Weber, J.-S. Meia, and S. Aubry

the other vixens. She was shot a few days before her capture, survived but never fully
recovered. Because of her weak body condition, she probably needed to distribute her
foraging effort over a longer period than other foxes.

Occurrence of rest during the activity period

Rest during the activity period (RAP) was related to AP length (r, = 0.69, n = 18, p< 0.01).
Longer activity periods led to more time spent resting. Variations between individuals also
occurred. Compared to other vixens, and whatever the season was, F12 rested much longer
during the activity period (Mann-Whitney U test, p< 0.05) (Tab. 2).

No correlation was found between foxes’ effective activity (EA) and the length of
activity period. Individual EA did not vary significantly throughout the year (Mann-
Whitney U test, p>0.05) (Tab.2), but in some cases (e.g. F19) a 90 minutes difference
could occur between the shortest and the longest EA.

Fox active period was generally interrupted by several inactive bouts. The average
number of such breaks per night did not differ from one vixen to another, except F12 who
used to take more breaks during her active period than F2, F8, F11 and F19 respectively
(Student t-test, p<0.05) (Tab. 3). The distribution of these breaks throughout the activity
period did not follow any individual routine as also was observed in urban foxes
(WOOLLARD and Harrıs 1990). Their varıable duration was classified into five groups
(Tab. 3). For every fox, most of the inactive bouts were short (< 15 minutes) (Wilcoxon’s
signed-ranks test, p<.0.05), and on average occurred from 1.0 per night (F19) to 2.3 per
night (F12). Intermediate (16-30; 31-45 minutes) and long resting periods (46-60; > 60
minutes) were less frequent. Only F8 and F12 used to rest on average once a night for more
than 60 consecutive minutes, which was not surprising considering the presence of cubs to
look after for F8 and the weak condition of F12. |

Table 3. Mean number of inactive bouts (IB) per night (X) according to their duration

N: Total number of inactive bouts per fox, n: number of nights

The average duration of inactıve bouts was shorter (except for F12) ın our area than ın
an urban environment. WOOLLARD and Harrıs (1990) recorded inactıve bouts in Bristol
averaging between 45.2 and 55.9 minutes, whereas our estimates ranged from 22.1 to 28.8
minutes (F12: 44.8 minutes). The comparatively higher diversity of potential food in the
urban habitat as well as the smaller sıze of their home range (Harrıs 1980) could lead
urban foxes to meet their daily energetic requirements more easıly than mountain foxes,
and accordingly to increase the duration of their resting periods.

Acknowledgements

We would like to thank Dr. R. Hewson for useful comments on the manuscript and Mr. N. FERRARI
for help with the field work. We acknowledge the support received from Prof. C. MERMmoD. The
study was funded by the Swiss National Fund, grant 31-27766.89/2.

Activity of foxes, Vulpes vulpes, in the Swiss Jura mountains 13

Zusammenfassung

Aktivität des Rotfuchses, Vulpes vulpes, im Schweizer Jura

Die Aktivität von 7 weiblichen, mit Sendern versehenen Rotfüchsen wurde im Gebiet des Schweizer
Jura studiert. Die Rotfüchse waren vor allem nachtaktiv und im allgemeinen azyklisch. Die Aktivitäts-
zeiten waren in den Jahreszeiten unterschiedlich lang, die kürzesten fielen in den Sommer. Die
effektive Aktivität erfuhr dagegen ım Laufe des Jahres keine merklichen Veränderungen. Die
Aktivitätsphasen der Rotfüchse wurden stets durch mehrere Ruhephasen unterbrochen. Die meisten
davon waren kurz (<15 Min.).

References

Asıes, E. D. (1969): Activity studies of red foxes in southern Wisconsin. J. Wildl. Manage. 33,
145-153.

ArTO1s, M. (1985): Utilisation de l’espace et du temps chez le renard (Vulpes vulpes) et le chat forestier
(Felis silvestris) en Lorraine. Gibier Faune Sauvage 3, 33-57.

ArToıs, M. (1989): Le renard roux (Vulpes vulpes Linnaeus, 1758). Encyclopedie des carnıvores de
France No.3. Paris: Societe Frangaise pour l’Etude et la Protection des Mammiferes.

Boyce, C. C. K. (1991): Water vole. In: The Handbook of British Mammals (3rd Ed.). Ed. by G. B.
CoRrBET and $. Harrıs. Oxford: Blackwell Scientific Publications. Pp. 212-218.

CapT, S.; STALDER, H. W. (1988): Untersuchungen zur Habitatnutzung von Rotfüchsen (Vulpes
vulpes L.) im schweizerischen Alpenraum. Ph.D. Thesis. Univ. Bern.

Esuchi, K.; Narazono, T. (1980): Activity studies of Japanese red foxes, Vulpes vulpes japonica
Gray. Jap. J. Ecol. 30, 9-17.

HaRrRrıs, S. (1980): Home ranges and patterns of distribution of foxes (Vulpes vulpes) in an urban area,
as revealed by radio tracking. In: A Handbook on Biotelemetry and Radio tracking. Ed. by C.].
AMLANER and D. W. Macponaıp. Oxford: Pergamon Press. Pp. 685-690.

MEIA, J.-S.; WEBER, J.-M. (1993): Choice of resting sites by female foxes Vulpes vulpes in a
mountainous habitat. Acta Theriol. 38, 81-91.

PrıLLıps, M.; CATLıng, P. C. (1991): Home range and activity patterns of red foxes in Nadgee Nature
Reserve. Wildl. Res. 18, 677-686.

WEBER, J.-M., Augry, S. (1993): Predation by foxes, Vulpes vulpes, on the fossorial form of the water
vole, Arvicola terrestris scherman, ın western Switzerland. J. Zool. (London) 229, 553-559.

WOOLLARD, T.; HARRIS, S. (1990): A behavioural comparison of dispersing and non-dispersing foxes
(Vulpes vulpes) and an evaluation of some dispersal hypotheses. J. Anım. Ecol. 59, 709-722.

Authors’ address: Dr. JEAN-MARC WEBER, JEAN-STEVE MEIA and STEPHANE AUBRY, Institut de
Zoologie, Universite de Neuchätel, Chantemerle 22, CH-2000 Neuchätel 7, Swit-
zerland

Z. Säugetierkunde 59 (1994) 14-20
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Cytogenetic analysis of autosomal polymorphism in
Graomys griseoflavus (Rodentia, Cricetidae)

By A. ZAMBELLI, Lıpra VIDAL-RıoJA, and R. WAINBERG

Instituto Multidisciplinario de Biologia Celular, La Plata, Argentina and Catedra de Biologia General,
Departamento de Ciencias Biolögicas, Facultad de Ciencias Exactas, Universidad Nacional de La
Plata, Argentina

Receipt of Ms. 1. 12. 1992
Acceptance of Ms. 8. 6. 1993

Abstract

South American phyllotine Graomys griseoflavns specimens were collected in eight localities of central
Argentina and cytogenetically analysed. These populations comprised the following karyomorphs: 2n
= 42,41, 38, 37, 36, 35 and 34. These chromosome polymorphisms resulted from Robertsonian fusions
(RFs). A pericentric inversion (Pl) in two different autosomal pairs are described. The numerical
karyotype variability is explained by successive RFs, starting from a karyotype with 2n=42.

Introduction

Graomys griseoflavus (Waterhouse, 1837) is aSouth American phyllotine rodent showing a
high degree of chromosomal polymorphisms. Previously, WAINBERG and FRONZA (1974a,
b) reported preliminary data in one population of Argentina that then was misdetermined
as Phyllotis griseoflavus griseoflavus. Later, according to a revision of the taxa (PEARSON
and Parron 1976) these specimens were recognized as Graomys griseoflavus (REIG pers.
comm.). WAINBERG and FronZzaA (1974a, b) described diploid numbers of 2n = 38, 37 and
36 for specimens collected ın Chasıcö (Buenos Aires province, Argentina) and suggested
RFs as a possible mechanısm for the karyotypic varıability found.

Some reports sustain the ıdea that rodents are in active speciation processes (GREEN-
BAUM et al. 1978; Hsu and ArrıcHı 1968; Reıs 1984). Supporting evidence are
chromosomal polymorphisms observed in many species. In Mus domesticus for instance,
there were defined more than 110 different Robertsonian fusions (RFs). The presence of
several Robertsonian populations ıs regarded as an example of stasipatric speciation (REDI
and CAPAnNA 1988). Cricetid rodents that show high chromosomal varıability have been
considered models for studies on chromosomal rearrangements, speciation and evolution
(Brancnhi et al. 1971, 1979; HooD et al. 1984; NACHMAN and Myers 1989; NACHMAN
1992). In the genus Ehigmodontia ZAMBELLI et al. (1992) described the occurrence of RFs
in a polymorphic system 2n = 34-33-32, where the 2n = 33 (heterokaryomorph) males
showed a trivalent in diakinesis-metaphase I. So far, in all the models described the
contributing acrocentric chromosomes have a random chance to fuse.

In the present study we describe the so far unknown 2n = 42, 41, 35 and 34
karyomorphs of G. griseoflavus and present G-banding analysis that confirms the occur-
rence of RFs and pericentric inversions (Pls). We suggest a hypothesis explaining the
karyotypie varıabilıty by successive RFs that occurred non-randomly.

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5901-0014 $ 02.50/0

Cytogenetic analysis of antosomal polymorphism in Graomys griseoflavus 15

Material and methods

We processed seventy-two anımals collected at the following eight localities of central Argentina
(Fig. 1): Chasicöo (Buenos Aires province); La Carrera (Catamarca province); Divisadero Largo
(Mendoza province); Salıcas and General Belgrano (La Rioja province) and Deän Funes, Laguna Larga
and Santiago Temple (Cördoba province). Table 1 shows the number of specimens caught in each
locality, and the diploid numbers (2n) found.

Bone marrow metaphase spreads were obtained by a modification of the technique of ROTHFELDS
and SıminovicH (1958). To enhance bone marrow cell proliferation all specimens were previously
treated with a yeast suspension (LEE and ELDER 1980). Meiotic preparations from testes were done as
indicated by Evans et al. (1964). Chromosomal G-banding was obtained according to SEABRIGHT
(1971). Homozygous and heterozygous terms were abbreviated Hm and Ht, respectively.

Fig. 1. Map showing the localities where specimens of G. griseoflavns were collected. 1: Chasicö; 2: La
Carrera; 3: Dean Funes; 4: Santiago Temple; 5: Laguna Larga; 6: General Belgrano; 7: Salicas; 8:
Divisadero Largo

16 A. Zambelli, Lidia Vidal-Rioja, and R. Wainberg

Table 1. Numbers of specimens of the different karyomorphs of G. griseoflavus collected
in each locality

Locality

. Chasicö
Bar@arrera

. Dean Funes

. Santiago Temple
. Laguna Larga

. General Belgrano
. Salicas

. Divisadero Largo

1
22
3
4
5
6
7
8

Results
Robertsonian fusions

The 2n = 42 karyomorph comprises twenty pairs of autosomes (Fig. 2a). Chromosomes
1-18 are acrocentric gradually decreasing ın size (large to small). The pair 19 ıs a medıum
sized submetacentric and pair 20 is a small submetacentric. In some anımals chromosome 4
can be submetacentric (see below pericentric inversions). The X ıs a large submetacentric
and the Y ıs a small acrocentrice chromosome.

The 2n = 38 karyomorph shows fourteen pairs of acrocentric and two additional pairs
of large submetacentric autosomes. These features constitute the main difference to the 2n
= 42 karyomorph (Fig. 2b).

The G-banding pattern analysis of 2n = 42 and 2n = 38 karyomorphs allowed us to
conclude that the two large submetacentric pairs observed in 2n = 38 anımals resulted from
RFs between the acrocentric chromosomes 15/17 and 16/18 pairs of 2n = 42 specimens
(RF15-17 and RF16-18, respectively) (Fig.2, Tab. 2).

It was previously proposed that the very large submetacentric chromosome of the 2n =
37 karyomorph was produced by a RF occurring in the 2n = 38 karyomorph. This
assumption was based on morphological chromosomal comparisons between these karyo-
morphs and on the presence of one trıvalent in diakinesis-metaphase I of 2n = 37 males. We
ıdentified by G-band comparison that the RF involves autosomes 1 and 6 (RF1-6) (Figs. 2,
3a, Tab.2). RF1-6 ıs also present as heterozygous in the 2n = 41 karyomorph and as
homozygous in the 2n = 36 karyomorph (Tab. 2, Fig. 3b). Diakinesis-metaphases I of 2n =
36 males showed 17 bivalents plus a symmerrical very large sızed RF1-6 bivalent (Fig. 3e).

In the 2n = 35 karyomorph there was a decrease of two acrocentric and the appearence
of one very large submetacentric chromosome produced which according to G-banding
arose by centric fusion of autosome 2 and 5 (RF2-5) (Tab. 2). Diakinesis-metaphases I of

Table 2. Different Robertsonian fusions found in each karyomorph of G. griseoflavus

RF15-17 RF16-18

(-) = absence of RF.

Cytogenetic analysis of autosomal polymorphism in Graomys griseoflavus 192

\

je)

g

HmPla 5

Ian

79%

19 20 RF 15-17 RF16-18 XX
Fıg. 2. G-banded karyotypes with 2n = 42 (a) and 2n = 38 (b) of G. griseoflavus. HmPI = homozygous
pericentric inversion

2n = 35 specimens showed 15 bivalents, one very large bivalent (homozygous RF1-6) and
one trivalent, which is formed by chromosomes RF2-5, 2 and 5 (Fig. 3f).
In the 2n = 34 karyomorph the RF2-5 was homozygous (Fig. 3c, Tab. 2).

Pericentric inversions

In our survey, we found two autosomal acrocentric pairs that underwent pericentric
inversion (PI), giving rise to submetacentric elements. G-band analysıs showed that this
rearrangement involved chromosomes 4 and 13 (PI4 and PI13, respectively) (Figs. 2a, 3d).

18 A. Zambelli, Lidia Vidal-Rioja, and R. Wainberg

Fig. 3. a: G-banded RF1-6 and its acrocentric homologues (1 and 6 autosomes) from a 2n = 37
karyomorph; b: G-banded RF1-6 chromosome pair from a 2n = 36 karyomorph; c: G-banded RF2-5
chromosome pair from a 2n = 34 karyomorph; d: Heterozygous pericentric inversion of pair 13 from
a 2n = 37 karyomorph; e-f: Diakinesis-metaphasel figures from 2n = 36 (e) and 2n = 35 (f)
karyomorphs. The arrows point to the very large RF-bivalent; the arrow head points to the trivalent

42
Ba HmRF15-17 and HmRF16-18
HtERF1-6 41 37 HtRF1-6

36 HmRF1-6
35 HtRF2-5

34 HmRF2-5

Fig. 4. Sequence of Robertsonian fusions explaining the karyotypic divergence of G. griseoflavus

The PI4 was found in 2n = 42, 38, 37 and 36 karyomorphs. The PI13 was found in 2n =38,
37 and 36 karyomorphs. Within a population both PIs may show a homozygous or hetero-
zygous state or even be absent.

Cytogenetic analysıs of autosomal polymorphism in Graomys griseoflavus 19
Discussion

The phyllotine rodent Graomys griseoflavus exhibits a high degree of chromosomal
polymorphisms produced by two pericentric ınversions and four Robertsonian fusions.
Cytogenetic findings suggest that the latter rearrangements occurred non-randomly and
can be summarized as follows: a. the RF15-17 and RF16-18 (present in 2n = 38-34
complex) were always found ın a homozygous state; b. all 2n = 37-36 individuals were
heterozygous and homozygous respectively tor RF1-6; c. all2n = 35-34 anımals examined
were homozygous for RF1-6 and heterozygous and homozygous respectively, for RF2-5.
These findings suggest a sequencial occurrence of Robertsonian fusions. Thus,
HmRF15-17 and HmRF16-18 (uncertain who was first) were followed by HtRF1-6,
HmRF1-6, HtRF2-5 and HmRF2-5. On these grounds, we propose a sequence of
Robertsonian events that lead to the karyotypic dıvergence observed, with the 2n = 42 as
the common ancestral karyomorph (Fig.4). This assumption agrees with the view of
GARDNER and PATToNn (1976) that karyotypic evolution ın Neotropical cricetids decreases
the chromosomal number vıa Robertsonian fusions.

To explain the karyotype divergence from 2n = 42 to 2n = 38 we assumed the existence
of the 2n = 41, 40 and 39 karyomorphs. At least, the 2n = 41 and 39 individuals should be
heterozygous for RF15-17 or RF16-18. Thus far, in fifty-one wild anımals studied only
individuals homozygous for these Robertsonian fusions were found. Thus, we assumme
that the heterozygous anımals may be not viable or that its frequency is low. To test these
hypotheses we performed experimental crosses to asses the segregation of Robertsonian
chromosomes. In matings between 2n = 42/41, 2n = 38/37, 2n = 38/36 and 2n = 37/37
individuals the Fl and F2 progenies showed normal meiosis and RF15-17, RF16-18 or
RF1-6 segregating in a Mendelıan fashion. On the other hand, when matings between 2n =
42/38 and 2n = 42/37 individuals were tried under the same laboratory breeding condi-
tions, no offsprings were obtained. Probably, the unsuccesful breedings were due to the
inviability of heterozygous RF15-17 and RF16-18 embryos.

There is evidence that Robertsonian fusions ‚may be involved in speciation processes
only when they cause reproductive failure, e.g. in the case of meiosis nondisjunction
(Kına 1987). In Graomys griseoflavus the inviability of heterozygous RF15-17 and/or
RF16-18 products may produce reproductive isolation of 2n = 42-41 and 2n = 38-37-36
(and probably also 2n = 35-34) karyomorphs, while heterozygous RF1-6 and RF2-5 (this
latter always present together wıth HmRF1-6) would not induce reproductive isolation.

According to the remarkable chromosomal polymorphism described in Graomys
griseoflavus we suggest that this species is evolving actively and therefore represents an
interesting model for speciation and chromosomal evolutionary studies.

Acknowledgements

We thank to Dr. NOEMf GARDENAL for providing some specimens included in this work and to
ANIBAL Ramos for the technical assistance. Our appreciation go to Dr. N. ©. BıancHi and Dr. K.
FrREDGA for critical reading and suggestions on this manuscript. This paper was supported by
CONICET grant PIDs 3085300/85 to R. W. and 3066400/88 to L.V.R.

Zusammenfassung

Cytogenetische Analysen von Autosomen-Polymorphismen bei Graomys griseoflavus

(Rodentia, Cricetidae)

Graomys griseoflavus aus Südamerika wurden an acht verschiedenen Orten gefangen und cytogene-
tisch untersucht. In diesen Populationen wurden Karyotypen mit folgenden Chromosomenzahlen
gefunden: 2n = 42, 41, 38, 37, 36, 35 und 34. Diese numerischen Karyotypvariationen werden auf
Robertsonsche Fusionen (RFs) zurückgeführt. Es wurden vier verschiedene RFs beschrieben, die acht
akrocentrische Autosomen als Fusionspartner betreffen. Weiterhin wurden pericentrische Inversionen

20 A. Zambelli, Lidia Vidal-Rioja, and R. Wainberg

an zwei verschiedenen Autosomen festgestellt. Die zahlenmäßigen Karyotypvariationen werden
durch aufeinanderfolgende RFs erklärt, wobei als Ausgangskaryotyp der mit 2n = 42 postuliert wird.

References

BıancHı, N. O.; MERANI, $.; LIZARRALDE, M. (1979): Cytogenetics of the South-American akodon
rodents (Cricetidae). VI. Polymorphism in Akodon dolores (Thomas). Genetica 50, 99-104.

BıancHı, N. O.; Reıc, ©. A.; DuLouT, F. N. (1971): Cytogenetics of the South American akodon
rodents (Cricetidae). I. A progress report of Argentinian and Venezuelan forms. Evolution 25,
12521353:

Evans, E. P.; BRECKTON, G.; FORD, C.E. (1964): An air-drying method for meiotic preparations
from mammalıan testes. Cytogenetics 3, 289-294.

GARDNER, A.L.; PATToNn, J. L. (1976): Karyotypic varıation in oryzomine rodents (Cricetidae) with
comments on chromosomal evolution in the neotropical cricetine complex. Occas. Papers Mus.
Zool., Lousiana State Univ. 49, 1-48.

GREENBAUM, ]. F.; BAKER, R. J.; RamsEy, P. R. (1978): Chromosomal evolution and its implication
concerning the mode of speciation in three species of deer mice of the genus Peromyscus. Evolution
32, 646-654.

Hoop, C. S.; RoBBıns, L. W.; BAkKER, R. ]J.; SHELLHAMMER, H. S. (1984): Chromosomal studies and
evolutionary relationships of an endangered species, Reithrodontomys raviventris. ]. Mammalogy
65, 665-667.

Hsuv, T. C.; ARRIGHI, F. E. (1968): Chromosomes of Peromyscus (Rodentia, Cricetidae). I. Evolutio-
nary trends in 20 species. Cytogenetics 7, 417446.

Kınc, M. (1987): Chromosomal rearrangements, speciation and the theoretical approach. Heredity
591-0.

LEE, M.; ELDER, F. (1980): Yeast stimulation of bone marrow mitosis for cytogenetic investigations.
Cytogenet. Cell Genet. 26, 3640.

NAcHMan, N. W. (1992): Geographic patterns of chromosomal varıation in South American marsh
rats, Holochilus brasiliensis and A. vulpinus. Cytogenet. Cell Genet. 61, 10-16.

NAcCHMan, M. W.; Myers, P. (1989): Exceptional chromosomal mutations in a rodent population are
not strongly underdominant Proc. Natl. Acad. Sci. USA 86, 6666-6670.

PEARsoN, O. P.; PATToN, J. L. (1976): Relationships among South American phyllotine rodents based
on chromosome analysıs. J. Mammalogy 57, 339-350.

Repı, C. A.; CAPANNA, E. (1988): Robertsonian heterozygotes in the house mouse and fate of their
germ cells. in: The Cytogenetics of Mammalıan Autosomal Rearrangements. Ed. by A. DanıEL.
New York: Liss. Pp. 315-359.

Reıc, OÖ. A. (1984): Geographic distribution and evolutionary history of South American muroids
(Cricetidae, Sigmodontinae). Rev. Brasil. Genet. 7, 333-365.

ROTHFELS, K. H.; SmminovicH, L. (1958): An air-drying technique for flattening chromosomes in
mammalıan cells grown in vitro. Stain Technol. 33, 73-77.

SEABRIGHT, M. (1971). A rapid banding technique for human chromosomes. Lancet 2, 971-972.

WAINBERG, R. L.; FRONZA, T. G. (1974a): Autosomic polymorphism in Phyllotis griseoflavus griseo-
flavus Waterhouse, 1837 (Rodentia, Cricetidae). Boll. Zool. 41, 19-24.

— — (1974b): Autosomic polymorphism in the South-American (Argentinean) rodent Phyllotis
griseoflavus griseoflauus Waterhouse, 1837 (Cricetidae) with karyotypes 2n = 38, 37 and 36. II.
The complement 2n = 36. First Intern. Theriol. Congr. Moscow. Vol. 2, 293-294.

ZAMBELLI, A.; DysENCHAUZ, F.; Ramos, A.; DE Rosa, N.; WAINBERG, R.; REIG, OÖ. A. (1992):
Cytogenetics and karyosystematics of phyllotine rodents (Cricetidae, Sigmodontinae). III. New
data on the distribution and variability of karyomorphs of the genus Eligmodontia. Z. Säugetier-
kunde 57, 155-162.

Authors’ addresses: ANDRES ZAMBELLI and Lıpıa VıDAaL-RıoJA, IMBICE, CC403 (1900) La Plata,
Argentina; RICARDO WAINBERG, Cätedra de Biologia General, Departamento de
Ciencias Biolögicas, Facultad de Ciencias Exactas 47 y 115, Universidad Nacional
de La Plata, (1900) La Plata, Argentina

Z. Säugetierkunde 59 (1994) 21-26
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Zur Altersstruktur von Apodemus flavicollis in einem Auwald
an der mittleren Elbe

Von J. HAFERKORN und M. STUBBE

Institut für Zoologie, Martin-Luther-Universität, Halle-Wittenberg, Deutschland

Eingang des Ms. 28. 1. 1993
Annahme des Ms. 21. 9. 1993

Abstract
Age structure of Apodemus flavicollis ın a floodplaın forest in the middle part of the river Elbe

Studied age structure of Apodemus flavicollis in a floodplain forest in the middle part of the river Elbe.
Age was determined using the dry mass of crystalline lenses of the eye. There exists a relationship
between age of animals and mass of lens. Three month old A. flavicollis had an average life expectancy
of 3.8 months. The average age of the population was 6.6 months, however, this changed throughout
the year with a maximum in April of 12.2 months and a minimum of 5 months in summer. The
turnover rate during the course of the year was 86.6 %. Sex ratio was nearly balanced with a slight
female bias.
The results indicate the typical high turnover rate and short life expectancy for small rodents.

Einleitung

Sichere Methoden zur Altersbestimmung bilden die Grundlage für die Erforschung der
Struktur und des Umsatzes von Populationen. Bisher war die Ermittlung des Alters
gefangener Gelbhalsmäuse nur begrenzt möglich. Aufgaben dieses Beitrages sind die
Darstellung einer Methode zur Altersbestimmung gefangener Gelbhalsmäuse und darauf
aufbauend Aussagen zur Altersstruktur.

Untersuchungsgebiet

Mit dem Lödderitzer Forst wurde ein naturnaher Auwald mit Hartholzauenbestockung ausgewählt.
Das Gebiet gehört zum Biosphärenreservat „Mittlere Elbe“ und liegt im Bundesland Sachsen-Anhalt.

Aufgrund des hohen technischen Ausbauungsgrades der Elbe und deren Nebenflüsse trat im
Untersuchungsgebiet letztmalig im Aprıl 1988 Hochwasser auf. In den Totalreservaten werden heute
keine forstwirtschaftlichen Maßnahmen mehr durchgeführt, frühere Eingriffe sind im Bestockungs-
bild noch sichtbar.

In der oberen Baumschicht dominieren Quercus robur, Fraxinus excelsior, Tılia cordata und
partiell anthropogen begründet Populus x canadensis. Eine zweite Baumschicht ist mit Acer campestre,
Carpinus betulus, Tiha cordata und vor allem Ulmus minor ausgebildet. Die ın den meisten
Waldbereichen gut strukturierte Strauchschicht setzt sich aus Crataegus laevigata, Corylus avellana,
Tılıa cordata und Ulmus minor zusammen. Typische Bodenpflanzen sind Anemone nemorosa,
Glechoma hederacea, Stellarıa holostea, Carex brizoides und Impatiens parviflora.

Material und Methode

Die Untersuchungen wurden von August 1989 bis Oktober 1991, mit 21 über den gesamten
Jahresverlauf verteilten Fangperioden, durchgeführt. Das Stellen der Schlagfallen erfolgte nach der
Fangquadratmethode von SykorA (1978). Die Fallen wurden täglıch kontrolliert und nachbeködert,
als Köder kamen Brot und Speck zum Einsatz.

Bei allen gefangenen Tieren (n = 268) wurde der Schädel freipräpariert. Unter dem Stereomikro-

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5901-0021 $ 02.50/0

DR J. Haferkorn und M. Stubbe

skop erfolgte bei ca. 25-facher Vergrößerung eine Einstufung in sechs Zahnabnutzungsklassen nach
STEINER (1968). Ferner wurden die Bulbi beider Augen herauspräpariert und mit einem Skalpell
angeschnitten. Unter leichtem Druck kamen Glaskörper und Augenlinse hervor, deren Trennung
durch Drehungen und leichten Druck erfolgte. Die Augenlinsen wurden 24 Stunden in 10%igem
Formalin fixiert und in 4%igem Formalin ca. 14 Tage gelagert, nach der Methode von Morkıs (1972).
Die Trocknung erfolgte im Exikator auf Blaugel im Trockenschrank bei ca. 80°C über 48 Stunden.
Beide Augenlinsen eines Individuums wurden paarweise auf 0,1 mg genau gewogen. Die Wägungen
wurden in Abständen zweimal wiederholt, die Daten dieser drei Messungen wurden gemittelt.

Über die Relation der Linsenmasse mit dem Alter der Tiere wurde eine Eichkurve erstellt, die zur
Alterseinstufung des Fangmaterials diente. Tiere bekannten Alters wurden in Standardversuchstierkä-
figen aus PVC-Schalen und Gitteraufsätzen mit den Maßen 50 x 30 x 19 cm auf Spänen gehalten. Als
Unterschlupf dienten umgestülpte Blumentöpfe mit ausgebrochenen Rändern. Wasser und Mäuse-
bzw. Rattenpellets standen ad libitum zur Verfügung. Dreimal wöchentlich bekamen die Tiere
Frischfutter, meist Apfel. Die Haltung erfolgte bei einer Temperatur von ca. 20°C unter einem
konstanten Langtaglichtregime (L:D = 16:8). Insgesamt wurden Linsenpaare von Tieren mit einem
Alter von 11 Wochen (n = 22), 4 Monaten (n = 12), 6 Monaten (n = 10), 8 Monaten (n = 7), 10
Monaten (n = 10), 16 Monaten (n = 9) und 18 Monaten (n = 3) verwendet. Statistische Unterschiede
wurden mit dem parameterfreien, univarıaten MAnn WHITNEY U-Test geprüft.

Ergebnisse

Bei A. flavicollis wurde eine funktionale Abhängigkeit der Augenlinsenmasse vom Alter
der Tiere registriert (Abb. 1). Die Gleichung der Regression auf y lautet y = 10,786 In x —
7,638 (r = 0,989). Die Augenlinsentrockenmassen der Tiere, getrennt in Zahnabnutzungs-
klassen nach STEINER (1968), unterschieden sich in jedem Falle signifikant (I-VI: U =
4,774; 8,147; 9,74; 6,625; & = 0,0001/U = 3,102; « = 0,005), obwohl sıch die Werte
besonders in den höheren Zahnabnutzungsklassen beträchtlich überschnitten (Tab. 1).

Einen Überblick über die Altersstruktur vermittelt die Lebenstafel (Tab.2). In ihr
wurden Tiere ab einem Alter von drei Monaten berücksichtigt, da Jungtiere in Freilandfän-
gen ın ihrer Nestlingsperiode und kurz danach unterrepräsentiert erfaßt werden. Die
mittlere Lebenserwartung stieg vom dritten zum fünften Lebensmonat und sank dann
mehr oder weniger kontinuierlich mit zunehmendem Alter. Jungtiere mit einem Alter von
drei Monaten hatten eine mittlere Lebenserwartung von 3,8 Monaten.

Augenlinsentrockenmasse in mg

257

205

1192

105

[6) T T T T T 1 1 | l | (ui T I | T 1 —
1 Dee re Ey ee Sr ee ee lee l8
Alter in Monaten

Abb. 1. Augenlinsentrockenmassen in Abhängigkeit vom Alter der Tiere

Zur Altersstruktur von Apodemus flavicollis in einem Auwald 23

Alter in Monaten

(0) I I I 1 T I I T I

I Ir Izl
ASONDJFMAMJJASONDJFMAM)J JASO

Abb. 2. Das mittlere Alter aller gefangenen A. flavicollis im Auwald (Balken gibt Standardabweichung
an)

40 %

30 %+:

20 % 7

10290,

HL A |

12STD 1702=18
Alter in Monaten

77 Männchen L|__| Weibchen

Abb. 3. Der prozentuale Anteil einzelner Altersklassen am Populationsaufbau

Im gesamten Fangmaterial betrug das durchschnittliche Alter aller Tiere 6,6 Monate.
Vom Spätherbst bis zum Beginn der neuen Reproduktionsperiode stieg das mittlere Alter
der Population mit Maximalwerten im Aprıl 1990 von 13,4 Monaten bzw. im April 1991
von 11,9 Monaten (Abb.2). Minımalwerte des mittleren Alters der Population wurden
jeweils im Sommer mit vier bis sechs Monaten registriert. Auffällig ist das geringe mittlere
Alter von fünf Monaten im November 1990, welches auf sechs gefangenen Individuen
basiert. Der Anteil vorjähriger Tiere in der Population sank bis August auf 17 bis 28 %.
Letztmalig wurden Tiere nach einer Überwinterung im darauffolgenden Jahr im Dezember
gefangen. Der Populationsumsatz fand ım Verlaufe eines Jahres zu 86,6 % statt und ist mit
eineinhalb bis zwei Jahren abgeschlossen.

24 J. Haferkorn und M. Stubbe

Tabelle 1. Mittlere Augenlinsentrockenmassen (x) und mittleres Alter (x) im Verhältnis zu den
Zahnabnutzungsklassen

(s = Standardabweichung, min. = minimales Alter, max. = maximales Alter)

Zahn-

abnutzungsklasse

Tabelle 2. Lebenstafel für Apodemus flavicollis

(I: Zahl der lebenden Tiere bezogen auf 1000
Individuen, d: Zahl der sterbenden Tiere inner-
halb der Altersklasse, q: Sterberate in % inner-
halb der Altersklasse, e: mittlere Lebenserwar-
tung in Monaten für Tiere zu Beginn der Alters-

Augenlinsen-IM

Alter in Monaten
x min.

=
=
=

Insgesamt wurde ein Männchenanteil
von 48% registriert. Das Zahlenverhältnis
der Geschlechter war in den einzelnen Al-
tersklassen mehr oder weniger ausgegli-
chen. Bei Jungtieren mit einem Alter bis zu
drei Monaten deutet sich ein leichter Weib-

klasse) chenüberschuß an (Abb.3).

Alter in
Monaten

Diskussion

Analysen zur Altersstruktur von A. flavı-
collis fehlen, da bisher keine genaue Alters-
bestimmung möglich war. Für die nahe
verwandte Art A. sylvaticus liegen bereits
zwei Untersuchungen vor, die sich mit dem
Zusammenhang der Linsenmassen und dem
realen Alter der Tiere befafßten (GURNELL
und KneE 1984; QUERE und VINCENT
1989). Bei Freilandfängen von A. flavicollis
bestimmten NaBaGLo und PACHINGER
(1979) die Augenlinsentrockenmassen. Da
ihnen das Alter der Tiere nicht bekannt war
und eine Eichkurve fehlte, konnten sie die
Linsenmassen nur mit dem Abkauungsgrad
der Backenzähne vergleichen. Der Abkauungsgrad der Zähne ist im starken Maße abhän-
gig von der Härte der aufgenommenen Nahrung, besonders vom Anteil der Oxalatkristalle
im Pflanzenmaterial. Dadurch nutzen sich die Zähne sehr unterschiedlich ab, beispiels-
weise schwankte das anhand der Augenlinsen ermittelte Alter der Tiere aus der dritten
Zahnabnutzungsklasse nach STEINER (1968) zwischen zwei und neun Monaten. Auf die
Ungenauigkeit von Altersabgrenzungen bei A. flavicollis anhand der Zahnabnutzung
wiesen schon ADAMCZEwSkA (1959) und EıcHsTÄDT (1987) hin.

Lebendfänge von A. flavicollis mit individueller Markierung führten Ranpa et al. (1969)
in Niederösterreich, YLÖNEN et al. (1991) in einem Feldgehölz bei Halle und Huco (1990)
im Nationalpark Berchtesgaden durch. Das Höchstalter markierter A. flavicollis gaben
RanDa et al. (1969) mit 18 Monaten und Huco (1990) mit 22 Monaten an. In der Elbaue
betrug der Anteil von A. flavicollis mit einem Mindestalter von 18 Monaten noch 4,9 %.

Zur Altersstruktur von Apodemus flavicollis in einem Auwald 25

Die mittlere Lebenserwartung von Jungtieren gaben Rappa et al. (1969) zwischen 3,5 und
3,9 Monaten an, für die Elbaue wurde ein ähnlicher Wert errechnet (3,8 Monate).

Entsprechend dem Fortpflanzungsgeschehen schwankte das Durchschnittsalter der
Population im Jahresverlauf mit einem Maximum im Aprıl zu Beginn der Reproduktions-
periode. Die darauffolgende starke Verjüngung wird bedingt durch die Natalität und den
Verlust von Alttieren, da bis zum Frühsommer viele vorjährige Tiere absterben. Auffällig
ist das geringe Durchschnittsalter der Population im November 1990. In dem Jahr
reproduzierte A. flavicollis bis Ende September nach einer geringen Fortpflanzungsrate im
trockenen Hochsommer (HAFERKORN 1992). Möglicherweise verursachten die September-
würfe eine zweite Populationsverjüngung im Spätherbst. Fangbedingte Abweichungen von
realen Verhältnissen können allerdings nie ausgeschlossen werden. ZEjpA (1976) stellte die
Altersgruppierung von A. flavicollis in einem südmährischen Auwald dar und erklärte
Unterschiede in der Altersstruktur zwischen den einzelnen Untersuchungsjahren mit dem
Zusammenspiel von der Dauer und der Intensität der Fortpflanzungsperiode.

FLOWERDEW (1985) beschreibt für die Art dichteabhängige Überlebensraten, die mit
geringeren Abundanzen steigen. Verbunden mit einem hohen Reproduktionsstoß müßten
die Populationen in partiell überfluteten Auwäldern die Möglichkeit zur schnellen Besied-
lung trockenfallender Waldbereiche haben. Für den Lödderitzer Forst konnte diese
Vermutung in den Hochwasserjahren 1987 und 1988 für Clethrionomys glareolus, nicht
aber für Apodemus flavicollis bestätigt werden (HAFERKORN et al. 1991).

Zusammenfassung

Die Altersstruktur von Apodemus flavicollis wurde in einem Auwald an der mittleren Elbe untersucht.
Zwischen dem Alter der Tiere und der Augenlinsentrockenmasse existiert ein positiv korrelativer
Zusammenhang, der zur Altersbestimmung genutzt wurde. Drei Monate alte A. flavicollis hatten eine
mittlere Lebenserwartung von 3,8 Monaten. Das Durchschnittsalter der Population betrug 6,6
Monate und schwankte im Jahresverlauf mit einem Maximum im April von 12,2 Monaten und einem
Minimum von fünf Monaten im Sommer. Der Populationsumsatz im Zeitraum eines Jahres betrug
86,6 %. Das Zahlenverhältnis der Geschlechter war relativ ausgeglichen mit einem leichten Weibchen-
überschuß.

Die Resultate zeigen die für Kleinnager typische hohe Populationsumsatzrate, verbunden mit einer
geringen individuellen Lebenserwartung.

Literatur

ADAMCZEWSKA, K. A. (1959): Untersuchungen über die Variabilität der Gelbhalsmaus, Apodemus
flavicollis flavicollis (Melchior, 1834). Acta theriol. 3, 141-190.

EıcHsTÄDT, H. (1987): Biometrische und ökologische Untersuchungen an der Gelbhalsmaus (Apode-
mus flavicollis) im Nordosten der DDR. Säugetierkdl. Inf. 2, 449-465.

FLOWERDEW, J. R. (1985): The population dynamics of wood mice and yellow-necked mice. In: The
ecology of woodland rodents. Bank voles and wood mice. Ed. by J. R. FLOWERDEW, ]J. GURNELL
and J. H. W. Gıpps. Symp. zool. Soc. London 55, 315-338.

GURNELL, J.; KNEE, C. (1984): Determining the age of wood mice (Apodemus sylvaticus). Zool. Listy
33, 339-348.

HAFERKORN, J. (1992): Populationsökologie von Kleinsäugern in Auwäldern Mitteleuropas. Diss.
Univ. Halle-Wittenberg.

HAFERKORN, J.; STUBBE, M.; HEIDECKE, D. (1991): Sukzession der Kleinsäugergesellschaft in einem
Auwaldbiotop. Populationsökologie von Kleinsäugerarten. Wiss. Beitr. Univ. Halle 1990/34 (P
42), 267-281.

Huco, A. (1990): Wiederfangerfolg bei Kleinsäugern mit einer neuen Markierungsmethode. Z.
Säugetierkunde 55, 421-424.

MoRrrıs, P. (1972): A review of mammalian age determination methods. Mamm. Rev. 2, 69-104.

NABAGLO, L.; PACHINGER, K. (1979): Eye lens weight as an age indicator in yellow-necked mice. Acta
theriol. 24, 119-122.

QUERE, J.-P.; VINCENT, J.-P. (1989): Determination de l’age le mulot gris (Apodemus sylvaticus L.,
1758) par la pesee des cristallins. Mammalia 53, 287-293.

RADDA, A.; PRETZMANN, G.; STEINER, H. M. (1969): Biometrische und ökologische Studien an

26 J. Haferkorn und M. Stubbe

österreichischen Populationen der Gelbhalsmaus (Apodemus flavicollis, Melchior 1834) durch
Markierungsfang. Oecologia 3, 351-373.

STEINER, H. M. (1968): Untersuchungen über die Variabilität und Biometrie der Gattung Apodemus
(Muridae, Mammalia) der Donau-Auen von Stockerau (Niederösterreich). Z. wiss. Zool. 177,
1-96.

SykorRA, W. (1978): Methodische Hinweise zur Kleinsäugerforschung. Abh. Ber. Naturk. Mus.
Mauritianum Altenburg 10, 1-33.

YLÖNEn, H.; ALTNER, H.-].; STUBBE, M. (1991): Seasonal dynamics of small mammals in an isolated
woodlot and its agricultural surroundings. Ann. Zool. Fennici 28, 7-14.

ZejDaA, J. (1976): The small mammal community of a lowland forest. Acta Sc. Nat. Brno 10, 1-39.

Anschriften der Verfasser: Dr. JÖRG HAFERKORN, Umweltforschungszentrum Leipzig-Halle GmbH,
Sektion Biozönoseforschung, Hallesche Str. 44, D-06246 Bad Lauchstädt;
Prof. Dr. MICHAEL STUBBE, Institut für Zoologie Martin-Luther-Univer-
sıtät, Domplatz 4, Postfach Universität, D-06099 Halle (Saale), BRD

Z. Säugetierkunde 59 (1994) 2741
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Island rodents: a new species of Octodon from Isla Mocha, Chile
(Mammalia: Octodontidae)

By R. HUTTERER

Zoologisches Forschungsinstitut und Museum Alexander Koenig, Bonn, Germany

Receipt of Ms. 10. 2. 1993
Acceptance of Ms. 22. 3. 1993

Abstract

A hitherto unknown species of Octodon occurs on Isla Mocha, a small coastal island in the Valdivian
rainforest zone of central Chile. The new Pacific degu (Octodon pacificus n.sp.) exhibits characters
such as soft and long fur, long and poorly tufted tail, homodont upper dentition, barely reduced third
lower molars, broad and asymmertrical teeth with long reentrant folds full of cement, all of which are
considered as plesiomorphic for the genus. It ıs suggested that geographic isolation led to the
preservation of primitive characters. The new species probably represents the sister taxon of Octodon
bridgesü, one of the three mainland species currently known from Chile and Argentina. The
phylogenetic significance of the new degu is discussed.

Introduction

Hystricognath rodents of the family Octodontidae occur in west-central South America,
from where sıx genera (Aconaemys, Octomys, Octodon, Octodontomys, Spalacopus, Tym-
panoctomys) with ten species are known from Bolivia, Chile and Argentina (PEARSON
1984; CONTRERAS et al. 1987; MAREs and OJEDA 1982; GALLARDO and Reıse 1992). The
systematic status and the contents of the family are rather controversial as no striking
synapomorphies are known to characterize the group (GLAnz and ANDERSON 1990); some
authors include the Ctenomyidae (Reıc 1970, 1986; REıG and QUINTANA 1991), others the
Abrocomidae (ELLERMAN 1940).

Confusion also exists at the species level. GALLARDO and REISE (1992) have recently
demonstrated that the genus Aconaemys comprises three species, not one or two, as
previously thought. Three species are generally accepted ın the genus Octodon: O. degus
(Molina, 1782), O. bridgesu (Waterhouse, 1844), and O. lunatus Osgood, 1943 (OsG00D
1943; MAREs and OJEDA 1982; PATTERSoN and FEıGL 1987; Reise and VENEGAS $. 1987;
GALLARDO 1992; REDFORD and EISENBERG 1992). CONTRERAS et al. (1987) questioned the
validity of O. lunatus on morphological grounds but the form has a distinctive karyotype
(SPOTORNO et al. 1988) and certainly ıs a valıd species. However, our knowledge on the
taxonomy, distribution and ecology of this genus ıs far from being complete, as this report
will show. New material already collected in 1959 from a small ısland off central Chile
demonstrates the existence of a fourth species of Octodon. Its characters considerably
enlarge the morphological diversity of the genus. The new degu ıs named and described
below and its significance discussed in the context of the family.

Material and methods

The specimens studied are stored in the collections of the Museum Alexander Koenig, Bonn (ZFMK),
the Staatliches Museum für Naturkunde Stuttgart (SMNS), the Senckenberg-Museum Frankfurt
(SMF), and the collection of W. von KoEnIGswALD, Paleontological Institute, University of Bonn
(VKB). The following specimens were used for illustrations and for comparison: Octodon degus, 9

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5901-0027 $ 02.50/0

28 R. Hutterer

(ZFMK 86.89, 86.701-3, 87.82, 87.361, 88.37; SMNS 1113, 33771), ©. bridgesü, 9 (ZFMK 88.63,
88.83, 92.309, 92.382; SMNS 534; SMF 11037; VKB 972), O. lunatus, 2 (SMNS 42960; SMF 862),
O. sp., 4 (ZFMK 92.383-6), Aconaemys fuscus, 1 (ZFMK 88.59), Spalacopus cyaneus, 1 (ZFMK
92.310). Other sources of information were the descriptions and figures in THomas (1920), ELLERMAN
(1940), Woon (1949, 1974), WALKER et al. (1964), LAnpry (1970), Reıcg (1970), Woops and
BoRARER (1975), GLANZ and AnDERSoN (1990), Nowak (1991), REIG and QuInTAnA (1991), and DE
Sanrıs et al. (1991). The terminology of the skull and the teeth follows Woon and WıLson (1936) and
Woops and HowLanD (1979). All measurements are in millimetres. A note is required about the use
of the name Aconaemys in this report. According to ReEıG (1986) the genus Pithanotomys, based on a
fossil, has priority over Aconaemys. However, evidence which the author announced to be presented
in a forthcoming paper was never published except for the same statement plus one figure of upper and
lower dentition ın Reıg and QuInTanaA (1991). Although the similarity between the molars of the
extant Aconaemys fuscus and the Pliocene Pithanotomys columnaris ıs strıking, the shape and structure
of the skull has not been described. I therefore refrain from following REIG and QuinTana (1991)
until a thorough comparison of both extant and fossil forms has been published.

Results and discussion
Octodon pacificus, new species

Holotype: ZFMK 92.384, skin and skull of an adult female, collected by Francısco BEHN
on 16 January 1959, field number L 6. The skin is in good condition, the cranıum (Fig. 2)
lacks the occipital and the bullae; the mandible is complete.

Paratypes: Skins and skulls of another adult female (ZFMK 92.383; Fig.1) and of two
juveniles (ZFMK 92.385-6), collected between 11 and 24 January 1959 by F. BEHn.

Measurements: See tables 1 to 3.

Type localıty: Isla Mocha (38°22’ S, 73°55’ W), Arauca Province, Chile. Mocha Island is
situated 31 km off the coast; its maximum extension from north to south is 13 km, and
5-7 km from west to east. The centre of the island forms a plateau of about 20 square
kilometres which is almost entirely covered by myrtle forest (Mrs. ErıkA BEHNn, ın litt.).
Two peaks ascend to an altitude of 323 m in the north and 390 m in the south of the island.
A lake ıs on top of the hills. Valleys run down from these peaks to the eastern coast. Large
meadows cover the coastal plains of the island. Dr. BEHN did not specify where he
collected the small mammals but a recent map shows two airstrips on the eastern side of the
island and one of them may have been the meadow where his expedition landed and
camped. BuLLock (1935) described the island as a hilly plateau covered with virgin forest,
almost inacessable except where the ıinhabitants cut tracks into the forest. A short
description of the ısland and its vegetation is also provided by ALMEYDA ARRoYyo (1955).
Translated from Spanish ıt reads: “To the eye of the seaman Mocha Island presents a
beautiful green aspect: the hills up to the peaks and the slopes towards the sea are covered
with large trees, providing easy access to wood.” Isla Mocha is near the northern limit of
the Valdivian rainforest zone, as outlined by OsGooDp (1943). Rocks and sediments of the
island are of Miocene, Pliocene and Quaternary ages (TAvVERA and VEYL 1958).

Diagnosis: Larger and heavier than the three other species of Octodon; fur uniformly dark
brown washed with orange, hairs soft and long; tail long (77% of head and body length)
with inconspicuous tuft. Skull long, particularly the diastema; zygomatic arches wide ın
dorsal and straight in lateral view; zygomatic process of squamosal inserting very high;
superior jugal process forming a characteristic spine. Upper cheek-teeth very broad and
uniform, the long reentrant folds filled with cement; lower cheek-teeth more or less
homodont, their folds running strongly oblique and almost parallel; reentrant folds also
cemented; the third lower molar is large and has the shape of an Arabic numeral 1.

Description: Although Octodon pacıficus n. sp. is larger and heavier than O. bridgesu, O.
lunatus and O. degus (Tab.1), its body appears to be more slender (Fig. 1). The overall

Island rodents: a new species of Octodon from Isla Mocha, Chile 29

colour of the pelage ıs brown-
orange, wıth the orange tips
being brighter on the hairs of
the underside. The dorsal
hairs are 22 mm long on aver-
age and very soft, scattered
long guard hairs being up to
27 mm. The basal 80 % of the
dorsal hairs are plumbous,
their tips are brown-orange.
The head of the anımal is uni-
formly coloured; no pale eye
rings are present. Ihe rela-
tively short dark ears appear
almost naked, bearing only
very fine short hairs. No
white or yellow ear tufts exist.
The snout bears about 30 vib-
rissae, some of which are
black and others white. The
forefeet have 4 digits with
claws and a tiny pollex wıth a
very reduced nail. Their dor-
sal surfaces are covered with
greyish-brown hairs; they
turn to white at the outer edge
of each forefoot. The ventral
surfaces are naked and show a
granulation which is typical
for the genus. Each palmar
surface has three interdigital
and two palmar pads. The
hairs on the dorsal surfaces of
the hindfeet are creamy-
brown. A group of long and
stiff white hairs sit on top of
each digit. Finely granulated
skin covers the space between
the five interdigital and two
plantar pads on the ventral
surfaces of the hindfeet. The
tail is long (77 % of head and
body length), thinly haired,
with the terminal 50 mm
bearing a tuft of slıightly
longer and darker hairs. The
more proximal part of the tail
is brown on the dorsal and

REN

4
.

Fıg.1. Octodon pacıficus n.sp., dorsal and ventral aspect of the
skin of paratype ZFMK 92.383; total length of the specimen is
390 mm

cream-brown on the ventral surface; the inconspicuous terminal tuft appears dark brown.

The cranıum of Octodon pacıficus n.sp. ıs shown in Fig.2. Unfortunately, the bullae
and the occipital have not been preserved, due to the initial preparation of the anımals in
the style of a bird skin. The skulls were subsequently removed from inside the skins and

30

R. Hutterer

Fıg. 2. Lateral view of a cranıum of
Octodon pacıficus n.sp. (based on
the holotype, missing parts of the
shaded area reconstructed), and of a
cranıum of O. degus (below,
adopted from Woops and BoRA-
KER 1975). Note the differences in
size, diastema, infraorbital fora-
men, and the position of the in-
ferıor rım of the zygomatic arch
(arrowed)

pacificus

zps

ip
» =
THAN
a nr
10 mm

bridgesii

ee

lunatus

Fig.3. A comparison of the zygomatic arch in the four species of Octodon. if: infraorbital foramen;
jjp: inferior jugal process; j: jugal; 1: lacrymal;.ljf: lateral jugal fossa; sjp: superior jugal process; ssp:
superior squamosal process; szr: superior zygomatic root; zpm: zygomatic process of maxillary; zps:
zygomatic process of squamosal

Island rodents: a new species of Octodon from Isla Mocha, Chile 31
the skins reworked as standard M?2 Mi P4
study skins in 1992. The skull is Ve
large and broad, wıth a long
diastema, a large infraorbıtal
foramen, and a broad interorbital \\\
constriction. The lateral ridges of
the frontals form a thin, translu-
cent roof which extends further
than in the other species. The ms
measurements for the zygomatic
width of both adult specimens = un
exceed all known measurements I mm
for the other species (Tab.2); the Fig.4. Isolated left upper and lower molars of Octodon
same holds for the length of the pacificus n. sp. (left side) and O. bridgesüi in lingual view
nasals. The tips of the nasals pro-
ject further than in the other species of the genus (Fig. 2). In its general configuration the
skull of Octodon pacıficus n.sp. resembles more that of O. bridgesü. In side view the
zygomatic arch is diagnostic: ıts jugal-maxillary part is very straight, as is the upper rim of
the lateral jugal fossa; the insertion of the zygomatic process of the squamosal is very high,
leaving considerable space between the inferior jugal process and the upper molar row
(Figs. 2, 3). The superior jugal process forms a characteristic spine, less developed in the
other species of the genus. The superior zygomatic root points slightly obliquely in
anterior direction, whereas thıs bone ıs upright or even points posteriorly in the other
species (Fig. 3). The superior squamosal process is poorly developed, much less so than in
O. lunatus and O. degus.

The upper incisors are thick and broad (Tab.2) and their position is opisthodont; the
anterior surfaces are stained orange, and the tips are deeply notched. The upper cheek-
teeth are rootless (Fig. 4) and larger and especially broader than in the other species (Tab. 3,
Figs. 5, 6). Both upper and lower teeth are more uniform in size and shape than in the
other species (Fig. 4). In fact, the upper P4 to M3 show only minor differences in size, not
in shape. All are highly asymmetrical with a heavy paracone and metacone and a long
reentrant fold which is almost completely filled with cement (Fig. 6). The same is true for
the lower cheek-teeth; here the two enamel folds run parallel to each other and strongly
oblique. The third lower molar ıs somewhat simplified but keeps the size and aspect of the

Table 1. External measurements and body mass of samples of Octodon; size data for species other
than pacificus n. sp. taken from REDFORD and EısEnBErG (1992), and body mass from BozınovIc
(1992)

O. pacıficus n. sp. O. bridgesii O. lunatus O. degus
holotype paratype n3—4 n4-6 n64-75

Total length 380 390 323.0 360.0 266.5

Tail length 165
Hindfoot c.u. 42

Ear 20

Body mass (g)

250-370

138.3
102-167

38.5
SA)

22.0
2023

176.1
(n23)

328-382

15729
152-161

40.7
40-42

28.0

1222
(n24)

200-307

U
81-138

SD)
a )

DET,
1931

195.9
(n52)

32

R. Hutterer

Table 2. Cranial measurements of some specimens of Octodon spp.

O. pacificus n. sp."

Greatest length
Zygomatic width
Interorbital width
Nasalia length

Nasalia width

Diastema length

Upper toothrow, crowns
Upper toothrow, alv.
Width P4-P4

Width of both upper I1

O. bridgesüi?

41.8—44.8

O. lunatus O. degus’

A
2

! holo-, paratype. — 6 skulls (ZFMK, SMNS, VKB). - * SMNS 42960. - * ZEMK 86.701.

Table 3. Width of upper (P4-M3) and lower (p4-m3) cheek-teeth of two adult specimens each of
Octodon pacificus n. sp. (holo- and paratype), O. bridgesii, and O. degus

O. pacificus n. sp.

DRSDIWDRSI
DOES
2207,42:62

DIA DEDS
DRG EDDD
DD NY
2,30,,2.45
2.05,62:35

O. bridgesü

2, LI
200229
I, Zoo!
1995 120%
DD 273)
520!
5 Zoll
110, 11074:

Te 5)
IND ONNGDONWVNO O

other molarıform teeth. In occlusal view, this tooth resembles the Arabic numeral 1
(Fig. 6).

The mandible is larger and heavier than in the other species of the genus, as are the
lower incisors. The condyloid process is particularly broad and heavy, corresponding to a
large glenoid fossa of the squamosal, as may be inferred from the development of the
posterior zygomatic root of the cranıum (Fig. 3).

= er

Fig. 5. Occlusal view of upper and lower

molars. Left: Octodon bridgesii (ZEMK 88.63), middle: O.

pacificus n.sp. (ZFMK 92.384), right: O. pacificus n.sp. (ZFMK 92.383); for measurements, see

table 2

Island rodents: a new species of Octodon from Isla Mocha, Chile 33

"Il

III

Fıg.6. Occlusal view of right upper and right lower molars of the four species of Octodon. A: O.
pacıficus n.sp. (holotype), B: O. pacıificus n.sp. (paratype), C: O. bridgesüi (ZFMK 88.63), D: O.
lunatus (SMNS 42960), E: ©. degus (ZFMK 86.701). The shading indicates cement

Notes on juveniles: As mentioned above, two juveniles were collected together with the
two adult females. Their general appearance matches that of the adults, except the fur is
duller. Especially the venter is more grey and not washed with orange. The weights of the
young were 50 and 55 g, respectively, and the upper and lower third molar were not
erupted. Their age may be estimated at two or three weeks. Assuming this age, they would
have been born around the last week of December. Taking a gestation time of 90 days as ın
Octodon degus (WEIR 1974), conception would have occurred in September of the
previous year. This coincides with the reproduction period of Octodon degus near
Santiago, Chile (FuLk 1976).

34 R. Hutterer

Comparisons: The three species of Octodon currently recognized can be distinguished
from O. pacıfıcus n.sp. as follows:

Octodon degus, type species of the genus, is smaller and has a stouter body; it has a
shorter tail with a thick terminal brush, a coarse, agouti-coloured fur, a lighter venter and
white ear tufts. In five specimens examined the pollex bears a small but clearly pointed
claw, which contrasts to the reduced nail of the new species. The statement of OsGooD
(1943) that the species of the genus Octodon bear a nail on the pollex is not always correct.
BEnnETT (1832), when diagnosing the genus Octodon and his new species cumingü (a
synonym of degus), correctly stated that “the thumb of the fore feet has a short obtuse
claw”, not a naıl. Nevertheless, Os6G00D’s statement ıs equally found in standard
references such as WooDs and BoRAKER (1975). It may be possible that the character is
variable but this has to be checked in a larger series. The observed differences between O.
degus (claw) and O. pacıficus n.sp. (nail), however, are very clear and rather suggest the
existence of a good diagnostic character. The skull of O. degus ıs smaller and the position
of the zygomatic arch is lower (Fig. 2). This species has the simplest molars of the genus
(Fig. 6): upper and lower molarıform teeth are fairly symmetrical and approach an 8-shape
more than any other Octodon.

Octodon lunatus ıs similar to O. degus externally and cranıally, although OsGooD
(1943) claimed that his new species was indistinguishable from O. bridgesi. The specimen
at hand matches perfectly the description of OsGooD, particularly with respect to the
lunaritorm third upper molar (Fig. 6). Externally, the specimen differs from degus by a
somewhat softer and longer pelage although the agouti colour is very similar. The
underside of the skin, however, ıs paler and washed with cream-white. Yellow ear tufts are
present. The dorsal surfaces of hands and feet are white. The pollex of the manus bears a
nail, not a claw. The tail ıs long with a prominent terminal brush, dorsally dark brown,
ventrally white on the first half and blackish-brown on the terminal part. Long whitish
guard hairs are present mainly on the posterior part of the fur. The skull ıs sımilar to O.
degus and O. bridgesü, but the M3 is diagnostic (Fig. 6).

Octodon bridgesii ıs similar to O. pacıficns n. sp. in the softer texture of the fur, and ın
its trend towards asymmetry of the teeth (Figs. 5, 6). In addition, the pelage is more
unıform and lacks white or yellow ear tufts. The colour of the body hairs is a mixture of
brown and yellow, not orange. The tail is shorter (75 % of head and body length) but is
similar to pacıficus n.sp. in not having such a prominent terminal tuft (WATERHOUSE 1844;
Mann 1958) as in ©. degus and O. lunatus. Averages for Octodon bridgesii are smaller for
all measurements (Tab. 2). Clear-cut differences to O. pacıficus n.sp. exist in the size and
shape of the skull, the shape and position of the zygomatic arch (Fig. 3), and in the size and
form of the molarıform teeth (Fig. 6), particularly in the upper and lower third molars.

History of discovery: The discovery of this new rodent has been a matter of enthusiastic
research effort and personal tragedy. One person involved is Dr. Francısco BEHN (11. 6.
1910-28. 5. 1976), ftormerly a Professor of Anatomy and Pathology at the University of
Concepciön, Chile, and a free-time ornithologist. Although occupied by his profession he
also spent his free time with the study of birds and such contributed to the ornithology of
Chile (BEHn and MıtLıe 1957). Together with his wife ERIKA BEHN he brought together
an important collection of birds from northern Chile to Antarctica. For many years he
corresponded with Dr. GÜNTHER NIETHAMMER (28. 9. 1908-14. 1. 1974), at that time
curator of birds at the Museum Alexander Koenig, Bonn, whom he knew personally from
a visit to Bonn in 1954, as evidenced by documents in the archives of the Bonn Museum. In
1959 Dr. BEHn and his family made an expedition to Isla Mocha where they camped for
two weeks. On the first of September, 1959, he wrote to G. NIETHAMMER [translated from
German]: “All of January I visited an ornithologically extremely interesting island, which
is situated some kilometres south of Concepciön off the mainland: the so-called Isla

Island rodents: a new species of Octodon from Isla Mocha, Chile 35

Mocha. It has been studied only once before in more detail by a bird watcher, since access
is only possible by aircraft which has to land on the beach or on one of the meadows.”

In his letter BEHN enthusiastically continued to talk about his observations on the
nesting habits of a seabird, Puffinus creatopus, which he planned to write up and publish as
soon as his busy professional life would allow. For unknown reasons he never did so and
all his experience was lost with his death in 1976. It remained also unknown that on his
expedition to Isla Mocha he had collected a small number of mammals which he sent to
G. NIETHAMMER’s then 22 year-old son, JOCHEN NIETHAMMER, who later became
Professor of Zoology at Bonn University. The specimens remained unstudied for more
than thirty years in his private collection and came to light only recently when it was
transferred to and curated at the Museum Alexander Koenig, a final consequence of his
tragic car accıdent which happened on an excursion with students ın July 1991. Severe
injuries terminated the career of this well-known mammalogıst, main editor of the
“Handbook of European Mammals” and author of numerous papers on Palaearctic and
African mammals, but also the discoverer of the extinct giant rat of the Galapagos Islands
(NIETHAMMER 1964).

Etymology: The species is currently known only from a small island in the Pacific Ocean,
hence its specific epithet.

Other mammals: Very little is known on the remaining mammal fauna of Isla Mocha.
PHurıppr (1900) described a new mouse as Mus mochae, which OsGooD (1943) assigned to
Akodon olivaceus (Waterhouse, 1837). Os6ooD (1943) also reported on three other rodent
species obtained by D.S. BuLLock in 1932 on this island. Apart from the new Octodon
and samples of Akodon longıpılis castaneus Osgood, 1943 (ZFMK 92.387-389), and
Akodon olivaceus mochae (Philippi, 1900) (ZFMK 92.390-393), the small collection
obtained by F. BEHn in 1959 contains a specimen of Rattus rattus Linnaeus, 1758 (ZFMK
92.394), a species which has not been reported from this ısland before. Table 4 summarises

Table 4. Mammals recorded from Isla Mocha, based on PrıLıpri (1990), OsGooD (1943), and the
present report

Collectors

PHILıPPI BULLOCK BEHN
(1900) (1932) (1959)

Octodon pacificus n. sp.

Akodon longipilıs
Akodon olivaceus
Geoxus valdivianus
Oryzomys longicaudatus
Rattus rattus

the small mammal species so far known from Isla Mocha. The cricetine rodents listed are
characteristic for the temperate Valdivian rain forest on the neighbouring mainland
(MESERVE et al. 1982). However, the material from Isla Mocha stored at the American
Museum, the British Museum, and the Bonn Museum deserves careful study of its own.
Some subspecies named for populations of Isla Mocha may in fact represent full species.
This seems probable for Akodon longıpilis castaneus, the skull of which differs markedly
from what Reıc (1987) figured as representing the typical mainland population of A.
longipihs.

36 R. Hutterer

Relationships within the genus Octodon

Looking at morphological characters, the four extant species of Octodon fall into two
groups. One includes O. bridgesii and the new species, the other O. degus and O. lunatus.
The first two species share the following characters: uniform colouration, soft pelage,
short ears, long but inconspicuously tufted tail, asymmetrical teeth with a long reentrant
fold. Photographs ot live O. bridgesu and O. degus in REısE and VENEGAs $. (1987) neatly
illustrate the external differences. O. degus and O. lunatus share the more vivid coloura-
tion, hairs of agouti-type, light eye marks, larger and tufted ears, pronounced black tail
tufts, less asymmetrical teeth, and highly reduced third molars. The author regards most of
the characters of the first group as primitive and those of the second group as derived for
the genus.

The supposed polarities are based on an outgroup comparison with fossil Octodontidae
(Woonp 1949; PascuAL 1967; PATTERSON and WooD 1982; REıG and QuINTAnA 1991),
Ctenomyidae (PascuAL et al. 1965; Reıc 1970; VERZI et al. 1991) and Echimyidae
(PATTERSoN and PascuAL 1968; LavocAT 1976), particularly with the Oligocene Platypit-
tamys brachyodon Wood, 1949, which is often taken as an ancestor model for the living
Octodontidae. This assumption is justified because fossıls of Platypittamys share a
characteristic enamel structure with the extant Octodontoidea (MARTIN 1992), but not
with other Caviomorpha. It should, however, be noted that REıG and QuinTanA (1991)
presented a different view of molar evolution in octodontids which will be discussed
below.

An ingroup comparison reveals that Octodon pacıficus n.sp. assembles more plesio-
morphic characters than the other three species. The high position of the zygomatic
process of the squamosal is only found in O. pacificus n.sp. (Fig.2); it ıs shared with
Platypittamys (Woop 1949). Platypittamys and some other fossil octodontids have
strongly asymmetrical and homodont molars (WooD 1949). Within genus Octodon, O.
pacificus n.sp. approaches these conditions more than the other species (Figs. 6, 8). A
reduction of the third molars, as in O. degus and O. lunatus, ıs certainly a derived feature
and may be a general evolutionary trend in octodontoid rodents.

In conclusion, Octodon pacıficus n.sp. may be regarded as the most primitive species of
the genus. Geographically ısolated and ın the absence of similar-sized competitors the
species may have lived on Isla Mocha since the Miocene, the geological age of the island
(TAvErA and VeyL 1958). Some morphological changes, however, must have occurred, as
the large infraorbital foramen and the backward position of the superior zygomatic root
(Fig.3) are derived characters. O. pacıficus n.sp. and O. bridgesü most probably had a
common ancestor; both share a similar morphology and possibly sımilar ecological
requirements. They have the southernmost distributions of the genus (Fig. 7) and seem to
be more restricted to forest (GREER 1968; MESERVE et al. 1982) than O. degus and O.
lunatus which are adapted to life in semiarid shrublands (Woonps and BoRAKER 1975;
CONTRERAS et al. 1987; MESERVE and LE BouLEnGE 1987; Bozınovic 1992).

How does this interpretation fit with the available chromosomal data? O. degus and
O. bridgesii have 58 chromosomes (GALLARDO 1992), while Octodon lunatus has 78
(SPOTORNO et al. 1988) (Tab. 5). GALLARDO (1992) discussed in detail the polarities of the
karyotypes concluding that 58 represents the plesiomorphic condition and the higher
number of 78 a derived condition. This view, which is in contradiction to SPOTORNO etal.
(1988), is in full congruence with the morphological conclusions made above.

Comparisons with other Octodontidae

It is a matter of curiousity that the type genus of the family, Octodon, does not show the
character for which it was named (Fig. 8). 8-shaped teeth are found in Aconaemys (MANN

Island rodents: a new species of Octodon from Isla Mocha, Chile 37,

Fig. 7. Maps of Chile and Argentina with the approximate distributions of the four species of Octodon
indicated; in part adopted from REDFORD and EISENBERG (1992), modified

1958; GLanz and QUINTANA 1991), and Spalacopus (REıc 1970), while Octodon (Figs. 6,
8) and Octodontomys (GLanz and ANDERSON 1990) have asymmertrical teeth; the latter
genus has no reentrant folds at all. If we look agaın at Platypıttamys (Fig. 8), we would
have to take Octodon as the most primitive genus of Octodontidae, with Octodontomys
perhaps as an offshoot of Octodon, and Aconaemys, Octomys, Tympanoctomys, and
Spalacopus as members of a more derived clade. This grouping coincides largely with
THomas (1920), who divided the then known genera in a group with “crescentic” and
another with “8-shaped teeth”. REıG and QuInTanA (1991) argued for the contrary. They
did not mention Platypittamys but stated that 8-shaped molars, as in Aconaemys, are
primitive and asymmertrical ones, as in Octodon, are derived, a view obviously taken from
the Miocene Pseudoplataeomys elongatus which has perfectly 8-shaped molars (REıG and
QuInTAanA 1991). They also described a new genus, Abalosia, from the Pleistocene of
Argentina, which in their interpretation may have been the ancestor of genus Octodon.
While the present author can concur that Pseudoplataeomys may have been ancestral to the
group of octodontids with 8-shaped molars (Aconaemys, Octomys, Tympanoctomys,
Spalacopus), it does not follow that Octodon is derived from the Pleistocene Abalosia. This
genus has very simple molars resembling Octodontomys (Fig. 8) but ıt also has very long
upper third molars which are quite unusual for the group. Also the skull of Abalosia, as
figured by REıG and QuinTanA (1991), ıs not similar to Octodon but instead recalls the
skull of Aconaemys with ıts short nasals, broad interorbital region, parallel molar rows,
and stout mandible. The phylogenetic position of ABarosıa should thus be regarded as
uncertain and its postulated relation to Octodon cannot be accepted. If we include the
Oligocene Platypittamys in this comparison then again Octodon would group next to it
(Fig. 8). The Pliocene Chasicomys octodontiforme Pascual, 1967 shows a somewhat
intermediate morphology of the upper molars (PascuAr 1967) and may be taken as support
for the supposed direction of molar evolution. It can be supposed that different groups of

38 R. Hutterer

A B C D

Fig. 8. Rıght upper molars of octodontid rodents. A: Platypittamys brachyodon (copied from WooD
1949, slightly simplified); B: Octodon pacificus n.sp. (paratype); C: Octodontomys gliroides (copied
from Reıg and QuinTanA 1991, reversed); D: Spalacopus cyaneus (ZFMK 92.310); E: Tympanoc-
tomys barrerae (copied from ReEıG and QuINTanA 1991, reversed); F: Aconaemys fuscus (ZFMK
88.59); G: Octomys mimax (copied from ELLERMAN 1940, modified). Not to scale, tooth rows
brought to approximately the same length for the purpose of comparison

octodontids already diverged in the Miocene, and that their phylogeny is at present
obscured by an incomplete fossil record. VERZI et al. (1991) also concluded from their
work on fossil Octodontoidea that octodontine and ctenomyine rodents radiated contem-
poraneously in the Miocene.

lf we neglect the molars and look only at the skull, then Tympanoctomys (LANDRY
1957; DE Santıs 1991) would be the candidate for the most primitive genus because of its
small infraorbital foramen and the extremelv anterior position of the superior zygomatic
root, which is shared by Platypittamys but

not by any of the other five genera. On the
other hand, Tympanoctomys and Octomys
(THomas 1920; WALKER et al. 1964) both
have hypertrophied auditory bullae, cer-
tainly derived structures related to their
gerbil-like habitus.

The chromosome complements, which
surprisingly are known for all species ex-
cept Octodon pacificus n.sp., present a
slightly different picture (Tab.5).
Aconaemys, Octomys, Spalacopus, and Oc-
todon have very sımilar karyotypes, whilst
Octodontomys has a lower (38) and Tym-
panoctomys a much higher (102) chromo-
some number. The latter two genera may
be taken as derived in this character.

None of the different data sets provides

Table 5. Chromosome numbers of the species

of Octodontidae; adopted from GALLARDO

(1992), GALLARDO and Reıse (1992), and refe-
rences cited therein

Species

Aconaemys fuscus
Aconaemys porteri
Aconaemys sageı
Octomys mıimax

Spalacopus cyaneus
Octodon degus

Octodon bridgesii
Octodon lunatus
Octodontomys gliroides
Tympanoctomys barrerae

a convincing solution of the phylogenetic relations between the six extant genera of
Octodontidae, nor does the work based on blood protein similarities of only three genera
(Woops 1982). At present it seems impossible to solve the phylogenetic relationships
within the Octodontidae with the given information. One reason may be that the sıx extant
genera represent only a small portion of the past diversity, a view held also by GALLARDO
(1992) and suggested by the fossils which are already known (Mones 1986; ReıG and
QuınTanaA 1991) although many more fossils may be expected. Nevertheless, study of the
fossil record together with the extant species allows the formulatıon of hypotheses on

Island rodents: a new species of Octodon from Isla Mocha, Chile 39

evolutionary trends which could be tested in the future against more complete data sets
based on other character complexes.

A note on island rodents and conservation

Since OsGoop’s comprehensive work (1943), the mammal fauna of Chile ıs regarded as
one of the best-studied ın South America (PATTERSON and FEıGL 1987). The latter authors
predicted that “further additions to the faunal list are apt to be those species with highly
restricted geographic ranges, especially those in remote areas”. This is perfectly true in the
case of Isla Mocha, and counts also for two other rodents discovered on islands in southern
Chile: Akodon markhamı (PınE 1973) and Akodon hershkovitzi (PATTERSON et al. 1984).
While those rodents were collected rather recently, the collection of the Pacific degu dates
already from 1959. No attempts have been made since to check the actual status of the new
species. It is not known whether it still forages in the meadows or forest of Mocha Island,
nor, in the positive case, how large the population may be. What seems certain is that the
natural range of the species covers only a few square kılometres and for this reason alone it
must be regarded as vulnerable. Soft fur, orange-brown colour and the long tail suggest
that this degu lives ın the forest, thickets or swamps, which are restricted habitats on the
island. In a report on the endangered mammals of Chile Mırer et al. (1983) classified one
of the mainland species, Octodon bridgesü, as vulnerable. They stated that the range of the
species had been much reduced, presumably by increased cultivation of ıts valley habıtat. If
this is true for a rather widespread species, then such threats may apply even more to a
small ısland population. The sample of Francısco BEHN also shows (Tab.4) that black
rats were already present on Isla Mocha ın 1959. Numerous examples show how fast
endemic mammal faunas on ıslands are destroyed by human occupation, just to mention
the Galapagos (STEADMAN and Ray 1982), the West Indies (MorGan and Woons 1986), or
the Canary Islands (Boye et al. 1992). In the case of Mocha Island, a considerate field
survey is suggested to test whether the Pacific degu still exists. Chilean mammalogists
should feel encouraged to study this aspect and propose conservation measures, if
necessary.

Acknowledgements

I owe gratitude to Mrs. ERIKA BEHN, who sent me valuable information on Isla Mocha and even a
short film taken during the 1959 expedition in which she participated. Dr. F. DIETERLEN, Dr. G.
STORcCH and Prof. Dr. W. von KOENIGSWALD provided important specimens from their collections
for study. Dr. K. Busse shared his knowledge of Chile and procured rare literature, and Mrs. H. von
ISSENDORFF, Dr. G. PETERS and Dr. G. STORCH critically read and improved the manuscript. The
photographs are the work of Mr. H. MEURER. I am grateful to all of them.

Zusammenfassung

Nagetiere auf Inseln: eine nene Octodon-Art von der Isla Mocha, Chile (Mammalıia: Octodontidae)

Auf der 31 km vor der chilenischen Küste gelegenen Insel Mocha lebt ein bisher unbekannter Degu,
der unter dem Namen Octodon pacıificus n.sp. beschrieben wird. Es handelt sich um eine große,
weichhaarige und langschwänzige Art von einheitlich braun-orangener Färbung, der helle Überaugen-
streifen oder Ohrbüschel fehlen. Weiterhin bemerkenswert sind die homodonte Backenbezahnung
und die asymmetrische Form der molarıformen Zähne, deren linguale Schmelzfalten vollständig mit
Zement ausgefüllt sind. Im Vergleich zu fossilen Octodontidae erweisen sich die meisten Merkmale
der neuen Art als plesiomorph; nächstverwandt dürfte Octodon bridgesii sein, während O. degus und
O. lunatus einer anderen abgeleiteten Gruppe zugerechnet werden. Beziehungen der 6 rezenten
Gattungen der Familie untereinander und zu fossilen Vertretern werden diskutiert. Der neue Pazifik-
Degu ist offenbar eine Reliktart, die auf der kleinen Insel Mocha überlebt hat. Das vorhandene
Belegmaterial wurde bereits 1959 von dem chilenischen Arzt Francısco BEHN gesammelt. Der
gegenwärtige Status der Art ist nicht bekannt.

40 R. Hutterer

References

ALMEYDA ARROYO, E. (1955): Geografia de Chile. 16. ed. Santiago de Chile: Impr. Talleres.

BERN, F.; MiLLie, G.R. (1957): Notes. In: Suplemento de Las Aves de Chile. Ed. by J. D. Goopaıı,
A. W. JoHnson, and R. A. PHıLıppı B. Buenos Aires, Argentina.

BENNETT, E. T. (1832): Characters of a new genus of Rodent Mammalia, presented by Mr. Cuming.
Proc. zool. Soc. London, 4648.

BoyE, P.; HUTTERER, R.; LÖPEZ-MARTINEZ, N.; MicHAux, J. (1992): A reconstruction of the Lava
mouse (Malpaisomys insularıs), an extinct rodent of the Canary Islands. Z. Säugetierkunde 57,
29-38.

Bozınovic, F. (1992): Rate of basal metabolism of grazing rodents from different habitats. ].
Mammalogy 73, 379-384.

BuLrock, D. $. (1935): Las aves de la ısla de la Mocha. Rev. Chil. Hist. Nat. 39, 232-253.

CONTRERAS, L. C.; TORRES-MURA, J. C.; YANEZ, J.-L. (1987): Biogeography of octodontid rodents:
an eco-evolutionary hypothesis. Fieldiana Zoology, N. S. 39, 401-411.

ELLERMAN, J. R. (1940): The families and genera of living rodents. Vol. I. Rodents other than
Muridae. London: The Trustees of the British Museum.

FuLk, G. W. (1976): Notes on the activity, reproduction and social behavior of Octodon degus. ].
Mammalogy 57, 495-505.

GALLARDO, M. H. (1992): Karyotypic evolution in octodontid rodents based on C-band analysıs. ].
Mammalogy 73, 89-98.

GALLARDO, M. H.; Reıse, D. (1992): Systematics of Aconaemys (Rodentia, Octodontidae). ].
Mammalogy 73, 779-788.

GLANZ, W. E.; AnDERsoN, $. (1990): Notes on Bolıvian mammals. 7. A new species of Abrocoma
(Rodentia) and relationships of the Abrocomidae. Amer. Mus. Novit. 2991, 1-32.

GREER, J. K. (1968): Mamiferos de la Provincia de Malleco. Publ. Mus. “Dillman $. Bullock”, Angol
12,.1-114.

LAnpry, $. ©. (1957): The relationship of Petromys to the Octodontidae. J. Mammalogy 38, 351-361.

LavocAT, R. (1976): Rongeurs caviomorphes de l’Oligocene de Bolivie II. Rongeurs du Bassın
Deseadien de Salla-Luribay (Collection Hoffstetter). Palaeovertebrata 7, 15-90.

Mann, F. G. (1958): Clave de determinaciön para las especies de Mamiferos Silvestres de Chile. Trab.
Centro Invest. Zool. Univ. Chile 3, 1-38.

MarEs, M. A.; OJEDA, R. A. (1982): Patterns of diversity and adaptation in South American
histricognath rodents. Special Publ. Ser. Pymatuning Lab. Ecol., Univ. Pittsburgh 6, 393432.
Marrın, T. (1992): Schmelzmikrostruktur in den Inzisiven alt- und neuweltlicher hystricognather

Nagetiere. Palaeovertebrata, Mem. extra., 1-168.

MESERVE, P. L.; LE BOULENGE, E. (1987): Population dynamics and ecology of small mammals in the
northern Chilean semiarıd region. Fieldiana Zoology, N. S. 39, 413431.

MESERVE, P. L.; MurVa, R.; LoPETEGu1, N. O.; Rau, J. R. (1982): Observations on the small
mammal fauna of a primary temperate rain forest in southern Chile. J. Mammalogy 63, 315-317.

MILLER, $. D.; ROTTMANN, J.; RAEDERE, K. ]J.; TABER, R. D. (1983): Endangered mammals of Chile:
status and conservation. Biol. Conserv. 25, 335-352.

Montes, A. (1986): Palaeovertebrata sudamericana. Catälogo sistemätico de los vertebrados fösıles de
America del Sur, parte I. Lista preliminar y bibliografia. Courier Forschungsinst. Senckenberg 82,
1-625.

Morcan, G. S.; Woops, C. A. (1986): Extinction and the zoogeography of West Indian land
mammals. Zool. J. Linn. Soc. 28, 167-203.

NIETHAMMER, J. (1964): Contribution & la connaissance des mammiferes terrestres de l’ile Indefatig-
able (= Santa Cruz), Galapagos. Mammalıa 30, 441-445.

Nowak, R. M. (1991): Walker’s mammals of the world. 5th ed. Baltimore: The Johns Hopkins
University Press.

Oscoop, W. H. (1943): The mammals of Chile. Field Mus. Nat. Hist., Zool. Ser. 30, 1-268.

PascvAr, R. (1967): Los roedores Octodontoidea (Caviomorpha) de la formacıön Arroyo Chasicö
(Plioceno inferior) de la Provincia de Buenos Aires. Revista Mus. de la Plata (N.S., Paleontol.) 5,
259-282.

PascuAL, R.; Pısano, J. A.; ORTEGA, E. J. (1965): Un nuevo Octodontidae (Rodentia, Caviomorpha)
de la formacıön Epecu£en (Plioceno medio) de Hidalgo (Provincia de La Pampa). Ameghinıana 4,
19-30.

PATTERsoN, B. D.; Feıcı, C. E. (1987): Faunal representation in museum collections of mammals:
Osgood’s Mammals of Chile. Fieldiana, Zoology, N. S. 39, 485496.

PATTERSoN, B. D.; GALLARDO, M. H.; Freas, K. E. (1984): Systematics of mice of the subgenus
Akodon (Rodentia: Cricetidae) in southern South America, with the description of a new species.
Fieldiana, Zoology, N. S. 23, 1-16.

PATTERSoN, B.; PascuAL, R. (1968): New echimyid rodents from the Oligocene of Patagonia, and a
synopsis of the family. Breviora 301, 1-14.

Island rodents: a new species of Octodon from Isla Mocha, Chile 41

PATTERSON, B; Woop, A. E. (1982): Rodents from the Deseadan Oligocene of Bolivia and the
relationships of the Caviomorpha. Bull. Mus. Comp. Zool. 149, 371-543.

Pıne, R. H. (1973): Una nueva especie de Akodon (Mammalia: Rodentia: Muridae) de la isla de
Wellington, Magallanes, Chile. Anales Inst. Patagonia 4, 423-426.

PHıLıppr, R. A. (1900): Figuras ı descripciones de los murideos de Chile. Anal. Mus. Nac. Chile, Ent.
14a, Zool., 1-70.

REDFORD, K. H.; EISENBERG, J. F. (1992): Mammals of the Neotropics. Vol. 2. The southern cone:
Chile, Argentina, Uruguay, Paraguay. Chicago: The University Chicago Press.

Reıc, O. A. (1970): Ecological notes on the fossorial octodont rodent Spalacopus cyaneus (Molina). ].
Mammalogy 51, 592-601.

— (1986): Diversity patterns and differentiation of high Andean rodents. In: High altitude tropical
biogeography. Ed. by F. VUILLEUMIER and M. MoNASTERIO. New York: Oxford University
Press. Pp. 404-439.

— (1987): An assessment of the systematics and evolution of the Akodontini, with the description of
new fossil species of Akodon (Cricetidae: Sigmodontinae). Fieldiana Zoology, N. S. 39, 347-399.

Reıc, ©. A.; QuUINTANA, C. A. (1991): A new genus of fossil octodontine rodent from the Early
Pleistocene of Argentina. J. Mammalogy 72, 292-299.

Reise, D.; VENEGAS $., W. (1987): Catalogue of records, localities and biotopes from research work
on small mammals in Chile and Argentina. Gayana, Zool. 51, 103-130.

SANTIS, L. J. M. DE; Roıc, V. G.; Justo, E. R. (1991): La anatomıa craneo-dentaria de Tympanoc-
tomys barrerae (Lawrence). Comparacion con Octomys mimax y consideraciones acerca de su
est[u]d[i]o taxonomico (Rodentia: Octodontidae). Neotropica (La Plata) 37, 113-122.

SPOTORNO, A.; WALKER, L.; CONTRERAS, L,; PINCHEIRA, J.; FERNÄNDEZ-DONoso, R. (1988):
Chromosomas ancestrales en Octodontidae y Abrocomidae. Arch. Biol. Med. Exper. 21, 527.

STEADMAN, D. W.; Ray, C. E. (1982): The relationships of Megaoryzomys curioi, an extinct cricetine
rodent (Muroidea: Muridae) from the Galäpagos Islands, Ecuador. Smiths. Contrib. Paleobiol.
51, 1-23.

TAVERA, J.; VEYL, C. (1958): Reconocimiento geolögico de la Isla Mocha. An. Fac. Cienc. Fis. Math.,
Univ. Chile 14-15, 157-188.

THomas, O. (1920): On mammals from near Tinogastra, Catamarca, collected by Mr. R. W. Hendee
near Lake Junin. Ann. Mag. nat. Hist. (9) 17, 313-318.

VErzı, D. H.; MONTALVo, C. 1.; VucETIcH, M. G. (1991): Nuevos restos de Xenodontomys simpsoni
Kraglievich y la sistematica de los mäs antiguos Ctenomyinae (Rodentia, Octodontidae).
Ameghinıana 28, 325-331.

WATERHOUSE, G. R. (1844): On the various skins of Mammalia from Chile, with notes relating to
them by Mr. Bridges. Proc. zool. Soc. London 12,.153-157.

WALKER, E. P.; WARNICK, F.; HAMLET, $. E.; LANGE, K. 1.; Davıs, M. A.; UlBLe, H. E.; WrıcHT, P.
F. (1964): Mammals of the world, vol. 2. Baltimore: The Johns Hopkins Press.

WEIR, B. J. (1974): Reproductive characteristics of hystricomorph rodents. In: The biology of
hystricomorph rodents. Ed. by I. W. Rowranns and B. J. WEIR. Symp. Zool. Soc. London 34.
London: Academic Press. Pp. 265-301.

Woop, A. E. (1949): A new Oligocene rodent genus from Patagonia. Amer. Mus. Novit. 1453, 1-54.

— (1974): The evolution of the Old World and New World hystricomorphs. In: The biology of
hystricomorph rodents. Ed. by I. W. RowLanns and B. J. WEır. Symp. Zool. Soc. London 34.
London: Academic Press. Pp. 21-60.

Woop, A. E., PATTERson, B. (1959): The rodents of the Deseadan Oligocene of Patagonia and the
beginnings of South American rodent evolution. Bull. Mus. comp. Zool. 120, 279-428.

Woop, A. E.; Wırson, R. W. (1936): A suggested nomenclature for the cusps of the cheek-teeth of
rodents. J. Paleont. 10, 388-391.

Woops, C. A. (1982): The history and classification of South American hystricognath rodents:
reflections on the far away and long ago. Special Publ. Ser. Pymatuning Lab. Ecol., Univ.
Pittsburgh 6, 377492.

Woops, C. A.; BORAKER, D. K. (1975): Octodon degus. Mammalıan Species 67, 1-5.

Woops, C. A., HowLann, E. B. (1979): Adaptive radıation of capromyid rodents: anatomy of the
masticatory apparatus. J. Mammalogy 60, 95-116.

Author’s address: Dr. RAINER HUTTERER, Zoologisches Forschungsinstitut und Museum Alexander
Koenig, Adenauerallee 162, D-53113 Bonn 1, Germany

Z. Säugetierkunde 59 (1994) 42-51
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Allozyme divergence and systematics of Common mole-rats
(Cryptomys, Bathyergidae, Rodentia) from Zambia

By Marıa G. Fırıppuccı, H. BURDA, E. NEvo, and J. Kocka

Department of Biology, University of Roma “Tor Vergata”, Roma, Italy; Department of Morphology,

Johann-Wolfgang-Goethe-University, Frankfurt am Main, Federal Republic Germany; Institute of

Evolution, University of Haifa, Haifa, Israel, and Surgery Department, University Teaching Hospital,
Lusaka, Zambia

Receipt of Ms. 3. 11. 1992
Acceptance of Ms. 28. 12. 1992

Abstract

Studied allozymic diversity encoded by 34 gene locı in African common mole-rats, Cryptomys sp.,
comprising two populations from Zambia and re-analyzed allozymic diversity encoded by 25 gene
loci in Cryptomys damarensis, C. hottentotus, and C. natalensis from South Africa. This is the first
genetic study of Cryptomys populations originating from outside the South African Subregion. The
dichotomy between C. damarensis on the one hand and C. hottentotus and C. natalensis on the other
hand revealed by previous studies was reconfirmed. Both Zambian populations are specifically distinct
from each other and both are distant from all the South African species. Zambian Cryptomys are much
closer to C. damarensis than to both other species. Interpretation of results of the allozymic study is
corroborated by data on other biological aspects. We show that currently used morphological criteria
for classification of Cryptomys are apparently wrong and that the genus Cryptomys requires urgently a
modern large-scale revision based on allozymes, karyotypes and phenotypic variations.

Introduction

“Asarule, systematic difficulties and doubts arıse largely from a paucity of specimens; but
Cryptomys ıs a genus which by comparison is already fairly abundantly represented in
museums, and every addition so far from making the situation clearer only seems to add to
the confusion” (ROSEVEAR 1969, p. 561).

The family Bathyergidae includes subterranean rodents endemic to Africa. It is agreed
that the systematic study of this famıly may elucidate many aspects of evolutionary biology
like historical biogeography of Africa, the evolution of eusociality, and the classification
and patterns of morphological evolution in rodents (HoNEYCcUTT et al. 1991). The famıly is
currently placed into the suborder Hystricognathi, yet even almost fifty years after
Sımpson’s (1945) statement that “Everyone agrees that (bathyergids) are extraordinarily
isolated among rodents”, a sister-group relationship between the Bathyergidae and any
single lineage within the Hystricognathi could not be established (HoneycuTtT et al. 1991).
The family is divided into five genera, the intergeneric relationships being far from clear
(see HONEYCUTT et al. 1991 for the most recent review).

Particularly interesting is the genus Cryptomys which, from the standpoint of sociobiol-
ogy, is considered by some authors as an intermediate link between solitary bathyergids
and eusocial naked mole-rats (Heterocephalus) (e.g. Jarvıs and BENNETT 1991; LovE-
GROVE 1991). Unlike other bathyergid genera, Cryptomys as a genus is rather eurybiomic
(sensu VRBA 1992), occurring from semi-arid to mesic habitats ın different soil types over a
wide geographical range from Ghana to the Cape (RosEvEAR 1969). Although it is not a
problem to recognize Cryptomys as Cryptomys, extreme variation in many morphological
traits makes taxonomic treatment of this genus very difficult. Thus for instance, 44 and 49
species of Cryptomys have been named by Arten (1939) and ELLERMAN (1940), respec-

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5901-0042 $ 02.50/0

Allozyme divergence and systematics of Common mole-rats from Zambia 43

tively. This number has been reduced to three species by later authors (cf. Nowak and
PArADIso 1983). Most recently, seven species have been recognized by HoNEYcUTT et al.
(1991). Different traits like the average size, pelage colouration, white head spots and other
markings, shape and size of the infraorbital foramen as well as some other cranıal
characters have been used for determination and classification by different authors.
However, many authors reported significant polymorphism in body size and pelage
colouration (which ıs age and socıal-dependent in Zambian Cryptomys, Burpa 1989,
1990), and in white markings even within a single colony. These traits thus cannot be used
for species diagnosis ıf only a small sample ıs available. Consequently, ROSEVEAR (1969)
and ALLEN (1978) suggested that, in addition to morphological characters, cytology and
serology should be brought into account before any reliable taxonomic opinion can be
expressed.

Four recent cytologic and genetic studies gave stimuli to revise phylogenetic relation-
ships among bathyergids: Nevo et al. (1986) analyzed karyotype differentiation, NEvo et
al. (1987) allozyme differentiation, HONEYCUTT et al. (1987) mitochondrial DNA restric-
tion-fragment varıation, and HoNnEYcUTT et al. (1991) mitochondrial nucleotide-sequence
variation. With respect to Cryptomys, ıt was demonstrated that two or perhaps even three
distinct species, possibly even fallıng into two genera, can be recognized among the South
African forms: ©. damarensis on the one hand, and ©. (hottentotus) hottentotus and C.
(hottentotus) natalensis on the other hand. However, the use of these data for reconstruc-
tion of the biogeographic history, evolution of socıality, and intrageneric relationships of
Cryptomys ıs limited by the fact that only South African populations were studied, while
the genus is much more widely distributed.

We analyzed allozyme diversity ın two populations of common (small) Cryptomys from
Zambia. The present study thus becomes the first genetic investigation of Cryptomys from
outside the South African Subregion.

Material and methods

Electrophoretic analysıs was carried out on 14 specimens of African common mole-rats, Cryptomys
sp. (Bathyergidae, Rodentia), representing two populations from Zambia, characterized by two
different diploid chromosomal numbers (Burpa et al. 1992). The karyotype 2 n = 68 originated from
Lusaka (localıty Chainda in eastern suburbs of Lusaka, vicinity of the University Campus), the
anımals of the karyotype 2 n = 58 were captured in Itezhi-Tezhi (locality Hot Springs; about 200 km
SWW of Lusaka). These two populations were also compared wıth samples of C. hottentotus, C.
natalensis and C. damarensis from South Africa, involved also in previous analyses by NEvo et al.
(1987).

Tissues of each specimen were preserved in the laboratory at -80 °C until processed. Homogenates
for electrophoresis were obtained from portions of muscle and kidney tissues crushed in distilled
water. Genic varıatıon of structural genes encoding for enzymatic and non-enzymatic proteins was
assessed using standard horizontal starch-gel electrophoresis. All gels were prepared using an 11 %
suspension of Connaught hydrolyzed starch.

Homogenates obtained from muscle were processed for the following enzymatic proteins: &-
Glycerophosphate dehydrogenase (E.C. 1.1.1.8; «-Gpdh), Lactate dehydrogenase (E.C. 1.1.1.27;
Ldh-1 and Ldh-2), Malate dehydrogenase (E.C. 1.1.1.37; Mdh-1 and Mdh-2), Malic enzyme (E.C.
1.1.1.40; Me-1 and Me-2), Isocitrate dehydrogenase (E.C. 1.1.1.42; Idh-1 and Idh-2), 6-Phospho-
gluconate dehydrogenase (E.C. 1.1.1.44; 6-Pgdh), Glucose-6-phosphate dehydrogenase (E.C.
1.1.1.49; G-6-pdh), Indophenol oxidase (E.C. 1.15.1.1; Ipo-1 and Ipo-2), Nucleoside phosphorilase
(E.C. 2.4.2.1; Np), Glutamate-oxalacetate transaminase (E.C. 2.6.2.1; Got-1 and Got-2), Hexokin-
asen (EIG 2 S1H1 EIk 1), @reatıne kinase (EC. 2.73.25Ck), Adenylate kinase (E.@. 27.43; Adk),
Phosphoglucomutase (E.C. 2.5.7.1; Pgm-1 and Pgm-2), Esterases (E.C. 3.1.1.1; Est-1, Est-2 and
Est-3), Acıd phosphatase (E.C. 3.1.3.2; Acph), Aminopeptidase (E.C. 3.4.1.1; Ap-1 and Ap-2),
Adenosine deaminase (E.C. 3.5.4.4; Ada), Fumarase (E.C. 4.2.1.2; Fum), Mannose phosphate
isomerase (E.C. 5.3.1.8; Mpi), and Glucose phosphate isomerase (E.C. 5.3.1.9; Gpi). Homogenates
obtained from kidney were processed for the following enzymatic proteins: Alcohol dehydrogenase
(E.C. 1.1.1.1; Adh), Sorbitol dehydrogenase (E.C. 1.1.1.14; Sdh), and Xanthine dehydrogenase
(E.C. 1.2.3.2.; Xdh).

44 Maria G. Filippucci, H. Burda, E. Nevo, and J. Kocka

The employed procedures were described earlier by NEvo et al. (1987) and FıLıppucct et al.
(1988). Isozymes were numbered in order of decreasing mobility from the most anodal one.
Allozymes were designated numerically according to their mobility, relative to the most common
allele (= 100) (<100 = slower mobility; > 100 = faster mobility) in C. (h.) hottentotus from South
Africa.

Allozymic data were analyzed as genotype frequencies with the BIOSYS-1 program of SvoFFORD
and SELANDER (1981). Intrapopulational genetic variation was estimated by the following genetic
indices: the mean heterozygosity per locus (observed, Ho, and expected, He), the proportion of
polymorphic locı in the population (P 1%: a locus is considered polymorphic if the frequency of the
common allele is not greater than 0.99), and the average number of alleles per locus (A). The amount
of genetic divergence between populations was estimated with the indices of standard genetic identity
(I) and distance (D) proposed by Net (1972).

The two populations from Zambia were analyzed for 34 loci. The comparison with the other South
African species was carried out on 25 shared locı; the following locı were excluded from this analysis:
Shd, Me-2, Ipo-1, Ipo-2, Ck, Adk, Ap-2, Fum, and Est-3.

A dengrogram of the genetic relationships among populations was obtained using the unweighted
pair group cluster analysis UPGMA (SokAL and SNEATH 1963).

Results
Biological divergence

Some of the relevant intrinsic (morphological, reproductive, sociobiological) as well as
extrinsic (ecological) characteristics of Zambian (and South African) Cryptomys and their
habitats are provided and compared in table 1. Representatives of both populations
(Lusaka and Itezhi-Tezhi) possessed eliptical infraorbital foramına (6 skulls of 2n=68,
3 skulls of 2n =58 were examined). Two mixed pairs consisting of females of the karyotype
2n=68 and males of the 2n=58 karyotype were kept for seven months. The anımals mated
normally and regular copulations (observed at least three times a week) were elicited by
constant re-pairing of anımals (cf. BurpA 1989). Both females conceived at least once but
resorbed or aborted the embryos within three to four weeks. Even after seven months
there was no offspring to any of both females. Subsequent pairing with conspecifics of the
same karyotype resulted within few weeks in conception, successfull pregnancy and
delivery, and thus confirmed the normal fertility of the involved anımals. Further
“hybridization” experiments are in progress. They are negative thus far.

Table 1. Some characteristics of Cryptomys and their habitats

Based on Jarvıs and BEnNETT (1991) and further studies of BENNETT and JARVvIS cited therein for
South African taxa; and on Burpa (1989, 1990) for Zambian mole-rats. The given weights refer to
mean weights of grown up breeding anımals

Species Weight (g) Gestation Eyes open Colony Rainfall
(days) (days) size (mm/year)

Lusaka 24 <25 840

Itezhi-Tezhi 58 ? 210) > 800

C. damarensis 74, 78 mp5 200-600
C. hottentotus 54 <14 200-500
C. natalensis 54 ?(2-3) 400-600

Pattern of variation

Twenty-three of the thirty-four locı analyzed were monomorphic and fixed for the same
allele in the two populations from Zambia: Adh, Sdh, Ldh-1, Ldh-2, Mdh-2, Me-1, Me-2,
Idh-1, Idh-2, Ipo-1, Ipo-2, Np, Got-2, Ck, Adk, Pgm-1, Pgm-2, Ap-1, Ap-2, Ada, Fum,

Allozyme divergence and systematics of Common mole-rats from Zambia 45

Table 2. Allelic frequencies observed at the polymorphic and/or discriminant loci for the analyzed
populations of the genus Cryptomys

Number of examined specimens ın parentheses

Alleles Lusaka Itezhi-Tezhi C. hott.
(12) (2) (4)

1.00 1.00 1.00

0.87

0.13

46 Maria G. Filippucci, H. Burda, E. Nevo, and J. Kocka
Table 2 (continued)

Alleles Lusaka Itezhi-Tezhi C. bhott.
(12) (2) (4)

0.05 0.88
0.95 0.12

0.90

0.10

0.88
O2

Mpi, Est-2. The allele frequencies of the polymorphic and/or discriminant locı in the two
populations from Zambia and in C. hottentotus, C. natalensis and C. damarensis from
South Afrıca are given ın table 2. For detailed allele frequencies in South African
populations see NEvo et al. (1987).

The population from Lusaka (2n=68) displayed polymorphism at the following locı:
«Gpdh, G6pdh, Got-1, Hk-1, Pgi, Acph, Xdh, and Est-3.

The population from Itezhi-Tezhi (2n=58) was polymorphic at the following locı:
a«Gpdh, Mdh-1, 6Pgdh, Pgı, and Est-1.

Genetic summary

The mean value of observed heterozygosity, based on 34 locı, for the populations from
Zambia was Ho=0.052 (Ho = 0.045 in Lusaka and Ho = 0.059 in Itezhi-Tezhi). The mean
value of expected heterozygosity was He=0.066 (He=0.054 in Lusaka and He=0.078 ın
Itezhi-Tezhi). The overall mean proportion of polymorphic locı (P1%) for the two
populations was P 1% =0.191 (P 1% =0.253 in Lusaka and P 1% =0.147 in Itezhi-
Tezhi). The overall mean number of alleles per locus was A=1.19 (A=1.24 in Lusaka and
A=1.15 in Itezhi-Tezhi).

Genetic differentiation

Two locı (6Pgdh and Acph) were found discriminant between Lusaka and Itezhi-Tezhiı,
displaying fixation of alternative alleles. Five locı («Gpdh, Mdh-1, Got-1, Pgı, Est-1)
partially discriminated the two populations.

Genetic distance

The values of genetic identity and distance (NEı 1972) between Lusaka and Itezhi-Tezhi
populations were I=0.871 and D=0.138 respectively.

The values of Ner’s genetic identity and distance observed on 25 shared genetic locı
between the three South African Cryptomys species and both Zambian populations are
given in table 3. An UPGMA dendrogram summarizing the genetic relationships found
between the populations studied is given in figure 2. In this comparison, the two
populations from Zambia displayed a higher value of genetic distance (D=0.196). The
populations from Zambia displayed very high values of genetic distance in comparison
with those from South Africa. The values of genetic distance ranged from 0.237 to 0.596.
Both populations showed higher affınity with C. damarensis (D=0.300, ranging from

Allozyme divergence and systematics of Common mole-rats from Zambia 7

Table 3. Values of NeEr’s genetic identity (I; above the diagonal) and distance (D; below the
diagonal) between populations of the genus Cryptomys from Zambia and South Africa, based on
25 loci

Population

1 Lusaka

2 Itzehi-Tezhi
3 C. hottentot.
4 C. natalensis
5 C. damarensis

0.237 in comparison with Lusaka to 0.364 in comparison with Itezhi-Tezhi populations).
Equivalent were instead the mean values observed comparing Cryptomys trom Zambia
with C. hottentotus (D=0.578) and C. natalensıs (D=0.584).

Discussion

According to GoRMAaNn and Renzı (1979), in populations wıth small sample size (Itezhi-
Tezhi), the heterozygosity could change by less than 2.5% as compared with a larger
sample size. The high number of locı analyzed compensates for the small sample size of
some populations. Values of heterozygosity and genetic distances are therefore reliable
with a reasonable margin of precision (SARICH 1977; NEı 1978; GoRMman and Renzı 1979;
SAGE et al. 1986).

The observed values of genetic varıation correspond to the values already observed in
South African species of Cryptomys by NEvo et al. (1987) and are within the range
generally reported for other Rodentia and even that of subterranean rodents which usually
display lower genetic varıation than above-ground rodents (NEvo et al. 1990).

Our findings on relationships of South African Cryptomys corroborate results of
previous studies by HoNEYCUTT et al. (1987, 1991) and NeEvo et al. (1987) in confirming
the dichotomy between C. damarensis and C. hottentotus, the latter taxon splitting again
into two distinct forms: hottentotus and natalensıs.

We suggest that the analyzed Zambian populations comprise two good biological
species which are related yet distinct from each other. The distinction is suggested not only
by a high value of Ner’s D (this study) but also by different karyotypes (BurDA et al.
1992), and (if not absolute then for sure relative) postmating reproductive barrier. Whether
there is also a premating barrier in nature is not clear. The boundary between both species
could not be determined thus far. Using NeEr’s (1975) criteria (based, however, on
presumption of neutrality) (cf. also Nevo et al. 1987), the divergence time (= 5x 10°D)
between the two Zambian populations would be about 700 000 (based on 34 loci) to
1 million years (based on 25 locı).

Based on allozymic data, Zambian Cryptomys are clearly distinct from all South African
taxa. This divergence is confirmed also by different karyotypes (NEvo et al. 1986; BURDA
et al. 1992) and by (combination of) some biological aspects (e.g. regular age-dependent
colour changing in both Zambian Cryptomys Burpa 1989, 1990, which is absent in C.
damarensis, LOVEGROVE pers. comm., and in fact was never noticed in literature dealing
with ontogenetic and social development in South African Cryptomys — cf. BENNETT et al.
1991 and literature cited therein).

Although the examined populations of Zambian Cryptomys for sure represent good
species, distinct from each other and from the South African taxa, it would be preliminary
to describe them formally as new species and to provide them with new specific names.

48 Maria G. Fihippucci, H. Burda, E. Nevo, and J. Kocka

Fig. 1. Map of Zambia. Typical localities of Cryptomys taxa described from (what is now) Zambia
(according to ALLen 1939; AnseıL 1978): A = C. amatus, D = C. damarensis micklemi, M = C.
molyneuxi. Distribution of different species of common Cryptomys across Zambia (according to
HoneEycuTT et al. 1991): 1 = ©. damarensis, 2 = C. bocagei, 3 = C. hottentotus (ssp. amatus). Note,
however, that according to AnseLL (1978), no Cryptomys occur east of the Luangwa river (asterisk).
Localities of Cryptomys described in this paper: I = Itezhi Tezhi (2n = 58), L = Lusaka (2n =68)

LUSAKA
ITEZHI-TEZHI
DAMARENSIS
HOTTENTOTUS
NATALENSIS

«6 =) .4 .3 .2 1 16)

Fig. 2. UPGMA dendrogram summarizing the genetic relationship among populations of the genus
Cryptomys from Zambia and South Africa, based on 25 loci. The cophenetic coefficient is 0.983

Allozyme divergence and systematics of Common mole-rats from Zambia 49

Our restraint is based on the fact that only from (what ıs now) Zambia, three taxa of
common (small) Cryptomys were described (cf. ALLEn 1939): C. amatus, C. damarensis
(micklemi), and C. molyneuxi. DE GRAAFF (1971), AnseL (1974), KınaDon (1974), and
SMITHERS (1983) considered all Zambıan common Cryptomys to be only subspecies of a
single species, Cryptomys hottentotus: C. h. hottentotus, C.h. damarensis (syn. micklemi),
C. h. whytei (syn. occlussus), and C. h. amatus (syn. molyneuxi) (see also Fig. 1). Accord-
ing to HonEvcuTTt et al. (1991), three species of common Cryptomys: C. hottentotus, C.
damarensis, and C. bocagei should occur in Zambia. The Lusaka population should belong
to C. hottentotus amatus, while the mole-rats from Itezhi-Tezhi should represent C.
damarensis (cf. Fig.1). As stated above this was not the case. In order to avoid further
nomenclatoric confusion, typical localities must be revisited and topotypical populations
must be reexamined by further methods.

In this study we omit giant mole-rats (Cryptomys mechowi/mellandi) from Zambia
which are for sure distinct from all other Cryptomys species (cf. also BurpA and KAwALIKA
1992). The animals which we called C. hottentotus in our previous studies belonged to the
Lusaka population (2n = 68).

The systematical classification of all Cryptomys outside the South African Subregion
was based on gross morphological traits only. HonEycUTT et al. (1991) divided the seven
recognized species into two groups according to the size and shape of the infraorbital
foramen. According to the authors, small circular foramına can be found in the central and
western African species (including C. damarensis) while eliptical foramına are typical of
the “hottentotus” group inhabiting southern and eastern Africa (incl. Zambia).

In fact, representatives of both populations (Lusaka and Itezhi-Tezhi) possessed
eliptical infraorbital foramına. However, ROSEVEAR (1969) (and previous authors quoted
by him) and Anseıı (1978) reported significant varıability in this trait and questioned its
usefulness for systematic diagnosis and classification.

In contrast to the classification based on the cranıal traits, our genetic findings show
higher affinity of Zambian Cryptomys to C. damarensıis rather than to C. hottentotus. This
relatedness ıs corroborated also by reproductive (slower pre- and postnatal development)
and sociobiological characters (larger families irrespective of diverse climatic and vegeta-
tional conditions), and by a more pronounced sexual dimorphism, which are sımilar ın C.
damarensis and Zambıan Cryptomys but different from those in C. hottentotus (cf. Tab. 1).
Apparently, the shape of the infraorbital foramen changed independently and parallelly in
diverse lineages. Being most probably neither adaptive nor strictly conservative, this trait
thus cannot be employed to elucidate sister relationships in Cryptomys.

Even only 200 km apart from each other, in comparable habitas, two distinct Cryp-
tomys species were found. Analogously, in Israel across 200 km (however, in different
climatic regions), four chromosomal species (considered good biological species) of Spalax
ehrenbergi occur (e.g. NEvo 1991). We may expect similar speciation in Cryptomys range.
A large-scale analysıs employing karyotypes, allozymes plus other inputs (morphological,
physiological, ecological, behavioural, reproductive, natural hybridization etc.) of the
genus Cryptomys over a broad range of its distribution ıs needed to provide a definite
answer to many interesting questions associated with historical zoogeography, adaptıve
radiation, and evolution of socıality of Cryptomys ın particular (and bathyergids and
subterranean mammals in general).

Acknowledgements

We thank Mr. M. KawaLıka, Mrs. E. Kocka and Mr. S. Krunıc for their assıstance in the field.
E. Nevo thanks to the Israel Discount Bank Chair of Evolutionary Biology and to the Ancell-Teicher
Research Foundation for Genetics and Molecular Evolution for their support. Partly supported by the
Ethologische Gesellschaft e. V.

50 Maria G. Fihppucci, H. Burda, E. Nevo, and J. Kocka

Zusammenfassung

Allozymatische Divergenz und Systematik der Graumulle (Cryptomys, Bathyergidae, Rodentia)
aus Sambia

Allozymatische Divergenz (34 Genloci) wurde bei zwei Populationen von afrikanischen Graumullen
(Cryptomys) aus Sambia untersucht. Diese Untersuchung stellt damit die erste genetische Studie von
Graumullpopulationen dar, die von außerhalb der südafrikanischen Region stammen. Zum Vergleich
wurden parallel Allozyme (25 Loci) bei Cryptomys damarensis, C. hottentotus, C. natalensis aus
Südafrıka neu analysiert. Eine in früheren Studien schon festgestellte Dichotomie zwischen C.
damarensis einerseits und C. hottentotus und C. natalensis andererseits wurde bestätigt. Die zwei
untersuchten sambischen Populationen stellen zwei gute biologische Arten dar, die von allen drei
südafrikanischen Arten spezifisch unterschieden sind. Sambische Cryptomys zeigen eine nähere
Verwandtschaftsbeziehung zu C. damarensis als zu beiden anderen Arten. Die allozymatischen
Befunde und deren Interpretation werden durch den Vergleich mit einigen anderen biologischen
Aspekten bestätigt. Es wird gezeigt, daß die morphologischen Kriterien, die zur Zeit für die
systematische Bewertung von Cryptomys benutzt werden, offensichtlich nicht ausreichend sind und
daß eine moderne, breit angelegte Revision der gesamten Gattung notwendig ist.

References

ALLEn, G.M. (1939): A Checklist of African Mammals. Harvard Coll., Cambridge, Mass. USA:
Bull. Mus. Comp. Zool.

AnseLL, W.F.H. (1978): The mammals of Zambia. Chilanga, Zambia: The National Parks and
Wildlife Service.

BENNETT, N. C.; JARVIS, J. U. M.; AGUILAR, G. H.; McDaip, E. J. (1991): Growth and development
in sıx species of African mole-rats (Rodentia: Bathyergidae). J. Zool. (London) 225, 13-26.

BurpA, H. (1989): Reproductive biology (behaviour, breeding, and postnatal development), in
subterranean mole-rats, Cryptomys hottentotus (Bathyergidae). Z. Säugetierkunde 54, 360-376.

— (1990): Constraints of pregnancy and evolution of socıalıty in mole-rats, with special reference to
reproductive and social patterns in Cryptomys hottentotus (Bathyergidae, Rodentia). Z. zool. Syst.
Evolut.-forsch. 28, 26-39.

Burpa, H.; FıLıppuccı, M. G.; MACHOLAN, M.; NEvo, E.; Zıma, J. (1992): Biological, allozyme, and
karyotype differentiation of African mole-rats (Cryptomys, Bathyergidae) from Zambia. Z.
Säugetierkunde Suppl. 57, 11-12.»

BurDa, H.; Kawauıka, M. (1992): Ecology and behaviour of giant mole-rats, Cryptomys mechowi
(Bathyergidae, Rodentia), from Zambia. Z. Säugetierkunde Suppl. 57, 12-13.

DE GRAAFF, G. (1971): Family Bathyergidae. In: The Mammals of Africa: An Identification Manual.
Ed. by J. MEESTER and H. W. SETZER. Washington: Smithsonian Inst. Press, pp. 1-5.

ELLERMANN, J. R. (1940): The families and genera of living rodents. London: Trustees of the Brit.
Mus. Nat. Hist.

Fırıppuccı, M. G.; Ropıno, E.; NEvo, E.; CAPANNA, E. (1988): Evolutionary genetics and systema-
tics of the garden dormouse, Eliomys Wagner, 1840. 2 - Allozyme diversity and differentiation of
chromosomal races. Boll. Zool. 55, 47-54.

GORMAN, G. C.; REnZIT, J. JR. (1979): Genetic distance and heterozygosity estimates in electrophore-
tic studies: Effects of sample size. Copeia 1979, 242-249.

HoneExcuTT, R.L.; ALLarn, M. W.; Epwaros, $. V.; SCHLITTER, D. A. (1991): Systematics and
evolution of the family Bathyergidae. In: The Biology of the Naked Mole-Rat. Ed. by P.W.
SHERMAN, J. U. M. Jarvıs, and R. D. ALEXANDER. Princeton, New Jersey: Princeton University
Press, pp. 45-65.

HoneEycuTT, R. L.; Epwaros, $. V.; Nerson, K.; NEvo, E. (1987): Mitochondrial DNA varıiation
and the phylogeny of Afrıcan mole rats (Rodentia: Bathyergidae). Syst. Zool. 36, 280-292.

Jarvıs, J. U. M.; BEnnETT, N. C. (1991): Ecology and behavior of the family Bathyergidae. In: The
Biology of the Naked Mole-Rat. Ed. by P. W. SHERMAn, J. U. M. Jarvıs, and R. D. ALEXANDER.
Princeton, New Jersey: Princeton University Press, pp. 66-96.

Kıncpon, J. (1974): East African Mammals. An Atlas of Evolution in Africa. London, New York:
Academic Press. Vol. II, Pt. B.

LOVEGROVE, B. G. (1991): The evolution of eusociality in molerats (Bathyergidae): a question of rısks,
numbers, and costs. Behav. Ecol. Sociobiol. 28, 37-45.

Neı, M. (1972): Genetic distance between populations. Amer. Natur. 106, 283-292.

— (1975): Molecular Population Genetics and Evolution. Amsterdam: North Holland.

— (1978): Molecular Evolutionary Genetics. New York: Columbia University Press.

Nevo, E. (1991): Evolutionary theory and processes of active speciation and adaptive radiation in
subterranean mole rats, Spalax ehrenbergi superspecies in Israel. Evolut. Biol. 25, 1-125.

Nevo, E.; BEn-SHLOMo, R.; BeıLes, A.; Jarvıs, J. U.M.; Hıckman, G.C. (1987): Allozyme

Allozyme divergence and systematics of Common mole-rats from Zambia 51

differentiation and systematics of the endemic subterranean mole rats of South Africa. Biochem.
syst. Ecol. 15, 489-502.

NEvo, E.; CAPANNA, E.; CoRTI, M.; JARVIS, J. U. M.; Hıckman, G. C. (1986): Karyotype differenti-
ation in the endemic subterranean mole rats of South Africa (Rodentia, Bathyergidae).
Z. Säugetierkunde 51, 3649.

NeEvo, E.; Fırippuccı, M. G.; BEILEs, A. (1990): Genetic diversity and its ecological correlates ın
nature: Comparison between subterranean, fossorial and aboveground small mammals. In:
Evolution of Subterranean Mammals at the Organısmal and Molecular Levels. Ed. by E. Nevo
and A. ©. Reıc. New York: Allan R. Liss. pp. 347-366.

Nowak, R. M.; PArADISO, J. L. (1983): Walker’s Mammals of the World. 4th ed. Baltimore, London:
The John Hopkins University Press.

ROoSsEVEAR, R. D. (1969): The rodents of West Afrıca. London: Trustees of the Brit. Mus. Nat. Hist.

SAGE, R. D.; CoNTRERAS, J. R.; Roıc, V.S.; PaTTon, J. L. (1986): Genetic variation in the South
American burrowing rodents of the genus Ctenomys (Rodentia: Ctenomyidae). Z. Säugetierkunde
51, 158-172.

SARICH, V.M. (1977): Rates, sample sizes and the neutrality hypothesis for electrophoresis in
evolutionary studies. Nature 263, 24-28.

Sımpson, G.G. (1945): The principles of classification and a classification of mammals. Bull. Amer.
Mus. Nat. Hist. 85.

SMITHERS, R. H.N. (1983): The mammals of the Southern African Subregion. Pretoria: University of
Pretoria Press.

SokAL, R. R.; SNEATH, P. N. A. (1963): Principles of numerical taxonomy. San Francisco: W.H.
Freeman.

SWOFFORD, D. L.; SELANDER, R. B. (1981): BIOSYS - 1: A Fortran program for the comprehensive
analysis of electrophoretic data in population genetics and systematics. J. Hered. 72, 281-283.

VRBA, E. S. (1992): Mammals as a key to evolutionary history. J. Mammalogy 73, 1-28.

Authors’ addresses: Marıa GRraZIA Fırıppuccı, Department of Biology, University of Roma “Tor
Vergata”, Via E. Carnevale, 1-00173 Roma, Italy; HynEk BURDA, Department
of Morphology, Johann-Wolfgang-Goethe-University, Theodor-Stern-Kai 7,
D-60596 Frankfurt am Main, FRG; EvıaTar NEVvo, Institute of Evolution,
University of Haifa, Mt. Carmel, Haifa 31999, Israel; Jırı KockA, Surgery
Department University Teaching Hospital, P.O. Box 50001, Lusaka, Zambia

Z. Säugetierkunde 59 (1994) 52-57
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Habitat segregation of three sympatric fossorial rodents in the
Spanish Pyrenees

By C. E. BoRGHI, STELLA M. GIANNONI, and J. P. MARTfnEZ-RıcA

Instituto Pirenaico de Ecologia, Jaca, Spain

Receipt of Ms. 16.4.1992
Acceptance of Ms. 2.12.1992

Abstract

Studied the microhabitat selection among three species of sympatric fossorial voles of the subgenus
Terricola above the timberline in the Spanish Pyrenees. The intensity of habitat use in each plot was
estimated by means of the capture data and by counts of mounds made by the fossorial activity of each
species. Environmental features were measured and related to the occupation level of each species by
means of correlations and correspondence analysis. Results reveal some segregation between the three
species in factors such as soil thickness and percentage of plant cover. Microtus pyrenaicus, an endemic
species, prefers deep soils and dense plant cover, while M. duodecimcostatus seems to avoıd plots with
many large rocks, high slopes and prefers dense plant cover, and M. Iusitanicus prefers low plant cover
and shallow soils.

Introduction

It is generally accepted that the diversity of subterranean mammals in a given area ıs low,
and that different species are distributed among particular habitats according to the
distribution of the limiting resources (NEvo 1979). From these premises and from several
considerations of the environmental features of the underground environment, NEvo
(1979) predicts that no substantial overlap occurs between fossorial mammals habiıtats; and
when two or more species coexist, the should have at least different food needs, decreasing
in this way interspecifice competition. Spatial coexistence among fossorial mammals would
be then possible only between herbivores and insectivores (Nevo 1979).

However, instances are known of the coexistence of fossorial mammals of the same
general diet, within the same order, and even the same family. For instance, fossorıal
mammals such as some Chrysochloridae and Bathyergidae may be found together, even in
the same burrows (HickMman 1990).

Many studies have investigated the microhabitat segregation of sympatric “above-
ground” small mammals (e.g. BROwn and LIEBERMANN 1973; M’CLoskEy and FIELDWICK
1975; DuEsER and SHUGART 1978; STAMP and OHMART 1978). However, to our know-
ledge, the only studies concerned with microhabitat segregation of subterranean mammals
are those of REICHMAN and Jarvıs (1989) and COMPARATORE et al. (1992). The main
conclusion of the former studies is that microhabitat segregation mainly depends on the
plant biomass above their burrows in the case of the three species of bathyergids
(REICHMAN and JARvIs 1989) and on the types of soil and vegetation in the case of the two
species of the genus Ctenomys (COMPARATORE et al. 1992). On the other hand, in the
subterranean environment on the mountains above timberline, some other environmental
resources may explain the microhabitat segregation between coexisting sympatric species
of the subgenus Terricola.

The objective of this study was to examine the coexistence of three very similar species
of fossorial rodents, Microtus pyrenaicus, M. Iusitanicus and M. duodecimcostatus. All of
them coexist in the supraforestal level, have similar body size, and are herbivorous species.

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5901-0052 $ 02.50/0

Habitat segregation of three sympatric fossorial rodents 53

Material and methods

Study area

The study plot is located in the Spanish Pyrenees, not far from the small town of Jaca, at an altitude of
2000 m above sea-level (masl), at a site where the three-mentioned species coexist. The area consists of
a plot, 100 m each side on a moderate slope with a grass cover of Trıfolium alpınum, Festuca rubra and
Nardus stricta. The area is covered with snow between November and June of every year.

Habitat selection

The plot was subdivided into 100 squares of 10 x 10. m. Pine voles were captured alive with traps of
the Sherman type and the environmental features were recorded. Data were obtained during the
summers of 1990 and 1991, periods when the study area was free of snow.

The intensity of habitat use in each subplot was estimated by means of direct methods: capture
data, and by indirect methods: counts of mounds made by the fossorial activity of every species.
Indirect methods of investigation are needed because of the difficulty in investigating fossorial rodents
in a complex habitat such as in a soil environment (REICHMAN and Jarvıs 1989). Environmental
varıiables recorded were: soil thickness, slope, plant cover, area covered by rocky outcrops, mean
stone diameter and the percentage of wild boar rooting.

Variables were analysed by means of univarıate and multivariate statistics. Correlation between
habitat varıables and the number of mounds at each sample plot were tested by Spearman’s rank
correlation coefficients (SIEGEL 1986) because distributions of values were not normal. We have
applied a correspondence analysis for the segregation of the different species. Habitat variables were
subdivided into a number of categories. Subdivision of a parameter was determined by arbitrary
evaluation of the width of the ecological gradient existing ın the study area. A contingency table was
created by expressing habıtat varıables as frequencies of occurrence of Microtus species. Calculation
was based on the program SYN-TAX IIV/PC (Popant 1988).

Results
Spearman correlations

Total number of captured voles was 107 (65 Microtus pyrenaicus, 27 M. duodecimcostatus
and 15 M. lusitanicus); 1153 soild mounds were found within the plot; from these, 393
were made by M. pyrenaicus, 640 by M. lusitanicus and 120 by M. duodecimcostatus.
Before any other analysis, we compared the capture level and the number of earth
mounds on each subplot, as a way of calibrating the two estimates of habitat use. Both

Spearman rank correlation coefficients between habitat variables and number of mounds at each
plot

(n = 1153)

Species M. pyrenaicus M. lusitanicus M. duodecimcostatus

rs pP Ts

Number of mounds of M. p. 1.0000 1.0000 -0.2694
Number of mounds of M. I. -0.2694 0.0073** 1.0000
Number of mounds of M. d. 0.0420 0.6760 -0.0733
Average soil depth 0.3527 0.0004*** 0.0695

Maximum soil depth 0.2267 0.0241” -0.0136
Minimum soil depth 090405. 0:00257 20.0757
% cover of exposed rock -0.1685 0.0936 0.0514
Average slope -0.1466 0.1447 0.0948
% cover of vegetation 0.30582.0:.0005 72 0.3679
% cover of wild boar rooting -0.0981 0.3293 0.5309
Average diameter of rocks 0.1452 0.1485 -0.0627

= 5 < 0.0001; *** p < 0.001; ** p< 0.01; *p < 0.05.

54 C. E. Borghi, Stella M. Giannoni, and J. P. Martinez-Rica

variables were highly correlated (r, = 0.48; p< 0.0001; n = 100). The low value of the
Spearman coefficient may be due to the different time scale of both variables: while capture
numbers reveal pine vole activity over a short span of time, the number of soil mounds is a
cumulative variable related to the average habitat use during a long period. The second
variable ıs preferred for habiıtat studies. The results of the correlation between variables are
summarized in the table.

Microtus pyrenaicus

Habitat use of this species ıs positively correlated to plant cover (r, = 0.37; p < 0.0001) and
to soil depth (r, = 0.35; p< 0.0001). Negative but not significant correlation exists
between slope (r, = -0.15; p = 0.14) and percentage of rocks in the subplot (r, = -0.17;
p = 0.09). It appears as if this species selected the best places to burrow, where soil is
deeper and flatter, but, although several varıiables seems linked to this species, not one of
them shows a correlation strong enough to allow prediction of the presence of the species
as a function of only this variable. Therefore, the distribution of Microtus pyrenaicus
should depend on an interaction between the four above-mentioned variables. Influence of
different variables can be clearly seen in the figure. While the species seems to prefer lower
slopes, it ıs found also in steep places, with slopes surpassing 40 %.

Microtus lusitanicus

The only positive correlation shown by this species is to the percentage of wild boar
rooting (r, = 0.53; p< 0.0001). A negative correlations exists between abundance of the
species and plant cover (r, = -0.37; p< 0.01).

Moreover, M. lusitanicus is negatively correlated to the abundance of M. pyrenaicus
(r, = 0.27; p < 0.01). It appears as ıf M. Iusitanicus chose, or was forced to select, marginal
habitats, not preferred by the former species. Association between this species and patches
of wıld boar actıvity ıs striking. Even where these patches do not coincide with active vole
colonies there are signs of an old deserted colony of M. lusitanicus. Wild boar could be a
predator of voles, and seems to prefer this species, due perhaps to accesibility offered by
the more superficial burrows that M. Iusitanicus makes.

M. lusitanicus inhabits sites with shallow soils, low proportion of rocky outcrops, high
slopes and medium to sparse plant cover (Fig.).

Microtus duodecimcostatus

This is a lowland species which in the study area almost reaches its altitudinal limit. It
occurs ın low abundance, and the number of both captures and mounds is small. The only
significant correlations are negative, and rather low: there is a negative relationship
between the abundance of the species and both the surface covered by rocks (r, = -0.25;
p = 0.012) and the average diameter of rock outcrops (r, = -0.25; p < 0.05). As seen in the
figure, the species seems to select soils of intermediate thickness, low presence of rocks,
moderate slope and dense plant cover.

Correspondence analysis

The first axis explains approximately 81 % of the variance, and the first two axes accounted
for the total inertia (100 %). The first axis seems to be linked to plant cover and soil
thickness; the second axis is not so clear as the former one, and seems to be linked to the
percent of rock outcrops and to gravel diameter. Wild boar rooting activity seems also
linked to the first axıs.

The first axis segregates the three species along the variables related with the slope, plant

Habitat segregation of three sympatric fossorial rodents 53

Hm. pyrenaicus [1 M. Iusitanicus®@ M. duodecimcostatus

12 14
10 12
g 10
2 8

6
4

4
2

2
0 0

0-20 21-40 41-62 0-75 76-95 96-100

Mean slope (%) Mean vegetation cover (%)

12 9
8
10
3
8 6
6 5
4
4 3
E 2
1
0 0
0-10 11-20 21-30 31-40 0-25 26-50, . 51-75 76-100
Mean soil depth (cm) Cover of exposed rocks (%)

14

12

0-20 21-70 71-250
Mean diameter of exposed rocks (cm)

Mean number of mounds in each plot, in Los Lecherines (2000 masl) related to the environmental
features recorded

cover and soil depth, pointing to M. pyrenaicus as the species that can live in intermediate
conditions, M. duodecimcostatus being associated with the deeper soils, and the highest
values of plant cover; and M. lusıtanicus associated with the opposite values of this axis.
The second axis segregates M. pyrenaicus from the other species, M. pyrenaicus being the
species that can lıve in plots wıth abundant and sızeable rocks, while the other species are
associated with the opposite values of this axis.

Discussion
Microtus pyrenaicus appears to be rather flexible ın its habitat use, being found at sıtes with

very different features. M. lusitanicus ıs relegated to the thinnest soils and to sites with
sparse plant cover, while M. duodecimcostatus ıs found only ın the opposite microhabiıtats,

56 C. E. Borghi, Stella M. Giannoni, and J. P. Martinez-Rica

associated with high scores of plant cover and soil thickness, and with low slopes. The first
species is also the only one that tolerates the use of rocky terrain.

Habitat segregation in sympatric fossorial mammals does not always rely on the
partitioning of food resources, as implied by NEevo (1979). The unexpected spatial
overlapping among three species of taxonomically related fossorial voles in the suprafores-
tal level of the Spanish Pyrenees, and the three sympatric species of molerats in the Cape
Province of South Africa (REICHMAN and Jarvıs 1989) are a clear contra-example.

Our results suggest that Microtus pyrenaicus selects advantageous sites for burrowing,
where soil is deep and flat. M. duodecimcostatus occurs at sites where competition with
other species is low, or where environmental features are still adequate (thick and flat soil
also), and M. Iusitanicus at sites where the plant cover is sparser and where there are fewer
interactions with M. pyrenaicus. M. lusitanicus ıs strongly associated with the activity of
the wild boar, probably because it is most preyed on by the wild boar, in turn, perhaps due
to its burrowing in the shallowest soils.

The highly complex topographic relief of the mountain environment produces an
increase in diversity of mammalian species, as has already been observed for mammals in
the United States (Stmpson 1964), and particularly the epigeal Microtus species in North
America (Rose and Bırney 1985). The coexistence of three sympatric fossorial rodents ın
the Spanish Pyrenees could, therefore, be the final product of a very complex subterranean
environment in the mountains, where soil depth, plant cover and slopes are very
heterogeneous. The underground environment is very different to that of Spalax ehren-
bergi studied by NEvo (1979) in arıd environments, where the underground habıtat is
structurally simple, and where the competition for food resources may lead to competitive
exclusion. Thus, an increase in complexity of the underground environment at high
altitudes allows the sympatry of the pine voles, while the most important variables
segregating the microhabitat of these species are soil depth and plant cover.

Acknowledgements

C. E. BoRGHI received a research grant within the framework of the CSIC-CONICET agreement
between scientific institutions of Spain and Argentina; $. M. GIAnNONI had a research grant from the
Instituto de Estudios Altoaragoneses, Spain; and the financial support was supplied from the project
“Erosion in deserted fields” of the Spanish CICYT. This study wıll be a portion of a doctoral thesis of
the first author. We would like acknowledge the help received from R. SORIGUER (Estaciön Biolögica
de Donana, Sevilla), R. AntoR, G. DEL BARRIO, D. GöÖMEZ, and J. Isern (Instituto Pirenaico de
Ecologia, Jaca). A. Danson assısted us with the English version. The critical comments of ].
TELLERIA, M. ALCANTARA, $. MONTALVO, J. POTTI and anonymous reviewers helped to improve the
manuscript greatly.

Zusammenfassung

Habitatwahl bei drei Arten von Kurzohrmäusen in den spanischen Pyrenaen

Drei Arten von Kurzohrmäusen, die in den spanischen Pyrenäen oberhalb der Baumgrenze syntop
vorkommen, wurden auf Unterschiede in der Wahl des Lebensraumes hin untersucht. Die Dichte
jeder Art ın den Probeflächen wurde anhand der Mittelwerte von Fängen und der Anzahl aufgeworfe-
ner Hügel abgeschätzt. In den gleichen Probefeldern wurden verschiedene Umweltmerkmale regi-
striert und mit den Vorkommenshäufigkeiten der drei Arten verglichen. Die Ergebnisse zeigen
deutliche Unterschiede, vor allem in den Variablen Bodentiefe und Pflanzenbedeckung. Microtus
pyrenaicus, eine endemische Art der Pyrenäen, bevorzugt tiefen, erdreichen Boden mit hoher
Vegetationsdichte. M. duodecimcostatus scheint in Gebiete mit groben Felsen, stark abschüssigen
Hängen und Gegenden mit hoher Vegetationsdichte auszuweichen, während M. lusitanicus anschei-
nend Flächen mit geringer Vegetationsdichte und flachgründigen Böden aufsucht.

Habitat segregation of three sympatric fossorial rodents 5%

References

Brown, J. H.; LIEBERMAN, G. A. (1973): Resource utilization and coexistence of seed-eating desert
rodents in sand dune habıtats. Ecology 54, 788-797.

COMPARATORE, V. M.; AGnusDEI, M.; BuscH, C. (1992): Habitat relations in sympatric populations
of Ctenomys australis and Ctenomys talarum (Rodentia, Octodontidae) in a natural grassland. Z.
Säugetierkunde 57, 47-55.

Dusser, R. D.; SHUGART, H. H. Jr. (1978): Microhabitats in a forest-floor small mammal fauna.
Ecology 59, 89-98.

HıckMmAn, G. C. (1990): The Chrysochloridae: studies toward a broader perspective of adaptation in
subterranean mammals. In: Evolution of Subterranean Mammals at the Organısmal and Molecular
Levels. Ed. by E. Nevo and O. Reıc. New York: Alan R. Liss. Pp. 23-48.

M’CLoskEy, R. T.; FIELDWICK, B. (1975): Ecological separation of sympatric rodents (Peromyscus and
Microtus). J. Mammalogy 56, 119-129.

Nevo, E. (1979): Adaptative Convergence and Divergence of Subterranean Mammals. Ann. Rev.
Ecol. Syst. 10, 269-308.

PoDant, J. (1988): SYN-TAX III, User’s Manual, Abstracta Botanica 12 (Sup. 1), 1-182.

REICHMAN, O. ]J.; JARVIS, J. U. M. (1989): The influence of three sympatric species of fossorial mole-
rats (Bathyergidae) on vegetation. J. Mammalogy 70, 763-771.

Rose, R. K.; Bırney, E. C. (1985): Community Ecology. In: Biology of New World Microtus. Ed. by
R. H. X. Tammarın. Spec. Publ. No. 8, The American Society of Mammalogists. Pp. 310-339.

SIEGEL, $. (1986): Estadistica no parämetrica. Mexico D. F.: Editorial Trillas.

Sımpson, G. G. (1964): Species density of North American Recent Mammals. Syst. Zool. 13, 57-73.

STAMP, N. E.; OHMART, R. D. (1978): Resource utilization by desert rodents in the lower Sonoran
desert. Ecology 59, 700-707.

Authors’ addresses: CarLos E. BoRGHI, STELLA M. GIannonI, IJADIZA-CRICYT, Casilla de
Correo 507, RA-5500 Mendoza, Argentina, and Juan P. MARTfNEZ-Rıca, Insti-
tuto Pirenaico de Ecologia. Apdo. 64, E-22700 Jaca, Huesca, Spain

Z. Säugetierkunde 59 (1994) 58-60
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

WBSENDSEEFIEIEEIEIENTE ISIN ZMITTEINLUNG

Reaction of a male Stone marten (Martes foina Erxleben, 1777)
to foreign faeces within its territory: a field experiment

By A. SEILER, H. H. KrÜGer and A. FESTETICS

Institut für Wildbiologie und Jagdkunde, Göttingen, Deutschland

Receipt of Ms. 26. 7. 1991
Acceptance of Ms. 29. 7. 1993

The use of faeces and urine for marking territories, revealing sex and reproductive status, is
widespread among carnıvores (GORMAN and TROWBRIDGE 1989; MACDONALD 1980).
Stone martens scent mark their territories by exposing special small droppings and
urinating on conspicuous locations (GRÜNWALD 1988). They are territorial against con-
specifics of the same sex (KRÜGER 1989) and increase their marking activity during the
matıng season (GRÜNWALD 1988). Here, we report on an experiment, where we sımulated
the intrusion of a foreign male stone marten into the territory of a resident male during the
mating season.

Examples of ranging movements of the male stone marten within its home range (convex polygon).
The dotted lines represent movements during the nights 5. (19 June) and 7. (25 June) before; the thick
solid line represents night 1. (26 June) after the exposition of conspecific faeces. The faeces were placed
along the “olfactory corridor” (hatched area). © = start, ®= end of one nıghtly movement

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5901-0058 $ 02.50/0

Reaction of a male Stone marten' to foreign faeces wıthin its terrıtory 59

The resident stone marten was trapped in June 1989 in the village of Hardegsen, near
Göttingen, German Federal Republik (51°57’ N, 9°60’ E). It was fitted with a radio collar
(Karl Wagener, Colonia) and tracked on foot and by bicycle continuously during its active
time for the next five weeks. Tracking data was complimented with direct observations.
Faeces of captive male stone martens (from the “ Arbeitskreis Wildbiologie, Giessen”) were
mixed with water and spread out on a line through the home range. The suspension was
renewed prior to every tracking night. A 25 m broad stripe along thıs line was defined as
the “olfactory corridor”, where the marten was regarded as having direct contact with the
faecal scent (Fig.). We compared the marten’s locomotoric activity (travel distance and
speed) within the corridor before and after placement of the faeces. A total of 14 nights of
observation were used for the analysıs, 7 nights to document movement pattern and home
range sıze and 7 nıghts for the experiment.

During the first period, the male ranged over nearly the whole village and moved often
along the outer limits of his territory (Fig., convex polygon, minimum area = 65.7 ha).
After placement of the faecal suspension, however, he mated with a female and spent a
significantly higher proportion of hıs activity ın the olfactory corridor (+75.4 %, one tailed
Mann-Whitney U-test: U = 39.5, df = 13, p = 0.027, Tab. 1). This reduced his range to the
inner and western part of hıs territory (Fig.), but did not affect his overall travel rate
(Tab.).

During the first nıght after the placement, 26 June, the male encountered the corridor
10 times, moved within it over about 1.6 km, but 4 times he turned around immediately
and ran back. In the second night, 27 June, after leaving his hiding place, he run directly to
the corridor, sniffed the faeces and deposed a scent mark himself, showing the typical
behaviour that GrRÜNDWwALD (1988) described for marking with urine. During the next 65
minutes he followed the corridor intensively over ca. 900 meters and sniffed several times

Travel distance (TD) and travel rate (TR) of a free living male stone marten throughout its
territory and within the olfactory corridor, before and after placement of the foreign conspecific
faeces

Observation time TD in total TD in corridor TR in total TD in corridor
(min.) (m) (m) (m/min.) /total TD

Before
1. 8 June
2. 10 June
3. 12 June
4. 14 June
5219, June
6. 22 June
7, 23 une

Average

After

. 26 June

27, June

. 28 June

. 30 June

2 July

4 July

5 July

Average ; So 1067.9

Nauzuns

* The difference between the periods is significant according to a one tailed Mann-Whitney U-
test: U = 39.5, df = 13, p = 0.027.

60 A. Seiler, H. H. Krüger and A. Festetics

along the outspread suspension. Then he returned to his hiding place and was observed
mating with a female, who probably lived in the western part of his territory. In the third
night, 28 June, the marten was especially active, moved over 8 km and investigated the
olfactory corridor very intensely (Tab.). In the following nights, the marten concentrated
his movements again on the western part of his home range, but encountered the corridor-
more seldom. No mating was observed during this time.

Oestrus females are considered to be the limiting resource for males of solitary
carnıvores (SANDELL 1989). Hence, prior to copulation, resident males should gain more
from keeping close to and defending their receptive females, than from patrolling their
territories. In our experiment, however, the male increased his activity along the olfactory
corridor, although he was mating in the second night of the experiment.

The stone marten was obviously attracted by the suspension of its conspecifics’ faeces.
A similar behaviour was described by GRÜNwALD (1988) with captıve stone martens. Her
animals showed a 11.3-fold increase in their exploratory behaviour and locomotoric
activity, when they encountered scent marks of foreign martens. This was interpreted as
agonistic and curiosity behaviour induced by the olfactory marks. Especially during the
mating season, the scent of a male conspecific within an occupied territory is likely to
initiate aggressive behaviour of the territory owner. At other times of the year, when other
resources are more prevalent, territorial scent marks might be less important (see PuL-
LIAINEN 1982).

References

GORMAN, M. L.; TROWBRIDGE, B. J. (1989): The role of odour in the social lives of carnivores. In:
Carnıvore behaviour, ecology and evolution. Ed. by J. L. GITTLEmAn. London: Chapman and
Hall. Pp. 57-88.

GRÜNWALD, A. (1988): Elemente des kommunikativen Verhaltens bei Baum-(Martes martes) und
Steinmarder (Martes foina) unter Gefangenschaftsbedingungen. Diplomathesis, Univ. Giessen.
KrÜGER, H. H. (1989): Home Ranges and patterns of distribution of stone and pine martens.

Transact. XIXth IUGB Congress, Trondheim, 348-349.

MacDonALD, D. W. (1980): Patterns of scent marking with urine and faeces amongst carnivore
communities. Symp. Zool. Soc. Lond. 45, 107-139.

PULLIAINEN, E. (1982): Scent-marking in the pine marten (Martes martes) ın Finnish Forest Lappland
in winter. Z. Säugetierkunde 47, 91-99.

SANDELL, M. (1989): The mating tactics and spacing pattern of solitary carnıvores. In: Carnivore
behaviour, ecology and evolution. Ed. by J. L. GITTLEMAn. London: Chapman and Hall. Pp.
164-182.

Authors’ addresses: ANDREAS SEILER, Swedish University of Agricultural Sciences, Department of
Wildlife Ecology, Grimsö Wildlife Research Station, S-73091 Rıddarhyttan,
Sweden; Hans-HEINRICH KrÜGer and Prof. Dr. AntaL FESTETICS, Institut für
Wildbiologie und Jagdkunde, Universität Göttingen, Büsgenweg 3, D-37077
Göttingen, Germany

NASEN IEENFDIERSGESIEIEIESGHIENVERT

Einladung

Auf Einladung von Frau Dr. F. SPITZENBERGER, Wien, findet die 68. Jahrestagung der
Deutschen Gesellschaft für Säugetierkunde e.V. von Sonntag, den 25. September, bis
Freitag, den 29. September 1994, im Naturhistorischen Museum ın Wien statt.

Sonntag, 25. September:

Montag, 26. September:

Dienstag, 27. September:

Mittwoch, 28. September:

Donnerstag, 29. September:

Vorläufiges Programm

ab 16.00 Uhr:

ab 19.00 Uhr:

9.00 Uhr:

9.30 Uhr:

14.00 Uhr:
16.00 Uhr:
17.00 Uhr:
19.00 Uhr:

9.00 Uhr:

14.00 Uhr:
16.00 Uhr:
17.00 Uhr:

9.00 Uhr:

13.30 Uhr:
14.15 Uhr:

14.45 Uhr:

9.00 bis

17.00 Uhr

9.00 bis
12.00 Uhr:

Anreise

Vorstandssitzung im Naturhistorischen Mu-
seum, Burgring 7

Zwangloser Begrüßungsabend ım „Gösser
Bräu“

Hörsaal 1 im Biozentrum der Universität:
Grußworte und Eröffnung der Tagung durch
den 1. Vorsitzenden

Hauptvortrag und Kurzvorträge zum The-
menschwerpunkt „Holozäne und spätpleisto-
zäne Säugetierfauna“

Kurzvorträge

Posterdemonstration

Mitgliederversammlung

Empfang im Naturhistorischen Museum

Hörsaal 1 ım Biozentrum der Universität:
Hauptvortrag und Kurzvorträge zum The-
menschwerpunkt „Akustische Kommunika-
tion und Gehör bei Säugetieren“
Kurzvorträge

Posterdemonstration

Gemeinsame Abfahrt zum geselligen Abend
beim Heurigen

Hörsaal 1 ım Biozentrum der Universität:
Hauptvortrag und Kurzvorträge zum The-
menschwerpunkt „Systematik und Ökologie
ost- und südosteuropäischer Säugetiere“
Posterdemonstration

Posterprämierung und Abschluß des wissen-
schaftlichen Programms

Gemeinsame Abfahrt zum Tiergarten Schön-
brunn. Führung und Empfang durch den
Tiergartendirektor Dr. H. PECHLANER.

Exkursion zum Neusiedlerseegebiet. Vogel-
zug und Zieselkolonie, Führung durch die
Biologische Station Illmitz

Kleiner Kurssaal im Naturhistorischen Mu-
seum: Symposium der „Arbeitsgruppe Biber“

62 Mitteilungen

Rahmenprogramm: Demonstration der Trainingsarbeit der Lipizzaner und Besichtigung
der Kaiserappartements.

Alle Interessenten, Mitglieder und Nichtmitglieder, sind zu dieser Jahrestagung 1994 der
Deutschen Gesellschaft für Säugetierkunde in Wien herzlich eingeladen. Falls eine persön-
liche Einladung gewünscht wird, wenden Sie sich bitte an den 1. Vorsitzenden der
Deutschen Gesellschaft für Säugetierkunde, Prof. Dr. U. SCHMIDT, Zoologisches Institut,
Poppelsdorfer Schloß, D-53115 Bonn (Tel. 02 28/73 54 68; Fax 02 28/73 54 58). Das
Programm mit der Vortragstolge wird den Mitgliedern - auf Anforderung auch Nichtmit-
gliedern - rechtzeitig vor der Tagung zugesandt.

Wir bitten um die Anmeldung von Tagungsbeiträgen. Außer Beiträgen zu den genann-
ten Themenschwerpunkten werden dieses Mal wieder verstärkt Kurzvorträge und Poster-
demonstrationen zu anderen Fachgebieten der Säugetierkunde berücksichtigt.

Bitte melden Sie Kurzvorträge (15 Min. + 5 Min. Diskussion) sowie Posterdemonstra-
tionen möglichst frühzeitig, spätestens jedoch bis zum 30. April (Ausschlußfrist) beim
Geschäftsführer der DGS, Prof. Dr. H. ERKERT, Zoologisches Institut, Auf der Morgen-
stelle 28, D-72076 Tübingen (Tel. 070 71/29 2958; Fax: 0 70 71/29 46 34) an. Der
Anmeldung ist eine maximal einseitige informative Kurzfassung (1,5zeilig) beizufügen.
Aus ıhr sollen die Fragestellung, Methoden, Ergebnisse und die daraus gezogenen Schluß-
folgerungen hervorgehen. Alle Kurzfassungen, die wieder in einem Sonderheft der Zeit-
schrift für Säugetierkunde publiziert werden sollen, sind nach dem folgenden, schon im
letztjährıgen Abstractheft eingeführten Schema abzufassen: Deutscher Titel, Leerzeile,
englische Titelübersetzung (kleine Anfangsbuchstaben im Text; bitte ggf. einen „native
speaker“ konsultieren), Leerzeile, Initialen und Familienname(n) des/der Autors(in) bzw.
der Autoren(innen) in Großbuchstaben, Leerzeile, Adresse, Leerzeile, Text (nicht forma-
tiert). Aus arbeitsökonomischen Gründen bitten wir dringend darum, zusätzlich zu
diesem ausgedruckten Abstract möglichst noch eine Fassung auf Diskette (5.25” oder 3.5”,
IBM-kompatibler DOS-PC) in Form eines Word- (5.0 oder 5.5) oder ASCII-Files
mitzuschicken. Die Maße für Poster werden im Juni-Rundschreiben der DGS bekannt-
gegeben.

Mit Fragen zum Tagungsort und zur Organisation wenden Sie sich bitte an Frau Dr.
F. SPITZENBERGER, Naturhistorisches Museum Wien, Postfach 417; A-1014 Wien, Öster-
reich (Tel. 0043/15217. 75 12; Fax: 0043717935254).

Internationales Symposium unter der Schirmherrschaft der DGS über
“Current Problems of Bat Protection in Central and Eastern Europe”

Auf Einladung von Prof. Dr. U. ScHMiDT, Bonn, findet vom 22. bis 25. Juli 1994 im
Zoologischen Institut der Universität Bonn ein internationales Symposium zum oben
genannten Thema statt. Alle Interessenten sind dazu herzlich eingeladen. Auskunft erteilt
Prof. Dr. U. SchmiDT, Zoologisches Institut, Poppelsdorfer Schloß, D-53115 Bonn (Tel.
02 28/73 54 68; Fax 02 28/73 54 58).

European Squirrel Group

Die “European Squirrel Group” ist ein Zusammenschluß von Forschern aus überwiegend
europäischen Ländern, die sich in ihren wissenschaftlichen Arbeiten mit Hörnchen
beschäftigen. Die Schwerpunktthemen decken ein recht breites Feld ab, welches sich von
Fragen zur Ökologie der verschiedenen Arten über morphologische und physiologische

Mitteilungen 63

Aspekte bis hin zur Populationsgenetik erstreckt. Auch Natur- und Artenschutzbelange
nehmen einen breiten Raum eın.

Die Gruppe existiert seit etwa zwei Jahren. Regelmäßige Treffen in ein- bis zweijähri-
gen Abständen sollen dem wissenschaftlichen Austausch dienen und die Kooperation
fördern. Die nächste offizielle Zusammenkunft ist im Anschluß an den “II. European
Mammal Congress“ in England 1995 geplant.

Weitere Informationen sind zu erhalten bei: European Squirrel Secretariat, Luc Wauters,
Dept. Biology, UIA, Universiteitsplein 1, B-2610 Wilrijk, Belgien, und Sibylle Münch,
Lusenstr. 41, D-94556 Waldhäuser, Bundesrepublik Deutschland, Tel. (08553) 6515.

Seventy-Fifth Anniversary Meeting of the American Society of Mammalogists

The National Museum of Natural History and National Zoological Park, Smithsonian
Institution, and the Biological Survey Section, National Biological Survey, are hosting the
75th Anniversary Meeting of the American Society of Mammalogists, to be held in
Washington, D. C., 18-23 June 1994. Due to the commemorative nature of this year’s
annual meeting, there will be special symposia and integrated workshops that thematically
address Biodiversity and Levels of Biological Organızation, in addition to the usual
schedule of technical sessions.

This ıs the final call for oral and poster presentations. There is space for a maximum of
97 oral presentations and 180 posters, allotted on a first-come, fırst-serve basis (1.e., data of
postmark). All abstracts must be submitted by 1 March 1994, and accompanıed by the
preregistration form and registration fee. Questions concerning registration should by
addressed to: MICHAEL D. CARLETON, Mammal Division MRC 108, National Museum of
Natural History, Washington, D. C. 20560, USA (202-786-2490) or 357-1920; Fax 202-
357-4779).

BUEEIDESIREIEHIUN GEN

UECKERMANN, E.: Das Sikawild. Vorkommen, Naturgeschichte und Bejagung. Hamburg
und Berlin: Paul Parey 1992. 103 S., 45 Abb., 30 Tab. Kart. DM 16,-. ISBN 3-490-08812-3

In zweiter, neubearbeiteter und erweiterter Auflage liegt nun „Das Sikawild“ als Heft 7 der
Schriftenreihe der Forschungsstelle für Jagdkunde und Wildschadenverhütung des Landes Nordrhein-
Westfalen vor. Ein wichtiger, auch im Untertitel genannter Aspekt dieser Abhandlung betrifft die
Bejagung. Dieser Teil, der in die Kapitel Abschußplanung, Abschußdurchführung und Abschußricht-
linien sowie Bewertung untergliedert ist, richtet sich besonders an den Jäger und Forstmann und soll
hier nicht weiter erörtert werden. Dagegen beanspruchen Vorkommen und Naturgeschichte dieser
Wildart auch das Interesse des Säugetierkundlers. Zwar gibt es in der Bundesrepublik nur etwa 2000
freilebende Sikahirsche in einigen wenigen, voneinander isolierten Gebieten, und somit ist diese Art
auf den ersten Blick nur von untergeordneter Bedeutung, doch als eingebürgertes Tier hat es
zweifellos einen besonderen Stellenwert, vor allem unter den Gesichtspunkten der Einpassung in
einen fremden Lebensraum und der Auseinandersetzung mit heimischen Arten. Hierzu fehlt noch
eine umfassende Darstellung, doch teilt der Autor zahlreiche Einzelheiten mit; das betrifft die
Beschreibung des natürlichen Areals der Art, ihrer Unterarten und deren Status sowie die Geschichte
der Einbürgerung des Sıkawildes sowohl in Deutschland als auch im weiteren europäischen Gebiet
und darüber hinaus. Des weiteren erörtert er die Art und Weise der Einbürgerung in Deutschland und
die Entwicklung der einzelnen Bestände unter Berücksichtigung ihrer Lebensweise; dabei geht er auch
auf Krankheiten, Verluste durch den Straßenverkehr, durch dieses Wild verursachte Schäden und
Bastardierungen ein.

Durch Abbildungen und Tabellen werden zahlreiche Sachverhalte veranschaulicht, ein Stichwort-
register erleichtert die Benutzung des Buches, und ein umfassendes Verzeichnis führt den Leser zu
weiterer Literatur über diese Hirschart. D. HEınrıch, Kiel

WIESEMÜLLER, W.; LEIBETSEDER, J. (Hrsg.): Ernährung monogastrischer Nutztiere.
Jena, Stuttgart: Gustav Fischer Verlag 1993. 308 S., 39 Abb., 166 Tab. DM 128,-.
ISBN 3-334-60428-4

Dieses Buch, an dem sieben Fachleute aus Deutschland und je einer aus Ungarn, Österreich und der
Schweiz mitarbeiteten, wendet sich an jene, die Nutzsäugetiere gesund, ausgewogen, wirtschaftlich
und mit möglichst geringer Belastung der Umwelt ernähren wollen. Es werden Leser angesprochen,
bei denen die Kenntnis der Grundlagen der Tierernährung vorausgesetzt werden kann, also Tiermedi-
ziner, Ernährungsfachleute, Landwirte und Futtermittelproduzenten.

Zunächst wird ein Überblick über Fragen der Futtermittelkunde vermittelt. Dabei wird nicht nur
kurz die Futtermittelanalyse, die energetische Futtermittelbewertung und das Futtermittelrecht
abgehandelt, sondern es werden auch durch Futtermittel verursachte Schadwirkungen besprochen.

Auf den folgenden Seiten werden Informationen zur Ernährung des Hausschweins geboten, wobei
nach einer Darstellung physiologischer Grundlagen gesondert Sauen, Ferkel, Zuchteber und Mast-
schweine berücksichtigt werden. Bei der folgenden Darstellung der Ernährung der Pferde werden
zunächst die ernährungsphysiologischen Grundlagen und dann Besonderheiten des Verhaltens der
Pferde gegenüber Mangel und Überschuß besprochen. Gesondert folgen dann praktische Hinweise
zur Fütterung von Sportpferden, Zuchtstuten, Fohlen, Deckhengsten und Ponys sowie Empfehlun-
gen zur Vermeidung ernährungsbedingter Krankheiten.

Die Ernährung des Kaninchens wird auf den folgenden Seiten abgehandelt, wobei in einem
einleitenden Abschnitt auf die bei den Lagomorpha auftretende Zäkotrophie, die Aufnahme eines
speziellen und im Blinddarm gebildeten Weichkotes, aufmerksam gemacht wird. Von vier ausgewähl-
ten Pelztierarten, wie Sumpfbiber oder Nutria, Nerz sowie Blau- und Silberfuchs wird die Ernährung
behandelt, ferner die von Hunden und Katzen. Am Ende der genannten Abschnitte findet sich je ein
Literaturverzeichnis. Diese Verzeichnisse sowie 166 außerordentlich materialreiche Tabellen, welche
meistens übersichtlich gestaltet sind, erlauben es, den vorliegenden Band als Nachschlagewerk bei
praktischen Fütterungsproblemen zu nutzen. Es erweist sich als sehr hilfreich und sollte anderen
Buchautoren und Verlagen zur Nachahmung empfohlen werden, daß sich auf den Vorsatzseiten am
Anfang des Bandes eine Liste der im Buch benutzten Abkürzungen befindet! Ein 16 Seiten langes
Sachregister schließt das Werk ab.

Neben den eingangs genannten Interessenten dürfte dieses praxisbezogene Buch für Tiergärtner
nützlich sein. Ferner macht der Band auch vergleichend interessierten Säugetierkundlern, Physiologen
und Anatomen eine große Fülle von Daten leicht zugänglich, die ın Spezialarbeiten und Mono-
graphien verstreut sind. P. LAnGER, Gießen

Zur Ökologie des Luchses Lynx Iynx
im Verlauf seiner Wiederansiedlung
in den Walliser Alpen

Von PD Dr. Heinrich Haller, Davos. 1992. 66 Seiten mit 26 Abbildungen,
davon 7 farbig und 11 Tabellen. Kartoniert DM 58,—

Die Wiederansiedlung des Luchses in den Schweizer Alpen ist ein Pionierwerk, vergleichbar
mit jener des Steinbocks. Über die ökologischen Folgewirkungen waren aber bei den
Freilassungen kaum Kenntnisse vorhanden.

Im Rahmen der vorliegenden Studie wurde versucht, die Wiederansiedlung des Luchses zu
dokumentieren, wobei der längerfristigen Integration der Katze in die Natur und den
Wechselbeziehungen mit Beutepopulationen besonderes Augenmerk galt.

Die Untersuchung bezieht sich aufausgeprägte Hochgebirgsverhältnisse. In einem Teilgebiet
mit überhegtem Schalenwildbestand konnte eine massive Prädationswirkung des Luchses
nachgewiesen werden, die in dieser Form bisher unbekannt war. Angesichts der verbreiteten
forstlichen Probleme mit überhöhten Huftierpopulationen in den Gebirgswäldern kommt
diesem Befund besondere Bedeutung zu.

Die vorliegende Publikation richtet sich an alle, die sich für den Luchs und seine Wirkungen
auf Beutepopulationen interessieren. Räuber/Beute-Beziehung, Wald/Wild-Problematik und
das Thema Wiederansiedlung sind Stichworte, von denen Biologen, Forstleute und im
Naturschutz tätige Personen speziell angesprochen werden. Preis: Stand 1. Januar 1994

Aus dem Verlagsprogramm Paul Parey jetzt bei

Blackwell Wissenschaft - Berlin

Erscheinungsweise und Bezugspreis 1994: 6 Hefte bilden einen Band. Jahresabonnement Inland:
DM 378,- zuzüglich DM 13,80 Versandkosten; Jahresabonnement Österreich: öS 2949,— zuzüg-
lich öS 164,- Versandkosten; Jahresabonnement Schweiz: str 364,— zuzüglich str 21,— Versandko-
sten; Jahresabonnement EG-Binnenmarkt-Länder mit USt-ID-Nr.: DM 353,27 zuzüglich
DM 19,63 Versandkosten; Jahresabonnement EG-Binnenmarkt-Länder ohne USt-ID-Nr. und
Drittländer: DM 378,- zuzüglich DM 21,- Versandkosten. Das Abonnement wird zum Jahresan-
fang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens
am 15. November eine Abbestellung ım Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhand-
lung oder bei der Verlagsbuchhandlung Paul Parey GmbH & Co. KG, Spitalerstraße 12,
D-20095 Hamburg, Bundesrepublik Deutschland, bestellt werden. Die Mitglieder der „Deut-
schen Gesellschaft für Säugetierkunde“ erhalten die Zeitschrift unberechnet im Rahmen des
Mitgliedsbeitrages.

Z. Säugetierkunde 59 (1994) 1, 1-64

Biologie
der Säugetiere

Von Prof. Dr. Walter Pflumm, Kaiserslautern, in Zusammenarbeit mit |
Margarete Pflumm-Eisbrenner. 1989. 565 Seiten mit 413 Abbildungen |
und 4 Tabellen. Kartoniert DM 19,80 h

Im Zeitalter der Molekularbiologie und Biochemie hat die Säugetierkunde |
nicht an allen deutschsprachigen akademischen Ausbildungsstätten, aber '
auch in den Leistungskursen der reformierten Sekundarstufe den Stellen- °
wert, derihrgeradein bezugaufdas direkt aufden Menschen übertragbare |
Wissen zukommt. Diesem Defizit will das vorliegende Buch abhelfen. Mit '
subjektiver Stoffauswahl und guter didaktischer Darbietung ist es um- |
fangreicher als die üblichen Hochschultexte und zugleich eines der am '
reichhaltigsten bebilderten Lehrbücher in der Biologie.

Ein Verzeichnis mit Erklärungen der zoologischen Fachwörter, besonders '
wichtig für Leser ohne Latein- oder Griechischkenntnisse, sowie zwei f
Tiernamenverzeichnisse, ein Sachregister und ein Literaturverzeichnis
machen das Buch von vielen Fragestellungen her zugänglich für einen |
großen Leserkreis. Dazu gehören Studierende der Biologie und Ober- |
schüler der Sekundarstufe II ebenso wie Biologielehrer, Ausbilder von '
Tierpflegern und alle Natur- und Tierfreunde, die an einer umfassenden,
in Wort und Bild leicht verständlichen Darstellung von Bau und Leben der | |
Säugetiere interessiert sind. Preis: Stand 1.-ULIZEIZTE

Aus dem Verlagsprogramm Paul Parey jetzt bei

Blackwell Wissenschaft - Berlin

01.59 (2), 65-128, April 1994 ISSN 0044-3468 21274 &
QL
1.700

7=-ITSCHRIFT FÜR
RUGETIERKUNDE

I Maria Lucia; Oliveira, J. A. de; Persson, Vanessa G.: Annual age structure and reproductive patterns in
' Marmosa incana (Lund, 1841) (Didelphidae, Marsupialia). — Jahreszeitliche Altersstruktur und Fortpflanzung bei

Marmosa incana (Lund, 1841) (Didelphidae, Marsupialia) 65
/hitaker, J. ©. Jr.; Shalmon, B.; Kunz, T. H.: Food and feeding habits of insectivorous bats from Israel. - Nahrung
und Ernährungsverhalten einiger insektivorer Fledermäuse aus Israel 74

\iernändez, Lucina; Delibes, M.: Seasonal food habits of coyotes, Canis latrans, in the Bolsön de Mapimi, Southern
Chihuahuan Desert, Mexico. — Jahreszeitliche Ernährungsgewohnheiten von Kojoten, Canis latrans, im Bolsön-

! de Mapimi-Reservat, südliche Chihuahua-Wüste, Mexiko 82
jeters, G.; Hast, M. H.: Hyoid structure, laryngeal anatomy, and vocalization in felids (Mammalia: Carnivora:
| Felidae). - Hyoidbau, Kehlkopfmorphologie und Lautgebung bei Feliden (Mammalia: Carnivora: Felidae) 87

tastelein, R. A.; Muller, M.; Terlouw, A.: Oral suction of a Pacific walrus (Odobenus rosmarus divergens) in air and
under water. — Orales Saugvermögen eines Pazifischen Walrosses (Odobenus rosmarus divergens) in Luft und
unter Wasser 105

4

runet-Lecomte, P.; Volobouev, V.: Comparative morphometry and cytogenetics of Microtus (Terricola) multiplex
(Arvicolidae, Rodentia) of the western French Alps. — Morphometrie und Cytogenetik von Microtus (Terricola)
multiplex (Arvicolidae, Rodentia) der westlichen französischen Alpen: im Vergleich 116

\
\

| issenschaftliche Kurzmitteilung
la, Bibiana L.: Use of dung piles by neighbouring vicufas. - Benutzung von Kotplätzen durch benachbarte

Temun,

| Vicufas SMITTEN 126

i

|

[erlag Paul Parey Hamburg und Berlin

r

|
|

HERAUSGEBER/EDITEORS

P. J. H. van BREE, Amsterdam -— W. FIEDLER, Wien — H. Frick, München - G. B.

HARTL, Wien - W. HERRE, Kiel - R. HUTTERER, Bonn - H.-G. Krös, Berlin - H.-].

Kunn, Göttingen -— E. KuLzer, Tübingen -— W. MAIER, Tübingen - J. NIETHAMMER,

Bonn —- OÖ. Anne E. Rasa, Bonn - H. ReıcHastein, Kıel - M. Rönrs, Hannover —

H. SCHLIEMAnN, Hamburg — D. STARcK, Frankfurt a.M. - E. THEnıus, Wien - P. Vo-
GEL, Lausanne - H. Wınkıng, Lübeck

SCHIRIETFEETEUNIGYE DI LORENZO EEIEE

D. Kruska, Kiel - P. LAnGER, Gießen

This journal is covered by Biosciences Information Service of Biological Abstracts, and by Current Con-
tents (Series Agriculture, Biology, and Environmental Sciences) of Institute for Scientific Information

Die Zeitschrift für Säugetierkunde veröffentlicht Originalarbeiten und wissenschaftliche Kurzmittei-
lungen aus dem Gesamtgebiet der Säugetierklinde, Besprechungen der wichtigsten internationalen
Literatur sowie die Bekanntmachungen der Deutschen Gesellschaft für Säugetierkunde. Verantwort-
licher Schriftleiter im Sinne des Hamburgischen Pressegesetzes ist Prof. Dr. Dieter Kruska.

Zusätzlich erscheint einmal ım Jahr ein Heft mit den Abstracts der Vorträge, die auf der jeweiligen
Hauptversammlung der Deutschen Gesellschaft für Säugetierkunde gehalten werden. Sie werden als
Supplement dem betreffenden Jahrgang der Zeitschrift zugeordnet. Verantwortlich für ihren Inhalt
sind ausschließlich die Autoren der Abstracts.

Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z. Hd. Prof. Dr. Dieter
Kruska, Institut für Haustierkunde, Biologiezentrum der Christian-Albrechts-Universität, Am
Botanischen Garten 9, D-24118 Kiel, Bundesrepublik Deutschland. Für die Publikation vorgese-
hene Manuskripte sollen gemäß den „Redaktionellen Richtlinien‘ abgefaßt werden. In ihnen
finden sich weitere Hinweise zur Annahme von Manuskripten, Bedingungen für die Veröffent-
lichung und die Drucklegung, ferner Richtlinien für die Abfassung eines Abstracts und eine

Korrekturzeichentabelle. Die Richtlinien sind auf Anfrage bei der Schriftleitung und dem Verlag
erhältlich.

Sonderdrucke: Anstelle einer Unkostenvergütung erhalten die Verfasser von Originalbeiträgen und
Wissenschaftlichen Kurzmitteilungen 50 unberechnete Sonderdrucke. Mehrbedarf steht gegen
Berechnung zur Verfügung, jedoch muß die Bestellung spätestens mit der Rücksendung der
Korrekturfahnen erfolgen.

Vorbehalt aller Rechte: Die ın dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich
geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks,
des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der
Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei-
ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch im Einzelfall grund-
sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines
Beitrages ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber-
rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 ın der Fassung vom
24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen
den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sınd
mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen.

Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of
the fırst page of an article in this journal indicates the copyright owner’s consent that copies of the article may
be made for personal or internal use, or for the personal or internaluse of specific clients. This consent is given
on the condition, however, that the copier pay the stated percopy fee through the Copyright Clearance
Center, Inc., 21 Congress Street, Salem, MA 01970, USA, for copying beyond that permitted by Sections 107
or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying
for general distribution, for advertising or promotional purposes, for creating new collective, or for resale.
For copying from back volumes of this journal see “Permissions to Photo-Copy: Publisher’s Fee List” of the
EC.

Fortsetzung 3. Umschlagseite

© 1994 Paul Parey. Verlag: Paul Parey GmbH & Co. KG, Hamburg und Berlin. Anschriften: Spitalerstr. 12,
D-20095 Hamburg; Seelbuschring 9-17, D-12105 Berlin, Bundesrepublik Deutschland. — Printed in Germany by
Westholsteinische Verlagsdruckerei Boyens & Co., Heide/Holst.

Z. Säugetierkunde 59 (1994) 65-73
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Annual age structure and reproductive patterns in
Marmosa incana (Lund, 1841) (Didelphidae, Marsupialia)

By Marıa Lucia LoRrInI, J. A. DE OLIVEIRA, and VANESSA G. PERSSON

Museu Nacional, Universidade Federal do Rio de Janeiro; and Musen de Historia Natural
“Capao da Imbuia”, Curitiba, Parana, Brazil

Receipt of Ms. 6. 1. 1993
Acceptance of Ms. 10. 2. 1993

Abstract

Investigated were the annual age structure and reproductive indicators in museum specimens of the
mouse opossum Marmosa incana. Monthly distributions of relative age classes, indexed by tooth
eruption and wear, suggest an almost total cohort turnover in an annual cycle (males: one year;
females: one year and half). Analysıs of tegumentary indicators of reproductive activity, gauged by
examination of internal reproductive tracts, demonstrated that all the sexually matured individuals
belong to the two oldest age classes. These populational features combined with a three-month
seasonal breeding period, detected in the greater part of the geographic range of M. incana, apparently
result in an unusual life history strategy for this species, characterized by only one “big-bang”
reproductive event in a lifetime.

Introduction

Central to the distinction between marsupials and eutherians are the profoundly different
morphologies and functions of the reproductive tracts (GRIFFITHS 1978). Likewise, there
are marked differences in the early phases of development, marsupials lacking a true
trophoblast and apparently not being able to provide prolonged protection for the
genetically foreign embryo against the mother’s immune system (LILLENGRAVEN et al.
1987). Such features determine a short gestation and a long lactation period, ın which a
major part of marsupial development is carried out. Together with the characteristically
lower rates of development in marsupials (LEE and CockBURN 1985), these reproductive
characters strongly affect aspects of their populational biologies, especially the age of
sexual maturation, litter size and litter frequency.

‚nhe greater part of what ıs known about the reproductive biology and life history of
marsupials ıs based on studies focusing on the Australian forms. Contrastingly, detailed
specific analyses of wild populations, such as those carried out upon Australian species
(e.g. NEwsoME 1965; GUILER 1970; WooD 1970; GEMMELL 1982), are still rare for the less
diversified South American marsupials. To date, most of the reproductive information for
the neotropical species comes from small mammal community mark-release studies (e.g.
Davıs 1946; Reıg 1964; O’ConnELL 1979; FOnsEcA and KIERULFF 1989; STALLINGS
1989). In this approach there are usually two difficulties to be faced: the ageing of living
specimens ın the field and the virtual impossibility of examining the internal reproductive
tracts of every individual, making it necessary to use indirect external evidence, such as
tegumentary gland activity, to infer the reproductive condition of specimens. The precise
correspondence between these indicators and the actual reproductive status, as well as the
interspecific variation in their expression, has not yet been investigated.

Herein we propose an alternative approach to the study of reproductive and life history
patterns of the didelphid marsupial Marmosa incana (mouse opossum), analysıng the
reproductive indicators in museum specimens for which real reproductive status and
relative age are, at least in part, available. Marmosa incana shows three pelage types (A, B

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5902-0065 $ 02.50/0

66 Maria Lucia Lorini, J. A. de Oliveira, and Vanessa G. Persson

and C), unevenly distributed over the year, and apparently related to age, sex and sexual
maturity (OLIVEIRA et al. 1992). Such results suggested a detailed investigation of the
monthly age structure and reproductive conditions ın this species. Notable aspects of the
reproductive biology and their bearing on the life history of this marsupial, so revealed, are
described in the present contribution.

Material and methods

The total sample analysed in this study is composed of 311 museum specimens from eastern Brazil
(Bahia, Minas Gerais, Espirito Santo, Rio de Janeiro and Paranä states), covering the greater part of
the geographic range of Marmosa incana. Localities, sample sizes and museum acronyms were listed
in OLIVEIRA et al. (1992). Date of collecting, sex, body length, weight and conditions of internal
reproductive tracts, when available, were taken from original labels. Specimens were aged on the basis
of molarıform tooth eruption (classes 1 to 6 as established by TRIBE 1990) and five consecutive stages
of M1-M4 wear, after complete positioning of PM3. The first two stages were allocated to age class 6
and the last three stages to age class 7 (OLIVEIRA et al. 1992).

As a first step toward understanding the age and seasonal distribution of M. incana, we tabulated
the occurrence of all individuals with sex and age class available (n = 225) by month of collecting.

Regarding the female reproductive condition, color change in the pouch or mammary region is
commonly used as an indicator of breeding activity ın living didelphids. In some species of Marmosa
an orange to rust-brown stain around the mammary area is produced during pregnancy and nursing
(BARNES 1977). We investigated the occurrence of this stained area and the degree of development and
degeneration of nıpples in M. incana and related these features to reproductive data available from
some specimens to infer the females’ reproductive condition.

The sternal gland area, a field of hypertrophied apocrine sudoriferous and sebaceous glands
(BARNESs 1977), was also investigated in each specimen. The activity of these glands has been referred
to in the literature as associated with territorial marking and reproductive activity in some species of
Marmosa (HUNSAKER Il and SHUPE 1977); sternal gland activity is characterized by an oily secretion
which dries to an amber to dark brown deposition attached to the skin, forming a concretion above
the gland field that often glues hairs together.

Results and discussion
Age structure

Analysıs of the age composition by month (Fig. 1) reveals that age classes are not
homogeneously distributed over the year. Young individuals (classes 3 and 4) are restricted
to between January and May ın all localities except Ilheus (state of Bahia), where one
additional specimen was collected in September. Class 5 appears in February and occurs
until May, with one exception in August for the Ilheus population. Disregarding these two
exceptional individuals from Ilheus, age classes 3, 4 and 5 are distributed in a restricted and
almost coincident period. From class 6 on, differences between sexes are noted in the
monthly distributions. Whereas males of class 6 are present from February to October,
with higher frequencies between April and May, females appear from April to November,
and are more frequent between July and August. Frequency of class 6 increases by the
middle of the year, when class 5 declines. Similarly, frequency of class 7 increases ın the
last months of the year, while frequency of class 6 declines.

Distribution of class 7 over the year extends for ten months for females and eight
months for males. Except for two individuals from Ilheus collected in April, males of age
class 7 are absent from February to May. Females of this age class do not occur between
July and August, except for one individual from Ilheus, trapped in July.

It is relevant to note that exceptional individuals reported above comprise only 8% of
the total sample with available data on sex, age and month of capture from Ilheus (n = 62).

Annual age structure and reproductive patterns in Marmosa incana 67

: Age. Class 3

Age: Class 7

JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
14 13 23 32 29 06 26 17 26 18 17 04

Fig. 1. Monthwise age distribution of Marmosa incana. Black bars: males; striped bars: females

Analysis of the mammary region

Analysis of the mammary conditions revealed that females of age classes 3, 4, 5 and 6 with
long and soft pelage (type A) did not show indications of reproductive activity. All re-
productive females belong to age classes 6 and 7, and have short and coarse pelage (type B).

Six different conditions of the mammary region were identified: a. not stained,
indistinguishable from the rest of ventral pelage; b. not stained, but showing glabrous
circles without teats; c. stained, showing conspicuous glabrous circles around incipient
teats; d. stained, with a homogeneous set of cylindrical nıpples up to 1 mm in diameter,
each one inside a well-defined glabrous area; e. stained, with a heterogeneous set,
containing some lax nipples (more than 1 mm in diameter) and others showing various
stages of degeneration, in some cases reduced to a scar; f. stained, showing only darkened
scars, with inconspicuous glabrous areas. Reproductive data described in the labels and the
analysis of the internal reproductive tracts of recently collected specimens permitted us to

68 Maria Lucia Lorini, J. A. de Oliveira, and Vanessa G. Persson

allocate the females, assigning each of the six conditions described above to their respective
stages: non-reproductive (a); pre-reproductive (b); pregnant (c), early lactant (d), late
lactant (e), and post-lactant (f).

All reproductive females presented a stain in the mammary region. However, even the
post-lactants showed the mammary stain, an indication that it remains in the integument as
a residual.

This feature has often been employed to determine the reproductive period in Marmosa
incana (FOnsEcA and KIELRUFF 1990; STALLINGS 1990). However, this indicator alone
does not provide a specific identification of the various reproductive phases of females,
since in marsupials the nursing period is extended. Marsupials are born in an exceptionally
altricial state, and a large part of their development takes place while fused to or dependent
on a nipple.

Monthly rates of reproductive females (Fig. 2) revealed that periods of pregnancy, early
and late lactancy, and post-lactancy are consecutive and seasonally distributed over the
year. Pregnant females are limited to November, except for one individual collected in

Fig. 2. Monthwise distribution of
female reproductive status. Fre-
quency of each status in relation
to the total of females showing
pelage type B. Bars marked with ı
denote exceptional individuals
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC from Ilheus

U] Pregnant Early lactant Late lactant MM Post- lactant

May in Ilheus. Early lactant females were captured between October and December, and
late lactants between January and April, except for three individuals from Ilheus collected
in September. Post-lactant females are restricted to the period from March to May in the
total sample.

Reproductive period

Analysıs of the annual distribution of female reproductive status, together with the
occurrence of offspring made it possible to identify the reproductive period for M. ıncana.
Distributions of pregnant and lactant females suggest that the mating period and subse-
quent teat attachment phase occur between October and December in the total sample
except for part of the Ilheus population, where a pregnant female was captured ın May.

The limited distribution of the youngest individuals in our sample (age class 3) ıs
independent evidence of a single and seasonal three-month reproductive period for the
total sample, except the Ilheus population, where an additional mating period is revealed ın
May. The distributions of juvenile ages (classs 3 and 4) are coincident wıth the January-
April distribution of late lactant females.

The annual distribution of post-lactant females is coincident with the last months of
occurrence of age classes 4 and 5. A comparison of our results with the development data
obtained for Marmosa robinsoni (EISENBERG 1981; O’ConNELL 1983) makes it possible to
relate the juvenile dental age classes 3, 4 and 5 with the rear cycle phases. Age class 3
individuals are probably in the nest phase, when the young of M. robinsoni begin to eat
solid food and are able to leave the nest alone or following the female (O’ConneLL 1983).
Classes 4 and 5 are correlated with the weaning and dispersion events. This conclusion is ın

Annual age structure and reproductive patterns in Marmosa incana 69

keeping with the attainment of a total crushing surface with four functional molarıform
teeth at these ages.

The reproductive period ıs probably more restricted ın a given locality than the three
months obtained ın our analysıs, and some asynchrony among localıties ıs not unexpected
at all, since reproductive patterns are influenced by local conditions. Among enviromental
factors that have been recognized in the determination of the reproductive patterns in
didelphid marsupials is the seasonality of rainfall (O’ConneıL 1979).

Our total sample comprises several series collected in varıous parts of the state of Minas
Gerais which show a comparable pattern of rainfall, with five to six drier months in the
year, and a sample from Ilheus, in the state of Bahia, a locality characterized by the absence
of a dry season (NIMER 1989). Additional records are from localities in Rio de Janeiro,
Espirito Santo and Paranä states, in which rainfall distributions show an intermediate
pattern as compared to Minas Gerais and Ilheus.

The reproductive period revealed by our analysıs corresponds very well to the rainy
season ın all localities of Minas Gerais, which compose the greater part of our sample.
Remaining localities do not show inconformity to the Minas Gerais pattern, except Ilheus.
In fact, more than one reproductive period in the Ilheus population is attested by the
occurrence of a pregnant female in May and exceptional young individuals of age classes 4
and 5 (August-September), together with a single late lactant female ın September.

Consequently, at least two reproductive periods ın the year are revealed in the analysıs
of the Ilheus sample. A major one, coincıdent wıth that shown by the total sample
(October-December), and another, revealed by a smaller number of individuals, occurring

from March to May.

Sexual maturity

Further interpretation of the results described above, together with that of pelage varıation
in Marmosa incana (OLIVEIRA et al. 1992), permits a determination of the sexually matured
individuals. Since TATE (1933), ıt has been wiıdely accepted that sexual maturity in
Marmosa ıs attaıned at an early stage of life, when the last molarıform tooth (third pre-
molar for M. ıncana) ıs beginning to erupt, a stage that corresponds to class five in our
analysıs. Our results do not confirm TATE’s assertion, at least for M. ıncana. As can be
inferred from figure 1, distributions of age class 5 and most of class 6 are not coincident
with the reproductive period. Indeed, among females, individuals with reproductive
indications were found only in the last two age classes (6 and 7), all showing the short and
dull pelage previously described as type B (OLivEira et al. 1992). Almost all males taken
during the three months of the reproductive period (October-December) belong to age
class 7. The attaıinment of sexual maturity in males was thought to occur in connection
with pelage type C (OLıveira et al. 1992). The monthly prevalence of males with
conspicuous type C pattern (modified hairs reaching the middorsum) relative to the total of
age class 7 and pelage type € (Fig. 3) demonstrates this relationship. Additional evidence of

%

60
Ele)

40

Fıg. 3. Frequency of males with
modified dorsal hairs reaching the
middorsum in relation to the total
age class 7 and pelage type C
males JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

70 Maria Lucia Lorini, J. A. de Oliveira, and Vanessa G. Persson

the strong correlation between the attainment of sexual maturity and pelage type C is
provided by histological analysıs of testes. Slides obtained from individuals of age class 7
and pelage type A did not reveal any spermatogenic activity (OLIVEIRA et al. 1992).

The only exceptions to the limited distribution of pelage type C males over the year
(July-January) are two individuals from Ilheus collected in April (OLıveıra et al. 1992),
precisely during the additional reproductive period at this locality.

Analysis of sternal gland activity

Our results show that the sternal gland is externally discernible in the skins of both sexes at
all ages analysed, contrary to TATE’s (1933) and HunsAkER and SHuPpe’s (1977) assertions
that the gland is not present ın Marmosa incana. Residuals of the oıly secretion were
detected in specimens of both sexes from class 4 onwards over the year. Frequencies of
individuals with sternal secretion by age classes attest to a similar pattern between males
and females, with a clear peak ın actıvity at age class 5 and a succeeding decrease at class 6
(Fig. 4). At age class 7, on the other hand, almost all males show glandular activity,
whereas females continue with a low percent of individuals showing sternal gland
secretion.

Fıg. 4. Frequency by age of indi-
viduals showing sternal gland ac-
3 4 5 6 Ta ENiCy,

These results do not point to an unequivocal correspondence between the occurrence of
sternal gland secretion and reproductive activity. The increasing frequency of sternal
secretion in both sexes at classes 4 and 5 (Fig. 4), probable ages of juvenile independence,
may be related, rather, to dispersion behavior.

Semelparity

The absence of class 7 males between February and May ın all localıties except Ilheus
cannot be accounted for by small sample sizes, as even at Alem Paraiba (Minas Gerais),
where a large sample from these months was assembled (n = 64), no males of age class 7
were obtained.

After the absence period (February-May), the only male of class 7 registered in June
and most of age class 7 males that were gathered in July show pelage type A. Mean weight
of specimens collected between June and August, the first three months after the absence
period, is lower than that from the three months before the males’ disappearance
(November-January). In the same way, molar wear is more accentuated among class 7
before the absence period, suggesting that the males which disappear after January and
those collected after June do not belong to the same generation.

In regard to females, the absence of age class 7 in July and August may also be evidence

Annual age structure and reproductive patterns in Marmosa incana 71
Descriptive statistics of body weight for samples of supposed cohorts in each sex

Within parenthesis, f-ratio and associated probability (ANOVA) for the null hypothesis of no
difference between samples from after (a) and before (b) adults’ disappearence

n

mean
sd. dev.
(F-ratio)
(P)

of a gap between two successive generations. The mean weight of females collected after
these months is also lower than that from the first semester, and the molar wear is less
accentuated.

Analysis of variance, employed to test the null hypothesis of no differentiation between
samples of these supposed cohorts, also revealed that differences in mean weight were
significant at the 1% level (P<0.01) for both sexes (Table).

A probable explanation for the males’ absence may be related to a general mortality
after the reproductive period. Sımilarly, the results set forth above also show that females
which had already reproduced are not present in a new reproductive period in the
following year. Since for females the gap between generations is shorter, it might be
supposed that occasıonally a female would reach the breeding season of the next year.
However, data from Marmosa robinsoni, a species of similar habits and size, do not
support the hypothesis that such a female would be reproductive, since females of that
species are no longer fertile after 17 months.

To date, M. robinsoni ıs the only mouse opossum for which comprehensive reproduc-
tive data ıs available. Young individuals of this species are completely weaned 65 days after
birth, and the total time that a female spends rearing a litter, from conception to weaning,
totals approximately 80 days (EISENBERG 1981; O’ConneLr 1983). Considering the three-
month breeding period of M. incana described above, and assuming a rearing cycle similar
to that of M. robinsoni, a female could only produce a single litter wıthin one reproductive
season. This hypothesis ıs corroborated ın our results by the restricted, consecutive and
unrepeated occurrence of the female mammary stages for the total sample, aside from the
Ilheus exceptions (Fig. 2). Furthermore, the limited breeding season shown by M. incana ıs
also an indication that the female estrous cycle is monoestric, or, at least, seasonally
polyestric, in a very limited ıinterval of time.

The above evidence suggests that in Marmosa incana each individual takes part ın only
one reproductive season except in the Ilheus population, where, occasionally, two
independent breeding seasons may occur within a year. Consequently, in localities where
there is no additional reproductive season, Marmosa incana reproduces only once in a
lifetime, which characterizes a semelparous way of life.

Rare among vertebrates, a semelparous strategy was originally demonstrated among
mammals ın the dasyurid marsupial Antechinus stnartüi. This species presents a total
turnover of the male cohort within consecutive generations owing to the death of all adult
males after the reproductive period. A monestrous cycle in the female and a synchronous
pattern of breeding within a short interval of time, in a local population, were also
documented.

BRAITHWAITE and LEE (1977) suggested that semelparity is the extreme expression of a
strategy characterized by an intense and highly synchronized reproductive effort where

v2 Maria Lucia Lorinı, J. A. de Oliveira, and Vanessa G. Persson

juvenile survival ıs consistently higher during one season of the year. Under this assump-
tion, the development of semelparity ın mammals would be favored among species
showing a maximum field longevity of approximately one year and an annual optimal
period for reproduction of sufficient duration for individuals females to raise successfully
one but not two litters (BRAITHWAITE and LEE 1977). These authors considered marsupials
weighting less than 1 kg and living in predictable, highly seasonal environments to be the
prime candidates for semelparity among mammals.

In keeping with these predictions, all presumptive examples of semelparous mammals
suggested since then are small marsupials confined to highly seasonal habitats. Indeed,
among New World mammals the only indication of semelparity was obtained in a
restricted population of the didelphid marsupial Monodelphis dimidiata at Balcarce,
Buenos Aires province, Argentina. In thıs localıty none of the adult individuals survive to
the first winter, or into the winter, after the reproductive period (Pine et al. 1985). It is
noteworthy that records of M. dimidiata to the north of Balcarce did not corroborate the
semelparous pattern.

Similar to Monodelphis dimidıiata, Marmosa incana also shows a small body size (total
weight is less than 120 g ın males and 80 g in females). As revealed by the present study
however, semelparity seems to be a rather widespread phenomenon in Marmosa incana.
This species is best known from the humid Atlantic forests of eastern Brazil (STREILEIN
1982), a relatively constant environment, where temperatures do not show extreme
varlations in the course of the year. Although there is some seasonalıty ın rainfall,
especially in the Minas Gerais localıties, thıs cannot be considered a predictable, strongly
seasonal environment, as the habitats seem to be of all semelparous mammals reported to
date. In the sample analysed, the semelparous pattern was corroborated ın all localıties
except the Ilheus sample, where it is not impossible that some females take part in more
than one breeding period. Although seasonality may constitute an important condition for
the evolution of semelparity in mammals, it appears, from our analysis, that this reproduc-
tive strategy may also exist in more constant environments than those previously hy-
pothesized.

Acknowledgements

We are greateful to the curators of mammal collections at Museu Nacional (UFR]J), Departamento de
Zoologia (UFMG), Museu de Biologia Mello Leitäo, and Museu de Histöria Natural “Capäo da
Imbuia”, who kindly allowed us to study the specimens under their care. We express our gratitute to
Drs. L.M. Pessöa and D.M. TeıxerraA for their help and use of facilities and critical revision of
preliminary versions of this manuscript. We thank G. REED for the revision of the English version and
M. PessöA for the drawings. Work by the authors was partially supported by graduate tellowships
from Conselho Nacional de Desenvolvimento Cientifico e Tecnolögico (CNPq).

Zusammenfassung

Jahreszeitliche Altersstruktur und Fortpflanzung bei Marmosa incana (Lund, 1841)
(Didelphidae, Marsupialia)

Jahreszeitliche Altersstruktur und Fortpflanzungsanzeichen wurden bei 311 Sammlungsexemplaren
des Mausopossums Marmosa incana studiert. Die monatlichen Häufigkeiten relativer Altersklassen,
die nach Zahndurchbruch und -abnutzung festgelegt wurden, deuten auf einen fast vollständigen
Wechsel der Population im Jahreslauf (1 Jahr bei Männchen, 1,5 Jahre bei Weibchen). Die Analyse des
Fortpflanzungszustandes zeigte, daß alle sexuell aktiven Tiere in die beiden höchsten Altersklassen
fallen. Diese Populationsmerkmale und eine im größten Teil des Verbreitungsareals von Marmosa
incana festgestellte dreimonatige saisonale Fortpflanzungsperiode deuten auf einen ungewöhnlichen
Lebenszyklus mit nur einmaliger Reproduktion hin.

Literature

BARNES, R. D. (1977): The special anatomy of Marmosa robinsoni. In: The biology of marsupials. Ed.
by Don Hunsaker Il. New York, London: Academic Press. Pp. 387—412.

Annual age structure and reproductive patterns in Marmosa incana 73

BRAITHWAITE, R. W.; Lee, A. K. (1977): A mammalian example of semelparity. Amer. Nat. 113,
151-155.

Davıs, D. E. (1946): The annual cycle of plants, mosquitoes, birds, and mammals in two Brazilian
forests. Ecol. Monographs. 15, 243-295.

EISENBERG, ]J. F. (1981): The mammalıan radiations: A study in evolution and adaptatıion. Chicago:
University of Chicago Press.

FonseEca, G. A. B.; KreruLır, M. A. M. (1989): Biology and natural history of Brazilian Atlantic
forest small mammals. Bull. Florida State Mus., Biol. Sci. 34, 99-152.

GEMMELL, R. T. (1982): Breeding bandicoots in Brisbane (/sodon macrourus: Marsupialıa.
Peramelidae). Australian Mammalogy 5, 187-193.

GRIFFITHS, M. (1978): The biology of monotremes New York: Academic Press.

GUILER, E. R. (1970): Observations on the Tasmanıan devil, Sarcophilus harrisu (Marsupialia:
Dasyuridae). II. Reproduction, breeding and growth of young. Australian J. Zool. 18, 63-70.
Hunsaker II, D.; SHuupe, D.V. (1977): Behavior of New World marsupials. In: The biology of

marsupials. Ed. by Don Hunsaker Il. New York, London: Academic Press. PP. 95-156.

LEE, A. K., CockBurn, A. (1985): Evolutionary ecology of marsupials. Cambridge: Cambridge
University Press.

LILLENGRAVEN, J. A.; THOMPSsoN, $. D.; McNas, B. K.; PATTon, J. L. (1987): The origin of eutherian
mammals. Biol. J. Linn. Soc. 32, 281-336.

NEwSOME, A. E. (1965): Reproduction in natural populations of the red kangaroo, Megaleia rufa
(Desmarest), in central Australia. Australian J. Zool. 13, 735-759.

NImER, E. (1989): Climatologia do Brasil. Rio de Janeiro: IBGE.

O’Conneı, M. A. (1979): Ecology of didelphid marsupials from northern Venezuela. In: Vertebrate
ecology in the northern Neotropics. Ed. by J. F. EISENBERG. Washington, D.C.: Smithsonian
Inst. Press. Pp. 73-87.

— (1983): Marmosa robinsoni. Mammalian Species 203, 1-6.

OLIVEIRA, J. A.; Lorinı, M.L.; PErsson, V.G. (1992): Pelage varıation in Marmosa incana (Didel-
phidae, Marsupialia) with notes on taxonomy. Z. Säugetierkunde 57, 129-136.

Pine, R. H.; Dargy, P. L.; Matson, J. ©. (1985): Ecology, postnatal development, morphometrics,
and taxonomic status of the short-tailed opossum, Monodelphis dimidiata, an apparently semel-
parous annual marsupial. Ann. Carnegie Mus. 54, 195-231.

Reıc, ©. A. (1964): Roedores y marsupiales del partido de general pueyrredon y regiones adyacentes
(provincia de Buenos Ayres, Argentina). Publ. Mus. Mun. Cienc. Nat., Mar del Plata 1, 203-224.

STALLINGS, J. R. (1989): Small mammal inventories in an eastern Brazilian park. Bull. Florida State
Mus., Biol. Sci. 34, 153-200.

STREILEIN, K. E. (1982): Behavior, ecology and distribution of the South American marsupials. In:
Mammalian Biology in South America. Ed. by M. A. Marzs and H. H. Grnowass. Special Publ.
Ser. Pymatuning Lab. Ecol., Univ. Pittsburgh 6, 231-250.

TATE, G. H. H. (1933): A systematic revision of the marsupial genus Marmosa. Bull. Amer. Mus.
Nat. Hist. 66, 1-250.

TRIBE, C. J. (1990): Dental age classes in Marmosa incana and other didelphoids. J. Mammalogy 71,
566-569.

Woop, D. H. (1970): An ecological study of Antechinus stuartii (Marsupialia) in a southeast
Queensland rainforest. Australian J. Zool. 18, 185-207.

Authors’ addresses: Marıa Lucia LoRINI and JoAO A. DE OLIVEIRA, Segäo de Mastozoologia, Museu
Nacional (UFR]J), Quinta de Boa Vista, s/n°, CEP 20940-040. Rio de Janeiro, R],
Brazil; Vanessa G. PERsson, Museu de Historia Natural “Capäo da Imbuia”,
Rua Prof. Benedito Conceicäo 407, CEP 82810-080, Curitiba, PR, Brazil

Z. Säugetierkunde 59 (1994) 74-81
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Food and feeding habits of insectivorous bats from Israel
By J. ©. WHITAKER, JR., B. SHALMoNn, and T. H. Kunz

Department of Life Sciences, Indiana State University, Terre Haute, USA,
Israel Mammal Information Center, Eılat Field School, Eilat, Israel, and Department of Biology,
Boston University, Boston, USA

Receipt of Ms. 14. 5. 1993
Acceptance of Ms. 21. 9. 1993

Abstract

Small numbers of fecal pellets were examined from nine of the 32 species of insectivorous bats from
Israel. Based on these small samples, Asellia tridens and Pipistrellus bodenheimeri were lepidopteran/
dipteran and lepidopteran/dipteran/coleopteran feeders, respectively; Myotis nattereri fed primarily
on Iygaeids and coleopterans; Otonycteris hemprichi fed exclusively on coleopterans; Pipistrellus
kuhli fed on hymenopterans and coleopterans; Pipistrellus rueppelli and Tadarida teniotis fed mostly
on lepidopterans and coleopterans; Plecotus austriacus fed exclusively on lepidopterans, and
Rhinolophus clivosus was a generalist feeder, taking several different kinds of insect prey. Information
on dietary analysis is supplemented with direct observations on the foraging habits of these bats.

Introduction

Knowledge of an animal’s diet is important for interpreting its ecological role as a predator
and its impact on local environments. Such knowledge is especially important as natural
habiıtats are being altered owing to increased urbanızation, modern agricultural practices
(intense irrigation and pesticide applications), and deforestation. In recent years, the use of
pesticides ın Israel was rated second in the world per capıta (Makın 1989). Thus,
knowledge of their insect prey is important for assessing the potential value of bats ın
controlling insect pests, especially in areas where urban and cultivated lands have replaced
natural habıtats.

There have been a limited number of observations on food habits and foraging behavior
of bats ın Israel (MAaxın 1987; BATEs and HARRISON 1989; HARRISON and BaTtzs 1991;
Yom-Tov 1993; Yom-Tov et al. 1992a, b), and only one substantial report on foraging
behavior (Pipistrellus kuhlıı, BaRAK and Yom-Tov 1989). Yom-Iov’s (1993) analysıs ot
morphological characters of insectivorous bats in the Dead Sea area ıs the most comprehen-
sive account of foraging characteristics of bats in Israel. The purpose of this study is to
present preliminary information on food and feeding habıts on selected insectivorous bats ın
Israel, based primarily on samples of fecal remains collected from bats captured at feeding
sites. These results are supplemented by observations ot foraging behaviour at sites where
bats were captured.

Material and methods

Fecal pellets were collected from nine species of bats, eight of which were captured in mist nets set at
feeding sites. These include Asellia tridens, Myotis nattereri, Pipistrellus bodenheimeri, Pipistrellus
kuhlü, P. rueppelli, Plecotus austriacus, Otonycteris hemprichi, Rhinolophus clivosus, and Tadarıda
tenıotis (feces from the latter species was collected at a roosting site). Following capture, individuals of
each species were placed in holding bags allowing feces to be deposited before releasing the bats at the
site of capture. Collection localities for each species are noted below. Most collection sites were
located in the vicinity of the Dead Sea. A summary of the geography, climate, and vegetation ın this
region is given in YOM-Tov et al. (1992b).

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5902-0074 $ 02.50/0

Food and feeding habits of insectivorous bats from Israel 75

We used fecal analysis to assess food habits so we could release the bats alive after collecting the
feces for analysis. For insectivorous bats, this method is comparable to stomach content analysis
(Kunz and WHITAKER 1983). Contents of fecal samples were identified to famıly and analyzed using
the methods described in WHITAKER (1988). Fach pellet was examined separately and the percent
volume of each food item was estimated visually in each pellet. Most results are presented as overall
percent volume (sum of individual volumes/total volume for the sample x 100) and percent frequency
(number of pellets of occurrence/total number of pellets in sample x 100). Observations of feeding
behavior were made at the time bats were captured and fortuitously at other times. These were
supplemented by unpublished observations and those reported by Yom-Tov et al. (1992a, b). Unless
otherwise noted, means and standard deviations for body masses of bats are from Yom-Tov et al.
(1992b).

Results and discussion

Data on nine species of insectivorous bats known from Israel are summarized below.

Asellia tridens (12 pellets, 2-5 mm long, x = 4.9 mm, from Navit Pools). This bat
ranges across northern Africa from Morocco and Senegal to Egypt and Ethiopia, Zanzibar,
Arabian Peninsula to Pakistan. In Israel bats were captured over a fresh water pool
surrounded by lush vegetation at an oasıs south of the Dead Sea. All contained Lepidoptera
(61,3 % volume, 100 % frequency) and all but one contained Diptera, mostly chironomids
(38.8 % volume, 91.7 % frequency). Based on an analysıs of 147 fecal pellets trom this bat
collected at an abandoned mine (North of Elat), the following prey items (expressed as
percent frequency) were taken: Coleoptera (59), Hymenoptera (21), Orthoptera (14),
Lepidoptera (3), Diptera (2), and Odonata (1).

This medium sized bat (x = 9.8 g + 1.9 g) is known to take large insects (Tettigoniidae,
Blattodea, and Lepidoptera) and transport them to underground shelters where they are
consumed. The diet of A. tridens, as determined in the present study, ıs consistent with ıts
habit of foraging over bodies of water and its use of Doppler-shift compensation
(GuUSTAFSON and SCHNITZLER 1979), which may allow it to use information on the wing-
beat frequency of certain insects to facılıtate prey capture. The echolocating calls have the
highest recorded frequency (around 121 kHz) among the bats ın Israel. One of us
(B. SHALMoN) observed A. trıdens flyıing around street lights in the settlements of Hazeva
and En Gedı. This bat reportedly is nearly extinct in the Mediterranean zone in Israel, and
is confined mostly to drier zones (Makın 1989). This bat ıs a highly desert-adapted species
that reportedly roosts (and feeds) in the vicinity of oases (GAISLER et al. 1972). In Oman it
has been observed flyıng near the ground and feeding in palm groves (Nowak 1991).

Myotis nattereri (2 pellets, both 4 mm long, from Zoarım Cave, Upper Galılee). This
9-11 g bat is a wıdespread species, ranging from northern Europe southward to Morocco,
across central Asia to Japan. In Israel, individuals were captured in dense Mediterranean
maquis, where it was also observed feeding. Four prey items were ıidentified with percent
volumes and frequencies as follows: Lygaeidae (42.5 %, 50%), Coleoptera (42.5 %,
100 %), spider (10 %, 50%) and Diptera (5 %, 50 %).

Compared with similar species of Myotis (e.g. M. capaccınü), the intensity of echoloca-
tion calls ıs very low. It has morphological adaptations for slow, maneuverable tlıght
(NORBERG and RAYNER 1987); ın Scandinavia this bat feeds along the contours of trees,
often among the branches (BAAG®E 1987). The relatively wide diversity of prey taken,
including Diptera, suggests that ıt also feeds near water.

Otonycris hemprichi (11 pellets, 3-7 mm, x = 5.0, from Sapır). Bats were captured over
a fresh water pool in arıd habitat, surrounded by vegetation. All fecal pellets examined
contained 100% Scarabaeidae. This relatively large (x = 19.0 g + 2.0g) long-eared,
desert-dwelling bat ranges from the arıd regions of Morocco and northern Niger through
Egypt and across the northern Arabıan Peninsula to Kazakhastan and Pakistan (Nowak
1991; Harrıson and Bates 1991). This bat flutters like a large butterfly over water and
nearby vegetation (Yom-Iov 1993). In southern Kirghizia (central Asia) this bat has been

76 J. ©. Whitaker, Jr., B. Shalmon, and T. H. Kunz

observed foraging close to rock surfaces (RyBın et al. 1989). Typical of gleaners, ©. hem-
prichi ı is a low- intensity echolocating or whispering bat. At times it probably relies on
vision and passive listening to locate its prey. Based on an analysis of its body mass and
wing morphology (low aspect ratio and low wing loading), FENTON and NORBERG (1988)
postulated this bat should be carnıvorous. In Israel it has been captured while foraging in a
date palm grove and, in nearby Jordan, it has been captured over a small pool in an arid
mountain gorge (BATEs and HARRISON 1989).

Pipistrellus bodenheimeri (14 pellets 2-4 mm long, x = 2.8). Fourteen pellets were
examined from three different localities (Tab. 1). Overall, the most important food items
included Lepidoptera, Coleoptera, and Diptera. Lepidopterans were the major food at two

Table 1. Food as indicated by 14 fecal pellets from Pipistrellus bodenheimeri at three localities in
Israel

Sapir En Gedi Neot Hakikar
18. 5. 1988 3. 3. 1988 3. 7. 1989
25. 4. 1988
Number of Pellets

5 7 2
% vol. % freg. % vol. % freg. % vol. % freg.

Lepidoptera
Chrysomelidae
Scarabaeidae
Insect
Coleoptera
Chironomidae
Diptera

Spider

localities followed by chrysomelid beetles at one and dipterans at the other. Dipterans were
the principal food item followed by coleopterans at the third localıty. This small bat (x =
2.78 + 0.2) is restricted to extreme desert areas in Israel, the Sınai Peninsula and the
southwestern Arabian Peninsula (NowAk 1991; Yom-Tov 1993). It appears to be an
opportunistic predator, although lepidopterans comprised most ot its diet in our sample. It
was collected in open desert habitats, around street lamps in the settlements of Elat, En
Gedi, and over freshwater pools in nearby arıd regions. Although this bat may hibernate
from October through April in Israel, it has been observed feeding in all months of the
year (Yom-Tov et al. 1992a). In winter, the number of bats feeding near street lıghts ıs
considerably smaller than during the summer months. In the Arabian peninsula, this bat
has been observed flying (foraging?) along rows of tamarısk and eucalyptus trees that were
planted around a cultivated area surrounded by desert and acacıa trees (HARRISON and
BREESEIISH):

Pipistrellus kuhlii (5 pellets 2-5 mm, x = 3.4, from Upper Galilee). Parts of winged ants
(alates) were the most important items collected from individuals captured near street
lamps at Sasa (56% volume, 100% frequency), followed by Coleoptera (24%, 60 %),
Chrysomelidae (8%, 20%) Lepidoptera (6%, 60%), Cerambycidae (4%, 20%) and
unidentified insect remains (2%, 20%). One of the most common bats ın Israel, thıs
small, x = 7.0g + 0.5, species is known from southern Europe, southwestern Asia,
northern, eastern, and southern Africa, and the Canary Islands (Kınapon 1974; NOWAK
1991). Its feeding activity occurs mostly within the first three hours following sunset and is
characterized by low, fast feeding flights in open areas. In urban areas it feeds on flyıng
insects attracted to street lights (HAFFNER and Sturz 1985-86) and over water (VERNIER
1989). In east Africa, Kınapon (1974) noted that this bat was attracted to insects swarming
around ripe fruits on a peach tree. SCHNITZLER et al. (1987) indicated that it relies on low

Food and feeding habits of insectivorous bats from Israel VL

frequency echolocation calls (35-40 kHz), which is typical of bats feeding in open areas
(NEUWEILER 1983).

BarAK and Yom-Tov (1989) suggested that the echolocation calls produced by P. kuhlii
caused some of the flying ınsects attracted to street lights (none were ıidentified) to disperse
and thus improve an individual bat’s abılity to capture prey. Evidence for this hypothesis
was based on the observation that feeding rates were highest at those sites whee foraging
group size increased from one to five individuals. Feeding rates of individual bats were
lower when fewer bats were present, suggesting that prey capture in P. kuhlii is enhanced
by feeding ın small groups.

Pipistrellus rneppelli (7 pellets 2-7 mm, x = 3.3, from Navit Pools). Six of seven pellets
examined contained 100 % lepidopterans. The seventh contained 65 % Lepidoptera and
35 % (volume) unidentified material. This 7.0 g # 0.5 g Afrotropical species ranges from
Egypt to Senegal and south to Angola and Botswana (HARRISON and BATEs 1991), and
southern Iraq (Yom-Tov et al. 1992b). In Israel thıs bat was captured while flying over
small freshwater pools. In East Africa, it is known from savannah and arid regions, where
it has been observed flying (feeding?) along rivers at dusk (Kınapon 1974; SKINNER and
SMITHERS 1990). LanG and CHapin (1917) reported P. rueppelli tlyıng (feeding?) adjacent
to oil palms ın riparian habitat. SKINNER and SMITHERS (1990) noted that P. rueppelli is also
associated with rivers and swamps.

Plecotus austriacus (5 pellets 3-6 mm, x = 4.0, from Sapır). Bats with long ears (e.g.,
Plecotus and some other genera) are often lepidopteran specialists (FENTON and NORBERG
1988). Judging from our findings, P. austriacus ıs indeed a lepidopteran specialist, as all
pellets examined contained 100% Lepidoptera. This low-intensity echolocating bat (x =
6.7 # 1.1 g) is known from southern Europe and Northern Africa, westward to Mongolia
and western China, Cape Verde Islands and Senegal (Nowak 1991). As with other
members of its genus, P. austriacus is known to hover and glean insects from vegetation
and other surfaces (ROBERTS 1977). Its short, broad wings, and low flight speed (NORBERG
and RAYNER 1987) presumably facilitate captures of resting insects in these situations.
Judging from its small sıze and weak dentition, ROBERTS (1977) suggested that this bat
probably feeds “mainly on smaller moths, spiders and lacewings (Planipennia sp.)”.

Rhinolophus clivosus (5 pellets 2-7 mm, x = 4.2, from En Gedi). Even though the
available fecal sample was extremely small, sıx separate prey items were taken: Coleoptera,
probably Chrysomelidae (53 %, 80%); Hymenoptera (31%, 20%); Lepidoptera (11%,
80%); Insect (2%, 40%); Scarabaeidae (2%, 20%) and Lygaeidae (1%, 20%). This
medium size bat (x = 11.0 g # 2.0 g) is known from central and southwestern Asia, and in
Afrıca from Liberia and Algeria eastward to Cameroon (Nowak 1991), and southward to
Southern Africa (HArrıson and BATEs 1991; SKINNER and SMITHERS 1990). When
foraging it flies below tree-top level (RAUTENBACH 1982) and low around trees and shrubs.
In Israel, individuals were captured as they flew low between trees in the desert oasis of En
Gedi. Elsewhere, the food items taken by this bat includes moths and small beetles
(SKINNER and Smithers 1990), although it sometimes eats large beetles (Kınapon 1990).
This bat establishes feeding roosts on branches of trees and beneath verandahs of houses
where individuals cull hard parts of insects before eating them (RAUTENBACH 1982).

Tadarida teniotis. Nine samples were available, all from Karmiel (Lower Galilee). Fifty
pellets were selected from each of these samples. Pellets were selected in order to include all
sızes, from largest to smallest. Pellets ranged from 3 to 10 mm long, but averaged about
6t0 7 mm. Prey items as indıcated by these nine samples are shown in table 2. Lepidopte-
rans were the most important food, ranging from 65.4 % to 87.6 % by volume of the total
material examined. Of the 450 pellets examined, 442 or 98.2 % of them contained moths,
whereas 215 of these pellets or 47.8% contained 100% of this material. Coleopterans,
especially ground beetles (Carabidae) and June beetles (Scarabaeidae), were the second
most important food items. The amount of beetles eaten ranged from 6.8 % to 27.3 %

J. ©. Whitaker, Jr., B. Shalmon, and T. H. Kunz

78

LT
ge

81
(72) (ZaEp)
9'/8 7'58

a EIETE CR e

2
FT
(a)
EZ

661 I ET

6'+
(8°9)
0'/8

661 € 81

eg

g']] F'sI
(677) (E/o)
759 9:69

EA

074

LO

as
Diß
07
(9'ST)
Y'78

661 € SI

'sasayyussed ul u9A1s ale s9aUum[oOA e191d03]09 TROLL,

0'8
(9/1)
€9L

1661 '6 "€

co’o

10)

L'8
(S’I7)
EC

l661 '8 ZI

(so1ru1) VNIAVOV
199sUJ paınuaprun

VAINIWIHIT
Jepıgo1swad]

VNALIOANAN
VYA.LdIA

AVaIDINYoJ
VYAZLIONAWAH
sepı[kg)
VYZLLIOHLAO
ersıdıupy 'pıun
prwogeIuag U9919
SERHISPID,
oepare3AT
VYALLIOWOH/VNALIIWIH
e19d0o9[05 'pruNn
9epruornsAnd
JepIpwosAlyy
JepTuOLIgaU2],
Jepıgeiey
Jeploegeiesg
„VYZLI09109
VYALd4OAIdIT

saıeq

sojep JU9IOJHP 6 UO yIe9 syajppd 0g JO sapdures 9UTU Aq p>yre9ıpul se “peis] "Pprwäeyy uo (JumjoA %) STJOIU2}J EPIIEPEL JO POOF 'Z A1gP 1

Food and feeding habits of insectivorous bats from Israel 79

volume. Crickets (Orthoptera: Gryllidae) also were regularly eaten, with consumption
ranging upwards to 7.3% volume.

This relatively large insectivorous bat (x = 18.5 + 2.0g) has a wide distribution,
ranging from Madeira, the Canary Islands, Morocco and the Iberian peninsula, eastwards
through North Africa and southern Europe to southern China, Tiawan and Japan
(HARRISON and BaTEs 1991). Tadarıda teniotis ıs a fast, high flyıng bat (GAISLER and
Kowauskı 1986) often observed 20-50 m above the ground where it feeds largely on
moths. In Israel it is commonly observed flyıng above settlements and cities, feeding on
such insects attracted to street lights. The high intensity, low frequency (12-14 kHz)
echolocation calls of thıs bat are audible to humans. It is one of the most beneficial bats to
farmers, which suffer heavy damage to crops from noctuid moths (e.g., Spodoptera
htoralıs, Argotıis ipsilon, and Earıas insulana). Other reports indicated that members of the
genus Tadarida are often generalist feeders, but consume large numbers of moths (e.g.,
Ross 1967; Kunz et al. 1994).

Based on our preliminary dietary analysis of nine species of insectivorous bats from
Israel, some appear to specialize either on coleopterans (beetles) or on lepidopterans
(moths), whereas others are generalist predators. As ın other species of insectivorous bats,
food habits can be expeced to vary depending on the locality, season, and ability of the bat
to detect (visually or accoustically) certain types of insects, and morphological characteris-
tics. Using an analysis of morphological characteristics, Yom-Tov (1993) placed 15 species
of bats known from the Dead Sea area into three feeding guilds as follows: Guild 1
included species with long wing tips, third digits and articulating metacarpals 1.3 times, or
more, longer than the forearm, and normal-size ears (forearm/ear ratio was 2.5 or larger).
Species in the present study that were included in Yom-Tov’s guild 1 include Asellia
tridens, Rhinolophus clivosus, Pıpistrellus bodenheimeri, P. rneppell, P. kuhli, and
Tadarıda teniotis. These bats feed in a variety of habitats, including high altitudes
(Tadarida) at medium to low heights (Pipistrellus), or forage amongst vegetation (Asellia,
Rhinolophus). Yom-Tov’s guild 2 included species with wing tips that are similar to species
listed in Guild 1, but each have exceptionally large ears (exceeding 30 mm), which are
twice the size of other bats included in his analysis. The two species which Yom-Tov
included in Guild 2, and that we also examined, include Otonyceris hemprichi and Plecotus
austriacus. Both species fly low and slowly and produce low-intensity echolocation calls
(1.e., whispering bats). None of the species listed in Yom-Tov’s guild 3 were included in
our study. Although morphological data may offer general insights into the feeding
ecology of an anımal (e.g., FREEMAN 1988; FENTON and NoRBERG 1988; Yom-Tov 1993),
it cannot be used to predict the behavioral or dietary variability such as we observed in the
present study.

Acknowledgements

We wish to thank D. Franko (Karmiel) for collecting fecal samples of Tadarida teniotis and
Y. Moskın for sharing observations on echolocation-call characteristics and feeding habits of Myotis
nattereri. Y. Yom-Tov (Tel Aviv University) kindliy commented on an earlier draft of this manuscript.
This project was facilitated by a travel grant from the Conanima Foundation to THK.

Zusammenfassung

Nahrung und Ernährungsverhalten einiger insektivorer Fledermäuse aus Israel

Kleine Proben von Kotpillen von 9 der 32 in Israel vorkommenden insektivoren Fledermausarten
wurden analysiert. Nach diesen Proben fraß Asellia tridens (n = 12) überwiegend Lepidoptera,
Diptera und Coleoptera. Myotis nattereri ernährte sich vorwiegend von Lygaeidae und Coleoptera;
Otonycteris hemprichi ausschließlich von Coleoptera; Pipistrellus kuhlii von Hymenoptera und Co-
leoptera; Pipistrellus rueppelli und Tadarida teniotis überwiegend von Lepidoptera und Coleoptera;
Plecotus austriacus ausschließlich von Lepidoptera, und Rhinolophus chvosus erwies sich als Generalist

80 J. ©. Whitaker, Jr., B. Shalmon, and T. H. Kunz

und nahm viele verschiedene Insekten als Nahrung. Ergänzend werden Direktbeobachtungen über
das Suchverhalten dieser Fledermäuse mitgeteilt.

References

BAAGBE, H. J. (1987): The Scandinavian bat fauna: adaptive wing morphology and free flight in the
field. In: Recent advances in the study of bats. Ed. by M. B. Fenton, P. Racer, and J. M. V.
RAYNER. Cambridge: University Cambridge Press. Pp. 57-74.

BARAR, Y.; Yom-Iov, Y. (1989): The advantage of group hunting in Kuhl’s bat Pipistrellus kuhli
(Microchiroptera). J. Zool. (London) 219, 670-675.

BATEs, P. J.; Harrıson, D. L. (1989): New records of small mammals from Jordan. Bonn. Zool.
Beitr. 40, 223-226.

FENToN, M. B.; NORBERG, U. M. (1988): Carnıvorous bats. Biol. J. Linn. Soc. 33, 383-394.

FREEMAN, P. W. (1988): Frugivorous and anımalıvorous bats (Microchiroptera): dental and cranial
adaptations. Biol. J. Linn. Soc. 21, 387408.

GAISLER, J.; KowaLskı, K. (1986): Results of the netting of bats in Algeria (Mammalıa, Chiroptera).
Vest. Ceckoslov. Spol. Zool. 50, 161-173.

GAISLER, J.; MADKOUR, G.; PELIKAN, J. (1972): On the bats (Chiroptera) of Egypt. Acta Sci. Nat.
(Brno) 6, 140.

GUSTAFSON, Y.; SCHNITZLER, H. (1979): Echolocation and obstacle avoidance in the hipposiderid bat,
Asellia tridens. J. Comp. Physiol. 131, 161-167.

HAFFNER, M.; Sturz, H. P. (1985/86): Abundance of Pipistrellus pipistrellus and Pıpistrellus kuhli
foraging at street lamps. Myotis 23-24, 167-168.

HarRrıson, D. L.; BATEs, B. J. (1991): The mammals of Arabia. London: Harrison Zool. Mus. Publ.

Kınsoon, J. (1974): East Afrıcan Mammals: insectivores and bats. Vol. IIA. Chicago: University of
Chicago Press.

— (1990): Arabian mammals, a natural history. London: Academic Press.

Kunz, T. H.; WHITARER, ]J. O., Jr. (1983): An evaluation of fecal analysis for determining food habits
of insectivorous bats. Can. J. Zool. 61, 1317-1321.

Kunz, T. H.; WHITAKER, ]. O., Jr.; WADonoL1, M. D. (1994): Dietary energetics of the insectivorous
Mexican free-tailed bat (Tadarıda brasiliensis) during pregnancy and lactation. Oecologia (in
press).

Lang, H.; CHapin, J. P. (1917): The American Museum Congo Expedition collection of bats. II: field
notes. Bull. Amer. Mus. Nat. Hist. 37, 476-496.

Maxın, D. (1987): [The insectivorous bats (Microchiroptera) of Israel: distribution and biology].
Re’em (Oryx) 6, 12-76 (in Hebrew).

— (1989): The status of bats ın Israel. In: European bat research 1987. Ed. by V. Hanak,
I. HoRACEK, and J. GAISLER. Praha: Charles University Press. Pp. 403-408.

NORBERG, U. M.; RAyneERr, J. M. V. (1987): Ecological morphology and flight in bats (Mammalıa;
Chiroptera): wing adaptations, flight performance, foraging strategy and echolocation. Phil.
Trans. Roy. Soc., Ser. B 316, 335427.

NEUWEILER, G. (1983): Echolocation and adaptivity to ecological constraints. In: Neuroethology and
behavioral physiology. Ed. by F. Huger and H. Markt. Berlin: Springer Verlag. Pp. 280-302.

Nowak, R. M. (1991): Walker’s Mammals of the World. Vol. I. 5th ed. Baltimore: Johns Hopkins
University Press.

RAUTENBACH, 1. L. (1982): The mammals of the Transvaal. Ecoplan Monograph 1, 111-211.

Roßerrs, T. J. (1977): The mammals of Pakistan. London: Ernest Benn.

Ross, A. (1967): Ecological aspects of the food habits of insectivorous bats. Proc. West. Found. Vert
Zool. 4, 205-264.

Rysın, $S. N.; HoRAcCER, I.; CERvEnY, J. (1989): Bats of southern Kirghizia: distribution and faunal
status. In: European bat research 1987. Ed. by V. Hanak, I. HoRACER, and J. GAISLER. Praha:
Charles University Press. Pp. 421441.

SCHNITZLER, H.; KALko, E.; MILLER, J.; SuLykk£e, A. (1987): The echolocation and hunting
behaviour of the bat, Pıpistrellus kuhlu. J. Comp. Physiol. 161, 267-274.

SKINNER, J. D.; SMITHERS, R. H. N. (1990): The mammals of the Southern African subregion.
Pretoria: University Pretorıa.

VERNIER, E. (1989): Ecological observations on the evening flights of Pipistrellus kuhlii in the town of
Padova (Italy). In: European bat research 1987. Ed. by V. Hanak, I. HoRACEK, and ]J. GAISLER.
Praha: Charles University Press. Pp. 537-541.

WHITAKER, ]J. O., Jr. (1988): Food habiıts analysis of insectivorous bats. In: Ecological and behavioral
methods for the study of bats. Ed. by T. H. Kunz. Washington, D. C.: Smithsonian Institution
Press. Pp. 171-189.

Yom-Tov, Y. (1993): Character displacement among the insectivorous bats of the Dead Sea area.
J- Zool. (London) 230, 347-356.

Food and feeding habits of insectivorous bats from Israel 81

Yom-Tov, Y.; Maxın, D.; SHALMoN, B. (1992a): The biology of Pipistrellus bodenheimeri (Micro-
chiroptera) in the Dead Sea area of Israel. Z. Säugetierkunde 57, 65-69.

— — — (1992b): The insectivorous bats (Microchiroptera) of the Dead Sea area, Israel. Israel
J. Zool. 38, 125-137.

Authors’ addresses: Prof. J. OÖ. WHITAKER, Jr., Department of Life Sciences, Indiana State Univer-
sıty, Terre Haute, Indiana 47809, USA; Dr. B. Suarmon, Israel Mammal
Information Center, Eilat Field School, Eilat, 88101, Israel, and Prof. T. H.
Kunz, Department of Biology, Boston University, Boston, MA 02215, USA

Z. Säugetierkunde 59 (1994) 82-86
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Seasonal food habits of coyotes, Canis latrans, in the Bolsön de
Mapimi, Southern Chihuahuan Desert, Mexico

By Lucına HERNANDEZ and M. DELIBES

Instituto de Ecologia, Durango, Mexico and Estacion Biolögica de Donana, Sevilla, Espana

Receipt of Ms. 01. 03. 1993
Acceptance of Ms. 16. 09. 1993

Abstract

Studied the seasonal foods of coyotes, Canis latrans, at the Biosphere Reserve of Mapimi in the
Southern Chihuahuan Desert, Mexico, by analysis of 508 faeces (scats) collected between March 1985
and November 1986. Lagomorphs were the most frequent food, occurring in 49% of the samples,
followed by fruits (33 %) and rodents (32 %). There were important seasonal and interannual changes
in food habits. As a rule, lagomorphs were the most frequent item in autumn and winter, but fruits
(mainly Opuntia sp.) predominated in summer and rodents (mainly Neotoma albigula) in spring.
Birds and reptiles were consumed to a low proportion and ungulates were eaten only when carcasses
were available. Our data suggest that the coyoe in this area behaves as a selective predator of rabbits,
occasionally using abundant alternative foods in an opportunistic way.

Introduction

The ecology of coyotes Canıs latrans is among the best known of all carnıvores (BEKOFF
1982). However, most investigations have been carried on ın the United States and
Canada. Thus, information about the food habits of coyotes ın the southern region of its
range is rather scarce. Studies on coyotes’ diet in Mexico have been conducted at cattle
ranches of Chihuahua, where the maın food was carrıon (PEREZ GUTIERREZ et al. 1982)
and vegetables (VELA 1985), and in pineoak forests of the Sierra Madre Occıdental, where
the spring-summer staple prey were rodents, arthropods and berries (DELI1BEs et al. 1989).
In addition, most ecological research on coyotes has been conducted in agricultural
regions, where this species ıs usually considered as a pest (VoıGT and BERG 1987). Thus,
studies in desert areas are lacking. In a previous investigation, HERNANDEZ et al. (1993)
reported on the autumn food habits of coyotes in some areas of the deserts of Sonora and
Chihuahua. The aım of this study ıs to report on seasonal food habits of this species ın the
Mapımi Biosphere Reserve of the Southern Chihuahuan desert.

Study area

The study was made in the surroundings of the field laboratory of the Mapımi Biosphere Reserve,
located in the Mapimi Bolsön area, close to the vertex formed by the Mexican states of Chihuahua,
Coahuila and Durango (ca. 26°%40’ N, 103°45’ W). The area ıs a flat plain (average altitude 1100 m
above sea level) with poor drainage. Vegetation ıs low and scattered, dominated by creosote bush
(Larrea tridentata), mesquite (Prosopis glandulosa), prickly-pear (Opuntia sp.) and agave (Agave sp.).
A recent vegetation map of the area has been published by MonTANA (1988). Climate is semiarıd, with
irregular summer rains (263 mm per year on average) and mean monthly temperatures ranging from
12°C ın January to 28°C in June and July (Corner 1988).

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5902-0082 $ 02.50/0

Seasonal food habits of coyotes, Canis latrans 83

Material and methods

By walking along a stretch of dirt roads of approximately 10 km in the reserve every month, from
March 1985 to November 1986, we collected a total of 508 coyote faeces. Faeces (scats) were identified
by sıze, scent and the nearby presence of coyote tracks. We used hairs, teeth, bones, feathers, scales,
any other hard structures, and seeds to identify prey species by comparison with a reference
collection. We were unable to differentiate the remains of black-tailed jackrabbits (Zepus californicus)
from those of desert cottontails (Sylvilagus auduboni). Number of occurrences and percent frequency
of occurrence (number of occurrences X 100/number of scats) were obtained for each food species or
group. These data dıd not accurately reflect the weight of ingested material, but they are usually
considered to yield a good representation of food habıts (e.g., NIEBAUER and RonGsTAD 1977).

Monthly samples were grouped into four seasons: spring (March to May), summer (June to
August), autumn (September to November), and winter (December to January). Seasonal differences
in the occurrence of particular items were determined by contingency-table analysis (G-test) (SOKAL
and RoHLF 1981).

Results

Lagomorphs were the most frequently occurring food item, followed by fruit and rodents.
Ungulates, birds, reptiles, arthropods and garbage (plastic material) were only rarely
ingested (Tab. 1).

By pooling data corresponding to seasons from 1985 and 1986, ıt was found that
lagomorphs were the most frequent item in autumn (45.3 % of occurrences) and winter
(57.3 %), while they occupied a second place in spring (34.1%) and summer (28.5 %).
However, interannual differences were significant (Table). Seasonal frequency of occur-
rence of lagomorphs was not related to estimations of lagomorph abundance in the field.

Fruits were the most frequent item in the scats collected in summer (50.3 % of
occurrences), but ranged on third place ın spring (18.4 %), autumn (16.1 %), and winter
(12.9 %). Prickly-pear was the most commonly consumed fruit, followed by mesquite.
Both were eaten in high proportions in summer, when they were riıpe. Turk head
(Hamatocactus hamattocanthus) fruit was consumed in winter, while lotebush (Ziziphus
obtusifoha) was frequent in summer (1985) and spring (1986) (the fructification of this
shrub shows a considerable plasticity in response to autumn precipitation; FOSTER et al.
1984). Althorn (Castella texana) fruıt was consumed during the first spring.

Rodents were the most frequent item in the scats collected in spring (35.7 %), and they
occupied a second place in autumn (32.8 %) as well as in winter (21 %), and the third place
in summer (13.5 %). Woodrats (Neotoma albigula) were the most commonly consumed
rodents, especially in spring (1985), followed by kangaroo rats (Dipodomys sp.), deer mice
(Peromyscus sp.) and pocket mice (Perognathus sp.). Ground squirrels (Spermophilus
spilosoma and Spermophilus mexicana) primarıly appeared during the first autumn.

Birds and reptiles were consumed to a low proportion, and there were no seasonal
trends in their utilization (Table). Arthropods (mainly beetles and grasshoppers) were
consumed especially ın spring and summer. Carcasses of calves were eaten during the
second spring, when there was a high mortality of cattle caused by disease. Remains of
mule deer (Odocoileus hemionus) appeared in the faeces mainly in winter, at the same time
when we observed some coyotes feeding upon two carcasses of deer killed by mountain
lions (Felis concolor).

Discussion

Our data suggest that lagomorphs were the staple prey for coyotes in Mapimi, as it occurs
in other areas (SHORT 1979). Nevertheless, a seasonally high availability of alternative prey
(mainly prickly-pears in summer) will produce a change to these temporally abundant
foods, even when leporids are present at a high level of population size (e.g., in summer
1986). As found by other authors (MurıE 1945; Jomnson and Hansen 1977, 1979), it

Lucina Hernandez and M. Delibes

84

su 6°9
SUSE
«YEl

soneA\ 9

so
RE

2
Sl
SE
el
6C
LEG
IST
9°8
L’S

N ON
oO

d
61
9
SC
II
98

(805)
TeoL

ovwvooooo0

-omvovo no

(eg)
98,LV

uesıpugis 10U = su °C 0 > d= x '100 >d= xx: '100 >d = x.
"pawnsuo9 Aaıd Jo uonıodo1d ayı ur sporiad UIIMIIG SIIUIADFFIP SOYEIIPUT 3599 I) '(S999eF Jo Jaquınu (97 x waı y9e9 JO SIIUFLINIIO FO J9quunu) J9U9LIM9I0O FO
Aauanba.ıy je303 = O, :uoseas y9e9 ul p>sAfeue s999eF Jo 1aquımN] = N '986] Ie9A = 98 !C86] IeaA = Gg -I9Ur N, = LA unıny = LV “owung = WS ‘3undg = JS

vowno%xvro
-D-OON OO
—
nnoomn«
NOoooo—-- oO

oO
—

anvoooou ı N
au oo ru.
TATnooxvro-
nNYNvo-o
nnnoTowmooeı«

oononN
—
nn Oo
N
—
N

OFEN;
N Oo
N

(02) (SZ) (TOT) (#9) (22)

I8ENS 98dS S8.LM S8LV S8WNS$

s93J0A09 JO S9IJ9EJ IJY} UT SPOOJ JU9I9JJIp JO SIUIILINIIO

on nanm«r

(89)
Ggds

sofge333A WO
vuvx91 v1]945v7
vuvXxa4 v1]918P7
vnofisnıgo sngdızız
snqzup9ogpwvg SnPP9owWpL]
(ds vrundog
vsojnpup3 sıdoso4]
ynag
spodosyuıy
sopnday
Sahel
PermiSpiun
"ds smIsdwuoAa.]
(ds sngavu8o4ag
‘ds snngdouuads
(ds sdwopodıq
pjnZ1qjp vwoI09N
sIU9PoYy
sydıowoseT
snuonuag sno1l090PO
snAnv] sog
sayepndun

N

uoseag

Seasonal food habits of coyotes, Canis latrans 85

appears that coyotes seek lagomorphs or some other abundant food and make it their basic
prey throughout the year. This fact suggests that they are selective rather than opportunis-
tic feeders. The high occurrence of leporids in the diet during the winter of 1985, when
being relatively scarce ın the study area, supports this statement. Conversely, changes in
the abundance of alternative prey were paralelled by seasonal variations in the frequency of
occurrence of these foods in samples, as noted for rodents in a similar area in Texas
(WINDBERG 1985). This foraging behaviour (a selective predator occasionally using abun-
dant alternative foods in an opportunistic way) is similar to that reported for the European
badger (Meles meles) by Kruuk and ParısH (1981).

Prickly-pears and woodrats were important buffer prey for coyotes in Mapimi. We
found that two radiotracked coyotes in the area showed a strong preference for a
structurally heterogeneous habitat wıth dominance of mesquite, creosote bush, galleta
grass (Hilarıa mutica) and prickly-pear, where woodrats (feeding on prickly-pears) were
especially abundant (GRENOT and SERRANO 1981). This habiıtat type ıs relatively rare ın the
Reserve, but it seems to be preferred by coyotes, probably because of the abundance of
these foods.

Our results differ from those of PEREZ-GUTIERREZ et al. (1982) probably due to the
high cattle raising actıvity in their study areas. They found carrıon as the main food item
for coyotes, followed by lagomorphs and rodents. VELA (1985) obtained results similar to
the present data ın one of three areas that she studied, vegetables being the main item
(followed by lagomorphs) in the other two.

In conclusion, lagomorphs are an important food for coyotes in the arıd zones of north-
central Mexico, their relative role ın the diet changing according to the abundance of buffer
foods. In areas such as the Mapimi Biosphere Reserve, where exploitation is very low,
lagomorphs are the basıc prey of coyotes, and changes in the density of this predator can
probably be expected to follow changes in leporıd abundance (CLAark 1972; ToDD et al.
1981). The effects of coyotes on large wild and domestic ungulates in these areas seem to be
scarce, their consumption being probably related to cattle diseases and mountain lion
predation on deer.

Acknowledgements

This research was supported by the Instituto de Ecologia within the project “Mapimi” and by an
agreement between the Mexican CONACYT and the Spanish CSIC. We gratefully acknowledge the
field assistance of A. L. HERRERA, A. HERRERA and F. HERRERA. R. Ru1z DE Esparza helped in the
identification of seeds and F. PiZARRO in the preparation of hair slides. Dr. E. EzcuRrA provided
helpful suggestions for improvement of the manuscript. N. BUSTAMANTE revised the English and
C. KELLER prepared the German summary.

Zusammenfassung

Jahreszeitliche Ernährungsgewohnheiten von Kojoten, Canis latrans, im Bolsöon- de Mapimi-Reservat,
südliche Chihnahna-Wüste, Mexiko

Die Ernährung von Kojoten, Canis latrans, wurde im Biosphäre-Reservat Mapimi in der südlichen
Chihuahua-Wüste untersucht anhand der Analyse von 508 Kotproben, die zwischen März 1985 und
November 1986 monatlich eingesammelt wurden. Lagomorpha kamen häufiger vor und wurden in
49% der Proben gefunden. Es folgten Früchte (33%) und Nagetiere (32%). Wichtige intra- und
interannuale Änderungen in den Ernährungsgewohnheiten wurden festgestellt. Normalerweise stell-
ten Lagomorpha den Hauptanteil der Nahrung im Herbst und Winter, während Früchte (hauptsäch-
lich Opuntia sp.) ım Sommer, und Nagetiere (vorwiegend Neotoma albigula) im Frühling gefressen
wurden. Vögel und Reptilien bildeten nur einen geringen Anteil der Nahrung, und Ungulaten wurden
nur als Aas gefressen. Unsere Daten lassen vermuten, daß Kojoten in diesem Gebiet eine selektive
Bevorzugung von Kaninchen zeigen, sich aber auch in opportunistischer Weise von zahlreichen
anderen Tieren und Pflanzen ernähren.

86 Lucina Hernandez and M. Delibes

References

BEKOFF, M. (1982): Coyote (Canis latrans). In: Wild Mammals of North America. Ed. by ]J. E.
CHAPMAnN and G. A. FELDHAMMER. Baltimore: John Hopkins Univ. Press. Pp. 447-459.

CLARK, F. W. (1972): Influence of jackrabbit density on coyote population change. J. Wildl. Manage.
36, 345-356.

CorNET, A. (1988): Principales caracteristiques climatiques. In: Estudio integrado de los recursos
vegetaciön, suelo y agua en la Reserva de la Biosfera de Mapimi. I. Ambiente natural y humano.
Ed. by C. Montana. Mexico: Instituto de Ecologia, Pp. 45-76.

DELIBEs, M.; HERNÄNDEZ, L.; HırAaLDo, F. (1989): Comparative food habits of three carnivores in
western Saszı Madre, Me Zr Säugetierkunde 54, 107-110.

FOSTER, M. A.; CIFRES, C. J.; JacoBy, P. W. (1984): Phenological development of lotebush, Ziziphus
obtusifoha (Rhamnaceae), in North Texas. Southwest. Nat. 29, 516-518.

GRENOT, C.; SERRANO, V. (1981): Ecological organızation of small mammal communities at the
Bolsön de Mapimi (Mexico). In: Ecology of the Chihuahuan desert. Ed. by R. BARBAULT and
G. HALFFTER. Mexico: Instituto de Ecologia. Pp. 89-100.

HERNANDEZ, L.; DELIBEs, M.; Hırardo, F. (1993): Role of reptiles and arthropods in the diet of
coyotes in extreme desert areas of Northern Mexico. J. Arıd Environm. (in press).

Jot#nson, M. K.; Hansen, R. M. (1977): Foods of coyotes in the lower Grand Canyon, Arizona.
J. Arız. Acad. Scı. 12, 81-83.

— — (1979): Coyote food habits on the Idaho National Engineering Laboratory. J. Wildl. Manage.
43, 951-956.

Kruuk, H.; ParısH, T. (1981): Feeding specialization of the european badger Meles meles in Scotland.
J. Anım. Ecol. 50, 773-788.

MOoNTANA, C. (1988): Las formaciones vegetales. In: Estudio integrado de los recursos vegetaciön,
suelo y agua en la Reserva de la Biosfera de Mapimi. I Ambiente natural y humano. Ed. by
C. MoNTANA. Mexico: Instituto de Ecologia. Pp. 167-197.

Muri, O. J. (1945): Notes on coyotes food habits in Montana and British Columbia. J. Mammalogy
26, 3340.

NIEBAUER, T. J.; RONGSTAD, O. J. (1977): Coyote food habits in northwestern Wisconsin. In:
Proceedings of the 1975 Predator Symposium. Ed. by R. L. PrırLıps and C. JonkeL. Missoula:
Montana Forest and Conservation Exp. Stn. Pp. 237-251.

PEREZ-GUTIERREZ, C.; FIERRO, L. C.; TREVINO, J. C. (1982): Derermnacion dela composiciön de la
dieta del coyote (Canis latrans Say) a traves del ano en la regiön central de Chihuahua por medio
del anälisis de cotnenido estomacal. Pastizales 13, 2-15.

SHORT, H. L. (1979): Food habits of coyotes in a semidesert grass-shrub habiıtat. Forest Service U.S.
Dept. Agriculture. Research Note RM-364, 14.

SoRKAL, R. R.; ROHLF, F. J. (1981): Biometry. 2nd ed. San Francisco: W. H. Freeman and Co.

Toop, A. W.; Kerth, L. B.; FiscHer, C. A. (1981): Population ecology of coyotes during a
fluctuation of snowshoe hares. J. Wildl. Manage. 45, 629-640.

Vera, C. E.L. (1985): Determinacıön de la composiciön de la dieta del coyote Canis latrans Say, por
medio del anälisis de heces en tres localidades del estado de Chihuahua. Tesis de Licenciatura,
Universidad Autönoma de Nuevo Leön, Monterrey, Mexico.

VoıcT, D. R.; Berc, W. E. (1987): Coyote. In: Wild Furbearer Management and Conservation in
North America. Ed. by M. Novak, J. A. BAkER, M. E. OBBARD, and B. MArLLocH. Ontario:
Ministry of Natural Resources., Pp. 345-357.

WINDBERG, L. A. (1985): Coyote-prey investigations in South Texas: Progress report. Predator
Ecology and Behavior Project. U.S. Fish and Wildlife Service. Denver Wildlife Research Center.

Authors’ addresses: Lucına HERNÄNDEZ, Instituto de Ecologia, Apdo. Postal 632, 34000-Durango,
Dgo., Mexico, and MıGUEL DeLIBEs, Estacıon Biolögica de Donana, CSIC,
Apdo. 1056, E-41080 Sevilla, Spain

Z. Säugetierkunde 59 (1994) 87-104
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Hyoid structure, laryngeal anatomy, and vocalization in felids
(Mammalia: Carnivora: Felidae)

By G. PETERS and M. H. Hast

Zoologisches Forschungsinstitut und Museum A. Koenig, Bonn, Germany, and
Northwestern University Medical School, Chicago, USA

Receipt of Ms. 8. 3. 1993
Acceptance of Ms. 17. 9. 1993

Abstract

Two types of hyoid structure are found in the Felidae. In five species it contains a cartilaginous
ligament, in all other species it ıs completely ossified. Tradıtionally, ıt has been hypothesized that the
presence or absence of roaring and purring in the acoustic repertoires of Felidae is correlated with
these differences in the structure of the hyoid. This character complex has been used as a major
criterion in the systematics of the Felidae. The present study brings together all currently available
data to test this hypothesis, and discusses new findings on laryngeal anatomy with respect to
vocalization and the hyoidean apparatus in the famıly Felidae.

In the Felidae an incompletely ossified hyoid does not automatically cause a species’ ability to roar
as this vocalization is restricted to only three of the five species wıth this hyoid type. All felid species
which have been proven to purr have a completely ossified hyoid, but definitive evidence of this
vocalization is still lacking in many species wıth this type of hyoid. Therefore, it is not possible at
present to decide whether a fully ossified hyoid automatically causes a species’ ability to purr. A
further type of vocalızation is restricted to those four felid species wıth a vocal fold morphology
differing from that of all other species of the family.

Introduction

RICHARD Owen (1834) was the fırst to hypothesize a correlation between hyoid structure
(see Figs. 1-3) and the occurrence of the specific type of vocalization called roaring in
certain species of the Felidae. Based on the differences he found in the structure of the
hyoid between the lion (Panthera leo) (felid taxonomy used in this publication follows
Honackt et al. [1982], unless otherwise stated) on the one hand, and other species
investigated (cheetah-Acınonyx jubatus, caracal-Lynx caracal, flat-headed cat-Felıs
planiceps, and domestic cat-F. silvestris f. catus) on the other, he pointed out (p. 129) that
for the lion:

“The larynx is consequently sıtuated at a considerable distance from the posterior
margin of the bony palate; but the soft palate is prolonged backwards to opposite the
aperture of the glottis, and the tongue is proportinately increased in length. Thus a
gradually expanding passage leads from the glottis, where the air is rendered sonorous, to
the mouth, and it ıs not unlikely that the strong transverse rıdges upon the bony palate may
contribute, with the preceding trumpet-like structure, to give that intonation which is so
aptly denominated ‘the roar of the lion’.”

Then, ın discussing his findings on hyoid structure in the other felids studied, he stated,
“From the difference in the voice, the feline animals might have been expected, ä priori, to
present some differences in that part of their anatomy which relates to it.”

Pocock’s (1916) comprehensive survey of the hyoıdean apparatus in the Felidae was
based on dissections of 23 species, extending the number of species studied to 24 of the 37
recent Felidae. Pocock confirmed Owen’s morphological findings and also adopted his
interpretation of the correlation between hyoid structure and the occurrence of roaring in

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5902-0087 $ 02.50/0

88

Fig. 1. The hyoid apparatus consists of a paired chain of bony/cartilaginous elements, connected by a
single median element, the body of the hyoid. Both chains consist of acranıal and a caudal cornu each.
Dorsal view of fully ossified hyoid apparatus of felid species a: domestic cat hybrid (Felıs s. silvestris
x F.s. f. catus) ad. ? (ZFMK 85.50), b: jungle cat (Felis chaus) ad. (ZFMK 86.123), and c: cheetah
(Acınonyx jubatus) ad. 2? (ZFMK 86.119). Scale 5 cm. Labeled structures (unilaterally only): 1 =
tympanohyoid (cartilage); bones: 2 = stylohyoid; 3 = ephihyoid; 4 = ceratohyoid; 5 = basıhyoid; 6 =
thyrohyoid; 7 = chondrohyoid (cartilage). 1-4 = cranial cornu, 5 = body, 6 and 7 = caudal cornu of
hyoid. (In athe border between body and the two cornua is marked on the bones for better visibility)

Hyoid structure, laryngeal anatomy, and vocalızation in felids 89

felids. According to Pocock, those species producing a roar are lion, leopard (Panthera
pardus), jaguar (P. onca), and tiger (P. tigris); and ın these the hyoid apparatus is not
completely ossified (cf. Fig. 3). The snow leopard (P. uncıa) is the only other felid species
sharing thıs anatomical character, but ıts voice was not known to Pocock. All remaining
species of this family very likely have a fully ossified hyoid (ct. Figs. 1, 2).

Pocock originally stated that the tympanohyal “... generally, if not always, remains
cartilaginous through life up to its point of attachment with the bulla.” (Pocock 1916:
222). However, recent textbooks on domestic cat anatomy like CroucH (1969) or FIELD
and TAayLor (1969) consider it to be a bone, or to consist of tight connective tissue (NICKEL

Fig. 2. Attachment of hyoid apparatus to skull and larynx in a felid species with a fully ossified hyoid
(schematic drawing by Anne DaHm). E = epiglottis; La = larynx; Tr = trachea; labeling of hyoid
bones as in Fig. 1

Fıg. 3. Hyoid apparatus (still attached to the larynx) of a leopard (Panthera pardus) ad. Q (ZFMK
89.478), a species in which the hyoid is not fully ossified. The epihyoid is replaced by an elastic
ligament (L). Other labeled structures as in Figs. 1 and 2; the tympanohyoid is cartilaginous; scale
5 cm. (Border between body of hyoid and its caudal and cranıal cornu is marked on the bones for

better visibility)

90 G. Peters and M. H. Hast

et al. 1977). The four hyoid specimens ın Figs. 1 and 3 clearly show that in these felids the
tympanohyal is not bony. In the present paper the statement of a fully ossified hyoid
pertains to the hyoid elements 2 to 6 as shown in figures 1 and 2.

To extend and consolidate the hypothesis of a correlation between hyoid structure and
vocalization in the Felidae, Pocock (1916: 229) stated: “Apart from the roar there is
another very distinctive feature about the voice of the cats with a normal (i.e. fully ossified
[G. P. and M. H. H.]) hyoid. This is the famıliar ‘purr”. Lions, tigers, leopards and jaguars
never purr; on the other hand, such widely different species as cheetahs, pumas, caracals,
jaguarondis, and others ... express ... that sound. These are interesting differences
correlated with the differences in the hyoidean apparatus described above... The species
in which the hyoid is provided with this ligament (instead of the bony epihyale [G. P. and
M. H. H.]) roar, but do not purr. All the other species of Felidae with normally
constructed hyoid purr, but never roar.”

In his later study on the classification of the existing Felidae, Pocock (1917b)
considered these differences so important that they are the major criteria which separate
two subfamilies of the Felidae, the Felinae and Pantherinae. The cheetah, as the sole species
of the third subfamily, Acinonychinae, is thought to be more closely related to the Felidae,
because it resembles them ın the structure of its hyoid (cf. Fig. 1c with 1a and b).

Many later studies presenting a classification of the Felidae or dealing in detail with felid
systematics largely followed Pocock’s (1917b) concept based on the division of “roaring”
and “purring” cats (e.g. NEFF 1982). Even very recent biochemical studies in this field (e.g.
WAYNE et al. 1989) include these criteria in their discussion. In a list of the classification of
the recent Carnıvora, WOZENCRAFT (1989) retained the Felidae and Pantherinae as
subfamilies of the Felidae but with a content quite different from Pocock (1917b), e. g.,
the Pantherinae also including the lynxes (genus Lynx) and the caracal, the marbled cat
(Felis marmorata) and clouded leopard (Neofelis nebulosa). WOZENCRAFT (1989) dıd not
go into detail on the decisive criteria for his system of classification but in the main he
followed CoLLiEr and O’BrRIEn (1985), O’BrıeEn et al. (1987) and WAYynE et al. (1989)
who based their classifications on karyological, albumin immunological distance and
isozyme genetic distance data. Yet the reference cited (WERDELIN 1983) for the inclusion of
the marbled cat ın the Pantherinae does not present any data to support thıs hypothesıs.

In his major study of felid skull morphology, HALTENORTH (1937: 235) expressly
questioned the significance of hyoid structure ın felid systematics and considered roaring as
a plesiomorphic vocalization character in species of this carnıvore family. In a very recent
study of felid skull morphology SaLLes (1992) included hyoid structure as a character and
considered a fully ossified hyoıd as the plesiomorphic state, but did not recognize
subfamilies within the Felidae. |

The current study tries to bring together all available data to test Owen’s (1834) and
Pocock’s (1916, 1917b) original hypothesis that there is a correlation between hyoid
structure and the presence of certain vocalızations in species of Felidae. Furthermore, this
study discusses the findings of Hast (1986, 1989) on laryngeal anatomy with respect to
vocalization and hyoid structure in felid species. In doing so, ıt will also provide a basıs for
evaluating the valıdity of using these character sets as criteria in felid systematics.

Material and methods

This study, being largely a review, is based on data published in the main papers cited on hyoid
structure (Owen 1834; Pocock 1916), laryngeal anatomy (Hast 1986, 1989) and vocalization (PETERS
1978, 1980, 1981, 1983, 1984, 1987) in the Felidae. In addition, supplemental data on hyoid structure,
laryngeal anatomy and vocalızation are also included. Table 1 lists the felid species for which
respective data are available.

Data on hyoid structure and vocalization are present for all species listed; laryngeal anatomy is

Hyoid structure, laryngeal anatomy, and vocalızation in felids 3

known in a smaller number. Although the sample size for hyoid structure and laryngeal anatomy in
some species is limited to only one specimen, it seems fairly safe to assume that the sample is
representative of each species and, more certainly, the family Felidae altogether with respect to these
character complexes. The amount of data on vocalızation ın the individual species is different. There is
no species for which a comprehensive structural and functional analysis of its complete acoustic signal
system exists. Therefore, statements on vocalization are made with reservations, especially for those
species for which only limited data are available (see Tab. 1). Nevertheless, the data on vocalızation
for the family Felidae, as a whole, are considered sufficient for a comparative survey.

Table 1. Data base on hyoid structure, laryngeal anatomy, and vocalization in the Felidae

Species Hyoid Laryngeal Vocalization
structure anatomy

Acınonyx jubatus
Felis aurata

F. bengalensıs
chaus

. colocolo

. concolor
geoffroyi
manu
margarıta
marmorata
nıigripes
pardalıs
planiceps
rubiginosa
serval

. sılvestris
temmincki
tıgrına
viverrina
wiedü

F. yagonaroundi
Lynx caracal

L. Iynx

L. rufus
Neofelis nebulosa
Panthera leo

P. onca

P. pardus

P. tıgrıs

P. uncıa

+++

+3

==

++ ++++++
++ 1 ++++1
=

FI

++++1
+1

EN

|
a
+++ + tt tt Hr HEHE HEHE HE

R.
JR
A
I.
JE
IE
JE
RB:
ER:
I
Mr.
R:
Jh
IB
I,
E.
IE

+

++ ++++++++++ +1
|

++ ++++ +1

+: character complex studied; -: not studied; (+): limited data [a qualification only used in the
“Vocalization’ column in species with poorly known vocal repertoire].

Results and discussion
Hyoid structure and vocalization

Since the early investigations of Owen and Pocock did not deal with vocalızation in a
technical manner, there have been few serious attempts to define “roaring” as a vocalıza-
tion of lions and other felid species, in appropriate technical terms. Therefore, it can be
assumed that the term “roaring” has usually been applied in ıts “common use” dictionary
definition, e.g. MERRIAM-WEBSTER (1986: 1963): “roar: to utter or emit a full loud heavy
prolonged sound ( the lions roared )” resp. “roar: the sound of roaring: the deep loud cry
of some wild beasts ( the roar of a lion )”, or OxFORD ENGLISH DICTIONarY (1933): “roar:
1. a full, deep, prolonged cry uttered by a lion or other large beast, ...” Thus, in the

92 G. Peters and M. H. Hast

context of a technical description of felid vocalızation, this is a circular definition and is
equivocal, in that lions can produce different types of sounds which fit this general
description. All the more, therefore, using the term “roaring” for vocalizations of other
felid species is equivocal. Definitions in some recent publications like “Ihe roar is a
distinet, specific vocalızation, very loud and resonant, which is produced by the pantheri-
nes — the lion, tiger, jaguar, and leopard.” (NEFF 1982: 20), obviously have the same
weakness as those given above, and give no criteria for an unequivocal identification of the
“roar” in any of these species.

Whereas Owen did not got into detail on roaring, and thus a critical evaluation of his
statement in this respect is not possible, Pocock’s (1916) denotation of roaring in different
Panthera species but the snow leopard is specific enough to identity those vocalizations for
which he used this term (1.e. vocalizations unique to the lıon, jaguar, leopard, and tiger).
Based on structural analyses (PETERS 1978), it has been shown that certain of the relevant
single calls of all four species and of the call types composing the structured call sequences
in the first three species, which Pocock lumped under ‘roaring’, actually represent
different types of vocalizations. Some of these vocalızations are not common to all four of
these species, while others are common to many more species of the Felidae. Thus,
Pocock’s use of the term “roar” was ambiguous and his hypothesis of a correlation
between hyoid structure and the presence of roaring in a species’ acoustic repertoire needs
to be retormulated and re-examined.

LEYHAUSEN (1950) was the first who attempted to deal with felid roaring in a more
technical way, but he did not go into much detail on vocalızation structure. However, he
pointed out differences in the sound ot roaring and its artıculation between lıon and tiger,
and the significance of roaring of hybrids between these two species for the elucidation of
differences between the equivalent calls of their parental species. RESCHKE (1960, 1966),
TEMBROcK (1962), SCHALLER (1972), and HEMMER (1966, 1968) published definitions of
roaring in the Felidae, with the former three authors basıng their definitions on bioacousti-
cal analyses.

Referring to SCHENKEL (1966), SCHALLER (1972: 105), in describing roaring ın lions,
stated that “Roaring represents a graded system, ranging from barely perceptible grunts to
full roars, ...” This definition of roaring includes different types of single calls, either
produced ın a loud structured call sequence, as roarıng “proper”, or singly or in series of
varıable composition and temporal structure at low to medium intensity. SCHENKEL’S
(1966: 617, 618) original text, in discussing different types of roaring, did not differentiate
between roaring as a structured call sequence and the single calls composing the sequence
and their specific character. However, he clearly stated that he did not regard all call types
involved as belonging to the same graded system. In an earlier treatment of vocalızatıon ın
the tiger, SCHALLER (1967: 258) held that “Moans and roars are lumped into one category
because the sounds represent varıations in intensity of the same basic vocalızation.”
Therefore, as in the lion, the term “roaring” in the tiger is applied to different types of
single calls as well as to call sequences composed of such calls. Using “roar” in this way,
SCHALLER very likely held that these vocalızations are fully equivalent ın both species; he
also listed (SCHALLER 1972: 452) the leopard as being able to roar. In effect, the ambıguous
use of the term roarıng was continued by SCHENKEL (1966) and SCHALLER (1967, 1972).

REscHkE (1960, 1966) and TEMBROcCK (1962) both used the German term for roarıng
(“Brüllen”), with qualifications, and pointed out its ambiguity. In defining different call
types and the structured call sequences of certain felid species, they presented unequivocal
characterizations of most relevant vocalizations in the Panthera species and some other
felids but did not comment at all on the postulated correlation of the presence of roaring ın
a species and its hyoid structure.

Though not based on proper bioacoustic analyses, HEMMER (1966) presented a rela-
tıvely clear definition of roaring in the lion, leopard, jaguar, and tiger. Because he was well

Hyoid structure, laryngeal anatomy, and vocalızation in felids 93

aware of the composite nature of the structured call sequences in the lion, leopard, and
jaguar and the absence of an equivalent call sequence in the tiger, he aimed at a definition of
roaring that would encompass all four species. Thus, the typical roaring sequence ot a lion,
as commonly understood (very likely also by Owen and Pocock in their relevant
publications), was not defined as roaring in its entirety but only its intense initial portion
(HEMMER 1966: 60-61). Under this object-defined definition of roarıng, HEMMER (1966)
held that the lion, leopard, jaguar, and tiger do roar, whereas the snow leopard does not
roar. In making such a statement this author neglected the differences between the former
four species in respect of temporal and intensity patterning of the relevant call series.
HEMMER (1966: 76) concluded that, ın felids, there ıs no direct correlation between hyoid
structure and vocalization. He based this conclusion on his observation that the snow
leopard is able to purr and unable to roar, despite the fact that ıts hyoid is not fully ossified;
this would predict just the opposite situation for these two vocalizations.

In a detailed technical study of vocalization in all Panthera species, the clouded leopard
and puma (Felis concolor), PETERS (1978) did not apply the equivocal term ’roar‘ as ıt had
been used for different types of vocalizations ın various species. For the lion, in which
“roarıng’ usually was applied to the loud species-specific structured call sequence (see Figs.
4, 5), PErers (1978) demonstrated that these sequences comprise at least two different
types of calls (Figs. 6, 7), one of which is definitely not found in the tiger. All three (very
rarely four) call types, however, which can appear ın a roarıng sequence of the lion may

5.088 kHz

8.888 Hz

5.808 kHz

8.888 Hz

5.888 kHz

8.868 Hz : 2
30.580 _s Kanal _1 Autospektrum 43.750

5.888 kHz

43.750 s Kanal 1 Autospektrum 57.0088 s

Fıg. 4. Continuous sonagram (53 s, partitioned over 4 sonagrams of equal duration) of a complete
roaring sequence (‘roaring proper’) of an adult d lion (Panthera leo). Frequency axis (y-axis)
represents 5 kHz, time axis (x-axis) 13.25 s in each sonagram. Labeled single calls no. 5 (see Fig. 6)
and 17-20 (see Fig. 7) are analyzed once again in more detail

94 G. Peters and M. H. Hast

2.088 V

-2. 889 y

5.888 kHz

8.8088 Hz

4.8088 s Kanal I Autospektrum 57.008 s

3.018 dB

-80.0088 dB

4.0908 s Kanal 1 Intensitätsverlauf 57.8088 s

Fig. 5. Oscillogram (top), sonagram (middle) and intensity graph (bottom) of the same roaring
sequence as in Fig. 4. X-axis (time) is the same for all (53 s duration), units and their respective
calibrations on the y-axes are given. The call types composing the sequence, their succession, the
general pattern of their change in intensity, duration and duration of the intervals between them are
species-specific for lion roaring

also be present in the equivalent sequence in the structured call series of the leopard and
jaguar. The usual form of these series in the latter two species, though, is different from the
roaring of lions in that they are equivalent to the second half of the lion roaring sequence.
Like HEMMER (1966), PETERS (1978) concluded that hyoid structure and the occurrence
of roarıing ın the Felidae (defined by reference to the lion’s structured call sequence) are not
correlated. Thus, with the relevant vocalızation of the lion - as very likely understood by
Owen and Pocock - as the standard of comparison, the leopard and jaguar have fully
equivalent vocalization sequences, whereas the tiger and snow leopard do not. These are
the only felid species with an incompletely ossified hyoid. Because of the established
differences in the relevant call types and call sequences (under the varıous definitions of
roarıng), and as none of these ıs shared by all five species, and at the same time ıs not
present in any other felid, an incompletely ossified hyoid in a felid species is not necessarily
associated with the presence of roaring in this species’ acoustic repertoire. Hence the
hypothesized direct correlation of roarıng and hyoid structure (1.e. an incompletely
ossified hyoid) can be refuted. For the first time, this statement was based on technical
analyses and definitions of all relevant sound types and call sequences (PETERS 1978).
Similar ambiguity and confusion exist for the definition of purring, the ‘counterpart’ of
roarıng within the felid acoustic repertoire, as postulated by Pocock (1916). The
MERRIAM-WEBSTER (1986) dictionary definition of purr as “a low vibratory murmur of a

Hyoid structure, laryngeal anatomy, and vocalization in felids 95

2.0888 V

-2.088 V

5.888 kHz

8.088 Hz

18.8008 s Kanal 1 Autospektrum 28.0808 s

Fig. 6. Oscillogram (top) and sonagram (bottom) of call no. 5 of the roaring sequence shown in Figs. 4
and 5. X-axis (time) is the same for both graphs (1.28 s duration), its subdivisions in the oscillogram
representing 0.1 s each. Units and their respective calıbrations on the y-axes are given. This call ıs a
main call with grunt element (cf. PETERS 1978)

cat that appears to indicate contentment or pleasure and is believed to result from the
streaming of air over the false vocal cords” gives no criteria for an unequivacal structural
identification of thıs felid vocalızation, although even making a statement on the supposed
mechanism of sound production. In principle, there ought to be less difficulty in correctly
identifying purring, because ıt is such a famıliar sound in the domestic cat. Moreover, some
of its unique structural charasteristics are already identifiable by careful observation
without proper bioacoustic analysıs. The articulation and structure of purring are so
singular that only a vocalization with all its pertinent characteristics (see FRAZER SIssoM et
al. 1991), established by bioacoustic analysis (cf. Fig. 8), ought to be so classified.
Anecdotally, this vocalızation has been reported in many felid species, even for most of
those that allegedly cannot purr, like lion or tiger, and other Panthera species (e.g.
HEMMER 1966, 1968; SCHALLER 1972; NEFF 1982). Several authors (e.g. HEMMER 1966,
1968; SCHALLER 1972) maintained that in some of these species purring ıs artıiculated
during exhalation only. LEYHAUSEN (pers. comm.) holds that very young cubs of the
Panthera species very likely are able to purr during in- and exhalatıon. It ıs highly probable
that different types of sounds were lumped under the term ‘purring’ in various species.
Verifiable evidence based on proper bioacoustical analyses for the existence of this
vocalızation was only presented for relatively few felid species: juveniles of margay (Felis
wiedii), little spotted cat (F. tigrina) and bobcat (Lynx rufus), juveniles and adults of the
domestie cat, jaguarundi (Fels yagomaroundi), puma and serval (F. serval), and adults of
the Eurasian lynx (Lynx lynx), cheetah, and the Indian desert cat (Felis silvestris ornata)

96 G. Peters and M. H. Hast

2.0888 V

5.888 kHz

8.8088 Hz
37.435 s$ Kanal 1 Autospektrum 417955

3.018 dB

-838.088 dB

37.435 s Kanal 1 Intensitätsverlauf 41.795 s

Fıg. 7. Oscillogram (top) and sonagram (bottom) of calls no. 17 to 20 of the roaring sequence shown
in Figs. 4 and 5. X-axis (time) is the same for both graphs (4.36 s duration), ıts subdivisions in the
oscillogram respresenting 1 s each. Units and their respective calibrations on the y-axes are given.
These calls are grunts (cf. PETERS 1978)

(PETERS 1981). All of these species have a fully ossified hyoid; unequivocal observations
and tape recordings of purring in one further species of the Felidae with this type of hyoid
morphology exist (see Tab. 2). However, a general statement on a direct correlation
between the presence ot a fully ossified hyoid in a felid species and its abılity to purr can
only be made with reservations, since unequivocal evidence of this sound is lacking in too
many species of the Felidae which have this hyoid morphology. On the basıs ot available
data, the existence of such a correlation also cannot be refuted, though. In deatiled analyses
of domestic cat purring (REMMERS and GAUTIER 1972; FRAZER SıssoM et al. 1991), no
mention was made of a role of the hyoid in articulating this sound. None of the five felıd
species with an incompletely ossified hyoid has been examined carefully enough to make a
scientifically substantiated statement as to whether they can or cannot purr.

So far, no other vocalizations have been technically described in the Felidae, the
distribution of which among the various species of the famıliy matches that of the two
hyoid structure types.

As detailed before, the postulated direct correlation of an incompletely ossified hyoıd ın
a felid species and its ability to roar has been refuted. In respect of purring, a definitive
statement cannot be made yet as to whether all species with a completely ossitied hyoid can
purr and whether this vocalization is restricted to these felids. Therefore, the hypothetical
correlation of purring and hyoid structure in the Felidae can neither be falsified nor verified
at the present state of analysis. Consequently, as this hypothesis is an attempt at explaining

Hyoid structure, laryngeal anatomy, and vocalızation ın felids 97

CEEKEIKEKECLECECLELEKCEEEEECEKECKECCECECECCECECEECKEECECECCECEECEECECCECEECECEECEECEECEEECEECEIEEITECEITECEETIECEEIEEETEICEEEEI

2.0088 V

-2.008 V
9.0800 Kanal 1 Zeitsignal 12.000 s

5.008 kHz

. Fur er N .
Kanal 1 Autospektrum

Fıg. 8. Oscillogram (top) and sonagram (bottom) of purring of an adult ? puma (Felis concolor). X-
axis (time) is the same for both graphs (3 s duration), its subdivisions in the oscillogram representing
0.5 s each. Units and their respective calibrations on the y-axes are given

the cause of major differences in the composition of acoustic signal systems in felids, it
seems to be scientifically more appropriate, for the present, not to maintain it.

Laryngeal anatomy and vocalization

Detailed comparative anatomical studies of the larynx as the sound producing organ had
been lacking ın the Felidae until very recently (Hast 1986, 1989). Earlier studies of felid
larynges by Nesus (1949), KELEMEN (1963) or SCHNEIDER (1964) dıd not deal with the
interdependence of larynx structure and the acoustic repertoire of a species in a specific
way. Ihey only arrıved at statements like “The roar of the lion is produced with a
comparatively simple vocal apparatus.” (KELEMEN 1963: 503) or “The larynx of the felids
in general is more primitively constructed than the larynx of the hare or of the antelope,
but in spite of this the vocal production of the latter ıs very much poorer.” (KELEMEN 1963:
514). This lack of information exists despite the fact that the hypothetical correlation of
hyoid structure and vocalızation in the Felidae already was a relatively common view.
The major findings of Hast (1989: 118, 119) are that, of the 14 felid species studied, the
larynges of lion, leopard, jaguar, and tiger are exceptional in having a “large pad of fibro-
elastic tissue which constitutes the rostral portion of each of the proportionately very large
undivided vocal folds”. The other species studied are not atypical for carnıvores in that
they have a “larynx with divided thyroarytenoid folds, with a depression between the
rostral and the caudal folds that varies from a slight fossa to a deep ventricle, and a vocal
fold with a sharp edge” (see Fig. 9 a-c). The specific structure of thick vocal folds in the

98 G. Peters and M. H. Hast

Fig. 9. Dorsal view of larynges of a: a jaguar (Panthera onca), b: a snow leopard (P. uncia), and c: a
puma (Felis concolor) cut coronally. The longitudinally very large vocal folds (VF) of the jaguar with
their large pad of fibro-elastic tissue (p) are readily distinguished from the proportionately shorter
vocal folds of the snow leopard and the even shorter vocal folds of the puma with their sharp edge (E =
epiglottis; C = crıcoid cartilage; T = thyroid cartilage). (Figures a and b from Hast [1989] with kind
permission of the Journal of Anatomy; c: reproduced at a larger scale than a and b)

four species mentioned has the consequence that they are able to produce sounds of high
acoustical energy and with lower frequency components than those species with thinner
vocal folds (Hast 1989). Additional felid species (Felis colocolo, F. nigripes, F. serval,
F. tigrina, F. yagonaroundi) studied subsequent to Hast (1986, 1989) confirm that only the
four species mentioned above have this special morphology of the vocal folds.

The incompletely ossified hyoid also enables lion, leopard, jaguar, tiger, and snow
leopard to move their larynx away from the oral cavity, and thus also from the mouth,
thereby extending the length of the tube (see Fig. 10 B), which results in an even lowered
pitch of vocalization. Consequently, the specific structure of their hyoid can amplify the
effect of the thick vocal folds ın the former four species quantitatively. However, it is not
possible at present to quantify the effect of their unusually shaped vocal folds in deepening
and intensitying the calls of these four species.

Generally, species of the Felidae vary considerably in the pitch of their equivalent calls,
and no uniform, direct size-correlation is manıfest in an interspecific comparison. On the
contrary, some felid species with sharp-edged vocal folds (and a completely ossified hyoid)
have strikingly high- and low-pitched calls in relation to their position within the over-all
size range of the Felidae (this statement also holds for the snow leopard with an
incompletely ossified hyoid). In relation to its size, this species’ main call ıs remarkably
high-pitched (PETErs 1978). This finding is a caveat that vocal fold morphology (and hyoid
structure) alone are not decisive parameters for the pitch of a felid species’ calls.

With the exception of a few measurements for lions (JAROFKE 1982), no quantitative
data on loudness of felıd vocalızations are available; therefore no interspecific comparisons
in this respect can be made. Generally, none of the morphological and physiological
parameters determining the pitch and intensity of a felid species’ vocalizations have been
quantified.

Hast (1986, 1989) observed that the specific structure of the vocal folds in the lıon,

Hyoid structure, laryngeal anatomy, and vocalızation ın felids 39

A

Tr > P M

Fig. 10. Schematic drawing of the vocal tract of A: a felid species with sharp-edged vocal folds and B:
a felid species with vocal folds with a large pad of fibro-elastic tissue. The open arrows ın A and B
show the direction of the expiratory air flow. The arrows in B indicate that a hyoid with an elastic
ligament instead of a bony epihyoid, which the species with this type of vocal fold morphology have,
allows for an elongation of the supraglottal vocal tract, ı.e. the distance of the sound generator from
the mouth, which results in lower pitch of calls (M = mouth; P = pharynx; Tr = trachea; VF = vocal

folds)

tiger, leopard, and jaguar has a qualitative influence on their vocal repertoires, enabling
them to roar (although not technically defined). Therefore, according to the above
discussions it is not clear to which vocalization(s) this term actually refers in these four
species. As detailed before, no matter how roaring was defined by any of the earlier
authors, it is not exclusively common to these four species among the Felidae. If the
hypothesis of a correlation between vocal fold morphology and vocalızation in the Felidae
is maintained, to the effect that these morphological characteristics have a qualitative
influence on the composition of the vocal repertoire of species, the distribution of
vocalization types must be examined. This distribution must be compared to that of the
different types of vocal folds ın the various species of this carnıvore family, especially lion,
leopard, jaguar, and tiger, ın contrast to the rest of the family. There is but one type of
vocalızation, the main call with grunt element (Fig. 6), which is common to these four
species but not present in any other felıd species (PETERS 1978). This call type ıs a
constitutive element in the structured call sequences of the lion (= roaring “proper’) (Figs.
4, 5) and may be present ın the equivalently structured call sequences of leopards and
jaguars. Tigers utter such calls as single calls or ın call series which are irregular with
respect to the call types that compose them, the calls’ intensity and their temporal sequence
(PETERS 1978). Parallel with this character complex, the larynges of these four species have
morphological characteristics in common, which distinguish them from those of all other
felid species, specifically a large pad of fibro-elastic tissue which constitutes the rostral
portion of proportionately very large vocal folds (Hast 1989). However, this concord-
ance/association of characters is not unequivocal evidence that the occurrence of this
specific type of vocalızatıon in only these felid species ıs caused by their special laryngeal
morphology. At the present state of analysıs no other type of vocalızation is known which
is shared only by the four species with specific vocal fold type, as opposed to the equivalent
character situation in the rest of the species of the family Felidae. Therefore, a qualitative

100 G. Peters and M. H. Hast

Table 2. Association of characters of hyoid structure, anatomy of vocal folds and vocalization in
species of the Felidae

Species Hyoid Vocal fold Vocalizations
structure type Purr

Acinonyx jubatus
Felis concolor

F. geoffroyi

F. serval

F. silvestris

F. tıgrına

F. wiedii

F. yagonaroundi
Lynx Iynx

L. rufus

Neofelis nebulosa
Panthera uncia
P. tıgrıs

Deo

P. onca

P. pardus

v+tt++t++++++

Dort m nm un vo vum wnmumumu

MC: main call with grunt element; os: hyoid completely ossified; e: hyoid with elastic ligament;
s: sharp edged vocal folds; p: vocal folds with thick pad of fibro-elastic tissue; character
+: present; —: absent; ?: no data/character state doubtful.

influence of vocal fold morphology on a species’ vocalization repertoire remains conjec-
tural in the Felidae.

Hyoid structure, laryngeal morphology, and vocalization as characters
in felid systematics

Since initial studies by Pocock (1917b), hyoid structure (and its correlation with vocalıza-
tion) has remained a major criterion in felid classification above the species level (e.g.
CorBET 1978; NEFF 1982). After discussing statements by varıous authors that lions,
tigers, leopards, jaguars and snow leopards (the felid species with an incompletely ossified
hyoid which allegedly ought to be unable to purr) can purr, NEFF (1982: 21) concluded
“. „there is still some doubt about the distribution of the ability to purr among the wild
cats.” Although she cited data that would refute the postulated correlation of hyoid
structure and the presence of roaring or purring in a felıd species, NEFF (1982) retained
Pocock’s criterion of hyoid structure to separate the subfamilies. Thus, she either
regarded the differences in hyoid structure as sufficient to separate the subfamilies of the
Felidae, irrespective of whether there is a correlation with vocalization or not - as verified
above there is not -, or she doubted reports of purring in species which allegedly are not
able to produce this sound because of their hyoid structure. Even very recent classıfıcations
of the Felidae based on character complexes like immunological distance or DNA-
hybridization data (WAvneE et al. 1989) refer to the differences in hyoid structure as
supporting evidence. HALTENORTH (1937), however, expressly denied any systematic
significance of hyoid morphology within the Felidae.

If the incomplete ossification of the hyoid is considered as a derived character within the
Felidae (cf. SarLes 1992), the monophyletic group characterized by this synapomorphy
would include lion, leopard, jaguar, tiger, and snow leopard, irrespective of the rank
attributed to this taxon and other characters allegedly/perhaps correlated with hyoid
structure. According to varıous authors, additional synapomorphic characters shared by
these five species are: 1. coat pattern on the head completely dissolved into spots (HEMMER

Hyoid structure, laryngeal anatomy, and vocalızation in felids 101

1981) (other authors like LEYHAusEn [1950] hold another view in respect of this character);
2. bridge of the nose covered with hair up to its anterior edge (HEMMER 1981); 3. simple
structure of the baculum without bifurcation in its basal portion (KrATocHviL 1976).

In addition to the above mentioned five species HEMMER (1978) included marbled cat
and clouded leopard in the ‘pantherine line’, based on morphological (see Pocock 1917a,
1932), ethological, physiological, and karyological data. However, ın a later publication on
the same topic (HEMMER 1981), the ‘pantherine line’ did not include the marbled cat. Based
on coat colour and pattern, WEIGEL (1961) grouped Panthera, Uncia, and Neofelis ın the
Pantherinae. In addition to the five species united in this subfamily by Pocock (1917b),
Staıns (1984) included the clouded leopard without giving criteria for doing so; WOZEN-
CRAFT (1989) included lion, leopard, jaguar, tiger, snow leopard, clouded leopard, the
marbled cat, the lynxes and the caracal ın the subfamily Pantherinae. Based on their
comparative biochemical studies CoLLIER and O’BrıEN (1985), O’Brıen et al. (1987) and
WAYNE et al. (1989) established a “Panthera lineage’ with the five roaring (sic) cat species
(lıon, leopard, jaguar, tiger, snow leopard), the lynxes, and the marbled cat, all of which
also share an identical karyotype different from the remaining felid genera (cf. WURSTER-
Hırı and CENTERWALL 1982). In addition, the “Panthera lineage’ of CoLLIER and O’BRIEN
(1985) includes - among others - such diverse species as the African golden cat (F. aurata),
serval, cheetah, and yagouaroundi (F. yagonaroundı).

However, illustrating again the conceptual confusion with regard to ‘roaring’, WAYNE
et al. (1989: 473) listed further characters supporting their classification of the Felidae ın
stating that “The most recent radıation led to the five species of roarıng cats, genus
Panthera, ...”, and then continue (473, 474) “The lıon, tiger, leopard, and jaguar have an
incompletely ossified hyoid that allows them to roar and thus unites the group. ... the
snow leopard (P. uncia), the only nonroaring member of the genus, ... has a hyoid with
structural similarıties to the pantherids.” (sic).

Hast (1986, 1989) considered the systematic significance of the distribution of the two
basıc types of vocal told morphology within the Felidae, especially as regards the content
of the genus Panthera and the generic association of the snow leopard. Shared acoustical
and morphological characteristics in lion, leopard, jaguar, and tiger (presence of main call
with grunt element, hyoid structure, vocal fold morphology), may support their joint
classification in the genus Panthera. The snow leopard shares only one of these three
characteristics wıth the latter four species, the incompletely ossified hyoid. These findings
may be taken to support the view of varıous authors (e.g. Pocock 1917b; Haltenorth
1937; HEMMER 1966, 1968, 1972, 1981; PETERS 1978, 1980; NEFF 1982) that the snow
leopard should be classified in ıts own genus Uncia and not in Panthera (Honackt et al.
1982; CorgBEr and Hırr 1991).

The continuing dispute on the systematics of the Felidae (cf. e.g. Stmpson 1945; EwWER
1973; HEMMER 1978; LEYHAUSEN 1979, 1992; KrAL and Zıma 1980; Honackt et al. 1982;
KrartochaviL 1982; CoLLIER and O’BrıEn 1985; WAYNe et al. 1989; WOZENCRAFT 1989;
CoRrBET and Hırr 1991; KITCHENER 1991; Nowak 1991; Sarıes 1992) is clear evidence
that conclusions drawn concerning this matter vary according to character complex(es)
studied and their weighting, in view of differences and conformities of character/character
state distributions found in species for various complexes.

Together with other morphological character complexes studied, data on hyoid struc-
ture, vocal fold morphology and especially vocalization support the hypothesis that
Panthera, Uncia and Neofelis form a monophyletic group (cf. HEMMER 1981; SaLLes
1992). These characters conflict with chromosomal (KrAL and Zıma 1980) and
immunological distance data (COLLIER and O’BrIEn 1985), which would ally the Iynxes
and the marbled cat to thıs group.

Systematic studies based on any set of genetically determined characters should be
congruent with other such studies based on different sets of characters in the same

102 G. Peters and M. H. Hast

organisms (Hırrıs 1987). The conflict ın the Felidae is very likely due to unrecognized
character convergencies, mistakes in the interpretation of shared characters/character states
as symplesiomorphic or synapomorphic, and the fact that the different character com-
plexes undergo different rates of evolution, with character evolution not generally reflect-
ing group phylogeny. We have not yet arrıved at a stage of analysis of the evolutionary
history of the Felidae in which congruence/consensus or complementarity of the various
character data sets are possible.

Acknowledgements

Both authors would like to express their sincere thanks to the staff members of the zoos and museums
who kindly provided postmortem specimens of cut larynges: Cheyenne Mountain Zoo, Field
Museum of Natural History, Florida State Museum, Jersey Wildlife Preservation Trust, Lincoln Park
Zoo, Little Rock Zoo, Milwaukee Zoo, National Zoological Park, Wildlife Safarı of Oregon,
Wuppertal Zoo, Zoological Society of San Diego. We also very much appreciate the time given by Dr.
RAINER HUTTERER, Professor Dr. PauL LEYHAUSEN, and Dr. BRUCE D. PATTERSoN to critically
review this paper prior to ıts submission for publication. G. P. thanks the preparators at his institute
for their careful and skilled work in preparing larynx and hyoid specimens, Mr. H. MEURER for doing
the photographs and Miss A. DanHm for the drawings.

Zusammenfassung

Hyoidban, Kehlkopfmorphologie und Lautgebung bei Feliden (Mammalia: Carnivora: Felidae)

Bei den meisten Arten der Feliden ist das Hyoid (Zungenbein) vollständig verknöchert, bei fünf Arten
der Familie jedoch nicht, sondern enthält ein elastisches Ligament. Dieser Unterschied im Hyoidbau
wurde von vielen Autoren als Ursache für die Ausbildung bzw. das Fehlen der beiden Lautformen
Brüllen und Schnurren ım Lautrepertoire der einzelnen Arten der Felidae angesehen. Hyoidbau und
damit hypothetisch korrelierte Lautgebung wurden in der Systematik der Felidae als wesentlicher
Merkmalskomplex bewertet. Die vorliegende Untersuchung diskutiert die postulierten Zusammen-
hänge zwischen Hyoidbau und Lautgebung auf der Grundlage eindeutig definierter Lauttypen und
bezieht dabei neue Ergebnisse zur Kehlkopfmorphologie der Feliden mit ein.

Es ist erwiesen, daß nicht alle fünf Felidenarten mit einem unvollständig verknöcherten Zungen-
bein brüllen können, ein solcher Hyoidbau bedingt also nicht automatisch die Ausbildung des
Brüllens. Andererseits findet sich diese Lautform aber auch bei keiner Art mit einem vollständig
verknöcherten Hyoid. Alle Arten, bei denen bisher die Fähigkeit zu schnurren gesichert ist, haben ein
vollständig verknöchertes Hyoid, für viele Arten mit diesem Hyoidtyp fehlt allerdings bisher noch der
eindeutige Nachweis dieser Lautform, so daß eine definitive Aussage darüber, ob ein solcher
Hyoidbau für die Ausbildung des Schnurrens notwendig ist und diese bedingt, bisher nicht möglich
ist. Das Vorkommen eines weiteren Lauttyps ist auf die vier Felidenarten beschränkt, deren
Stimmlippen eine von allen übrigen Arten der Familie abweichende Form aufweisen. Die Aussagefä-
higkeit dieser Merkmalskomplexe für eine Rekonstruktion der stammesgeschichtlichen Beziehungen
innerhalb der rezenten katzenartigen Raubtiere wird erörtert.

References

COLLIER, G. E.; O’Brıen, $. J. (1985): A molecular phylogeny of the Felidae: Immunological
distance. Evolution 39, 473-487.

CoRBET, G. B. (1978): The mammals of the Palaearctic region: A taxonomic review. London: British
Museum (Natural History).

CoRBET, G. B.; Hırı, J.E. (1991): A world list of mammalian species. 3rd ed. London: British
Museum (Natural History).

CROUcH, ]J. E. (1969): Text-atlas of cat anatomy. Philadelphia: Lea and Febiger.

EwER, R. F. (1973): The carnıvores. London: Weidenfeld and Nicolson.

FıeLp, H. E.; TavyLor, M.E. (1969): An atlas of cat anatomy. 2nd ed. Chicago: University of Chicago
Press.

FRAZER SıssoM, D. E.; Rıcz, D. A.; PETERS, G. (1991): How cats purr. J. Zool. (London) 223, 67-78.

HALTENORTH, T. (1937): Die verwandtschaftliche Stellung der Grofßkatzen zueinander 1. Z.
Säugetierkunde 12, 97-240.

Hast, M. H. (1986): The larynx of roarıng and non-roaring cats. J. Anat. 149, 221-222.

— (1989): The larynx of roaring and non-roaring cats. J. Anat. 163, 117-121.

HEMMER, H. (1966): Untersuchungen zur Stammesgeschichte der Pantherkatzen (Pantherinae). Teil I.
Veröff. zool. StSamml. Münch. 11, 1-121.

Hyoid structure, laryngeal anatomy, and vocalızation in felids 103

— (1968): Untersuchungen zur Stammesgeschichte der Pantherkatzen (Pantherinae). Teil. I: Studien
zur Ethologie des Nebelparders Neofelis nebulosa (Griffith 1821) und des Irbis Uncia uncia
(Schreber 1775). Veröff. zool. StSamml. Münch. 12, 155-247.

— (1972): Uncia uncia. Mammalıan species 20.

— (1978): The evolutionary systematics of living Felidae: Present status and current problems.
Carnivore 1, 71-79. E

— (1981): Die Evolution der Pantherkatzen — Modell zur Überprüfung der Brauchbarkeit der
HENNIGschen Prinzipien der phylogenetischen Systematik für wirbeltierpaläontologische Stu-
dien. Paläont. Z. 55, 109-116.

Hırııs, D. M. (1987): Molecular versus morphological approaches to systematics. Ann. Rev. Ecol.
Syst. 18, 23-42.

Honackır, J. H.; Kınman, K. E.; Koeppr, J. W. (1982): Mammal species of the world - a taxonomic
and geographic reference. Lawrence: Allen Press.

JAROFkE, D. (1982): Messungen der Lautstärke des Chorgebrülls der Löwen (Panthera leo) ım Zoo
Berlin. Bongo 6, 73-78.

KELEMEN, G. (1963): Comparative anatomy and performance of the vocal organ in vertebrates. In:
Acoustic behaviour of animals. Ed. by R.-G. BusneL. Amsterdam: Elsevier. Pp. 489-521.

KITCHENER, A. (1991): The natural history of the wild cats. London: Christopher Helm.

KrAL, B.; Zıma, J. (1980): Karyosystematika celedi Felidae. Gazella 2/3, 45-53.

KraTocHvit, ]J. (1976): Os penis der Gattung Panthera und das System der Felidae (Mammalıa).
Zool. Listy 25, 289-302.

— (1982): Karyotyp and System der Familie Felidae (Carnıvora, Mammalıa). Folia Zool. 31,
289-304.

LEYHAUSEN, P. (1950): Beobachtungen an Löwen-Tiger-Bastarden, mit einigen Bemerkungen zur
Systematik der Großkatzen. Z. Tierpsychol. 7, 46-83.

— (1979): Katzen, eine Verhaltenskunde. Hamburg: Paul Parey.

— (1992): Ein neues Verfahren der Schädeluntersuchung zur Anwendung in der Säugetiersystematik.
Bonn. zool. Beitr. 43, 215-228.

Nerr, N. A. (1982): The big cats - the paintings of Guy Coheleach. New York: Abrams.

Nesus, V. E. (1949): The comparative anatomy and physiology of the larynx. New York: Grune and
Stratton.

NickEL, R.; SCHUMMER, A.; SEIFERLE, FE. (1977): Lehrbuch der Anatomie der Haustiere. Bd. 1.
Bewegungsapparat. 4th ed. Berlin: Paul Parey.

Nowak, R.M. (1991): Walker’s mammals of the world. Volume I. 5th ed. Baltimore: Johns Hopkins
University Press.

O’BRIEN, S. J.; COLLIER, G. E.; BENVENISTE, R. E.; NAsH, W. G.; NEwMAn, A. K.; SIMONSoNn, J. M.;
EICHELBERGER, M. A.; SEAL, U. S.; Janssen, D.; ‚Bush, M.; WıLpr, D. E. (1987): Setting the
molecular clock in the Felidae: The great cats, Panthera. In: Tigers of the world. Ed. by R. L.
Tırson and U. S. Sear. Park Ridge: Noyes. Pp. 10-27.

Owen, R. (1834): On the anatomy of the cheetah, Felis jubata, Schreb. Trans. zool. Soc. London
1835, 1,, 129-136.

PETERS, G. (1978): Vergleichende Untersuchung zur Lautgebung einiger Feliden (Mammalia, Feli-
dae). Spixiana, Suppl. 1.

— (1980): The vocal repertoire of the snow leopard (Uncia uncia, Schreber 1775). Internat. Pedigree
Book of Snow Leopards 2, 137-158.

— (1981): Das Schnurren der Katzen (Felidae). Säugetierkdl. Mitt. 29, 30-37.

— (1983): Beobachtungen zum Lautgebungsverhalten des Karakal, Caracal caracal (Schreber, 1776)
(Mammalıa, Carnıvora, Felidae). Bonn, zool. Beitr. 34, 107-127.

— (1984): On the structure of friendly close range vocalızations in terrestrial carnivores (Mammalia:
Carnivora: Fissipedia). Z. Säugetierkunde 49, 157-182.

— (1987): Acoustic communication in the genus Lynx (Mammalia: Felidae) - comparative survey and
phylogenetic interpretation. Bonn. zool. Beitr. 38, 315-330.

Pocock, R. I. (1916): On the hyoidean apparatus of the lion (F. leo) and related species of the Felidae.
Ann. Mag. nat. Hist., 8th ser., 18, 222-229.

— (1917a): On the external characters of the Felidae. Ann. Mag. nat. Hist., 8th ser., 19, 113-136.

— (1917b): The classification of existing Felidae. Ann. Mag. nat. Hist., $th ser., 20, 329-350.

— (1932): The marbled cat (Pardofelis marmorata) and some other Oriental species, with the
definition of a new genus of the Felidae. Proc. zool. Soc. Lond. 1932, 741-766.

REMMERS, J. E.; GAUTIER, H. (1972): Neural and mechanical mechanisms of feline purring. Resp.
Physiol. 16, 351-361.

I B. (1960): Untersuchungen zur Lautgebung der Feliden. Dipl.-Arbeit, Humboldt-Univ.,
Berlin.

— (1966): Vergleichende Untersuchungen an Lautfolgen der Pantherinen. Wiss. Z. Karl-Marx-Univ.
Leipzig 15, 499-505.

104 G. Peters and M. H. Hast

Sarıes, L. ©. (1992): Felid phylogenetics: Extant taxa and skull morphology (Felidae, Aeluroidea).
Am. Mus. Novit. No. 3047.

SCHALLER, G. B. (1967): The deer and the tiger: A study of wildlife in India. Chicago: Chicago Univ.
Press.

— (1972): The Serengeti lion: A study of predator-prey relations. Chicago: Chicago Univ. Press.

SCHENKEL, R. (1966): Zum Problem der Territorialität und des Markierens bei Säugern- am Beispiel
des Schwarzen Nashorns und des Löwen. Z. Tierpsychol. 23, 593-626.

SCHNEIDER, R. (1964): Der Larynx der Säugetiere. Handb. Zool. 5(7), 35, 1-128.

Stars, H. J. (1984): Carnıvores. In: Orders and families of recent mammals of the world. Ed. by S.
ANDERSON and J. Knox JONES, Jr. New York: John Wiley and Sons. Pp. 491-521.

TEMBROCK, G. (1962): Methoden der vergleichenden Lautforschung. In: Symposium Theriologicum.
Proceedings of the International Symposium on Methods of Mammalogical Investigation. Ed. by
J. KratocHvır and J. PEerıkan. Prague: Czechoslovak Academy of Sciences. Pp. 329-338.

WAYNE, R. K., BENVENISTE, R. E.; JanczEwskı, D. N.; O’BrıEn, $. J. (1989): Molecular and
biochemical evolution of the Carnıvora. In: Carnivore behavior, ecology, and evolution. E.d. by.
J. L. GiTTLEman. Ithaka: Cornell Univ. Press. Pp. 465494.

WEIGEL, I. (1961): Das Fellmuster der wildlebenden Katzenarten und der Hauskatze in vergleichender
und stammesgeschichtlicher Hinsicht. Säugetierkdl. Mitt. 9, Sonderheft.

WERDELIN, L. (1983): Morphological patterns in the skulls of cats. Biol. J. Linn. Soc. 19, 375-391.

WOZENCRAFT, W. C. (1989): Classification of the Recent Carnıvora. Appendix in: Carnivore
behavior, ecology, and evolution. Ed. by J. L. GıiTTLeman. Ithaka: Cornell Univ. Press. Pp.
569-593.

WURSTER-HiıLL, D. H.; CENTERWALL, W. R. (1982): The interrelationships of chromosome banding
patterns in canıds, mustelids, hyena, and felids. Cytogenet. Cell Genet. 34, 178-192.

Authors’ addresses: Dr. Gustav PETERS, Zoologisches Forschungsinstitut und Museum Alexander
Koenig, Adenauerallee 160, D-53113 Bonn, Germany; Prof. Dr. Marcoım H.
Hast, Northwestern University Medical School, 303 East Chicago Avenue,
Chicago, Illinois 60611, USA

Z. Säugetierkunde 59 (1994) 105-115
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Oral suction of a Pacific walrus (Odobenus rosmarus divergens)
in air and under water

By R. A. KastELein, M. MULLER, and A. TERLOUW

Harderwijk Marine Mammal Park, Harderwijk, Holland

Receipt of Ms. 1. 4. 1993
Acceptance of Ms. 21. 6. 1993

Abstract

Walruses mainly eat sessile bentic prey. Of bivalve molluscs, usually only the siphons and feet are
found in walrus stomachs, and it is thought that walruses use oral suction to separate the molluscs
from their shells. Low pressure in the buccal cavity is caused by retraction and depression of the
tongue which acts like a piston. The pressure in the oral cavity of a female walrus was measured during
several in-air and underwater suction tests. The lowest pressure recorded in aır was -87.9 kPa (-0.879
Bar, almost vacuum) when the walrus sucked on the pressure transducer. The lowest pressure
recorded under water was -118.8 kPa (-1.188 Bar) when the walrus was sucking on a mackerel. The
walrus has good control over its tongue muscles and over both the pressure and the duration of
suction.

Introduction

In contrast to most pinnipeds which prey on fish and squid, walruses mainly eat sessile
bentic prey (Fay 1982). Of bivalve molluscs, usually only the sıphons and feet are found in
walrus stomachs. The bodies may be digested so quickly that they are difficult to detect.
VıBE (1950) suggested that walruses use suction to separate the molluscs from theır shells.
Evidence to support this hypothesis was given by OLivEr et al. (1983, 1985) who found
intact empty shells on both sides of furrows in 'the ocean floor in walrus foraging areas.
KASTELEIN and MosTERD (1989) observed walruses feeding on bivalve molluscs in a sandy
substrate in a pool and leaving the empty shells on the bottom. FAy (1982) suggested that
low pressure in the buccal cavity could be caused by retraction and depression of the
tongue which could act like a piston. KASTELEIN and GERRITS (1990) showed that the
buccal cavity of the walrus ıs relatively large due to the curvature of the maxilla and hard
palate, and KAsTErein et al. (1991) described the well-developed tongue muscles which are
involved in producing oral suction.

The retraction speed of feet and sıphons of molluscs depends on the species but the
process ıs also temperature dependent. There is good evidence that Serripes (Cardıum)
groenlandicus is comparitively slow in retracting its feet. This may allow the walrus to
remove the feet, and possıbly the attached body, before they have retracted into the shells
(MAnSFIELD 1958). The suction force of the walrus, required to separate the body or body
parts from the shells, probably depends on the degree of retraction and closure of the clam.
It is probable that beyond a certain state of retraction, the Walrus is unable to extract the
edible parts.

After detecting an object on the ocean floor, a walrus has a limited amount of time to
identity (KASTELEIN and van GAALEN 1988; KASTELEIN et al. 1990), in certain cases
excavate (OLIVER et al. 1983; KASTELEIN and MoSTERD 1989) and position the prey item
between its lips (KASTELEIN etal. 1991), if it is to use the suction technique succesfully with
a clam. This foraging technique has to be efficient because adult walruses in oceanaria eat
about 50 kg of food per day (KASTELEIN pers. obs.). This would be about 3000 adult sand
gapers (Mya arenaria) with an average soft body weight of 17 g. Born and KnuTtsen

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5902-0105 $ 02.50/0

106 R. A. Kastelein, M. Muller, and A. Terlouw

(1990) even found 6401 individual prey items in a walrus stomach. A bivalve may detect the
vibrations or current caused by an approaching walrus sooner than the walrus detects the
clam, so the retraction of foot and siphon may have begun by the time the walrus touches
the prey (KrıstEensen 1957). So, the amount of time available for food processing is
governed by the detection distance and retraction speed of the mollusc, as well as by the
identification and excavation speed and suction power of the walrus. The prey identifica-
tion speed of walruses was studied in a psychophysical study by KasTELEIN and van
GAALEN (1988). The present study investigates the parameters determining suction force.

Material and methods

Anımal

The study was done with a 10-year-old female Pacific walrus (Odobenus rosmarus divergens, code
OrZH004) which was born in the wild, arrıved at Harderwijk in 1985 and which has participated in
educational performances since then.

Study area

A 50 cmx50 cm square hole was made ın a door between the walrus quarters and an adjacent room. In
this room, a50 cmXx50 cmX50 cm water trough was placed on the floor beneath the hole. The walrus
was trained to put her head through the hole in the door and into the water in the trough on
command.

Experimental procedure

To measure the pressure changes in the walrus’ mouth a Millar PC 350 catheter pressure transducer
was used. This pressure transducer was chosen, because it ıs small (1.67 mm diameter), it has a linear
relationship between pressure and output voltage, a broad bandwidth (about 3 kHz) and it is
stabilized for temperature effects (VAn LEEUWEN and MULLER 1983). The output voltages were
amplified with a differential amplifier (AD 610 K). The signals were stored on a Bell and Howell
recorder (speed 30 inch/s, bandwidth 0-10 kHz), played back to be visualized on a Tektronix digital
storage scope (type 2211) and plotted on a HP 7475A plotter (Hewlett-Packard).

The pressure transducer was inserted into a thawed fısh. To protect the sensor, a hollow metal tube
with a pointed end was first inserted through the body of the fısh from the anus to the mouth. Then,
the wirelike pressure transducer was threaded through the tube until the tip became visible. The tube
was carefully removed and the sensor was placed so that its tip stuck out of the mouth of the fish by
about 1 cm.

The pressure changes in the walrus’ mouth were measured under various circumstances. When a
fish containing the pressure transducer was held in front of the walrus’ mouth, the anımal gripped the
rostral third part of the fish in its mouth, and then tried to suck it from the hand of the trainer (Fig. 1).
The trainer kept a firm hold of two thirds of the fish until it broke or slipped from his hands. This
could be done both in air and under water. Herring (Clupea harengus) and mackerel (Scomber
scombrus) 20 to 25 cm in length were used.

In air, the pressure transducer was also offered while held along the trainer’s finger, the walrus
having been trained to suck the finger. In air, the transducer could also be held perpendicular to the
cheek of the trainer. The walrus had been trained to “kiss” the trainer, and thus to suck the transducer,
which extended about 6 cm into the walrus’ mouth cavity. Both sucking the trainer’s finger and
“kissing” his cheek were known behaviours to the walrus because they were used in educational
performances. In other experiments, the pressure transducer alone was offered to the anımal. In all
experiments, about 6 cm of the catheter tıp was inside the walrus’ mouth during suction.

The suction curves produced in this study consist of a zero level which is equal to the ambient aır
pressure, a descending part in which the pressure is dropping due to the depression and retraction of
the tonge, a section in which maximum pressure is exerted, and an ascending part in which the
pressure is returning gradually to ambient pressure because air or an object slips, or water flows, into
the mouth cavity. For an example see figure 2. The following parameters were calculated:

Amp 0 = ambient pressure.

Amp 99 = 90% of the maximum amplitude.

Amp 10 = 10% of the maximum amplitude (10% below ambient pressure).
Amp df = Amp 90 - Amp 10.

Amp max = maximum amplıtude.

T10 = Duration of the suction event at Amp 10.

Oral suction of a Pacific walrus in air and under water 107

Fig. 1. The walrus sucking a fish containing the pressure transducer. The hole in the door ıs 50x 50 cm
(Photo: HENK MERJENBURGH)

44.2

Ampo

pressure (kPa)
o

- 132.6

0.2 0.4 0.6 0.8 1.0 1.2 1.4
time (sec)

Fig. 2. An example of a curve showing the pressure changes in a walrus oral cavity during oral suction.
(For abbreviations see text)

108 R. A. Kastelein, M. Muller, and A. Terlouw

T90 = Duration of the suction event at Amp 90.
Tde = Time needed to decrease the pressure from Amp 10 to Amp 9.
Tin = Time needed to increase the pressure from Amp 90 to Amp 10.

Results

Table 1 shows the suction curve parameters for the different test situations which will be
described in detail below.

Finger in air

On average the shortest T90 and smallest Amp 90 occurred when the walrus sucked the
finger. A typical example of a suction curve is shown in figure 3A.

Cheek in air

On average the shortest T10, Tde and Tin and the largest Amp 90 occurred when the
walrus “kissed” the trainer’s cheek in air. Using this method, the suction parameters were
less variable in each trıal which is apparent from the relatively small standard deviations. A
typical example of a suction curve ıs shown in figure 3B.

Table 1. Average values (x) and standard deviations (S.D.) of suction parameters during a number
of trials (N) in a 10-year-old female Pacific walrus

Situation In air Under water

Fish

u|
z*
m

Transducer

—r

Qı

5
6
A
5
1
A
40
5
4
1
6
4

Oral suction of a Pacıfic walrus in air and under water 109

Pressure transducer alone

The T10, T90, Tde and Tin were the longest when the walrus sucked on the pressure
transducer alone. The suction curves were quite irregular. An example is shown in figure
3@:
Fish in air

In each trial the animal usually sucked a few times without being able to move the fish. The
final successful suck either pulled the entire fish from the trainer’s hand, or the fish was
pulled apart, leaving about half of it in the trainer’s hand. In both events, the fish or a part
of it was sucked over the sensor shortly after the suction was created. The values of the
parameters measured during trials with herring and mackerel did not differ statistically,
and therefore are analysed together.

Compared to the other test situations (except when the anımal was sucking on the
pressure transducer alone), the T90, Tde and Tin were on average long when the walrus
sucked on fish in air. A typical example of a suction curve is shown ın figure 3D.

Fish under water

The most conspicuous differences between under water and in-aır suction on fish were the
on average shorter T10 and Tde, the longer Tin and larger Amp 90 under water. The lines
of the suction curves were smoother than when fish was sucked on ın air. An example of a
suction curve is shown in figure 3E.

Discussion
Physics

Fay (1982) reports on an anecdotal observation of a walrus which produced a pressure of
around -91.4 kPa (-0.914 Bar) when sucking a tube which was connected to a mechanical
pressure gauge. At the time of measurement, the anımal was pulling aır along the mouth
piece. In the present study, using more sophisticated equipment, the lowest pressure
recorded (Amp max) ın air was -87.9 kPa (-0.879 Bar) when the walrus sucked on the
pressure transducer. The minimum pressure reached while sucking under water was lower;
-118.8 kPa (-1.188 Bar) when the walrus was sucking a mackerel.

The buccal cavity of a walrus can be regarded as a cylinder with a piston (the tongue). In
rest, the tongue fills the buccal cavity almost entirely (KAsSTELEIN et al. 1991), so the initial
volume of the cavity is practically zero. When the tongue is withdrawn to a caudal
position, the volume of the buccal cavity is enlarged.

If the mouth cavity is filled with air, pressure (p) and mouth volume (V) are related by
Boyle’s law: p;V; = pr V; (1 = initial, f = final). As V5>V; and p; = the atmospheric
pressure = 1 atm., it follows that p; ıs close to 0 atm.; so the mouth cavıty ıs close to
vacuum. In air, the pressure cannot reach a value below O atm., so that a pressure
transducer may record a pressure of maximally 1 atm. below the “baseline” of ambient
pressure.

In water, the situation is different, and Boyle’s law does not apply. When the piston is
withdrawn to generate pressure of 0 atm., the water column will break after a certain short
period and the cavity will be filled with water, water vapour and gas that was originally
dissolved in the water. This phenomenon is unstable and is called “cavitation”. At
sufficiently low pressure, cavitation always occurs for a time, the length of which depends
on the pressure and on the concentration of particles and dissolved gas in the water.

On a shorter time scale, the force applied to the piston may be transformed directly to

1110) R. A. Kastelein, M. Muller, and A. Terlouw

S 0) RS 0
& &
5 SID 5 SD28]
- 44.2 - 44.2
- 66.3 - 66.3
0.1 0.3 0.2 0.4
time (sec) time (sec)
0
SS
E21
®
2
&
- 44.2
- 66.3
0.2 0.4 0.6 0.8 1.0 11.2 1.4 1.6 2.0
time (sec)
Fig. 3 A-C

pressure. So, a walrus may theoretically be able to generate a pressure of more than 1 atm.

below ambient pressure for a short time.

In the above explanation, ıt is assumed that the flow through the mouth aperture is
insignificant. In reality, food will be sucked into the oral cavity during tongue retraction
and the pressure changes in the buccal cavity will be more complex.

Finger in air

The short T90 and small Amp90 are probably due to the fact that this was a trained
behaviour in which the anımal was rewarded for producing the “suction” sound, and not
for the maximum suction power, or duration. From experience the animal probably knew
it would not be able to suck in the object (= finger).

Cheek in air

The suction parameters between trials were rather similar when the anımal sucked on a
cheek in air. This is probably because other than the walrus’ tongue, nothing moved into

Oral suction of a Pacific walrus in air and under water an

- 22.1

pressure (kPa)

- 44.2

- 66.3
0.2 0.4 0.6 0.8 1.0
time (sec)

223]

pressure (kPa)

- 44.2

- 66.3

- 88.4

- 110.5 |

0.2 0.4 0.6 0.8 1.0 12 1.4 |
time (sec)

Fıg. 3 (see also left side). Typical examples of suction curves on the same scales of time and pressure of
A: Finger ın air, B: Cheek ın air, C: Pressure transducer alone, D: Fish in air, and E: Fish under water

the oral cavity and because the walrus controlled the entire situation. It started the trial by
putting its lips against the cheek with a certain pressure, retracted its tongue, and allowed
some air to enter the oral cavity to produce the “kiss” sound it was trained to produce in
performances. The lips did not surround an object, but were pushed against the cheek of
the trainer. This apparently also closed off the oral cavity tightly, resulting in a high
Amp90. However, this parameter was not as high as when sucking fish under water. Air
flows faster into the oral cavity than water, thus the pressure amplitudes in air are less
negative than in water.

112 R. A. Kastelein, M. Muller, and A. Terlouw

Pressure transducer alone in air

The long duration of the suction event when sucking the transducer alone in air can be
explained by the fact that the transducer only filled a minute portion of the oral cavity,
leaving a large volume of aır to be expanded. This allowed the tongue to retract completely,
which took more time. The relatively large Amp90 indicates that the walrus is able to close
its lips tightly, even around such a small diameter pressure transducer. However, some air
probably leaked into the oral cavity causing the ırregularities in the suction curves.

Fish in air
The suction curves produced on fish in air are irregular. Pressure changes are probably
caused by air leaking around the fish while it was sliding into the oral cavity. The suction
events are long probably because the walrus attempted to swallow the fish. The irre-

gularities at the end of the suction event in figure 3D are caused by the fish slipping over
the pressure transducer.

Fish under water

During suction on fish under water the Tde was shorter, the Amp90 much higher, and the
Tin longer than ın air, probably because water flowed into the oral cavity more slowly than
air. During the Tin the fish was pulled from the hand quickly and some water probably
flowed into the oral cavity. The irregularity at the end of the suction event ın figure 3E is
caused by the fish slipping over the pressure transducer.

Correlations between suction parameters

To determine the level of control the walrus has over the different suction parameters, the
correlations between these parameters were calculated. The correlations between the
different suction curve parameters ın all situations in air and under water are shown in table
2. The parameters within the following pairs are positively correlated both in aır and under
water: T10-Tde, T10-IT90 and T10-Tin. This means that the longer the entire suction event
(T10), the longer each of ıts 3 time components (Tde, T90 and Tin). Under water the
parameters of the following 2 parameters pairs are positively correlated: Amp90-T90 and
Tin-Tde. These are not correlated in air. This is probably due to the high density of water

Table 2. The correlation between the different suction parameters in air (A) and under water (B)

A - indicates no significant correlation. A + indicates significant correlation (p <0.05)

A. Suction ın air (N = 29, significance at r > 0.306)

T90

Tde

Tin Tin
Amp9%0 +

B. Suction in water (N = 17, significance at r > 0.412)

T9O
Tde
Tin
Amp90

Oral suction of a Pacıfic walrus in air and under water 118)

compared to air. Water passes between the object and the lips with more difficulty, causing
a higher force to retract the tongue which causes a slow sliding movement of the prey. In
air, probably too much air leaks into the oral cavity, and the anımal may not be able to
maintain a large Amp90 for a long period without air leaking into the oral cavity. In air, the
parameters of the following 2 pairs are positively correlated: Tde-T90 and Tin-T90. These
are not correlated under water. This is probably due to the different shapes of the suction
curves; ın air, they are often V-shaped, while under water they are usually U-shaped.

Ecological significance

The present study shows that at least 3 parameters influence the shape of the suction curve
(Tab. 3). The Tde is influenced by the retraction speed of the tongue, the tightness of the
lips on the food item, and the presence of sealing mucus. The T90 is influenced by the time
the anımal can keep or decides to keep its tongue retracted, the tightness of the lips on the
food item, and by the toughness and slipperiness of the food item and the strength of the
trainer. The Amp90 ıs influenced by the strength and volume of the tongue, the initial
position of the tongue in the buccal cavıty, the fit of the tongue in the buccal cavity, the
tightness of the lips around the food ıtem and also by the shape, firmness and toughness of
the food item and the presence of sealing mucus.

Table 3. Parameters which influence the shape of the suction curve

Tde T9O Amp90

Walrus 1. Retraction speed of . Time the anımal can . Strength of tongue
tongue keep or decides to . Inıtial position of
2. Tightness of lips on keep its tongue re- tongue
food item tracted . Fit of tongue in buccal
. Tightness of lips on cavity
food item . Volume of tongue
. Tightness of lips on
food item

Food item Sealing mucus . Toughness . Shape
2. Slipperiness . Firmness
. Toughness
. Sealing mucus

Trainer Strength

In air, the walrus normally uses its capacity for oral suction mainly during its suckling
period. Ray (1960) describes a walrus calf which emptied a 225 ml baby bottle in 15
seconds, and often sucked the plastic container flat. Walruses have a relatively long
suckling period of at least 15 months (Fay 1982), and suckling probably occurs both on
land and under water (MıLLEr and Boness 1983). Some walruses sometimes eat seals and
use in air and under water oral suction to process their prey. Only strips of skin and
blubber are found in walrus stomachs, indicating that these parts only are sucked off
without mastication (CoLLıns 1940; BrEsHIn 1958; PErRY 1967; Lowry and Fay 1984;
Fay et al. 1990; TimosHENKo and Popov 1990).

When a walrus encounters a clam in the sea bed, whether the clam will be eaten or not
depends on the behaviour of both organisms. For the clam, its shape, firmness and
toughness and the ability to detect a walrus are of importance. Shape and firmness are fixed
properties of a clam, but the toughness of the siphon may depend on its retraction state.
The more retracted, the more difficult it is to suck the clam out of its shell. For the walrus

114 R. A. Kastelein, M. Muller, and A. Terlouw

the volume, strength and retraction speed of the tongue and the firmness of the lips on a
food item are of importance. The volume seems to be a fixed property, unless the walrus
can retract the tongue partly during a suction event. The walrus can probably determine
the strength and retraction speed of its tongue and the pressure on the clam with its lips.
The pressure of the funnel-shaped lıps should be sufficient to prevent water from flowing
around the prey into the oral cavity, but low enough to prevent the clam’s shells from
breaking. Depending on the toughness of its prey, the walrus may retract its tongue faster,
or use more muscle bundles. Possibly, the walrus has an expectation of the toughness of its
prey before it sucks. If that is true, the Tde ıs mainly consciously determined by the
walrus. This explains also part of the correlations in table 2. If a sıphon is slowly stretched
during a suck, the toughness ıs slowly decreased until the sıphon breaks off the clam’s
body. This might explain the function of the long 'T'90’s found in the present study (in air
625 ms and under water 658 ms); because the walrus was eager to swallow the fish, she
kept her tongue retracted for a longer time.

The present study provides insight into the control a walrus has over its oral suction
power, speed and duration. Whether the walrus can process bıvalve molluscs at all stages of
sıphon and foot retraction remains to be determined.

Acknowledgements

We thank trainer PIET MOSTERD for his practical help with the experiment, Dr. BERNHARD NEUROHR
(Duisburger Zoo) for translation of the summary, and Prof. Jan Osse and Prof. PIET WIEPKEMA
(Agrıcultural University of Wageningen), and Nancy VAUGHAN for their comments on the manu-
script.

Zusammenfassung

Orales Saugvermögen eines Pazifischen Walrosses (Odobenus rosmarus divergens)
in Luft und unter Wasser

Walrosse fressen hauptsächlich sessilen tierischen Benthos. Von zweischaligen Weichtieren werden
normalerweise nur die Siphonen und Füße in Walroßmägen gefunden. Es wird allgemein angenom-
men, daß Walrosse durch ihr Saugvermögen Weichtiere von deren Schalen trennen können. In der
Mundhöhle kann Unterdruck dadurch erzeugt werden, daß die wie ein Kolben funktionierende
Zunge zurück- und heruntergezogen wird. Während mehrerer Über- und Unterwasser-Saugtests
wurde der Druck in der Mundhöhle einer Walroßkuh gemessen. Durch Saugen am Druckübermittler
konnte über Wasser als niedrigster Druck - 87,9 kPa (-0,879 bar; fast Vakuum) gemessen werden.
Beim Ansaugen einer Makrele unter Wasser konnte als niedrigster Druck -118,8 kPa (-1,188 bar)
registriert werden. Da das Walroß seine Zungenmuskeln präzise kontrollieren kann, sind Druck und
Dauer des Saugaktes gut regulierbar.

References

Born, E. W.; Knutsen, L. ©. (1990): Satelite tracking and behavioural observations of Atlantic
Walrus (Odobenus rosmarus rosmarus) in NE Greenland in 1989. Teknisk rapport-Gronlands
Hjemmestyre, Afdelingen for Levende Ressourcer. 20 Oct. 1990, 1-68.

BresHin, A. (1958): Bloodthirsty walrus. Ogonek (Moskow) 41 (1634), 30.

Coins, G. (1940): Habits of the Pacific Walrus (Odobenus divergens). J. Mammalogy 21, 138-144.

Fay, F. H. (1982): Ecology and biology of the Pacific walrus (Odobenus rosmarus divergens llliger).
North American Fauna 74, 1-279.

Fav, F. H.; SEASE, J. L.; MERRICK, R. L. (1990): Predation on ringed seal, Phoca hispida, and a black
guillemot, Cepphus grylle, by a Pacific Walrus, Odobenus rosmarus divergens. Mar. Mamm. Scı.
6, 348-350.

KASTELEIN, R. A.; GAALEN, M. A. van (1988): The sensitivity of the vibrissae of a Pacific Walrus
(Odobenus rosmarus divergens). Part 1. Aquatic Mammals 14, 123-133.

KASTELEIN, R. A.; GERRITS, N. M. (1990): The anatomy of the Walrus head (Odobenus rosmarus).
Part 1: The Skull. Aquatic Mammals 16, 101-119.

KASTELEIN, R. A.; GERRITS, N. M.; DUBBELDAM, J. L. (1991): The anatomy of the Walrus head
(Odobenus rosmarus). Part 2: Description of the muscles and of their role in feeding an haul-out
behaviour. Aquatic mammals 17, 156-180.

Oral suction of a Pacific walrus in air and under water 115

KAsTELEIN, R. A.; MosTErD, P. (1989): The excavation technique for molluscs of Pacific Walruses
(Odobenus rosmarus divergens) under controlled conditions. Aquatic Mammals 15, 3-5.

KASTELEIN, R. A.; Paasse, M.; KLINKHAMER, P.; WIEPKEMA, P. R. (1991): Food dispensers as
occupational therapy for the Walrus (Odobenus rosmarus divergens) at the Harderwijk Marine
Mammal Park. Int. Zoo Yearb. 30, 207-212.

KASTELEIN, R. A.; STEVENS, $.; MOSTERD, P. (1990): The tactile sensitivity of the mystacial vibrissae
of a Pacific Walrus (Odobenus rosmarus divergens). Part 2: Masking. Aquatic Mammals 16,
78-87.

KRISTEnSEn, I. (1957): Differences in density and growth in a Cockle population in the Dutch
Waddensea. Diss. Leiden State Univ., Holland.

LEEUWwEn, J. L. van; MULLER, M. (1983): The recording and interpretation of pressures in prey-
sucking fısh. Neth. J. Zool. 33, 425475.

Lowry, L. F.; Fay, F. H. (1984): Seal eating by Walruses in the Bering and Chukchi Seas. Polar Biol.
3, 11-18.

MansFIELD, A. W. (1958): The biology of the Atlantic walrus, (Odobenus rosmarus rosmarus)
(Linnaeus) in the eastern Canadian Arctic. Fish. Res. Board of Canada. Biol. Ser. 653.

MiLLEr, E. H.; Boness, D. J. (1983): Summer behavior of Atlantic walruses (Odobenus rosmarus
rosmarus) (L.) at Coats Island, N.W.T. (Canada). Z. Säugetierkunde 48, 298-313.

OLIVER, J. S.; KvITEk, R. G.; SLATTERY, P. N. (1985): Walrus feeding disturbances scavenging habits
and recolonization of the Bering Sea Benthos. J. Exp. Mar. Biol. Ecol. 91, 233-246.

OLIVER, J. S.; SLATTERY, P. N.; O’Connoß, E. F.; Lowry, L. F. (1983): Walrus (Odobenus rosmarus)
feeding in the Bering Sea: a benthic perspective. Fishery Bulletin 81, 501-512.

PERRY, R. (1967): The world of the Walrus. London: Cassel.

Ray, C. (1960): Background for a baby Walrus. Anım. Kingd. 63, 120-124.

TIMOSHENKO, Iu, K.; Popov, L. A. (1990): On the predatory habits of the Atlantic Walrus. In: The
ecology and management of Walrus populations. Ed. by F. H. Fay, B. P. Kerry and B. A. Fay:
Report T68 108 850 for the U.S. Marine Mammal Commission. Washington.

VisE, C. (1950): The marıne mammals and the marine fauna in the Thule District (North-west
Greenland) with observations on ice conditions in 1939-1941. Medd. om Grenl. 150, 1-115.

Authors’ addresses: Ron A. KASTELEIN, Harderwijk Marine Mammal Park, Strandboulevard-oost 1,
NL-3841 AB Harderwijk, Holland; MEes MuLLER and ArıE TERLOUW, Depart-
ment of Experimental Anımal Morphology and Cell Biology, Agricultural Uni-
versity, Zodiac, Marijkeweg 40, NL-6709 PG Wageningen, Holland

Z. Säugetierkunde 59 (1994) 116-125
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Comparative morphometry and cytogenetics of
Microtus (Terricola) multiplex (Arvicolidae, Rodentia)
of the western French Alps

By P. BRUNET-LECOMTE and V. VOLOBOUEV

Centre des Sciences de la Terre, Universite de Bourgogne, Dijon, France and Institut Curie,
Parıs, France

Receipt of Ms. 25. 11. 1991
Acceptance of Ms. 11. 2. 1993

Abstract

The western French Alpine populations of Microtus (Terricola) multiplex (Arvicolidae, Rodentia)
occupy the marginal distributional area of the species. Analysis of the characters of the Relative
Length of the Anterior Part (RLAP) and of the Pitymyan Rhombus (PR) of the first lower molar (M,)
of these populations and subsequent comparison with other populations from the Central Alps of
France and Switzerland permits to distinguish the populations of the western French Alpine area
(Saint-Martin-de-la-Cluze and La-Chapelle-en-Vercors) from the populations of the Central Alps
belonging to the subspecies multiplex, fatior and druentins. Karyological analysis of the population
from Saint-Martin-de-la-Cluze confirms the cytogenetic differentiation of populations from the
western French Alpine area which are characterized by a subtelocentric X chromosome and a
metacentric Y chromosome. The synthesis of these results leads to the proposal of a new subspecies
for the populations from the western French Alpine area, Microtus (Terricola) multiplex niethammeri
nov. Ssp. /

Introduction

European ground-voles of the subgenus Terricola constitute an advantageous group for
study of speciation because of their geographic chromosomal varıation (MEyYLAN 1970,
1972; WınkınG 1976; STORCH and WInkInG 1977; GRAF and MEyLAn 1980) and their
discontinuous distribution (NIETHAMMER and Krapp 1982). The Alpıne ground-vole
M. (T.) multiplex (Fatıo 1905), a species of the Middle European phyletic group (BRUNET-
LECOMTE 1990) is characterized by chromosomal polymorphism of 2n = 46 to 48 and NF =
52 to 54 (NIETHAMMER and KrapPp 1982). Since the species was first described by FATIo ın
1905, six subspecies have been identified: multiplex (Farıo 1905) from Lugano (Ticino,
Switzerland), fatior (MoTTAz 1990) from Zermatt (Wallis, Switzerland), druentius (in
Mırrer 1912) from Terres-Plaines near Barcelonnette (Alpes-de-Haute-Provence,
France), orientalis (Dar Pıaz 1924) from Madonna dı Campiglio (Trentino, Italy),
liechtensteini (WETTSTEIN-WESTERSHEIM 1927) from the top of Malı Rainac mountain near
Krasno (Croatia, Yugoslavia), petrovi (KrysTUFEK 1983) from Socerga near Koper
(Slovenia, Yugoslavia). In France the species occurs in the southern Central Alps (Mercan-
tour, Übaye, Briangonnais), northern Central Alps (Vanoise), western Pre-Alps (Vercors,
Chartreuse) and perhaps further west as far as the Rhone Valley and eastern Massıf Central
(HEIM DE BaLsac and BEAUFORT 1966; BRossET and HEIM DE Bausac 1967; NIETHAMMER
and Krarp 1982; FAyarD 1984). Cytogenetic and electrophoretic studies of the western
and Central Alpine subspecies multiplex, fatioi and druentinus (GRAF and MEyLAan 1980)
underline the differentiation of the population of La-Chapelle-en-Vercors from the
western Pre-Alps of Vercors (Tab. 1). This population ıs characterized by a subtelocentric
X chromosome (GRAF and MEyLAn 1980) and a high genetic distance of Nei in comparison
with the other studied populations (between 0.09 and 0.18) (GrAar and MEyLAn 1980),

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5902-0116 $ 02.50/0

Comparative morphometry and cytogenetics of Microtus multiplex 117
Table 1. Chromosome data of Microtus (Terricola) multiplex

Locality Karyotype References
2n FN Autosomes

1 Gudo (E), Meride (F), 48 50 2IST+2mM+42A GRAF and
Varenzo (H), Zermatt (I) MEyLAn (1980)

2 La Cayolle (]) 48 50 2IST+2mM+42A * GRarFand
Le Lautaret (C) MEYLAN (1980)

3 La-Chapelle-en 48 50 2IST+2mM+42A GRAF and
en-Vercors (B) MEyLANn (1980)

4 Saint-Martin de 48 50 2IST+2mM+42A present study
la-Cluze (A)
5 Fivizzano (L) 46 48 2IST+2mM+40A SM A GRAF and
MEYLAN (1980)

6 Arsıe (M) 46 46 21A+2mM+40A ISM IA STtorcHand
Wınkıng (1977)

7 Calliano (N) 46 4 21A+2mM+40A ISM ISM STorcH and
Wınkıng (1977)

Locality: letters in brackets denote the labels of populations from figure 1. Karyotype: A =
acrocentric, ST = subtelocentric, SM = submetacentric, M = metacentric, m = medıum, | = large.
*: although the authors suggest that the unpaired medium-sized metacentric may be the Y
chromosome, definitive interpretation is made by us.

corresponding to an inter-subspecies distance, whereas the genetic distances calculated
between the populations from Ticino (subspecies multiplex), Zermatt (subspecies fatior)
and Brianconnais (subspecies druentins) are smaller (between 0.02 and 0.06).

The aim of this study is to present the karyotype of a population of M. (T.) multiplex
from Saint-Martin-de-la-Cluze, ın the Drac Valley, which lies near, although outside, the
Pre-Alps of the Vercors; to compare the karyotype and the first lower molar morphology
of this population with those of the other populations of the Alps and to clarıfy the
systematics of the populations from the western French Alpine area.

Material and methods

Nine populations were studied, each presented by 10 teeth (M,). The populations were from: A:
Saint-Martin-de-la-Cluze (Isere, France); B: La-Chapelle-en-Vercors (Dröme, France); C: Col du
Lautaret (Hautes-Alpes, France); D: Les Vigneaux (Hautes-Alpes, France); E: Gudo (Ticino,
Switzerland); F: Meride (Ticino, Switzerland); G: Bioggio (Ticino, Switzerland); H: Varenzo (Ticino,
Switzerland); I: Zermatt (Wallis, Switzerland).

Localities are mapped in figure 1.

Morphology of the first lower molar (M,)

The analysıs of the characters Relative Length of the Anterior Part (RLAP) and Pytymyan Rhombus
(PR) was based on the comparison of means between the populations according to the analysis of
varıance method completed by Scheffe’s method.

Chromosomal study

Chromosome preparations were obtained from primary fibroblast cultures from tail biopsies of four
specimens (238 and 2?) all from Saint-Martin-de-la-Cluze (Isere, France). Explants and a portion
of the cells of studied specimens are routinely kept in liquid nitrogen in the cell and tissue collection of
the Laboratoire de Structure et Mutagen&se Chromosomiques (Institut Curie, Paris, France).

118 P. Brunet-Lecomte and V. Volobouev

GERMANY
FRANCE
SWITZERLAND NInTeE
P= H = =?
=> P
16,0 = =
2 > „= x
—g RR
FAZC = ITALY >> YUGOSLAVIA
DES
= Q

|
„N

Fig. 1. Localities of the populations studied. Morphological study. A: Saint-Martin-de-la-
Cluze (Isere, France); B: La-Chapelle-en-Vercors (Dröme, France); C: Col du Lautaret (Hautes-
Alpes, France); D: Les Vigneaux (Hautes-Alpes, France); E: Gudo (Ticino, Switzerland); F: Meride
(Ticino, Switzerland); G: Bioggio (Ticino, Switzerland); H: Varenzo (Ticino, Switzerland); I:
Zermatt (Wallis, Switzerland). Karyological study. A;B; GC; E; F; H; 1; J: Col de la Cayolle
(Alpes-de-Haute-Provence, France); L: Fivizzano (Toscana, Italy); M: Arsıe (Italy); N: Calliano

(Italy).

Type localities: ©: Lugano (Ticino, Switzerland) = multiplex; I = fatioi; K: Terres-Plaines
near Barcelonnette (Alpes-de-Haute-Provence, France) = druentius; P: Madonna di Campiglio
(Irintino, Italy) = orientalis; Q: Top of Mali Rainac mountain near Krasno (Croatia, Yugoslavia) =
liechtensteini; R: Socerga near Koper (Slovenia, Yugoslavia) = petrovi.

Hatching: the Alps

Mitotic chromosomes were studied with RHG R-banding and CBG C-banding (HARNDEN and
KLinGEr 1985) after, CARPENTIER et al. (1972) and SUMMER (1972) respectively. Replication banding
(RBG) was studied using the method of VIEGAs-P£EQUIGNOT and DUTRILLAUX (1978). At least 20
metaphase plates were analysed for each specimen.

Results
Morphological analysis of M, characters

a. RLAP: The mean and the standard deviation of the RLAP of each population are given
in table 2. The distribution of means of each population is given in figure 2. The RLAP ıs
significantly less developed in the populations A and B (Saint-Martin-de-la-Cluze and La-
Chapelle-en-Vercors) than in the populations C and D (Brianconnais) and E to H (Ticino);
population I (Zermatt) having an intermediate RLAP between those of these populations.
b. PR: The mean and the standard deviation of the PR of each population are given in table
3. The distribution of means of each population is given in figure 3. The PR is significantly
greater (less inclined) in the populations C and D (Brianconnais) than in the other
populations.

Chromosomal study

Karyotypes of all four specimens from Saint-Martin-de-la-Cluze are similar and charac-
terized by 2n = 48 and FN = 54. All pairs of autosomes and both sex chromosomes were
precisely identified by high resolution R-bandıng (Fig. 4). Among autosomes two pairs are
biarmed, one of them, the largest in the karyotype, is subtelocentric, while the other N° 6
is a medium sized metacentric. The X chromosome is subtelocentric and sımilar ın size to

Comparative morphometry and cytogenetics of Microtus multiplex IS

Table 2. Mean and standard deviation (SD) of the Relative Length of the Anterior Part (RLAP)
of the M, (no units)

Analysis of variance and Scheffe’s test. Populations (Pop.) A-I as shown in figure 1

I
°

Group N SD Scheffe’s test”

0.019
0.025 AandB 20 0.505 0.021
0.025
0.014 @fand!D7 20 0.523 0.020
0.011
0.013
0.012 EtoH 40 0.523 0.012
0.012
0.015 I 10 0.513 0.015
Analysis of variance p = 0.0007

A
B
€
D
E
je

G
H
I

* Means with the same number are not significantly different.

Table 3. Mean and standard deviation (SD) of the Pitymyan Rhombus (PR) of the M,
Analysis of variance and Scheffe’s test. Populations (Pop.) A-I as shown in figure 1 (unit 10°” mm)

SD Scheffe’s test”

e
I

an DB el)@er

3a

Analysis of variance p<0.0001

* Means with the same number are not significantly different.

the largest acrocentric pair, the Y chromosome is metacentric and similar in size to 14-15th
pairs of autosomes. C-bandıing analysıs (Fig. 5) revealed that all the autosomes possess a
small centromeric block of C-heterochromatın. The Y chromosome is almost entirely C-
positive. The short arms of the X chromosomes are late replicating and thus heterochroma-
tic (Fig. 4) although C-heterochromatın was detected in their proximal parts (Fig. 5). No
obvious varıation of C-heterochromatin was noticed either among the cells or among the
anımals studied.

Discussion

The analysıs of the characters RLAP and PR shows that populations A and B (Saint-
Martin-de-la-Cluze and La-Chapelle-en-Vercors) can be clearly distinguished, on the one
hand, from populations C and D (Briangonnais) of the inner French Alps by their more
steeply inclined PR and by their less developed RLAP, and on the other hand, from
populations E to H of Ticino by their less developed RLAP. This same analysıs of
characters shows that populations A and B are close to population I of Zermatt (Wallis)
despite the geographical isolation of this latter population (GRAF and MEYLAN 1980). In
comparison with other existant European species, Microtus (Terricola) multiplex ıs charac-

120 P. Brunet-Lecomte and V. Volobouev

RLAP
0.530
0.520
0.510
0.500
A B (€ D E F G H I

POPULATIONS

Fıg. 2. Distribution of the mean +/- standard error of the mean of the Relative Length of the Anterior
Part (RLAP) of the M,. For populations’ A-I distribution see figure 1 (no units)

A B € D E F G H I
POPULATIONS

Fig. 3. Distribution of the mean +/- standard error of the mean of the Pitymyan Rhombus (PR) of the
Mı. For populations’ A-I distribution see figure 1 (unit 10°” mm)

terized by a poorly inclined PR (BRUNET-LECOMTE 1990). Therefore the inclined PR of
populations A and B leads us to believe that the populations from the western French Alps
are derived from a Central Alpine subspecies (druentius or multiplex).

Microtus (Terricola) multiplex has been relatively well studied karyologically: to date, at
least 25 populations have been studied over most of its distributional area (see Zıma and
Krar 1984; ZAGORODNYUR 1990). The published data show that all cases of chromosomal

Comparative morphometry and cytogenetics of Microtus multiplex 121

zu.»
Eu 22%
Pe ua
anaew
= u
er
I >

SQ ma!

2 - 4
’$

a.
u...

© az»

— 6
2:

8 AL Pr

a us T AR ö8
16 17 18 19 20
s4 56 8: r

21 22 23 y

Fig. 4. R-banded (RBG) chromosomes of Microtus (Terricola) multiplex. Female sex chromosomes are
given in the insert. One of the X chromosomes and both short arms are late replicating

varıation but one (see below) have been observed between populations, thus, pointing to
chromosomal polytypy and not chromosomal polymorphism. All known chromosome
forms including the new one from Saint-Martin-de-la-Cluze are presented in table 1. As
can be seen from these data, there are two types of karyotypes among these forms, with 46
and 48 chromosomes. The Italian karyotypic forms with 2n = 46 are different from each
other. The form from Fivizzano (GRAF and MEYLAN 1980) differs from the 48 chromo-
some forms by the loss of a pair of acrocentric autosomes, while the sex chromosomes are
similar to those in form 1 (Tab. 1). The next two forms with 2n = 46 (differing from each
other by morphology of the Y chromosome) both belong to subspecies hechtensteini and
are different from all 48 chromosome forms, as well as that wıth 2n = 46 from Fivizzano.
These differences concist in a translocation of both sex chromosomes into a pair of large

122 P. Brunet-Lecomte and V. Volobouev

P}
. »* . © .. £>»

u

1 2 3 & 5
.. “ . 0 . +

23

XY

Fig. 5. C-banded chromosomes of Microtus (Terricola) multiplex. The sex chromosomes of a female
are given in the insert

acrocentric autosomes, and an acrocentric state of the largest pair of autosomes in
comparison with a submetacentric state in all other forms and the presence of an additional
subtelocentric pair of autosomes. In our opinion these rearrangements are sufficient to
ensure cytogenetic isolation of this form, and we subscribe to PETROV and ZıvKovic’s
(1971) conclusion that hechtensteini ıs a distinet species although this was queried in
subsequent studies (STORCH and WInKInG 1977; GRAF and MEYLAN 1980; ZAGORODNYUK
1990). The differences between karyotypic forms with 2n = 48 are caused by different
morphology of the sex chromosomes. Therefore, the X chromosome may be submetacen-
trıc or subtelocentric and the Y chromosome acrocentric or metacentric. All four possible
combinations of different varıants of the X and the Y chromosomes are found in nature,

Comparative morphometry and cytogenetics of Microtus multiplex 103

always in different populations. The only exception is a female heterozygous specimen
featuring submetacentric and subtelocentric varıants of the X chromosome. It was col-
lected in a population of Brianconnais not far from the population of La-Chapelle-en-
Vercors (Tab. 1) and was probably of hybrid origin (GrAr and MEYLAN 1980).

The karyological differences between 48 chromosome forms are hardly capable of
providing cytogenetic isolation by themselves but they might serve as indications as to the
level of differentiation and origin of the chromosome forms. Therefore, the occurrence of
the same variant of the X chromosome in two neighbouring populations of M. (T.)
multiplex (Saint-Martin-de-la-Cluze and La-Chapelle-en-Vercors) not found elsewhere
undoubtedly means they have a common origin. The close relationships between these
populations also follow from morphological analysıs of Mı. On the other hand, the
population of La-Chapelle-en-Vercors is genetically more distant from all the other
studied populations (GRAF and MEYLAn 1980). These observations are in agreement with
paleontological data on the isolation of western Alpine populations of M. (T.) multiplex
during the Upper Pleistocene cited by Grar and MEYLAN (1980).

Thus, the data provide new evidence concerning the particular taxonomic position of
populations in the western French Alps.

The karyological, genetical and morphometrical analyses made by GrAr and MEYLAN
(1980) and our research lead to the following conclusion: the populations from Saint-
Martin-de-la-Cluze and La-Chapelle-en-Vercors are sufficiently differentiated to be clas-
sified as belonging to at least a new subspecıies.

Microtus (Terricola) multiplex niethammeri nov. ssp.

Holotype: Adult male (skın and skull), Centre des Sciences de la Terre de l’Universite
de Bourgogne N° DIMMUL891001.

Type locality: Saint-Martin-de-la-Cluze, Isere, Rhöne-Alpes, France.

Diagnosis: Subspecies characterızed by the following karyotype: Autosomes: 2 large
subtelocentrics, 2 medium metacentrics, 42 acrocentrics; X chromosome: subtelocentric,
Y chromosome: metacentric.

Morphological diagnosis of Mı: RLAP not very well-developed for the species Microtus
(Terricola) multiplex: mean +/- standard error of the mean = 0.507 +/- 0.006 (no units).
PR inclined for the species Microtus (Terricola) multiplex: mean +/- standard error of the
mean = 1.40 +/- 1.1 (unit 10°” mm).

Distribution: Known from Saint-Martin-de-la-Cluze. The population from La-
Chapelle-en-Vercors which differs by ıts Y acrocentric chromosome can be include in thıs
subspecies.

Etymology: In honour of Prof. Dr. Jochen NIETHAMMER, Bonn, for his work on
European voles.

Acknowledgements

We are indebted to F. CATZEFLIS and A. NADACHOwsKI for helpful comments and suggestions, to
T. VoGEL and V. PArısoT for translation, to J. D. Grar, A. MEyLAn, L. DE Rocuın and M. SALvIoNI
for the loan of the M, from La-Chapelle-en-Vercors, Gudo, Meride, Bioggio, Varenzo, Zermatt; to
F. Sprrz for the loan of the M, from Col du Lautaret and Les Vigneaux; to J. F. ConsTanr for the help
in trapping voles at Saint-Martin-de-la-Cluze.

This research was supported by the French C.N.R.S.: Programme “Modalites evolutives” de
’URA 157 and A.S.P. “Evolution” as well as “Approches interdisciplinaires et developpements
methodologiques” de la Direction de la Recherche et des Etudes Doctorales, grant n° EU 1969 (resp.
J- CHALINE).

124 P. Brunet-Lecomte and V. Volobouev

Resume

Morphometrie et cytogenetique comparees de Microtus (Terricola) multiplex (Arvicolidae, Rodentia)
des Alpes occidentales frangaises

Les populations de la partie occidentale des Alpes frangaises de Microtus (Terricola) multiplex
(Arvicolidae, Rodentia) occupent une position marginale dans l’aire de distribution de l’espece. La
comparaison des caracteres longueur relative de la partie anterieure et rhombe pitymyen permet de
separer les populations de la partie occidentale des Alpes frangaises des autres populations des Alpes
internes. L’analyse cytogenetique de la population de Saint-Martin-de-la-Cluze (Isere, France)
confirme la differenciation chromosomique des populations de la partie occidentale des Alpes
frangaises, lesquelles sont caracterisees par un chromosome X subtelocentriqueet un chromosome Y
metacentrique. La synthese de ces resultats et de ceux de la litterature conduit ä proposer une sous-
espece nouvelle pour les populations de la partie occidentale des Alpes frangaises, Microtus (Terricola)
multiplex niethammeri nov. ssp.

Zusammenfassung

Morphometrie und Cytogenetik von Microtus (Terricola) multiplex (Arvicohidae, Rodentia) der
westlichen französischen Alpen im Vergleich

Die Populationen von Microtus (Terricola) multiplex (Arvicolidae, Rodentia) im westlichen Teil der
französıschen Alpen nehmen eine Randposition bei der räumlichen Verteilung der Art ein. Der
Vergleich der Merkmale relative Länge des Vorderteils des Mı und Pıtymys-Rhombus am M, erlaubt
eine Abtrennung der Populationen des westlichen Teils der französischen Alpen von den Populatio-
nen der inneren Alpen. Die Analyse der Chromosomensätze der Population von Saint-Martin-de-la-
Cluze (Isere, France) bestätigt die chromosomale Differenzierung der Populationen des westlichen
Teils der französischen Alpen, die gekennzeichnet sind durch ein subtelozentrisches X-Chromosom
und ein metazentrisches Y-Chromosom. Diese Ergebnisse und die aus der Literatur führen zu dem
Vorschlag einer neuen Unterart in den Populationen des westlichen Teils der französischen Alpen:
Microtus (Terricola) multiplex niethammeri nov. ssp.

References

BrosseET, A.; HEım DE Barsac, H. (1967): Les micromammiferes du Vercors. Mammalıa 31, 325-356.

BRUNET-LECOMTE, P. (1988): Les campagnols souterrains (Terricola, Arvicolidae, Rodentia) actuels et
fossiles d’Europe occidentale. These Doctorat, Univ. Bourgogne.

— (1990): Evolution morphologique de la premiere molaire inferieure des campagnols souterrains
d’Europe (Arvicolidae, Rodentia). Z. Säugetierkunde 55, 371-382.

CARPENTIER, $.; DUTRILLAUX, B.; LEJEUNE, J. (1972): Effet du milieu ionique sur la denaturation
thermique menageee des chromosomes humains. Ann. Genet. 15, 203-305.

Dar Pıaz, G.B. (1924): Studie delle Arvicole trentine. Studi Trentini di Scienze Naturali 4, 1-17.

Farıo, V. (1905): Campagnols et musaraignes suisses. Quelques formes peu connues. Importance
variable de certains caracteres. Arch. Sci. Phys. Nat. Geneve 4, 182-206.

FAvarD, A. (1984): Atlas des mammiferes sauvages de France. Paris: Soc. Frangaise pour l’Etude et la
Protection des Mammiferes.

GRAF, J.D.; Meyran, A. (1980): Polymorphisme chromosomique et biochimique chez Pitymys
multiplex (Mammalia, Rodentia). Z. Säugetierkunde 45, 133-148.

HARNDEN, D.G.; KLinGer, H.P.; (eds.) (1985): An international System for Human Cytogenetic
Nomeneclature. Basel: S. Karger.

HEIM DE Bausac, H.; BEAUFORT, F. DE (1966): Regime alımentaire de l’effraie dans le Bas-Dauphine.
Application ä l’etude des vertebres. Alauda 34, 309-324.

KrysTUFER, B. (1983): New subspecies of Pitymys liechtensteini Wettstein, 1927 from Yugoslavia
Bioloski Vestnik. 31, 73-82.

MEyLAN, A. (1970): Caryotypes et distribution de quelques Pitymys europ&ens (Mammalia, Rodentia)
(Note preliminaire). Rev. suisse Zool. 77, 562-575.

— (1972): Caryotypes de quelques hybrides interspecifiques de Pıtymys (Mammalıa, Rodentia).
Experientia 28, 1507-1510.

MILLER, G.S. (1912): Catalogue of the Mammals of the western Europe (Europe Exclusive of Russia).
London. British Mus. Nat. Hist.

Morraz, C. (1909): Quelques interessantes captures de petits mammiferes. Bull. Soc. Zool. Geneve 1,
178-180.

NIETHAMMER, J.; Krapp, F. (eds.) (1982): Handbuch der Säugetiere Europas. Bd. 2/I, Nagetiere II.
Wiesbaden: Akad. Verlagsges.

PETRov, B.; Zıvkovic, $. (1971): Zur Kenntnis der Pitymys liechtensteini Wettstein, 1927 (Rodentia,
Mammalia) in Jugoslawien. Arh. biol. nauk. Beograd 23, 31-32.

Comparative morphometry and cytogenetics of Microtus multiplex 125

STORCH, G.; Wınkıng, H. (1977): Zur Systematik der Pitymys multiplex-Pitymys liechtensteini-
Gruppe (Mammalıa, Rodentia). Z. Säugetierkunde 42, 78-88.

SUMMER, A.T. (1972): A sımple technique for demonstrating centromeric heterochromatin. Exper.
Cell. Res. 75, 304-306.

VIEGAS-PEQUIGNOT, E.; DUTRILLAUX, B. (1978): Une methode simple pour obtenir des prophases et
prometaphases. Ann. Genet. 21, 122-125.

WETTSTEIN-WESTERSHEIM, ©. (1927): Beiträge zur Säugetierkunde Europas. II Arch. Naturgesch. 92,
64-146.

Wınking, H. (1976): Karyologie und Biologie der beiden iberischen Wühlmausarten Pitymys mariae
und Pıtymys duodecimcostatus. Z. zool. Syst. Evolut.-forsch. 14, 104-129.

ZAGORODNYUK, 1.V. (1990): Karyotypic varıability and systematics of the Arvicolini (Rodentia).
Species and chromosomal numbers. Vestn. Zool. 2, 26-37.

Zıma, J.; Krar, B. (1984): Karyotypes of European mammals. II Acta Sci. Nat. Brno. 18, 1-62.

Authors’ addresses: PATRICK BRUNET-LECOMTE, Centre de paleontologie analytique et geologie
sedimentaire, URA CNRS 157, Laboratoire de pr£histoire et pal&oecologie du
Quaternaire de ’EPHE, Centre des sciences de la terre, 6 Bd Gabriel, F-21000
Dijon, France and VıraLy VOLOBOUEV, Institut Curie, Structure et mutagenese
chromosomiques, URA CNRS 620, 26 rue d’Ulm, F-75321 Paris Cedex 05,
France

Z. Säugetierkunde 59 (1994) 126-128
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

WISSENS CHATZBEILEEIEIKEU RTZNDIERSTIAUIN TE

Use of dung piles by neighbouring vicunas
By BısıanA L. VırA

Wildlife Conservation Research Unit, Department of Zoology, University of Oxford, Oxford,
United Kingdom

Receipt of Ms. 9. 2. 1993
Acceptance of Ms. 1. 10. 1993

Dung-piling behaviour is typical among the South American camelids, and is notable in the
wild vicuna (Vicugna vicugna) where both males and females, adults and young urinate
and defecate in dung piles. Vicuna socıal organization is based upon stable family groups
that live in an area defended year round, and bachelor groups which are more variable in
composition and location. Bachelors and “passing vicunas” usually use dung piles in family
territories ıf the family ıs absent (drinking water or walking) (KororD 1957; FRANKLIN
1974, 1980, 1983; BoscH 1984) suggesting that the marks “keep insiders in” instead of
“keeping outsiders out“. The aim of this study ıs to analyse the use of dung piles between
adjacent familıes.

This study was conducted at the Abrapampa Experimental Station of the National
Institute of Agricultural Tecnology (INTA). The station ıs located in dry grassland 3475 m
above sea level in the Puna region of Jujuy Province, NW Argentina. For details of the
study area see VırA (1992). The vicuna stock of the station consisted of 600 anımals living
in a 400 ha area limited by sheep fence, containing natural pasture with a narrow river
flowing through the area. No management techniques were used on these animals. The
field work was carrıed out during March 1989. Observations were made from an
observatory hut (6.5 m high) with binoculars.

The 16 dung piles that were located close to the observation hut were numbered. They
were under view sımultaneously (Fig.). The vicunas belonging to the three families that use
this area intensively were recognized individually. These families were: “NCla” with
composition 1:3:1 (male:female:offspring), “Gde” 1:7:5 and “Mar“ 1:4:2. When an
anımal (member of these families, member of a bachelor group, or solo) used a dung pile, a
record was made of his/her identity and the individual number of the pıle used. “Use” was
considered when the anımal defecated and/or urinated in the pile and not when they only
smelled the pile.

In this study, vicuhas’ behaviour in relation to the dung pile prior to elimination
(smelling, kneading, turning and positioning) was the same as that described for the species
ın Perü (KororD 1957; FRANKLIN 1980).

Use of dung piles: 140 defecation-urination events were observed. For each dung pile,
the percentage of use by anımals belonging to each family or no family groups and/or solo
animals was calculated (Table). As is clear from the table, some dung piles were used
exclusively by some families, while others were shared between families. The Figure shows
the distribution of the dung piles and the location of families which used them. In almost
all cases animals not belonging to these families made use of the piles. Among members of
the family groups, males used the dung piles more than three times as often as females
(t test 3.16 p<0.05).

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5902-0126 $ 02.50/0

Use of dung piles by neighbouring vicunas 127

ANGED
3GB

Distribution of the dung piles under observation and the families which shared them. The observation
hut ıs marked as OÖ. H.

Vicunas territorial behaviour has been ‚described as havıng very rıgid boundaries, and
the mating system has been described as “resource defence polygyny”. (FRANKLIN 1974,
1983). Some evidence of males defending females and retaining them in their territories has
been found (BoscH 1984; BoscH and SvENDSEN 1987; VırA 1990) (suggesting in
Abrapampa a mixed mating pattern of territory and females defence, VırA 1992) and also
social organization has been found to differ according to wet/dry conditions, pasture and
season (MENARD 1982; FRANKLIN 1983; VırA and Roıc 1992).

This study showed that although families used an area almost exclusively there was
some tolerance between neighbours, and adjacent famiılies used border areas at different

Percentage use of dung piles by animals belonging to different families and non-familiar ones

Families

GDE No-families

125556
25 %
25 %
20 %

16.6 %

la 76

28.6 %
Ya Mo

6.3 %
100 %
I 70

25070
33.3, 90
I Ko
30.7 %

128 Bibiana L. Vıla

times. This territory overlap allowed one family to use the dung piles of another family.
KororDp (1957) showed that vicunas used the nearest dung pile to defecate, and this might
explain the pattern found here. In a study which analysed the relationship between scent
marking and resource holding of some antelopes, GosLinG (1990) discussed the possible
function of scent-marking as status advertisement, reducing the cost of agonistic encoun-
ters. This hypothesis requires that males have to be “sufficientely sedentary” (GosLInG
1990), which is the case in the vicunas year-round defended areas. The scentmark as part of
a compound (dung pile-male) causes bachelors to move out of the zone. The frequent use
of the piles on the boundaries may mark the “possibilities” of expanding the territory
(GosLinG 1987), that is the tolerance male vicunas have with their neighbours. This
hypothesis can also explaın why in the absence of the territorial male and the family, other
animals use the piles; again the scentmark and the scent-marker form a compound and
without both components the message is not the same. Thus, the data presented here are
consistent with both “using the nearest” and “compound dung-pile/male” explanations.

Although preliminary, this work shows how a knowledge of the individual animals can
help to discover differences ın relation to previous information.

Acknowledgements

Many thanks to Ing. JORGE BERTONI, Director of the INTA-Abrapampa. This study was written
while visiting the Wildlife Conservation Research Unit, University of Oxford, as a post-doctoral
academic visitor supported by an external grant of the Argentine Research Council (CONICET) and
British Council/Antorchas Fellowship. I thank Dr. Davıp MAcDoNALD, head of the “Wildcru”. Dr.
RosıE WOODROFFE corrected the English. I am very grateful to Dr. MARcELO Cassınt for his
criticism and help with this work.

References

Bosch, P. C. (1984): Parental investment by a territorial ungulate, the vicuna (Vicugna vicugna
Molina, 1782). Ms thesis, College of Arts and Sciences of Ohio University, USA.

BoscH, P. C.; SvENDSEN, G. E. (1987): Behaviour of male and female vicuna (Vicugna vicugna
Molina, 1782) as it relates to reproductive effort. J. Mammalogy 68, 425-429.

FRANKLIN, W. L. (1974): The social behaviour of the vicuna. In: The behaviour of ungulates and its
relation to management. Ed. by V. Geist and F. WALTHER. JUCN: Morges, Switzerland. Pp.
477A37.

— (1980): Territorial marking behaviour by the South American vicuna. In: Chemical Signals.
Vertebrate and Aquatic Invertebrates. Ed. by D. MULLER-SCHWARZE and R. M. SILVERSTEIN.
New York: Plenum Press. Pp. 53-66.

— (1983): Contrasting socioecologies of South America’s wild camelids: The vicuna and the guanaco.
In: Advances in the study of mammalıan behaviour. Ed. by S. F. EISENBERG and D. G. KLEINMAN.
Special Publ. No 7. Am. Soc. Mammalogists. Pp. 573-629.

GosLing, L. M. (1987): Scent marking in an antelope lek territory. Anım. Behav. 28, 172-184.

— (1990): Scent-marking by resource holders: alternative mechanisms for advertising the costs of
competition. In: Chemical signals in vertebrates 5. Ed. by D. W. Macponaıp, D. MULLER-
SCHWARZE, and $. E. NATynczuk. Oxford: Univ. Press. Pp. 315-328.

Kororp, C. B. (1957): The vıcuna and the Puna. Ecol. Monog. 27, 153-219.

MENARD, N. (1982): Quelques aspects de la socioecologie de la vicogne Lama vıcugna. Terre vie 36,
1533.

VırA, B. L. (1990): El comportamiento de la vicuna durante la temporada reproductiva. PhD. Thesis,
Sciences Fac., Univ. Buenos Aires, Argentina.

— (1992): Vicunas (Vicugna vicugna) agonistic behaviour during the reproductive season. In:
Ungulates 91. Proc. Intern. Symposium. Ed. by F. Spıtz, G. JANEAU, G. GONZALEZ and S.
AULAGNIER. Toulousse: Institut de Recherche sur les Grands Mammiteres. Pp. 475482.

VırA, B. L.; Roıc, V. G. (1992): Diurnal movements, family groups and alertness of vicuna (Vicugna
vicugna) during the late dry season in the Laguna Blanca Reserve (Catamarca, Argentina). Small
Rumin. Res. 7, 289-297.

Author’s address: Dr. Bısıana L. VırA, GEMA-Sur, Estanislao del Campo 1260, (1602) Florida,
Pcia. de Buenos Aires, Argentina

Walter de Gruyter
Berlin +» New York

HANDBUCH DER ZOOLOGIE / HANDBOOK OF ZOOLOGY
Eine Naturgeschichte der Stämme des Tierreiches /

A Natural History of the Phyla of the Animal Kingdom
Band / Volume VIII Mammalia

Herausgeber / Editors: J. Niethammer, H. Schliemann, D. Starck
TEILBAND / PART 60

Karl F. Koopman: Chiroptera: Systematics

1994. 29,7 x 21 cm. VI, 217 pages. With 19 figures.
Hardcover DM 435,-/ 6S 3.393,-/ sFr 412,- ISBN 3-11-014081-0

The systematics of Chiroptera and its subdivisions from the ordinal to the specific
level. Characters, distribution, and subordinate taxa (to subspecies) are given.
Photographs of skulls (in ventral view) together with tables of forearm lengths and
dental formulae for genera are included. The first attempt to provide a systematic
description of all bats since 1878.

Price is subject to change

Erscheinungsweise und Bezugspreis 1994: 6 Hefte bilden einen Band. Jahresabonnement Inland:
DM 378,- zuzüglich DM 13,80 Versandkosten; Jahresabonnement Österreich: öS 2949, zuzüg-
lich 6S 164,- Versandkosten; Jahresabonnement Schweiz: sfr 364,— zuzüglich sfr 21,- Versandko-
sten; Jahresabonnement EG-Binnenmarkt-Länder mit USt-ID-Nr.: DM 353,27 zuzüglich
DM 19,63 Versandkosten; Jahresabonnement EG-Binnenmarkt-Länder ohne USt-ID-Nr. und
Drittländer: DM 378,- zuzüglich DM 21,- Versandkosten. Das Abonnement wird zum Jahresan-
fang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens
am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhand-
lung oder bei der Verlagsbuchhandlung Paul Parey GmbH & Co. KG, Spitalerstraße 12,
D-20095 Hamburg, Bundesrepublik Deutschland, bestellt werden. Die Mitglieder der „Deut-
schen Gesellschaft für Säugetierkunde“ erhalten die Zeitschrift unberechnet im Rahmen des

Mitgliedsbeitrages.
Z. Säugetierkunde 59 (1994) 2, 65-128

Lemuren ım /00

Aktuelle Forschungsergebnisse, Artenschutz, Perspektiven

Herausgegeben von Vaclav Ceska, Saarbrücken, Hans-Ulrich Hoffmann, Dudweiler, und Dr.
Karl-Heinz Winkelsträter, Brebach-Fechingen. Mit Beiträgen von Verena Dillenburger, Wolf-
gang Dreier, Lutz Dues, Michael Flügger, Hans-Ulrich Hoffmann, Henrike Hultsch, Wolfgang !
Köhler, Jean-Marc Lernould, Volker Löschcke, Wolfgang Maier, Bernhard Meier, Jean-Jacques |
Petter, Jörg Pfeiffer, Gerhard Pflughöft, Rosemarie Plätke, Gabriele Ros, Yves Rumpler, Maria-
Elisabete Santini, Christel Schulz, Hans R. Siegismund, Ian Tattersall, Claude Thouvenin, '
Dietmar Todt, Jürgen Tomiuk, Ansgar Vössing und Karl-Heinz Winkelsträter. h

1992. 336 Seiten mit 216 Einzeldarstellungen in 101 Abbildungen, 24 Farbabbildungen i
auf 4 Tafeln und 10 Tabellen. Kartoniert DM 68,- |

In gleichem Maße, in dem die natürlichen Lebensräume schwinden, wächst den Zoologischen
Gärten die Aufgabe zu, verstärkt zum Schutz bedrohter Tierarten beizutragen. Die Halbaffen
Madagaskars, die Lemuren mit vielen unmittelbar vom Aussterben bedrohten Arten, liefern i
hierfür ein extremes Beispiel.

Die prekäre Situation in Madagaskar, die Rolle der Zoos im Artenschutz sowie die tiergärtnerisch
relevanten Aspekte der Lemuren werden in diesem Buch unter der Herausgeberschaftvon Vaclav |
Ceska, Hans-Ulrich Hoffmann und Karl-Heinz Winkelsträter ausführlich dargestellt, wobei N
aktuelle Forschungsergebnisse ebenso berücksichtigt werden wie genetisches Managementund ;
künftige Perspektiven. Zahlreiche Autoren haben zu diesem Buch beigetragen, unter ihnen |
prominente Wissenschaftler, Primaten- und Lemurenforscher wie Jean-Jacques Petter, Yves ,
Rumpler, Ian Tattersall und Bernhard Meier. Sie haben für das mit 4 Farbtafeln und zahlreichen :
Abbildungen ausgestattete Werk auch schwer zugängliche Quellen und weit verstreute Litera- ‘
tur, oftin internationaler Zusammenarbeit, ausgewertet. Allein die Bibliographie bietetmit mehr \
als 800 Zitaten die umfangreichste Sammlung, diebisher über die Lemuren publiziert worden ist. '

The keeping of lemurs in zoological gardens is the subject of this book, emphasizing the
engagement for the protection of endangered animal species. In particular, the precarious
situation of lemurs in Madagascar is considered and explained by recent resultsinresearchand
genetical management. Further, there are paedagogical, ethological and veterinary aspects also
regarded as well as future prospects. Devoted to the practical work in zoological gardens, thebook '
applies to a great audience of zoologists, biologists, geneticists, veterinarians and to the students
of these disciplines.

Preis: Stand 1. März 1994 ''

Blackwell Wissenschafts-Verlag - Berlin

Vol. 59 (3), 129-192, Juli 1994 ISSN 0044-3468 C 21274
7

== ITSCHRIFT FÜR
SAUGETIERKUNDE

INTERNATIONAL JOURNAL
FMAMMALIAN BIOLOGY

Organ der Deutschen Gesellschaft für Säugetierkunde

|

"Wenhold, Brigitte A.; Rasa, ©. Anne E.: Territorial marking in the Yellow mongoose Cynictis penicillata: sexual
advertisement for subordinates? — Territoriumsmarkierung bei der Fuchsmanguste Cynictis penicillata: sexuelle
„Reklame“ für untergeordnete Tiere? 129

|Feer, F.: Observations &thologiques sur Bubalus (Anoa) quarlesi Ouwens, 1910 (Ruminantia, Bovidae) en captivite.
- Ethological observations on Bubalus (Anoa) quarlesi Ouwens, 1910 (Ruminantia, Bovidae) in captivity. —

Ethologische Beobachtungen an Bubalus (Anoa) quarlesi Ouwens, 1910 (Ruminantia, Bovidae) im Zoo 139
Petit, P.; De Bois, H.; De Meurichy, W.: Chromosomal reduction in an Okapi pedigree (Okapia johnstoni). -
Chromosomenreduktion in einer Okapi-Familie 153

Dannelid, E.: Chromosome polymorphism in Sorex alpinus (Mammalia, Soricidae) in the western Alps and the
_ Swiss Jura. - Cromosomenpolymorphismus von Sorex alpinus (Mammalia, Soricidae) in den Westalpen und im

Schweizer Jura 161
Lindström, E. R.: Placental scar counts in the Red fox (Vulpes vulpes L.) revisited. - Zählungen von Implantations-
narben beim Rotfuchs (Vulpes vulpes L.)., eine Revision 169

Giacometti, M.; Ratti, P.: Zur Reproduktionsleistung des Alpensteinbockes (Cypra i. ibex L.) in der Freilandkolonie
Albris (Graubünden, Schweiz). — On the reproductive performance of the free-ranging alpine ibex population
(Capra i. ibex L.) at Albris (Grisons, Switzerland) 174
|Arlettaz, R.; Beck, A.; Güttinger, R.; Lutz, Miriam; Ruedi, M.; Zingg, P.: Oü se situe la limite nord de repartition
geographique de Myotis biythii (Chiroptera: Vespertilionidae) en Europe centrale? — Where is the northern border
ofthe distribution range of Myotis blythii (Chiroptera: Vespertilionidae) in Middie Europe? — Wo liegt die nördliche
| Verbreitungsgrenze von Myotis blythii (Chiroptera: Vespertilionidae) in Mitteleuropa? 181

Wissenschaftliche Kurzmitteilung

Lode, Th.: Feeding habits of the Stone marten and environmental factors in western France. — Ernährungsgewohn-
heiten des Steinmarders und Umweltfaktoren in Westfrankreich 189

mer mausser.r
——n gen.

Mitteilungen der Gesellschaft THSONIA N 132
v

REIN NV I ANY
» “
km int \
E \
N \ L u‘ ;
Ve | hi | Bar >
ar } { 7 „ss =

|
Voraa Paul Parey Hamburg
|
|

|

HERAUSGEBER/EDITORS

P. J. H. van BrEE, Amsterdam — W. FIEDLER, Wien -— H. Frick, München - G. B.

HaARTL, Wien -— W. HERRE, Kiel - R. HUTTERER, Bonn - H.-G. Krös, Berlin - H.-].

Kunn, Göttingen — E. KuLzer, Tübingen -—- W. MAIER, Tübingen — J. NIETHAMMER,

Bonn - ©. Anne E. Rasa, Bonn — H. ReEıcHstein, Kiel -— M. Rönrs, Hannover —

H. SCHLIEMANN, Hamburg — D. STARcK, Frankfurt a. M. - E. Thuenıus, Wien - P. Vo-
GEL, Lausanne - H. Wınkıng, Lübeck

SCHRIFTLEITUNG/EDIEORITATOERICE

D. Kruska, Kiel - P. LAnGeEr, Gießen

This journal ıs covered by Biosciences Information Service of Biological Abstracts, and by Current Con-
tents (Series Agriculture, Biology, and Environmental Sciences) of Institute for Scientific Information

Die Zeitschrift für Säugetierkunde veröffentlicht Originalarbeiten und wissenschaftliche Kurzmittei-
lungen aus dem Gesamtgebiet der Säugetierkunde, Besprechungen der wichtigsten internationalen
Literatur sowie die Bekanntmachungen der Deutschen Gesellschaft für Säugetierkunde. Verantwort-
licher Schriftleiter im Sinne des Hamburgischen Pressegesetzes ist Prof. Dr. Dieter Kruska.

Zusätzlich erscheint einmal im Jahr ein Heft mit den Abstracts der Vorträge, die auf der jeweiligen
Hauptversammlung der Deutschen Gesellschaft für Säugetierkunde gehalten werden. Sie werden als
Supplement dem betreffenden Jahrgang der Zeitschrift zugeordnet. Verantwortlich für ihren Inhalt
sind ausschließlich die Autoren der Abstracts.

Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z. Hd. Prof. Dr. Dieter
Kruska, Institut für Haustierkunde, Biologiezentrum der Christian-Albrechts-Universität, Am
Botanischen Garten 9, D-24118 Kiel, Bundesrepublik Deutschland. Für die Publikation vorgese-
hene Manuskripte sollen gemäß den ‚„Redaktionellen Richtlinien“ abgefaßt werden. In ihnen
finden sich weitere Hinweise zur Annahme von Manuskripten, Bedingungen für die Veröffent-
lichung und die Drucklegung, ferner Richtlinien für die Abfassung eines Abstracts und eine
Korrekturzeichentabelle. Die Richtlinien sind auf Anfrage bei der Schriftleitung und dem Verlag
erhältlich.

Sonderdrucke: Anstelle einer Unkostenvergütung erhalten die Verfasser von Originalbeiträgen und
Wissenschaftlichen Kurzmitteilungen 50 unberechnete Sonderdrucke. Mehrbedarf steht gegen
Berechnung zur Verfügung, jedoch muß die Bestellung spätestens mit der Rücksendung der
Korrekturfahnen erfolgen.

Vorbehalt aller Rechte: Die ın dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich
geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks,
des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der
Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei-
ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch im Einzelfall grund-
sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines
Beitrages ıst auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber-
rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom
24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen
den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind
mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen.

Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of
the first page of an article in this journal indicates the copyright owner’s consent that copies of the article may
be made for personal or internal use, or for the personal or internaluse of specific clients. This consent is given
on the condition, however, that the copier pay the stated percopy fee through the Copyright Clearance
Center, Inc., 21 Congress Street, Salem, MA 01970, USA, for copying beyond that permitted by Sections 107

or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying
for general distribution, for advertising or promotional purposes, for creating new collective, or for resale.
For copying from back volumes of this journal see “Permissions to Photo-Copy: Publisher’s Fee List” of the

Fortsetzung 3. Umschlagseite

© 1994 Paul Parey. Verlag: Paul Parey GmbH & Co. KG, Hamburg. Anschrift: Spitalerstr. 12, D-20095 Hamburg,
Bundesrepublik Deutschland.
Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., Heide/Holst.

Z. Säugetierkunde 59 (1994) 129-138
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Territorial marking in the Yellow mongoose Cynictis penicillata:
sexual advertisement for subordinates?

By BrıGITTE A. WENHOLD and O. Anne E. Rasa

Department of Zoology, University of Pretoria, Pretoria, RSA and Zoologisches Institut,
Universität Bonn, Bonn, FRG

Receipt of Ms. 24. 9. 1993
Acceptance of Ms. 25. 2. 1994

Abstract

In the yellow mongoose, a colonial territorial viverrid, territory marking is performed by all group
members. Subordinate adults, both male and female, have higher marking rates than the dominants
and juveniles of both sexes and their marks carrying identity cues are deposited especially in border
areas and outside the territory. They are also the anımals most involved in territorial defence. Within
the territory, marking was concentrated around the burrow systems and dominants either did not
mark (dominant male) or did not visit (dominant female) the territory borders. Subordinate males,
which have low reproductive success in their natal colonies, disperse usually to neighbouring colonies
and subordinate females cross territory borders when in oestrus and are mated by males from
neighbouring groups. The classes of anımal most likely to seek mating opportunities outside the
colony thus mark most often where their marks will be encountered by strangers. Apart from its role
in territory familiarisation and territory owner assessment, marking in this species ıs hypothesised as
being a means of sexual advertisement for subordinate anımals.

Introduction

Marking behaviour has long been regarded as an important component of anımal social
communication (THIESSEN and Rıce 1976), although its function ıs still obscure. A number
of hypotheses to explain why anımals mark have been put forward ranging from deterrence
of intruders (HEDIGER 1949) to establishment of a familiar smell within the territory
(JoHnson 1973), the latter attempting to explain why anımals often mark as frequently
inside their territories as on the borders. A more recent hypothesis proposes that scent
advertises the identity of a territory inhabitant to an intruder, enablıng the latter to assess
the holder’s status (GosLıng 1982), thus avoiding costly fights. Marking is also considered
to be associated with dominance in many species (Ras 1971; STODDART 1976; BROWN
and McDonALD 1985; KAPPELER 1980; SOMERS et al. 1990). Marking behaviour and its
relation to dominance is investigated here in a group-living viverrid, the yellow mongoose,
Cynictis penicıllata.

Cynictis ıs amonotypic genus ın the family Viverridae, limited to the southern African
subregion (SMITHERS 1983). Feeding habits and reproduction have received some attention
(MicHAELIs 1972; HERZIG-STRASCHIL 1977; LyncH 1980), together with the species’
implied role as a rabies vector (ZumpT 1969, 1976) but little is known of its behaviour
(EARLE 1977, 1981). Cynictis exhibits a degree of sociality intermediate between the highly
social and solitary mongooses (WENHOLD 1990). Group sızes varying from single or paired
anımals to 10 or more have been reported (RowE-RoweE 1978; ZumPpTr 1976; PRINGLE 1977;
Du Toır 1980; STUART 1981; EARLE 1981; SMITHERS 1983). Anımals inhabit a communal
burrow system but forage alone. Colonies are territorial, male intruders being chased away
while females are tolerated, especially during the mating season (WENHOLD 1990).
Detenders are in most cases male residents. Territories are marked by all group members

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5903-0129 $ 02.50/0

130 Brigitte A. Wenhold and O. Anne E. Rasa

and this study investigates the various marking modes employed from temporal, spatial
and individual aspects.

Material and methods

The study was conducted on Big Island in the Vaal Dam (26° 52” S, 28° 11” E), Transvaal, a temporary
island of approximately 200 ha which is a peninsula for most of the year. At least 11 mongoose
colonies were present on the island. The study colony consisted of 13 individuals (4 adult 84, 3 adult
22,5 juvenile dd, 1 juvenile ?) and was habituated to observer presence during an initial period of
5 months. Adults were anımals over 1 year old, juveniles 1 year old or less. The data on which the
following analyses were based were collected on a daily basis during the 9 month period subsequent to
habituation (April-December 1987) for a total of over 900 h. Mongooses were identified individually
by means of natural markings and scars.

Observations were made using 10 x 50 Bencon binoculars. Since marking activities were of short
duration, absolute frequency of occurrence of varıous marking modes were recorded using all events
sampling when the anımals were in groups and focal anımal sampling during foraging (ALTMANN
1973). Observations were timed to cover the entire day from first emergence from the burrow system
in the morning to return to the burrow at midday or night. Focal animals were followed for as long a
period as possible until they were lost to sight. Focal observations on individuals of less than 15 min
duration were not included in the analysıs since they would tend to bias the results in favour of areas
around the burrow systems. A different focal anımal was selected each day to prevent continuity
effects. Data were entered into a computer using a standard database program and further analyses
were conducted using standard statistical packages. Statistical tests used are mentioned at appropriate
points in the text and the level of probabılity given ıs two-tailed in all cases.

Marking frequency

Since the time periods over which individuals were observed varıed, and different focal animals were
observed on different days, rates of marking were calculated (mean rate/h/day/animal) to allow
comparison between individuals. For statistical analyses, the daily rates of occurrence were considered
independent since observations on the same anımal were separated by at least seven days. Individual
marking rates were calculated separately for the five different modes of marking observed.

Location of marking sites

The 21,300 m? territory was divided into 10 x 10 m grid squares to determine marking frequency in
various areas. The rate of occurrence of marking in each grid square was then calculated ie.mean
marking frequency/h that the anımal was observed in a particular square. Grid squares were then
allocated to four categories: a. inside the territory but not in the vicinity of a burrow, b. within the
territory and containing a burrow, c. on the territory border, d. outside the territory (Figure). The
territory border was considered as the approximately 20 m wide fringing strip which was as far as an
animal would either chase an intruder out of its territory or, conversely, how far it was chased after
intruding in the neighbouring territory. The rates of occurrence of varıous marking behaviour patterns
were determined for the four areas. The marking rates of various classes of anımal in the same area
were also calculated.

Results
Description of marking activities

Cynictis marks objects by four means. Anal marking, using secretions of the peanut-sized
anal glands on either side of the anus, is the most common. The anımal squats with the tail
arched and presses the glands briefly on the substrate. Sniffing the site usually precedes
marking and may follow it. In general, only a single mark is deposited on a low object,
usually a stone or branch. A handstand posture typical of several other viverrids was not
observed and anal dragging was rare.

Objects are also marked with a glandular area on the cheek using a wiping motion,
sometimes repeated with the same or the alternate cheek. The object ıs stroked ın a
continuous movement from snout to eye. Typical objects are branches and patches of bare
earth, small objects being steadied with the forepaws. The action is often preceded by

Territorial marking in the Yellow mongoose Cynictis penicıllata IS!

The territory of the study colony divided into 10 x 10 m? grid squares. The shoreline of the island is
on the right. Shaded squares contain burrows, open squares are territory areas without burrows,
hatched squares indicate the border of the territory and dotted squares show the areas outside the
territory that were visited by group members

sniffing, and frequently followed by wiping the entire side of the body on the ground,
termed side-wiping. The functional significance of the latter is not clear, since there does
not appear to be any epidermal glandular tissue in this region of the body. However, side-
wiping sites are often covered with dried mongoose faeces and it is possible that this
behaviour is a form of “self-anointing”, faecal odour being transferred by this means from
one site to another. Frequently used side-wiping sites, which are usually situated next to a
bush or a clump of grass, are recognisable as depressions in the ground devoid of any
debris except the faeces deposited there at intervals.

Defaecation and urination can be considered as secondary ways of marking a territory.
Specific defaecation sites (middens) are present, at least one being situated near every set of
burrows. On emerging from the burrows in the morning, one of the first actions ıs to
approach the midden, sniff at the faeces present, and then defaecate. This is performed
from a squat with the tail arched as ın anal marking, but without the anus touching the
ground. Anımals also defaecate at communal sites scattered throughout the territory and
usually associated with secondary burrows. These are burrows used, usually by single
animals, to sleep in during the heat of the day or occasionally overnight. The faeces,
especially, have a strong and distinct odour, possibly from secretions of the anal gland
covering them.

The same squatting posture is used by both sexes during urination, which also takes
places primarily at middens, although dominant females may also urinate wıth one hind leg
lifted. Urination also occurs at sites throughout the territory during foraging excursions.

Marking frequency

The daily rates/h of various marking modes were calculated for each anımal and their
means determined. The mean rates for different sex and age classes are shown in table 1.

132 Brigitte A. Wenhold and O. Anne E. Rasa
Table 1. Mean marking rates per hour for different sex and age classes

Mean marking rate/hour

ur def

Dominant male ; ’ \ i 0.0 0.39
Subordinate : \ Ä £ 0.06 0.47
adult males
Juvenile males 5 $ 0.14 0.61
All adult males Ä a N ä 0.04 0.45
All males e 5 e Ä 0.10 0.54
Dominant female N N h i 0.0 0.68
Subordinate : : s i 0.11 0.38
adult females
Juvenile females 0.0 0.0 0.56
All adult females i ö 0.13 ; 0.07 0.50
All females s i 0,112 h 0.06 0.50
All Adults 5 i 0.30 \ 0.05 0.47
All Juveniles 0.38 0.29 .00 0.13 0.61

am = anal marking, sw = side wiping, cw = cheek wiping, sm = scent marking (sum of am + sw +
cw), ur = urination, def = defaecation.

Anal marking (am), cheek wiping (cw) and side-wiping (sw) rates were added to give a
mean scent marking (sm) rate. The data following were tested for significant differences
using the Mann-Whitney U-test.

Subordinate adult males have a significantly higher anal marking rate than the dominant
male (p< 0.05). Males in general anal mark significantly more than females, whether only
adults (p < 0.05) or all age groups of both sexes are compared (p < 0.01). Subordinate adult
males also side-wiped significantly more than the dominant (p<0.05) and adult males
more than adult females (p< 0.05). All males side-wiped significantly more often than all
females (p<0.02). For cheek wiping, the only significant difference found was for
subordinate males, which have a higher rate than the dominant (p<0.05). Although not
significant, all males tend to cheek wıpe more than all females (p < 0.07).

Subordinate males also had a higher rate of scent marking ın general than the dominant
male (p<0.0005) and subordinate females marked more than the dominant female
(p<0.05). Adult males had a higher marking rate than adult females (p<0.02) and
juvenile males a higher one than the juvenile female (p< 0.02). All males together thus
scent mark more than all females (p < 0.002).

No significant differences were found between any of the age and sex classes for
urination and defaecation (p> 0.05) but subordinate adult males had significantly higher
defaecation rates than the dominant male (p< 0.05).

Location of marking sites

The rate of occurrence per hour of various types of marking was determined for each
individual in each grid square in which it was observed. The mean data for age and sex
classes in each of the four areas are shown in tables 2-5.

In grid squares containing a burrow system (Tab. 2), the only significant difference
found was between adult and juvenile males, adult males having a significantly higher rate
of scent marking in general than juveniles (p<0.001; p>0.1 ın all other cases).

No significant differences were found in marking rates between any age or sex category
in grid squares within the territory which did not contain a burrow system (Tab. 3),
(p>0.1 ın all cases).

Adult males anal marked and side-wiped in the border area significantly more than

Territorial marking in the Yellow mongoose Cynictis penicillata 133

juvenile males (p<0.05, Tab. 4). Their rate of all scent marking activities combined was
also significantly higher than that of juveniles (p< 0.01). The dominant male was never
observed marking on the border although he visited it. The dominant and the juvenile
females were also never seen on the border of the territory throughout the study period.
The data for females ıs therefore based only on observations from subordinate adults.
Subordinate adult females anal marked the border area significantly more than juvenile

Table 2. Mean marking rates in territory grid squares containing a burrow system

Mean marking rate/hour

def

Dominant male , h A S 0.0

Subordinate s } S i 0.06
adult males

Juvenile males ß | 0.07

Dominant female } 3 ; h 0.09 0.03

Subordinate } N : i 0.01 0.10
adult females

Juvenile females 1 ö 0.0 0.0

am = anal marking, sw = side wiping, cw = cheek wiping, sm = scent marking (sum of am + sw +
cw), ur = urination, def = defaecatıon.

Table 3. Mean marking rates in territory grid squares without a burrow system

Mean marking rate/hour
sm

Dominant male } i { 2.50

Subordinate ! i N 1417.
adult males

Juvenile males Ä 1.74

Dominant female { i ä 0.0

Subordinate : } 3 0.23
adult females

Juvenile females 1 0.0

am = anal marking, sw = side wiping, cw = cheek wiping, sm = scent marking (sum of am + sw +
cw), ur = urination, def = defaecation.

Table 4. Mean marking rates for grid squares on the territory border

Mean marking rate/hour

Dominant male
Subordinate

adult males
Juvenile males
Dominant female
Subordinate

adult females
Juvenile females 1

am = anal marking, sw = side wiping, cw = cheek wiping, sm = scent marking (sum of am + sw +
cw), ur = urination, def = defaecation.

134 Brigitte A. Wenhold and O. Anne E. Rasa

(p<0.05) but not subordinate adult males and were never observed side-wiping or
urinating there.

Subordinate adult males had a significantly higher rate of scent marking in outside the
territory than did juvenile males (p< 0.001, Tab. 5). Amongst females, only the juvenile
had a significantly higher rate of defaecation outside the territory than the subordinate
adults (p<0.001). Females were never observed side-wiping outside the territory.

Table 5. Mean marking rates for grid squares outside the territory

Mean marking rate/hour

Dominant male
Subordinate

adult males
Juvenile males
Dominant female
Subordinate

adult females
Juvenile females 1

am = anal marking, sw = side wiping, cw = cheek wiping, sm = scent marking (sum of am + sw +
cw), ur = urination, def = defaecatıon.

Comparison between areas

To determine whether different classes of anımals showed significant differences in
marking modes ın various areas of the territory, rates were compared using the Kruskal-
Wallıs test (SIEGEL 1956). The level of significance ın all cases ıs 0.05. Since this test
compares the mean of the ranks given to the data points in the samples, the results in some
cases do not appear to reflect the differences between the mean rates shown in the tables.

For grid squares containing a burrow system vs those ın the remainder of the territory,
the following significant differences were found: adults anal marked and cheek wiped
significantly more often in the vicinity of the burrows than elsewhere in the territory,
juveniles showing the opposite tendency. There were no differences between side-wiping
rates, except for juvenile males which performed this behaviour significantly less often near
burrows. For all classes of anımals, defaecation and urination take place more often near
burrows than in the remainder of the territory.

Anal marking, side-wiping and cheek wiping rates were all higher on the border than ın
areas of the territory away from the burrow systems for all classes of anımal. Males
defaecated more on the border than did females. Only juvenile males urinated more on the
border than inside the territory away from the burrows. All other differences were non-
significant.

A comparison of border areas with areas around burrow systems showed a significantly
higher rate of anal marking at the border for subordinate males and females while the
opposite was true of juvenile males and the two dominant anımals. Subordinate adult males
also had a higher rate of side-wiping at border vs burrow areas than females. Juvenile males
showed no significant difference. The reverse was true for cheek wiping. Here, juvenile
males cheek wiped more at the border than near burrows, other classes of anımal showing
no difference. No differences were found for defaecation in the two areas. Subordinate
adult males, however, had higher urination rates near burrows.

Comparison of marking rates between grid squares on the border and those outside the
territory in which group members were observed showed that subordinate adult males and
females had a higher rate of anal marking on the border than outside it. Other classes of

Territorial marking in the Yellow mongoose Cynictis penicıllata 135

animals, with the exception of the dominant female who did not visit the border, anal
marked equally frequently in both areas. No female side-wiped outside the territory,
otherwise all classes had significantly higher side-wiping and cheek wiping rates on the
border than outside. Defaecation was significantly higher outside the territory for subordi-
nate adult males while subordinate adult females and juvenile males showed no significant
difference. Juvenile males, however, urinated more outside the territory than on the

border.

Marking and territorial defence

The frequency with which individuals marked/10 h and were observed attacking and
chasing intruders in the territory/10 h shown in table 6. There was a highly significant
relationship between marking and territorial defence (Spearman’s rho correlation, r, =
0.7701, N = 12, p = 0.03). Of all 22 only subordinate 22 was observed to attack
strangers, the majority of the defence activities being performed by the adult and older
juvenile males.

Table 6. Mean rates/10 hours of attacking intruders and marking for each group member

Adult d& Adult 2?
1 2 2

Attacks

Da 2.4

Marking
VE 25:3

Discussion

Probably the most interesting aspect of this study ıs that all modes of territorial markıng
are performed more frequently by subordinate adults than by dominant ones, in contrast
to EARLE’s (1977) statement. This ıs not typical for mammals (see Raııs 1971; EISENBERG
and Kreıman 1972 and Brown and MacDonarn 1985 for reviews) where dominant
males, especially, are the ones most involved ın territorial advertisement and defence.
However, in the yellow mongoose, ıt is the subordinate adult males which are more
involved ın territorial defence than the dominant male and have their highest marking rates
at the territory border and in fringing areas.

The finding that males mark more ın general than females has its parallels ın oh.
mammalıan species (JOHNSON 1973; THIEssEn and RıcE 1976). Although a tendency for
adults to mark more than juveniles has been reported in the literature (THIESSEn and RıcE
1976), this was not substantiated for the yellow mongoose. Adult males marked more than
juvenile males, but this was not true for adult females. There was a strong correlation
between marking and active territorial defence in this species, females playing little part in
the latter. The study period coincided with puberty onset in the juveniles and for a related
species, the dwarf mongoose Helogale undulata, juveniles were found to be amongst the
most active group members in territorial defence (Rasa 1977) and also to have high
marking rates (Rasa 1973). Thus the finding that juvenile males mark as often as adults
may reflect their active role in territory protection.

The study has also shown that the areas most frequently marked by subordinate males
and females are sites at the territory border and in fringing areas while the dominant female
and juvenile females never visit the border, confining their marking to the territory interior
around the burrows. The dominant male also marks predominantly within the territory.

136 Brigitte A. Wenhold and O. Anne E. Rasa

Adult subordinate males anal mark and sidewipe most frequently at the border and outside
the territory, subordinate adult females anal mark most frequently there. These marking
modes are the ones most lıkely to carry ıdentity cues, as has been found in both dwarf
mongooses (RasA 1973) and small Indian mongooses Herpestes auropunctatus (GORMAN
1976). Juveniles urinate and cheek wipe most frequently in border areas, juvenile males
extending the increased urination rate to sites fringing the territory. Cheek wiping and
urination carried no identity cues in dwarf mongooses (Rasa 1973) but cheek wiping was
indicative of high excitement. Juvenile marking in border areas is thus likely to have other
connotations than that of the subordinate adults.

The data suggests that both Jorunson’s (1973) famıliarıty hypothesis and GosLıng’s
(1982) assessment hypothesis may apply to the yellow mongoose. The high marking rates
observed in territory areas containing a burrow system as opposed to those without one
suggest that odour is preferentially deposited near major resting sites. The concentration of
marks around burrows is more likely to be associated with provision of a familiar odour
for ınhabitants rather than with intruder deterrence, concurring with JOHNSoN’s
hypothesis. GosLing’s hypothesis, however, suggests that marks within a territory should
be evenly distributed to increase an intruder’s likelilhood of encountering them and
enabling it to assess the territory owner. The tendency to mark outside the territory using
modalities carrying identity cues, with the highest scent marking rate occurring on the
border, may establish an odour gradient external to the territory itself, indicating to an
intruder the perimeter within which it can expect attack. Since yellow mongooses from
different groups frequently encounter each other outside their territory borders, assess-
ment may take place before the territory ıs penetrated. These marks have no deterrent
effect, intruders sniffing them and then continuing on into the territory, as found for other
carnıvores (LEYHAUSEN 1965; SCOTT 1967), HEDIGER’s (1949) deterrence hypothesis does
not appear to hold for thıs species.

The surprising finding that it ıs the subordinate adult males and subordinate females
that mark more at the border than within the territory, and that the dominant anımals
rarely visit these areas, suggests that these marks may serve a second advertising purpose
for subordinates. Subordinate males disperse from their natal group when adult, usually to
neighbouring groups (WENHOLD 1990), as do subordinate females. Females also cross
territory borders when in oestrus and are mated by males from a neighbouring group
(WENHOLD 1990). The deposition of anal gland scent ın areas where it ıs most likely to be
encountered by neighbours may serve to advertise the age, sex and reproductive state of the
marker, as well as familiarising members of neighbouring groups with the marks of
potential immigrants and possibly facilıtating the transfer of individuals between colonies.

Animals unlikely to disperse or mate or be mated outside the colony either never visited
the territory border (the dominant and juvenile females), did not mark there (the dominant
male), or had low rates of identity carrying marks in this area (juvenile males). Since the
marking frequency of subordinate females does not appear to correlate with territorial
defence activities as it does in males, another hypothesis to explaın this high female
marking rate on the border is suggested. By depositing the majority of their identity
carrying marks in border areas, subordinate females may utilise odour cues to advertise
their presence to neighbouring males for future mating purposes. For subordinate males,
which are likely to have no reproductive success with females in their natal colonies
(WENHOLD 1990; Rasa et al. 1992), high marking rates on the border may also be a means
of advertising their existence to neighbouring females, with which they have been observed
to mate. Marking in this species may thus not only play a role in assessment for intruders
and familiarıty for territory inhabitants but may also serve a secondary purpose as a major
means of sexual advertisement for subordinates, especially temales, to attract prospective
mates from neighbouring territories.

Territorial marking in the Yellow mongoose Cynictis penicillata 137

Acknowledgements

The authors would like to thank the Rand Afrikaans University for permission to work on their
property and the Foundation for Research and Development for a grant (to A. Rasa) for partial
support of the fieldwork. Special thanks are due A. van ZyL for compling the computer programmes.

Zusammenfassung

Territorinmsmarkierung bei der Fuchsmanguste Cynictis penicıllata: sexuelle „Reklame“ für
untergeordnete Tiere?

Bei der Fuchsmanguste (Cynictis penicıllata), einer in Kolonien lebenden Viverridenart, wird Territo-
rıumsmarkierung von allen Gruppenmitgliedern durchgeführt. Adulte untergeordnete Tiere, sowohl
Männchen als auch Weibchen, markieren häufiger als dominante und juvenile Tiere beiderlei
Geschlechts. Die Duftstoffe, die Identitätsmerkmale beinhalten, werden besonders an Territoriums-
grenzen und außerhalb des Territoriums abgelegt. Adulte subordinate Gruppenmitglieder sind zudem
ın der Verteidigung des Territoriums am aktivsten. Innerhalb des Territoriums findet Markierung
hauptsächlich um die Wohngänge herum statt, und dominante Tiere markieren entweder nicht an den
Territorıiumsgrenzen (dominantes Männchen) oder besuchen das Grenzgebiet nicht (dominantes
Weibchen). Untergeordnete Männchen, die einen geringeren Fortpflanzungserfolg innerhalb der
Gruppe aufweisen, wandern gewöhnlich in benachbarte Territorien ab. Untergeordnete Weibchen
überqueren die Territoriumsgrenze, wenn sie ım Oestrus sind, und werden auch von Männchen der
benachbarten Gruppen begattet. Folglich markieren gerade die Tiere, die am ehesten Fortpflanzungs-
möglichkeiten außerhalb der Geburtskolonie suchen, häufiger in einem Gebiet, wo der Duft von
Koloniefremden angetroffen werden kann. Es wird die Hypothese aufgestellt, daß Markieren bei
dieser Art außer als Vertrautheitszeichen und Besitzanspruch des Territorıuminhabers darüber hinaus
als Mittel zur sexuellen „Reklame“ für untergeordnete Tiere eingesetzt wird.

References

ALTMAnN, J. (1973): Observational study of behaviour: Sampling methods. Behaviour 49, 227-267.

Brown, R. E.; MacDonarp, D. W. (1985). Social odours in mammals. Oxford: Clarendon Press.

Du Toır, C. F. (1980): The yellow mongoose Cynictis penicillata and other small carnıvores in the
Mountain Zebra National Park. Koedoe 23, 179-184.

EARLE, R. A. (1977): Die soziale- en voedingsgedrag van Cynictis penicıllata. Unpubl. Honours
thesis, Univ. Pretoriıa.

— (1981): Aspects of the social and feeding behaviour of the yellow mongoose, Cynictis penicıllata
(G. Cuvier). Mammalıa 45, 143-152.

EISENBERG, J. F.; KLEIMAN, D. G. (1972): Olfactory communication in mammals. Ann. Rev. Ecol.
Syst. 3, 1-32.

in, M.L. (1976): A mechanısm for individual recognition by odour ın Herpestes auropunctatus
(Carnivora: Viverridae). Anım. Behav. 24, 141-146.

GosLing, L. M. (1982): A reassessment of the function of scentmarking in territories. Z. Tierpsychol.
60, 89-118.

HEDIGER, H. (1949): Säugetier-Territorien und ihre Markierung. Bijd. Dierk. 28, 172-184.

HERZIG-STRASCHIL, B. (1977): Notes on the feeding habits of the yellow mongoose Cynictıs
penicıllata. Zool. Afr. 12, 225-229.

JoHnson, R. P. (1973): Scent marking in mammals. Anım. Behav. 21, 521-535.

KAPPELER, P. M. (1990): Social status and scent marking behaviour in Lemur catta. Anım. Behav. 40,
774-788.

LEYHAUSEn, P. (1965): The communal organisation of solitary mammals. Symp. Zool. Soc. Lond. 14,
249-263.

LyncH, C. D. (1980): Ecology of the Suricate, Suricata suricatta and yellow mongoose, Cynictis
penicıllata with special reference to their reproduction. Mem. Nat. Mus. Bloemfontein 14, 1-145.

MicHaELIs, B. (1972): Die Schleichkatzen Afrikas. Säugetierkundl. Mitt. 20, 1-110.

PRINGLE, J. A. (1977): The distribution of mammals in Natal. Part 2. Carnivora. Ann. Natal. Mus. 23,
93-115.

Rats, K. (1971): Mammalıan scent marking.. Science 171, 443-449.

Rasa, ©. A. E. (1973): Marking behaviour and its social significance in the African dwarf mongoose,
Helogale undulata rufula. Z. Tierpsychol. 32, 293-318.

— (1977): Differences in group member response to intruding conspecifics and potentially dangerous
stimuli in dwarf mongooses (Helogale undulata rufula). Z. Säugetierkunde 42, 108-112.

Rasa, O. A. E.; WENHOLD, B.; HowARD, P.; Maraıs, A. M.; PALLETT, J. (1992): Reproduction in the
yellow mongoose revisited. $. Afr. J. Zool. 27, 192-195.

138 Brigitte A. Wenhold and O. Anne E. Rasa

Rowe-Rowe, D. T. (1978): The small carnıvores of Natal. Lammergeyer 25, 148.

ScoTT, J. P. (1967): The evolution of social behaviour in dogs and wolves. Amer. Zool. 7, 373-381.

SIEGEL, $. (1956): Nonparamertric statistics for the behavioral sciences. New York: McGraw-Hill.

SoMmERS, M.; Rasa, ©. A. E.; Apps, P. J. (1990). Marking behaviour and dominance in suni antelope
(Neotragus moschatus). Z. Säugetierkunde 55, 340-352.

SMITHERS, R. H.N. (1983): The mammals of the southern African subregion. Pretoria: Publ. Univ.
Pretorıa.

STODDART, D. M. (1976): Mammalıan odours and pheromones. London: Edward Arnold.

STUART, T. T. (1981): Notes on the mammalıan carnıvores of the Cape Province, South Africa.
Bontebok 1, 1-58.

THIEssEn, D.; Rıce, M. (1976): Mammalıan scent gland marking and social behaviour. Psychol. Bull.
83, 505-539.

WENHOLD, B. A. (1990): The ethology and sociology of the yellow mongoose Cynictis penıcıllata.
M.Sc. thesis, Univ. Pretoria.

ZuMPT, 1. F. (1969): Factors influencing rabies outbreaks: the age and breeding cycle of the yellow
mongoose, Cynictis penicillata (G. Cuvier). J. S. Afr. vet. med. Ass. 40, 319-322.

— (1976): The yellow mongoose (Cynictis penicillata) as a latent focus of rabies ın South Africa.
]ES-Afez ver Ass47, 21 2:

Authors’ addresses: Prof. Dr. ©. A. E. Rasa, Abt. Ethologie, Zoologisches Institut, Universität
Bonn, D-53115 Bonn, Germany, and B. WENHOLD, Dept. of Veterinary Science,
University of Pretoria, Onderstepoort, R. $. Africa

Z. Säugetierkunde 59 (1994) 139-152
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Observations ethologiques sur Bubalus (Anoa) quarlesi
Ouwens, 1910 (Ruminantia, Bovidae) en captivite

By F. FEER

Museum National d’Histoire Naturelle de Paris, Brunoy, France

Reception du Ms. 1. 3. 1993
Acceptation du Ms. 10. 12. 1993

Abstract

Ethological observations on Bubalus (Anoa) quarlesi Ouwens, 1910 (Ruminantia, Bovidae) in captıvity

Described individual and social behaviour of captive Bubalus (Anoa) quarlesi for comparison with
known Bovinae and other Bovidae. A group of 3 males and 3 females was observed during a total of
110 hours over two periods. Many behavioural characters like horning, pawing the ground, frontal
pushing of partner, frequent social licking, frontal fighting and male mating posture were common
with known Bovinae, but others, typical of the group like herding, were absent. Coupling pawing the
ground with urination and elements of male hierarchical and sexual display were similar to other
Bovinae. The paucity of demonstrative and visual effects relates the Anoa to small forest ruminants.
Except for the privileged relations between adult female and her young daughter, the social life of the
Anoa appeared rudımentary as for species living in similar habitat.

Introduction

Les Bovines (tribu des Bovini) sont phylogenetiquement divers (BOHLKEN 1958), se
repartissent sur plusieurs continents et occupent un large Eventail de milieux, depuis la
foret tropicale dense jusqu’a la savane (WALKER 1983). Le genre Anoa ne se trouve que
dans les forets pluviales et il est endemique de Sulawesi (Iles Celebes). Tres proche du
Buffle d’eau asıatique (Bubalus bubalıs), ıl est actuellement considere comme un sous-
genre de Bubalus (HALTENORTH 1963; GROVES 1969) bien que certains auteurs en aient fait
un genre distinet. Doran (1965) considere qu’il n’y a qu’une espece et 3 sous-especes
(depressicornis, fergusoni et quarlesi) mais GRovEs (1969), dont j’ai adopte le point de vue,
distingue deux especes: l’Anoa de plaine, B. (Anoa) depressicornis (H. Smith, 1827) et
l’Anoa de montagne, B. (Anoa) quarlesı (Ouwens, 1910) qui est l’objet de cette etude. Le
manque de specimens dans les collections et de donnees provenant de la nature fait que la
discussion sur leur taxonomie reste encore ouverte. B. (Anoa) quarlesi est le plus petit
Bovine actuel, (hauteur au garrot environ 70 cm). DorAan (1965), GrRovEs (1969) et
FrÄDRICH (1973) decrivent les 2 especes et les varıations morphologiques possibles. Les
animaux que j’aı observ& presentent des taches claires sur les joues et sous le menton
nettement moins etendues que celles de B. (Anoa) depressicornis.

Il n’existe que quelques donnees parcellaires sur le comportement et la reproduction des
2 especes (DoLan 1965; POURNELLE 1965; FRÄDRICH 1973). Des etudes comportementales
sont en cours en Allemagne, mais les observations dans la nature manquent totalement. Ce
travail ne pr&sente pas un ethogramme de l’Anoa de montagne mais ıl decrit les principaux
comportements pour situer cette espece au sein des Bovines et par rapport aux Bovides.

Materiel et methodes

Les observations ont &tE faites au Zoo de Krefeld (Allemagne) durant les mois de mai 1989 et
septembre 1990. Durant la premiere periode les anımaux (Tab. 1) occupaient un enclos de 150 m? avec

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5903-0139 $ 02.50/0

140 F. Feer

Tableau 1. Caract£ristiques des Anoa de montagne observ&s au zoo de Krefeld
Sexe Date et lieu de naissance Denomination Observations

Mäle 20.1.1975 Berlin Mäle adulte Pelage noir

Mäle 8.10. 1987 Krefeld Jeune mäle Fils de la femelle adulte,
brun

Femelle 31.521977, Berlin Femelle adulte Pelage brun allaitante
en 1990

Femelle 13. 9. 1986 Krefeld Jeune femelle Fille de la femelle adulte,

brune

des gradins de pierre et des arbres et, durant la deuxicme, un enclos de 600 m?, plante d’arbustes et
decore de rochers. Les observations totalisent 86 heures en mai 1989 et 24 heures en septembre 1990
durant les periodes d’activite maximum des anımaux, c’est & dire le matin des leur sortie des boxes oü
ils passaient la nuit et en fin d’apres-midi. Les comptages des interactions sociales pacifiques et
agonistiques ont Et£ faıts sur les 4 anımaux sımultanement, puisqu’ils Etaient aisement ıidentifiables et
pouvaient s’observer ensemble pendant plus de 95 % du temps. Les descriptions des comportements
ont et completees de mesures chronometrees et de photographies. La comparaison des frequences a
ete faite par le test G avec la correction de William.

Resultats
Comportement individuel

Miction et defecation: les postures de miction different selon les sexes mais celles de
defecation sont semblables. Les mäles urinent en reculant legerement les membres poste-
rieurs (Fig. 1A). Les femelles abaissent la croupe en flechissant les membres post£rieurs et
levent la queue pendant la miction (Fig. 1B). Contrairement & ce qui se passe chez la
plupart des Ruminants, la croupe est plus abaissee lors de la defecation qu’elle ne l’est au
cours de la miction des femelles. Les membres posterieurs sont plus flechis et leurs
extremites sont Ecartees (Fig. 1C). Il n’a pas Et€ observe de varıation notable selon les
individus.

Lever de tete et du cou: la tete est levee et tournee dans le m&me temps sur son axe. Le
mouvement varie en intensite depuis le relevement de la ligne du menton ä environ 45°
accompagne& d’une faible inclinaison de la tete jusqu’au lever a la verticale avec une rotation
d’un quart de tour de la tete se presentant de profil vers l’avant, les cornes en contact avec le
cou (Fig. 2). Les jeunes indıvidus font ce mouvement en tournant sur eux-memes ou en
avancant la tete renversee en arrıere. Il semble que le mouvement initial du menton soit
dirige vers l’origine de la stimulation qui l’a provoque (congen£re, personne proche, bruit
ou odeur insolite) mais cela reste difficile A contröler pour les jeunes. Tandis que le lever de
tete est frequent chez les autres anımaux, il est tres rare chez le mäle adulte; ıl est seulement
ebauche et debouche rapidement sur la posture de parade hierarchique. FRÄDRICH (1973)
evoque brievement un comportement analogue au lever de tete dans un contexte de
menace.

Frottement de la tete: le front, la base des cornes, les cötes de la tete, les joues et plus
rarement le bord du museau sont frottes avec des mouvements de va et vient sur divers
supports fixes. Durant la deuxicme periode d’observation, les anımaux passaient la t£te, le
front ou la joue, en un seul mouvement, dans les branches de petits arbustes et le mäle
adulte frottait son front sur le sol gazonn& (Fig. 3A), stade intermediaire entre le travail du
sol avec les cornes et le passage des cornes dans un arbuste.

Räclement des cornes: le räclement des cornes sur un support rigide resulte le plus
souvent d’une amplification des mouvements de frottement avec la tete. Les traces les plus

Observations ethologiques sur Anoa en captivite 141

visibles des frottements de tete et du
räclage des cornes apparaissaient sur
l’arbre situ& au centre du premier
enclos, dont l’Ecorce &tait arrach&e
sur une grande surface jusqu’au ni-
veau du sol. Durant la deuxieme pe-
rıode, le mäle adulte, qui faısait a luı
seul 48 % de ces comportements (n =
29), repetait le mouvement avec
force, en alternance avec un grattage
du sol avec la patte avant dans plus
de la moitie des cas. Une amplıfica-
tion du mouvement conduisait par-
fois au frottement du front puis au
creusement du sol avec les cornes
devant l’arbuste. Le travail du sol
avec les cornes, effectu& parfoıs en
appui sur un metacarpe, a ete ob-
serve ä 2 reprises chez la femelle
adulte. L’action ‘de passer des
branches coupees entre les cornes et
de secouer la tete ou de les transpor-
ter sur quelques metres (Fıg.3B)
S’apparente A un jeu, le plus souvent
observe chez les jeunes anımaux
(87%, n = 63, G = 10,55, P< 0.01).
Le jeune present ä la deuxieme pe-
rıode soulevait avec les cornes les
mottes de gazon arrachees par sa
mere.

Graniage du@ssolg} Apres un
contröle olfactif de la place, le sol est gratt€ avec une patte anterieure, plus rarement avec
les 2 successivement. Le mouvement de la patte est peu accentu& et la tete est basse, le
museau proche du sol (Fig. 3C). Le mäle adulte est le seul anımal ä effectuer le grattage du
sol dont ıl accompagne 54 % (n = 165) de ses mictions: ıl avance de deux pas et urine sur la
place grattee. La grande majorite des grattages est liee aux mictions (77%, n = 117, G =
37,79, P<.0,001). Les autres ont lieu soit seuls (14 %), soit avant un räclage des cornes

Fig. 1. a: Miction du mäle; b: miction de la femelle; c:
defecation du mäle (dessins d’apres photographies)

Fıg. 2. Etapes du lever de la t&te par la femelle adulte (de gauche & droite)

142 F. Feer

dans un arbuste (7%) et plus rarement accompagnes du frottement du front ou de la
defecation. Le grattage du sol n’a jamais lieu ä proximite immediate d’un individu
particulier ou en relation avec un comportement social. Il a ete constate cependant qu’il se
produisait tres souvent apres une interaction agressive Intense avec un membre du groupe
ou des manifestations du m&me ordre dirigees vers l’exterieur. Il semble donc qu’il soit lie A
une demonstration hierarchique ou agonistique adressee A l’ensemble du groupe et/ou aux
personnes connues de l’exterieur.

Comportement social
Interactions non agressives

Flairages: les contröles olfactifs les plus frequents ont lieu dans la region ano-g£nitale et sur
la tete. Le plus jeune mäle flaire regulierement les femelles a la naissance des membres
anterieurs, sur le ventre et autour des cuisses. Les frequences des flairages different entre les
individuse (Gi SZ R528
PS, 7° 153)Tayeczune
plus grande activit€e du jeune
male (lab.D)EllEleszdirisei:
87 % sur les 2 adultes. Le flaı-
rage du penis du mäle adulte
Parle>jeunesmalezewspazalı
0% femelle adulte a Ete reguliere-
ment observe.

Contröle urinaire: en de-

A hors du contexte strictement
sexuel du mäle contrölant l’u-
rine d’une femelle, le lechage
de l’urine pendant la miction
ou sur le sol a te observ& chez
tous les autres individus du
groupe. Les 2 adultes lechent
occasıonnellement leur propre
urine.

Contact frontal: le front
est pose sur le flanc, l’epaule
ou le cou du partenaire; dans

b 60% des cas (n= 41), ilya
une breve poussee (Fig. 4A).
Ce comportement peut prece-
der une interaction pacıfique
et ne declenche en general au-
cune reaction defensive. Du-
rant la premiere periode, il est
surtout observ& chez la fe-
melle adulte (59 %, n = 46, G
= 146,34, P< 0,001) principa-

c lement sur le mäle adulte (G =
22,00, P<0,01,n=27).

Frottement de la t£te: la

Fig. 3. a: Frottement du front sur le sol; b: jeu avec les bran- JOUE, plus rarement le cöte du
chages; c: grattage du sol par le mäle adulte front, le cou ou le menton

Observations ethologiques sur Anoa en captıvite 143

Tableau 2. Nombre de flairages (a), de contact frontaux (b), et de frottement de la tete (c) observes
dans le groupe d’Anoa durant la premiere periode

Animal Anımal receveur

initiateur d adulte ? adulte d jeune ? jeune

10
& adulte b 1
3

—

2 adulte

155)

oo oONW ON
N

DUE@:S TS ang

no

d jeune

? jeune

sont frottes de bas en haut, rarement de mani£ere r&petee, sur le haut des cuisses et la region
anale du partenaire (Fig. 4B). Durant la premiere periode, la majorite des observations a ete
faite chez la jeune femelle (70 %, n = 136, G = 68,06, P< 0,001), ä& 97 % sur sa mere.
Lechage social: le lechage est la ftorme de contact pacifique la plus prolongee entre
individus. Nous ne considererons dans ce chapıtre que les lechages non sexuels, le plus
souvent pratiques sur la tete, le cou et une zone qui va du garrot A la croupe. L’animal qui
veut lEcher approche le partenaire par le cöte ou par devant, la t£te basse et le museau tendu
en avant. L’invitation au lEchage consiste en une immobilisation de la tete legerement

Fıg. 4. a: Poussee du front par la femelle adulte sur le mäle adulte; b: frottement de la joue par la jeune
femelle sur sa mere

144 Tiokeen

abaissee, museau dirige vers le
sol, a proximite de la tete du par-
tenaire (Fig.5) puis la tete est
parfois legerement tournee, de
facon & presenter la joue, ou rele-
vee pour montrer le cou. Face &
un individu qui menace, l’atti-
tude d’invitation au lechage a un
effet inhibiteur immediat. Le le-
chage est le plus souvent unilate-
ral; quand 2 anımaux se lechent,
Fig. 5. Posture d’invitation au lEchage par l’anımal de gauche c’est toujours en alternance et ja-

mais sımultanement. De longues
seances de lechage sont observees, mettant parfois en jeu 3 anımaux. Les frequences des
lechages sont differentes entre les individus (G = 14,74, P< 0,01, n = 284) avec une plus
grande activit€ du jeune mäle (Tab. 3). Les lechages entre mäles et entre femelles sont plus
frequents que les lechages intersexuels (respectivement G = 6,42, P<0,02,n = 156 et G =
11,05, P<0,001, n = 128). Les lechages du mäle adulte par le jeune mäle sont plus
nombreux que l’inverse (G = 6,02, P<0,02, n=78),.

Pose du menton: le menton est pose sur la partie posterieure du corps du partenaire ou
plus rarement sur le garrot. Ce comportement qui fait partie du comportement sexuel
normal du mäle adulte (cf plus loın) a ete observe& chez les jeunes, le plus souvent le mäle
(G = 5,55, P< 0,002, n=46).

Tableau 3. Nombre de lechages entre les individus du groupe d’Anoa durant la premiere periode

Animal Animal receveur

inıtiateur & adulte 2 adulte d jeune

d adulte 24 20
? adulte 19
d jeune 30

Q jeune 30 25

Association au repos: les frequences des repos solitaires par rapport aux repos en
association avec un autre anımal du groupe sont differentes (G = 40,27, P<0,001, n
= 205). Le mäle adulte est plus souvent seul (71 %, n = 28) que la femelle adulte (G = 26,99,
P< 0,001, n = 93), le jeune mäle (G = 39,39, P< 0,001, n = 97) ou la jeune femelle (G =
29,94, P<.0,001, n = 71). Ces 3 derniers montrent un certain allomimetisme et prennent
leur repos ensemble dans 15 % des cas de repos commun (n =81). Le mäle adulte suit un
rythme relativement different et occupe preferentiellement quelques emplacements qui lui
sont exclusifs.

Interactions agressives

Menace t£te basse: la tete est abaissee en direction du partenaire, la ligne du front proche de
la verticale et le museau pres du sol. Quand le partenaire adopte une posture ıdentique, les
anımaux se font face, front bas et perpendiculaire au sol (Fig.6A). Ce comportement,
egalement observ& par FRÄDRICH (1973), est peu frequent (18% des menaces, n = 70) et se
produit le plus souvent entre les adultes chez qu’il semble correspondre ä& une lutte avortee.
Menace des cornes: la t&te, en position basse, est flechie de maniere ä& diriger les pointes
des cornes vers l’adversaire (Fig. 6B). C’est une menace plus serieuse que la precedente.

Observations ethologiques sur Anoa en captivite 145

Fig. 6. a: Menace tete basse reciproque; b: presentation des cornes; c: parade hierarchique; d: coup de
cornes

Parade: le cou est porte horizontalement et le museau est tendu en avant, la tete Etant
tournee d’environ 45° sur son axe. Cette position est maintenue pendant l’approche du
partenaire en marche rapide ou au trot (Fig. 6C) parfois jusqu’a ce que les 2 anımaux soient
en parallele. La tete est le plus souvent tourne&e de facon a montrer les cornes au partenaıre.
Le mäle adulte ne parade pas en direction des autres anımaux du groupe, mais toujours en
reaction A une perturbation venant de l’exterieur. Ce comportement a Egalement ere
observ&, mais tres rarement, chez la femelle adulte et le mäle vivant ısole. FRÄDRICH (1973)
publie une photographie d’un mäle en attitude de parade, les cornes dirigees vers un anımal
d’une autre espece sıute dans l’enclos voisin.

Lever hierarchique: il se compose de poussee du front ou de coup de cornes sur le flanc
ou la croupe, d’appui du menton ou de la joue sur la croupe ou le garrot, d’appui d’une
patte anterıieure (rarement les deux) sur le corps. En reponse aux coups de cornes, l’anımal
couche peut allonger le cou et le menton sur le sol.

Charge agressive: l’agresseur galope droit sur son adversaire, la tete en position normale
a hauteur du garrot ou redressee avec occasionnellement les oreilles rabattues en arrıere. De
brefs meuglements peuvent &tre entendus. La charge se termine par un coup violent si le

146 F. Feer

partenaire ne fuit pas. Le mäle adulte, tres agressif envers l’observateur et certains
soigneurs, faisait frequemment des charges dans leur direction, puis les redirigeait vers les
anımaux du groupe sıtues pres de lui. Dans des moments d’excitation avant ou apres la
charge, il marchait en bloquant brievement en l’air une patte anterieure ou posterieure, une
attitude egalement observee chez les autres individus quand ils &taient inquiets ou excites.

Coup de front: ä la difference du simple contact et de la poussee, le coup de front sur
l’epaule ou le flanc est porte avec Elan. Il est peu brutal & courte distance, mais peut-£tre
violent quand il suit la charge.

Coup de cornes: apres flexion de la tEte, le coup de la pointe des cornes est donne de bas
en haut, le plus souvent au flanc ou au ventre de l’adversaire (Fig. 6D). Des mouvements de
tete repetes de bas en haut A vide ont Et€ observes chez le mäle et la femelle adulte.

Les comportements agonistiques sans contact corporel (menace, charge) (Tab. 4) s’ob-
servent pour l’essentiel entre les 2 adultes (95 %, n = 150, G = 83,45, P< 0,001, n = 300).
Le mäle adulte ne les dirige vers aucun partenaire en particulier (G = 0,02, N S, n = 142)
tandıs que la femelle adulte les concentre sur le mäle adulte (G = 55,83, n = 142). La
frequence des comportements agressifs directs est la plus Elevee chez le jeune mäle (G =
70,53, P< 0,001, n = 444) qui montre d’ailleurs peu d’autres comportements agonistiques.
Le jeune mäle les dirige principalement vers la femelle adulte qui y repond relativement
peu, ce qui fait supposer qu’il existe une part de jeu dans ces &changes. Au contraire, les
relations interadultes sont tendues, caracterisees pour la femelle et pour le mäle par une
faıble frequence relative de comportements directs par rapport aux comportements indi-
rects (respectivement G = 5,35, P< 0,05, n = 196 et G = 6,53, P< 0,02, n=56).

Tableau 4. Nombre de comportements agonistiques indirects (menaces, charges: a) et directs (coup
de front, de cornes, lever: b) dans le groupe d’Anoa observe durant la premiere periode

Anımal Anımal receveur

initiateur d adulte ? adulte d jeune Q jeune

d adulte 23 31 17
5 22 13
2 adulte

d jeune

? jeune

Hierarchie: durant la premiere periode la femelle adulte marque sa dominance par
davantage de comportements indirects ou directs que le mäle adulte (respectivement G =
10,63, P<0,01, n = 1976 et G = P<.0,001) qui cede le passage quand elle approche.
Pendant la deuxieme periode la femelle adulte ne manifeste aucune agressivite envers le
mäle adulte tandıs que celui-ci montre une frequence de comportement agonistique
semblable a celle de la premiere periode (respectivement 0,46/h et 0,33/h). Je n’aı observe
aucun combat serieux entre adultes mais d’apres les soigneurs ils ont periodiquement lıeu et
sont tres intenses. Le mäle adulte manifeste davantage son agressivite vis ä vis de l’exterieur
et particulierement envers un autre mäle subadulte dans un enclos voisin durant la
deuxieme periode. Ceci correspond bien aux observations de DoLan (1965) sur l’Anoa de
plaine. Alors que les agressions entre femelles sont rares durant la premiere p£riode, elles
deviennent plus frequentes pendant la deuxieme, particulierement de la part de la femelle

Observations ethologiques sur Anoa en captivite 147

adulte qui allaite encore son nouveau jeune (14 observations contre 2 chez la jeune femelle
qui connait son premier oestrus).

Soumission: l’anımal domine se couche, cou et menton sur le sol. Cette attitude
observ&e chez le jeune mäle permet egalement d’eviter les coups de cornes portes sous le
corps par la femelle adulte.

Jeu

Le jeu solitaire se manifeste par des courses au galop & travers l’enclos ou des charges vers
des objets ou des oiseaux. Un anımal provoque un autre au jeu en galopant dans sa
direction avec des bonds desordonnes (Fig.7A) ou avec des hochements de tete avec
presentation des cornes. FRÄDRICH (1973) decrit les m&mes comportements ludiques
parfois tres prolonges chez les plus jeunes. Le jeu de combat consiste en poussees, front
contre front ou cornes engag&es, avec les tetes complement flechies et les museaux rabattus
vers l’arrıiere. Les corps sont face a face (Fig. 7B) ou en position parallele, les tetes tournees
l’une contre l’autre (Fig. 7C). Pour resister aux poussees de son adversaire ou A ses coups de
cornes donnes sous le corps, l’un des partenaires se tient occasıonnellement sur les
metacarpes (Fig. 7D). Pendant l’affrontement, les anımaux cherchent a se degager pour
donner des coups de front ou de cornes dans le tlanc de l’adversaire. Comme ils cherchent

Fıg. 7. Jeu social: a: galop d’invitation au jeu et poursuite A 3 animaux; b: combat de jeu, face & face; c:
combat de jeu en position laterale; d: appui sur les metacarpes

148 F. Feer

egalement a derober leur propres flancs aux coups, ıl en resulte un mouvement circulaire
des deux participants parfois en position parallele inversee. Le jeu de combat est entre-
coupe de brefs arr£ts face A face, tEtes basses, et de poursuites au cours desquelles l’individu
poursuivi rue de maniere desordonnee vers son rıval. Des meuglements se font entendre
dans ces moments d’excitation intense. D’apres les soigneurs, le combat de jeu serait tout ä
fait sımilaire au combat serieux.

Durant la premiere periode d’observation, la majorite des jeux avaient lieu entre les 2
jeunes (82 %, n = 79, G = 47,02, P< 0,001, n = 158). Le jeune mäle en £tait le plus souvent
l’initiateur (73 % des invitations observe&es, n = 46, G = 6,36, P< 0,02, n = 112). Il invitait
principalement la jeune femelle (66 %, n = 41) mais aussi le mäle adulte que ne repondait
que rarement, tandis que la femelle adulte, pourtant moins sollicitee, jouait plus souvent
avec Juı (15 %, n = 79). Durant la deuxieme periode d’observation, la grande majorite des
jeux se deroulaient entre la jeune femelle et le nouveau jeune de la femelle adulte.

Comportement sexuel

Contröle urinaire: les mäles contrölent l’etat genital des femelles en lechant la vulve et ils
provoquent le plus souvent la miction (Fig. 8A). Ils laissent couler l’urine sur le museau et
la goütent en sortant la langue. Le Flehmen suit frequemment ce comportement qui peut
ensuite Etre immediatement renouvele ou suivi du lechage des talons et des cuisses
mouillees d’urine.

Flehmen: au cours du Flehmen la levre sup£rieure est retroussee, la bouche entrouverte
et la tete un peu relevee, le front proche de l’horizontale; la tete est parfois un peu tournee
de cöte (Fig. 8B). Chez le mäle adulte qui fait la grande majorite des Flehmen durant la
premiere periode (91 %, n = 129), ce comportement est presque toujours unique et dure
peu longtemps (7,9 s +/- 1,8; n = 19). Le Flehmen des mäles a lieu essentiellement & l’urine
des femelles mais quelques observations ont Ete faıtes du mäle adulte et du mäle isole
reagissant A leur propre urine apres s’etre lEche le penis. La femelle adulte a effectue un
Flehmen apres avoir provoque la miction de la jeune femelle, et 3 autres en reaction A
diverses stimulations olfactives provenant de l’exterieur.

Parade sexuelle: le mäle suit la femelle, le cou etire horizontalement A hauteur du garrot,
le museau tendu en avant dans le prolongement (Fig. 8C). Cette posture est analogue & celle
de la parade hierarchique mais la rotation de la tete ne l’accompagne que rarement. Quand
le mäle est proche de la femelle, ıl donne des coups de langue & vide et cherche & luı lecher la
vulve pour provoquer la miction. Quand le mäle approche pour tester la receptivite de la
femelle, il avance brusquement derriere elle avec un trepignement des pattes anterieures, en
relevant le cou et le menton ä environ 45°, le dessous du cou venant en contact avec la
croupe de la femelle (Fig. 8D). Quand les anımaux sont tres excites, ils font entendre de
brefs meuglements.

Monte sexuelle: apres la parade sexuelle et/ou le lechage vulvaıre, le mäle pose son
menton sur la croupe de la femelle ou monte directement (Fig. 8E). Pendant la monte le
cou et le menton sont appuyes sur le dos et le garrot de la femelle (Fig. 8F); les pattes
anterieures raidies enserrent ses flancs.

Discussion

Le comportement d’Anoa pr&sente de nombreux traits communs avec differents Bovines
connus, ce qui est en accord avec son appartenance A ce taxon, basee sur les criteres
morphologiques. La miction et la defecation s’accomplissent separ&ment et sans postures
particulierement accus&es, comme chez la plupart des Bovides. Le grattage du sol comme
comportement d’expression est tres repandu dans tous les groupes de Bovides ä l’exception

Observations ethologiques sur Anoa en captivite 149

Fıg. 8. Comportement sexuel: a: lechage vulvaire par le mäle provoquant la miction de la femelle; b:
Flehmen du mäle; c: parade sexuelle du mäle; d: comportement pr&copulatoire avec trepignement des
pattes anterieures; e: pose du menton et tentative de monte; f: monte sexuelle

des Tragelaphines et de Boselaphus. Chez Anoa cependant il est restreint A quelques
comportements particuliers du mäle alors qu’il est beaucoup plus gen£ralise chez la plupart
des Bovines (SCHLOETH 1958). Remarquons que les mouvements de grattage du sol du
mäle Anoa sont beaucoup moins accuses que ceux de Bos taurus (SCHLOETH 1958). Le
grattage avec les cornes, le frottement de la tete sont des comportements de confort qui ont
egalement une fonction d’expression dynamique et sont tres repandus chez les autres
Bovides (WALTHER 1968). Les contacts frontaux entre adultes peuvent &tre consideres

150 F. Feer

comme des attitudes agressives ritualisees servant au contact social mais il ne semble pas
exister de comportement semblable au «Hornen» decrit par SCHLOETH (1961) chez 2.
taurus. Bos javanicus faıt un frottement de joue (HALDER 1976) semblable 3 celui de la
jeune femelle Anoa. La frequence des lechages sociaux entre adultes rapproche davantage
l’Anoa de son groupe et des Tragelaphines (SCHLOETH 1961; WALTHER 1964) que d’autres
Bovides (WALTHER 1979). La valeur apaisante de l’invitation au lEchage est connue chez 2.
taurus (SAMBRAUS 1969). L’initiateur du leEchage a souvent un rang inferieur A son
partenaire comme chez B. taurus (SCHLOETH 1961; Bouıssou 1974) et B. ganrus (SCHAL-
LER 1967).

L’imposition laterale en parallele ou antiparallele, tres courante chez les Bovines, n’a pas
ete observee chez Anoa mais il est possible qu’elle existe entre 2 mäles adultes. Les menaces
et le combat frontal ne le distinguent pas de nombreux Bovides. La position en appui sur
les metacarpes sert ä Eviter le soulevement par l’adversaire mais elle n’est pas une attitude
typique comme chez Boselaphus, Tetracerus et Alcelaphus. La pose du menton sur la
croupe de la femelle avant la monte est presente chez les autres Bovines (SamBraus 1971;
Lorr 1974; HALDER 1976) mais sans mouvement lateral. La posture de monte est
contorme ä celle de la plupart des Bovines, Boselaphus, Tragelaphus.

Certains comportements typiques des Bovines connus sont absents chez Anoa, tandis
que d’autres le rapprochent de groupes voisins. Le couplage preferentiel du grattage du sol
avec la miction chez le mäle rappelle le comportement des petites esp£eces territoriales
d’Alcelaphines et d’Antilopines mais ıl est difficıle de l’interpreter en captivite. Le
frottement et grattage de supports laissent des traces visibles qui pourraient avoir une
valeur de marquage optique. Il est probable que chez une esp£ce forestiere a tendance
probablement solitaire comme l’Anoa, ils aient la m&me fonction que chez les Ruminants
forestiers (DuBosT 1983; FEER 1984). Le lever de tete et du cou semble &tre absent chez les
autres Bovines, alors qu’un comportement assez similaire a et€ observee chez Capra ıbex et
Hemitragus jemlahicus, et chez les jeunes Taurotragus euryceros (HAMANN 1979). Il existe
aussi chez les femelles de Tragelaphines (WALTHER 1958).

La parade hierarchique d’Anoa n’a pas EtE clairement observee chez les Bovines. La
rotation de la tete en particulier, ressemble au comportement de Ovis canadensis (GEIST
1966). Au cours de la poursuite sexuelle Anoa fait des coups de langue a vide comme les
Caprınes, Aepyceros et Boselaphus et sa posture de parade est rare chez les Bovines. Le
«Hüten» qui caracterıse de nombreux Bovines (WALTHER 1979) est absent. Le relevement
du cou contre la femelle a la fin de la parade sexuelle, couple A un trepignement des pattes
anterieures n’a Ete observ& que chez B. javanıcus (HALDER 1976) et rappelle le Laufschlag
rudimentaire et rıtualise de Gazella granti (WALTHER 1965) et Antilope cervicapra (DUBOST
et FEER 1981).

Une partie du comportement d’Anoa a pu &tre model& par son mode de vie, comme par
exemple le bloquage d’une patte en l’air qui caracterise les especes de milieux denses
(WALTHER 1979). De m&me, la pauvrete des Elements demonstratifs visuels et la forme de la
parade hierarchique et sexuelle rapproche Anoa des petits Ruminants forestiers comme
Muntiacus, Cephalophus, Neotragus (Dusost 1983). Ces derniers se montrant dans
l’ensemble moins gregaires que ceux des milieux ouverts (EstEs 1974; WALTHER 1979;
Dusosrt 1983) on peut s’attendre A trouver chez Anoa les indices d’une vie sociale
relativement rudimentaire. Malgr& P’exiguite des conditions de captivite qui risquent de
biaiser les observations, il apparait des possibilites de vie en groupe entre femelles et
subadultes (allomimetisme, frequents lEchages, jeux sociaux, association au repos). Le mäle
adulte serait plus solitaire et intolerant envers un autre adulte. La femelle adulte garde des
relations privilegiees avec sa fille tant qu’elle n’est pas en oestrus ou qu’elle n’a pas elle
m&me un autre jeune. La vie sociale d’Anoa serait proche de celle des petites especes
forestieres comme Muntiacus et Cephalophus (Dusosrt 1980, 1983a, 1983b). Des donnees
sur les Bovines plus ou moins lies ä la fort comme Bos banteng, Bos sauveli et

Observations ethologiques sur Anoa en captivite 151

paticulierement B. mindorensis (TALBOT et TALBOT 1966) manquent pour discuter de cette
question.

Remerciements

Je remercie le Dr. W. EnckE, Directeur du Zoo de Krefeld et le Dr. P. VoGT pour leur accueil et leur
aide sur place. Ce travail a Et€ finance par le laboratoıire d’Ecologie Generale du Museum National
d’Histoire Naturelle de Paris. Je remercie G. DuBosT, A. HEYMER et un critigue anonyme pour leur
commentaires sur le manuscript.

Zusammenfassung

Ethologische Beobachtungen an Bubalus (Anoa) quarlesı Ouwens, 1910 (Ruminantıa, Bovidae) im Zoo

Zwischen Mai und September 1990 wurde im Zoo Krefeld eine Gruppe von Berg-Anoas (3
Männchen, 3 Weibchen) während insgesamt 110 Stunden ethologisch beobachtet. Individual- und
Sozialverhalten werden beschrieben und die Ergebnisse, soweit möglich, mit anderen Bovinae und der
Gesamtheit der Bovidae verglichen. Das Vorderlaufscharren ist ein meist mit der Miktion assoziiertes
Verhaltenselement und wird nur vom erwachsenen Bullen ausgeführt. Es begleitet aber auch das
Bodenhornen, Stirnreiben und Aufsteilen. Die häufigsten sozialen Kontaktformen sind das gegensei-
tige Kopfstoßen, das Kopfreiben auf der Kruppe des Partners und das soziale Lecken. Drohen mit
gestrecktem Kopf oder mit den Hörnern sowie effektive Angriffe charakterisieren die ständig
gespannten Beziehungen zwischen den Erwachsenen, während die Jungtiere häufiger den frontalen
Kopfstoß anwenden oder mit dem Gehörn direkt angreifen. Drohverhalten mit gestrecktem Hals und
seitlich verdrehtem Kopf ist das häufigste Element der Bullen. Soziales Spielverhalten zwischen den
Jungtieren besteht aus Verfolgungen und langen Stößen Kopf gegen Kopf oder in einer sich um die
Achse drehenden Antiparallelstellung. Bei einer brünftigen Kuh zeigt der Bulle dann ein Werbeverhal-
ten, das der Rangdemonstration sehr ähnlich ist. Der hochgestreckte Kopf geht dem Aufspringen
voraus und wird von Tretelbewegungen mit den Vorderläufen begleitet. Das Verhalten der Berg-
Anoas zeigt zahlreiche Gemeinsamkeiten mit den übrigen Bovinae, jedoch sind deren Ausdruckswei-
sen weniger markant und weniger differenziert. Die Anoas besitzen unter anderem einige besondere
Verhaltensweisen wie das Treten mit den Vorderläufen als ritualisierten Laufschlag und das Kopfdre-
hen beim Drohen, die sie mit anderen Bovidae verbinden. Sie haben auch Gemeinsamkeiten mit
einigen kleinen Waldarten wie Muntiacus und Cephalophus, so etwa in ihrem Sozialverhalten.

Bibliographie

BoHLkEn, H. (1958): Vergleichende Untersuchungen an Wildrindern (Tribus Bovini 1945). Zool. Jb.
68, 113-202.

Bousssou, M. F. (1974): Etablissement des relations de dominance-soumission chez les bovins
domestiques. I: Nature et Evolution des interactions sociales. Ann. Biol. anımale Biochem.
Biophys. 14, 383—410.

Doran, J. M. (1965): Breeding of the lowland anoa, Bubalus (Anoa) d. depressicornis H. Smith, 1827)
in the San Diego Zoological Garden. Z. Säugetierkunde 30, 241-248.

Dusost, G. (1980): L’ecologie et la vie sociale du Cephalophe bleu (Cephalophus monticola
Thunberg), petit ruminant forestier afrıcain. Z. Tierpsychol. 54, 205-266.

— (1983): Le comportement de Cephalophus monticola Thunberg et C. dorsalıs Gray, et la place des
cephalophes au sein des ruminants. Mammalia 47, 141-177 et 281-310.

DusosT, G.; FEER, F. (1981): The behavior of the male Antilope cervicapra L., ıts development
according to age and social rank. Behaviour 76, 62-127.

— — (1988): Variabilite comportementale & l’interieur du genre Cephalophus (Ruminantia, Bovidae),
par l’exemple de C. rufılatus Gray, 1846. Z. Säugetierkunde 53, 3147.

Estzs, R. D. (1974): Social organization of the African Bovidae. In: The behaviour of ungulates and its
relation to management. IUCN new series 24, 166-205.

FrRÄDRICH, H. (1973): Einige Bemerkungen über den Anoa. Z. Kölner Zoo 16, 101-105.

FEER, F. (1984): Observations ethologiques sur Pudu pudu (Molina, 1782) en captivite. Zool. Garten
54, 1-27.

GEIST, V. (1966): The evolutionary significance of mountain sheep horns. Evolution 20, 558-566.

Grovzs, C. P. (1969): Systematics of the Anoa (Mammalıa, Bovidae). Beaufortia 17, 1-12.

HALDER, U. (1976): Okologie und Verhalten des Banteng (Bos javanicus) in Java. Eine Feldstudie.
Mammalıa depicta 10, Hamburg: Parey.

HALTENORTH, T. (1963): Klassifikation der Säugetiere: Artiodactyla. Hdb. Zool., Berlin 8 (32),
1-167.

Hamann, U. (1979): Beobachtungen zum Verhalten von Bongoantilopen (Tragelaphus euryceros
Ogılby, 1836). Zool. Garten 49, 319-375.

152 F. Feer

LoTT, D. F. (1974): Sexual and agressive behavior of adult male American bison (Bison bison). In: The
behaviour of ungulates and its relation to management, IUCN, new series, 24, 382-394.

Ouwens, P. A. (1910): Contribution a la connaissance des mammiferes de Celebes. Bull. Dept. Agric.
Indes Neerl. 38.

POURNELLE, G. H. (1965): A breeding herd of Lowland anoas, Anoa d. depressicornis. Int. Zoo
Yearbook 5, 56-57.

SAMBRAUS, H. H. (1969): Das soziale Lecken des Rindes. Z. Tierpsychol. 26, 805-810.

SCHALLER, G. B. (1967): The deer and the tiger. A study of wildlife in India. Chicago: Chicago
University Press.

— (1976): Aggressive behaviour of domestic yak. J. Bombay Nat. Hist. Soc. 73, 385-389.

SCHLOETH, R. (1958): Cycle annuel et comportement socıal du taureau de Camargue. Mammalia 22,
121-139.

— (1961): Das Sozialleben des Camargue-Rindes. Z. Tierpsychol. 18, 574-627.

TALBOT, L. M.; TALBoT, M. H. (1966): The tamarau (Bubalus mindorensis (Heude)) observations and
recommendations. Mammalia 30, 1-11.

WALTHER, F. (1958): Zum Kampf- und Paarungsverhalten einiger Antilopen. Z. Tierpsychol. 15,
340-380.

— (1965): Verhaltenstudien an der Grantgazelle (Gazella granti Brooke 1872) im Ngorongoro-
Krater. Z. Tierpsychol. 22, 167-208.

— (1979): Das Verhalten der Hornträger (Bovidae). Hdb. der Zoologie, 8.(54), 1-184.

Adresse de Pauteur: Dr. FRANcOIS FEER, Laboratoire d’Ecologie Generale, MNHN, CNRS, 4,
Avenue du petit Chäteau, F-91800 Brunoy, France

Z. Säugetierkunde 59 (1994) 153-160
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Chromosomal reduction in an Okapi pedigree
(Okapia johnstoni)

By P. Perırt, H. De Boıs, and W. DE MEURICHY

Division of Human Genetics, Department of Human Biology, University Hospital, Leuven and Royal
Zoological Society of Antwerp, Belgium

Receipt of Ms. 1. 2. 1993
Acceptance of Ms. 17. 11. 1993

Abstract

The karyotype of a female okapı showing 2n = 44 has been investigated by G-R-C and Ag-NOR
banding methods. This anımal is the offspring of a captive female and a wild-caught male, which are
both heterozygotes showing 2n = 45 with centric fusion between two unequal-sized acrocentric
chromosomes. The okapı karyotype was arranged according to the cattle standard karyotype in
consideration of the high degree of banding homologies found between the two species. The reduction
from 2n = 46 to 2 = 45 and 2n = 44 ıs the result of a Robertsonian translocation involving cattle
equivalent chromosomes 4 and 26. Other autosomal rearrangements, like centric fusions and a tandem
translocation, as well as structural changes in the X and the Y chromosomes, are tentatively identified.
The location of centromeric heterochromatin and of nucleolus organızer regions in the okapi
karyotype are described and discussed in relation to the karyotype ın other species of Bovidae.

Introduction

The place of the Giraffidae, including the okapı, in the mammalıan systematics is still
controversial (GiJzEn 1959; Taylor et al. 1969; ROMER and Parsons 1986). Before the
advent of the banding techniques, the chromosomes of the okapı (Okapia johnstoni), one
of the two surviving species ın the family Giraffidae, were first examined by ULBRICH and
ScHMITT (1969) showing 2n = 46. Subsequently, Hösrı and Lang (1970) studied two other
anımals with 2n = 45. Later BENIRSCHRE et al. (1983) confirmed the fusion between two
acrocentric autosomes as a common cytogenetic event referred to as “Robertsonian”
without phenotypical changes ın anımals with reduction from 46 to 45 chromosomes.
Using various banding methods, PETtTT and MEURICHY (1986) studied two other anımals
showing 2n = 46 and 2n = 45 ın a female and a male okapi, respectively. We were also able
to notice the centric fusion between two unequal-sized acrocentrics in the male heterozy-
gote anımal with 2n = 45. We report here the cytogenetic studies of a fertile female anımal
with 2n = 44 (Studbook No 328) showing two translocated elements as the chromosomal
origin of her homozygote status and of her male son with 2n = 45 (Studbook No 403),
respectively. The okapı “nombre fondamental” (NF) of 60 is very close to the most
common bovid NF of 60, therefore, karyotypes of these anımals were performed accord-
ing to standardized cattle banded karyotype (ISCNDA 1989).

Material and methods

Figure 1 illustrates the pedigree prepared from the International Studbook of the Okapi (Puijen-
BROECK and Boıs 1991). Anımals No 257 and No 273 were previously reported on by PErrr and
MEuRIcCHY (1986). The diploid chromosome number of okapı No 283 was established in a North-
American zoo. To our knowledge, animals No 311 and No 343, have not yet been karyotyped.
Chromosome studies of the remaining anımals (Studbook Nos 219, 403) were routinely performed by

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5903-0153 $ 02.50/0

154 P. Petıt, Hl. De Bois, and W. De Meurichy

45
257

46 44 ?
273 328 343

46
| 283
394

Fig. 1. Pedigree of okapı with studbook numbers in lowercase and chromosome numbers in
uppercase. 46 indicates diploid number with absence of t(4;26); 45 and 44 heterozygote and
homozygote for t(4;26), respectively indicates male stillborn and ? that chromosome number is
unknown

us. The metaphases were observed from cultured skin biopsies, applying to medium and high
resolution G and R coloration banding methods (Par and THomAs 1980; Yunıs et al. 1978). C-bands
were stained applying to the method of SumNER (1972). Nucleolus Organizer Regions (NORSs) have
been located using the silver staining technique (GOODPASTURE and BrooMm 1975). The okapı
karyotypes were constructed from at least ten well banded metaphase spreads in each case and were
prepared according to the standardized cattle banded karyotypes (ISCNDA 1989).

Results

The G-banded karyotypes of two anımals demonstrating the reduction from 2n = 45 to 2n
= 44 are illustrated in figure 2. In the female with 2n = 44 seven pairs of submetacentric and
fourteen pairs of acrocentric chromosomes are numbered according to the cattle nomencla-
ture (ISCNDA 1989). As a consequence of this, the seven biarmed chromosomes are
identified as follows: t(27;2;23), t(4;26), t(5;29), t(10;24), t(14;25), t(17;18), t(20;22). The
(2752523) ıs a Robertsonian type 2;23 translocation and 2;27 a tandem translocation. The
centromeric region of the chromosome 27, fused at the telomeric region of 2, could not be
identified. This t(27;2;23) has been identified previously as t(2;22) by PETIT and MEURICHY
(1992). The translocation event which concerns the chromosomal homozygosity in the 2n
= 44 female okapı is ıdentified as t(4;26). The remaining acrocentric okapı chromosomes
appear similar in G- and R-banding to the banding patterns of cattle homologues (Fig. 3).

Chromosomal reduction in an Okapı pedigree 155

F-4
| „rs no;
nd main 7
1 27/2 [23 3 4/26 5 29 6 7

8 9 10/24 11 12 13 14 /25

„ u
- „m 6° v "ss
1 el „Wo Hu UN ı- [)
15 16 17 fe 19 20/22 21 28 xx

% |
ı\kınin

2. 202./23 3 ajae A 5/20 6 7

%% | |
ee‘

8 9 10/24 11 12 = 14/25

15 16 17/18 19 2022 >23 26 28 xY

Fıg. 2. a: G-bandes karyotype of the 2n = 44 female okapi (Studbook No 328) showing the two t(4;26)
elements. b: G-banded karyotype of 2n = 45 male heterozygote (Studbook No 403)

In the male animal with the modal number of 2n = 45 heterozygosity for t(4;26) ıs
demonstrated in figure 2. After using C-banding, small amounts of centromeric hetero-
chromatın (HC) are observed in all the biarmed chromosomes as in t(27;2;23), t(5;29),
t(10;24), t(17;18), respectively, whereas in the two remaining translocation elements
t(4;26) and t(20;22) much larger HC blocks are demonstrated (Fig. 4). After NOR banding
silver dots are observed located in the centromeric region of chromosomes 3, 6, 11, and 28,
respectively (Fig. 5).

P. Petit, H. De Bois, and W. De Meurichy

156

87

jewrue 44 = uz ayı Jo odAroAıey propdey papueg-H-y pourquon 'g "317

LG

6

9

sı

Chromosomal reduction in an Okapı pedigree 157

“
; ‚ .
‚N E,
| % 2 i r an " ® Mi
% ur i yo . .» =
- |.

Fig. 4. C-banded metaphase chromosomes of the 2n = 44 anımal showing large heterochromatın
blocks in two t(4;26) and t(20;22) elements (large arrows: t) and an additional interstitial C-band close
to the centromeric C-band in the X chromosomes (small arrows). Note also reduced amounts of HC
in the other biarmed chromosomes except for t(4;26) and t(20;22) (arrowheads)

Discussion

From a survey of 21 okapi, BENIRSCHRE et al. (1983) reported that 8 anımals had 46
chromosomes and 13 had 45 chromosomes. These authors demonstrated that the reduction
from 46 to 45 chromosomes was the result of a Robertsonian translocation between two
unequal acrocentrics without deleterious effect in the carriers. Recently 43 anımals have
been karyotyped, among a captıve population of 74 anımals, showing that 18 anımals
possess 46 and 25 only 45 chromosomes (PuIJENBROECK and Boıs 1991). However, the
geographical origin of the t(4;26) remains unknown and was most likely imported from the
wild as suggested by BENIRSCHRE et al. (1983). Thus, the conception of an okapı anımal
with 2n = 44 resulting from mating of a captive mother (No 249) with a wild-caught father
(No 257) would occur at a frequency of 25 % when two 2n = 45 anımals are crossed. In thıs
report a 2n = 44 fertile female (No 328) gave birth to a male stillborn (20 kg) with 2n = 45
(No 394) wıthout visceral anomalies at autopsy, and later to a second healthy male (No
403) with 2n = 45. Furthermore, the history of this 2n = 44 female okapı did not reveal
either miscarriages or aborted malformed fetuses as a result from chromosomal malsegre-
gation. According to FosseE (1978) neonatal mortality rate in the captive okapı population
was high but has been virtually eliminated ın the United States in contrast to the European
okapı population (Bois et al. 1988).

Cytogenetically, the distribution of HC, as revealed by C-banding, has demonstrated
interesting features in the okapi karyotype. At first, the well marked amount of HC in the
acrocentric autosomes is conform with observations ın cattle (BUCKLAND and Evans
1978b). Secondly, ın the pericentromeric regions of the fused elements t(4;26) and t(20;22)
large blocks of HC contrast with the small blocks in the 5 other translocated autosomes.
Similarly, large blocks of HC on t(15;25) in contrast with a small block on t(1;29), has been

158 P. Petit, H. De Bois, and W. De Meurichy

Fig. 5. a: The seven bi-armed okapı chromosomes after C- and G-banding. Arrows indicate the
breakpoints in the composite t(27;2;23). b: Acrocentric autosomes 3, 6, 11, and 28 after consecutive
G-banding (left) and NOR staining dots (arrows). c: Representative okapi sex chromosomes obtained
by G-R-C-banding from left to right

reported ın a 59,XX Portuguese Barrosa cow (IANNUZzzi1 et al. 1992). These findings imply
that translocated chromosomes, ıinvolving fusion of different bovid acrocentric autosomes,
may ınitially contain large blocks of HC which are reduced ın sıze with time. Thus, recent
or “new” Robertsonian translocations should progressively loose blocks of HC in the
biarmed chromosomes during further evolution.

With regard to the Robertsonian fusions, the presence of the 2523 translocation in the
rıver buffalo (2n = 50) and in the anoa (2n = 46) was also demonstated but without
additional translocated chromosome 27 at the telomeric region of chromosome 2 (Ian-
NuUzZI et al. 1990). As a consequence, the resolution of the composite t(27;2;23) charac-
terizes the okapı karyotype. Comparison of the okapı karyotype with a recent cytogenetic
survey of 12 bovid species, has indicated considerable monobrachial G-band homologies,
but few biarmed chromosome homologies (GALLAGHER and WOoMACcK 1992). Interestingly
we have found that the okapı not only shares t(14;25) and t(17;18) with the Roosevelt
gazelle (2n = 30) but has the t(20;22) in common with the topi (2n = 36). These findings
emphasize that certain biarmed homologies are ıdentical among the translocation events
which have arısen independently in three Artiodactyla groups, i.e. the Giraffidae, the
Antilopinae and the Alcelaphinae (BuckLanp and Evans 1978a; GALLAGHER and
WOoMAck 1992). In contrast, no homologous biarmed chromosomes have been observed in
a comparative study between the okapi and the giraffe showing 2n = 30 with NF = 58
(PErrT and MEURICHY 1992). However, the metacentric morphology of the X chromo-

Chromosomal reduction in an Okapı pedigree 159

some in these related anımals is strikingly different from submetacentric X chromosomes
in cattle (Perrtt and MeurıcHyv 1992; Tannuzzı 1990). The large okapı Y chromosome is
similar in size to those of the rıver buffalo and the anoa (lannuzzi et al. 1990). From
previous studies, heterochromatisation of the okapı long arms is evident after C-banding
(Perıt and MeurıcHy 1992). According to MATTHEwS and REED (1991) the large size of
the Y chromosome could be considered as resulting from a class of repeated DNA
sequences that is represented in the male bovine genome. This could be related to the
phenomenon of acquisition of large amounts of HC material, leading to the morphology of
the okapı Y chromosome.

Usually, NORSs are observed in telomeric positions in the Bovidae (Mayr et al. 1985;
Meo et al. 1991). In contrast, we have identified four acrocentric autosome pairs numbered
3,6, 11 and 28 demonstrating NORSs close to the centromeres. Although pericentromeric
location of these NORs have been observed by us in several okapis, it remains unclear why
the location of NOR in the okapı is centromeric, while in the Bovidae no other examples
have yet been described.

Acknowledgements

We wish to thank Prof. K. BENIRSCHkE and Dr. L. Iannuzz1 for encouragement and Prof. F. Van
LEUVEN for reading the manuscript.

Zusammenfassung

Chromosomenreduktion in einer Okapi-Familie

Der Karyotyp eines weiblichen Okapıi, das 2n = 44 aufwies, wurde mittels der G-R-C und Ag-NOR-
Bänderung untersucht. Dieses Tier ist das Resultat einer Paarung zwischen einem aus Tierhaltung
stammenden Weibchen und einem in Freiheit aufgewachsenen Männchen, welche beide heterozygot
sind und 2n = 45 mit einer zentrischen Fusion zwischen zwei ungleich großen akrozentrischen
Chromosomen zeigen. Der Okapı-Karyotyp wurde anhand eines Rinderstandards unter Berücksich-
tigung des hohen Grades an Bandenhomologie zwischen den beiden Spezies angefertigt. Die Reduk-
tion von 2n = 46 auf 2n = 45 und 2n = 44 ist das Resultat einer Robertson’schen Translokation, die das
Rinderäquivalent von Chromosom 4 und 26 einbezieht. Abgesehen von den autosomalen Verände-
rungen wurden auch Tandemtranslokationen und strukturelle Anderungen bei X- und Y-Chromoso-
men identifiziert. Die Anordnung von zentromerischen Heterochromatin und Nucleolus organisie-
renden Regionen im Okapı-Karyotyp werden unter Bezugnahme auf den Karyotyp anderer Arten der
Bovidae beschrieben und diskutiert.

References

ISCNDA (1989): International system for cytogenetic nomenclature of domestic animals. Cytogenet.
Cell Genet. 53, 65-79.

BENIRSCHKE, K.; KuMAMOTO, A. T.; Cousin, E. F. H. M.; Boer, L. E. M. pe (1983): Further
observations on the chromosomes of the okapi, (Okapia johnstoni). Verhber. 25. Intern. Symp.
Erkrankungen der Zootiere, Wien. Rp. 363-372.

Boıs, H. DE; PuUIJENBROECK, B. Van; DHONDT, A. A. (1988): The Studbookpopulation of the okapi -
Okapia johnstoni - some remarks on the current demographic and population genetic status. Acta
Zool. Path. Antv. 80, 53-64.

BuckLAnND, R. A.; Evans, H. J. (1978a): Cytogenetic aspect of phylogeny in the Bovidae I. G-
banding. Cytogenet. Cell Genet. 21, 42-63.

— — (1978b): Cytogenetic aspects of phylogeny i in the Bovidae II. C-banding. Cytogenet. Cell. 21,
64-71.

Foss£, T. J. (1978): Demographic and genetic models and management for the okapı (Okapia
johnstoni) ın captivity. Acta Zool. Path. Antv. 73, 119-195.

GALLAGHER, JR., D. S.; WOMAcK, J. E. (1992): Chromosome conservation in the Bovidae. J. Hered.
83, 287-298.

GijzEn, A. (1959): Der Platz des Okapıs im System. Das Okapı Okapia johnstoni (Scalter).
Wittenberg Lutherstadt: A. Ziemsen Verlag: Pp. 40-43.

GOODPASTURE, C.; BLOoMm, $. C. (1975): Visualisation of nucleolus organizer regions in mammalian
chromosomes using silver staining. Chromosoma 53, 37-50.

Hösıı, P.; Lang, E. M. (1970): A preliminary note on the chromosomes of the giraffidae: Giraffa
camelopardalıs and Okapia johnstoni. Mammal. Chromos. Newsl. 11, 109-110.

160 P. Petit, H. De Bois, and W. De Meurichy

Iannuzzı, L. (1990): An improved characterization of cattle chromosomes by means of high-
resolution G- and R-band comparison. J. Hered. 81, 80-83.

IANNUZZI, L.; RANGEL-FIGUEIREDO, T.; MEO, G. P. Di; FERRARA, L. (1992): A new Robertsonian
translocation in cattle, rob (15-25). Cytogenet. Cell Genet. 59, 290-293.

MATTHEws, M. E.; REED, K. C. (1991): A DNA sequence that is present in both sexes of Artiodactyla
is repeated on the Y chromosome of cattle, sheep, and goats. Cytogenet. Cell Genet. 56, 40-44.

MAYR, B.; MENDELAK, M.; KRUTZLER, K.; SCHLEGER, W.; KALAT, M.; AUER, H. (1985): LEVELS OF
CONSERVATION AND VARIATION OF HETEROCHROMATIN AND NUCLEOLUS ORGANIZERS IN THE
BOVIDAE. Can. J. GENET. CyYToL. 27, 665-682.

Meo, G. P. Di; IANNUZZI, L.; FERRARA, L.; RugIno, R. (1991): Identification of nucleolus organizer
chromosomes in goat (Capra hircus). Caryologia 44, 309-316.

Pa1, G. S.; THomas, G. H. (1980): A new R-banding technique ın clinical cytogenetics. Hum. Genet.
54, 41-45.

Pertr, P.; MEuRICHY, W. DE (1986): On the chromosomes of the okapi (Okapia johnston:). Ann.
Genet. 29, 232-234.

— — (1992): Comparative cytogenetic studies in the family Giraffidae. Proc. 10th European Coll.
cytogenetics of domestic anımals. Utrecht: Utrecht University (in press).

PUIJENBROECK, B. Van; Boıs, H. De (1991): Studbook of the Okapıi, Okapia johnstoni (Scalter).
Antwerp: Royal Zool. Soc. Antwerpen.

RoMER, A. S.; Parsons, T. S. (1986): Who’s who among the vertebrates. The Vertebrate Body.
Philadelphia, London, Toronto: W. B. Saunders Comp.

SUMNER, A. T. (1972): A simple technique for demonstrating centromeric heterochromatin. Exp. Cell
Res. 75, 304-306.

TAyLor, K. M.; HUNGERFORD, D. A.; SNYDER, R. L. (1969): Artiodactyl mammals: the chromosome
cytology in relation to patterns of evolution. In: Comparative mammalıan cytogenetics. Ed. by K.
BENIRSCHRE. Berlin, Heidelberg, New York: Springer Verlag. Pp. 346-356.

ULBRICH, F.; SCHMITT, J. (1969): Die Chromosomen von Okapia johnstoni (Scalter, 1901). Acta Zool.
Pathol. Antv. 49, 123-124.

Yunıs, J. J.; SAwvER, ]J. R.; Barı, D. W. (1978): The characterization of high-resolution G-banded
chromosomes of man. Chromosoma 67, 293-307.

Authors’ addresses: P. Perıt, M. D., Centre for Human Genetics, U.Z., Gasthuisberg, Herestraat
49, B-3000 Leuven, Belgium; H. De Boıs and W. DE Meurichy, Royal
Zoological Society of Antwerp, Koningin Astridplein 26, B-2018 Antwerpen,
Belgium

Z. Säugetierkunde 59 (1994) 161-168
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Chromosome polymorphism in Sorex alpinus (Mammalıa,
Soricidae) in the western Alps and the Swiss Jura

By E. DANNELID

Institut de Zoologie et d’Ecologie Animale, Universite de Lausanne, Lausanne, Switzerland

Receipt of Ms. 19. 7. 1993
Acceptance of Ms. 29. 12. 1993

Abstract

The karyotype (chromosome morphology and G-banding pattern) was examined ın 15 individuals of
Sorex alpınus, the alpine shrew, collected in southwest Switzerland and neighbouring France. All
three recorded chromosome numbers, 2N = 58, 56 and 54, occur within Switzerland. It was found
that this difference in chromosome number is due to Robertsonian polymorphism. Polymorphism in
the number of acrocentric versus subtelocentric chromosomes may also occur.

Introduction

The shrew Sorex alpıinus ıs distributed ın the mountainous areas of Central and Southern
Europe: ı.e. the Alps, the Carpathians, the mountains of the western Balkan Peninsula, the
Swiss Jura, and some mountainous areas further north in Germany (Schwarzwald, Harz,
Fichtelgebirge etc.). There ıs also one doubtful record for the Pyrenees. The records of
distribution are taken from SPITZENBERGER (1990). Morphologically, Sorex alpınus seems
to be very distinct from other Sorex species and it occupies a rather isolated position within
the genus, even though comparisons between S. alpinus and the East Asıan species S.
mirabılis have been made. These similarıties were mostly due to the shape of the glans penis
(HUTTERER 1982). Allozyme data also sets $. alpinus clearly apart from other Sorex species
(CATZEFLIS et al. 1982; CATZEFLIS 1984).

The karyotype of Sorex alpınus ıs not well known, REUMER and MEyLANn (1986) give 2N
= 54-56 (CATZEFLIS et al. 1982), 56 (Zıma and KrAr 1984) and 58 (Meyran 1964, 1966).
More recently, Zıma and KrAr (1990) again recorded 2N = 56 for the species. This ıs a
higher chromosome number than in any other Eurasıan Sorex (except for the members of
the Sorex cinereus-group in extreme northeast Siberia). Also most chromosomes seem to be
acrocentric although no definite statement of the NF-number has been made. This
karyotype differs greatly from others known in Eurasian shrews. To the author’s know-
ledge no banding of any kind has ever been published on the chromosomes of this species.
In the present study a description of the chromosome set of S. alpinus (including G-
banding) is given.

Material and methods

The material used consisted of 15 shrews (13 from the Swiss and the French Alps, 2 from the Swiss
Jura), 6 males and 9 females, trapped between November 1980 and December 1992. For details
regarding sites and dates, see figure 1 and table 1. The chromosomes were prepared by the direct
method from spleen and bone marrow. Colchicine was applied for a period varying from 45 to 60
minutes. The preparations were left for 3 days, after which they were digested with trypsin, using a
modified form of the method described by SEABRIGHT (1971). G-banding was performed on 11 of the
individuals. The terminology for the chromosomes follows LEvan et al. (1964), thus acrocentric
chromosomes are termed t1, t2 etc.

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5903-0161 $ 02.50/0

162 E. Danneld
Results

All three reported diploid numbers of chromosomes, 2N = 58, 56 and 54, respectively,
were found among the investigated animals. Four anımals (all females) had a diploid
number of 58 chromosomes. They were trapped in the Swiss Alps, east of Lake Geneva
and north of the Rhone Valley, in the cantons Vaud, Valais and Bern (Fig.1). Three
animals (one male and two females) had a diploıd number of 56 chromosomes. Two of
these anımals were trapped in the Swiss Jura, the third ın the Swiss Alps southeast of Lake
Geneva, in canton Valais, south of the Rhone Valley (Fig. 1). Eight anımals (five males and
three females) had a diploıid number of 54 chromosomes. These anımals were trapped in
the Swiss and French Alps, south of Lake Geneva (Fig. 1).

Description of the chromosome formula

The 58-karyotype: There was only one pair of large metacentric chromosomes, two pair of
medium-sized subtelocentric chromosomes, one pair of small biarmed chromosomes and
24 pairs of acrocentric chromosomes The X-chromosome was a large submetacentric.
Since all investigated anımals were females, the Y-chromosome was not known. One
individual had one acrocentric pair less, but two pairs of small biarmed chromosomes
instead of one paır.

The 56-karyotype: This karyotype differed from the 58-type in that there was one
additional pair of large metacentric chromosomes and only 22 pairs of acrocentric
chromosomes. The Y-chromosome was possibly acrocentric.

The 54-karyotype: This karyotype differed in that there were three pairs of large

m 2N=54 % 2N =56 @ 2N=58

Fig. 1. Location of trapping sites for the animals used in the study

Chromosome polymorphism in Sorex alpınus 163

Table 1. Data of the animals used

VASEN Date of Locality
number capture

989 female 56 . 11. 1980 Bassıns, Vaud

999 female 56 oo Le Vaud, Vaud
3051 female 54 . 10. 1987 Les Houches, France
3053 male 54 EZ Les Allamands, France
3300 female 58 . 08. 1988 Bärfel, Oberwald, Valais
3421 male 54 . 12. 1988 Les Houches, France
3422 female 58 . 12. 1988 Innertkirchen, Haslital, Bern
3780 male 54 . 11. 1989 Les Houches, France
4330 male 54 9911991 Champery, Val d’Illiez, Valais
4331 female 54 291991 Champery, Val d’Illiez, Valais
4705 female 58 . 09. 1992 Innertkirchen, Haslital, Bern
4709 male 56 . 09. 1992 Mase, Val d’Herens, Valais
4710 female 58 . 10. 1992 Pont-de-Nant, Vaud
4713 male 54 1021992 Champery, Val d’Illiez, Valais
4732 female 54 ‚12,10% Les Houches, France

Note: the numbers given for the two first anımals are not IZEA-numbers, they refer to a
collection made by FRAnco1Is CATZEFLIS, and storend at the Institut de Zoologie et d’Ecologie
Anımale in Lausanne.

biarmed chromosomes (two metacentric and one submetacentric) and only 20 pairs of
acrocentric chromosomes. The Y-chromosome was small and probably biarmed.

The G-banding pattern

The G-banded pattern of the chromosomes is shown in figures 2 and 3. The G-banding
showed that of the two large metacentric pairs, the one here called m2 was present in all
anımals. The pair ml occurred ın 54- and 56-anımals and was formed by a fusion between
the acrocentric chromosomes called t9 and til in 58-anımals. Finally, the large sub-
metacentric pair (sm1) occurring only in 54-anımals was formed by a fusion between the
acrocentric pairs called t4 and t20 in other animals. The relative length of the different
chromosomes is given in table 2.

Discussion
Robertsonian polymorphism

The different diploid numbers of chromosomes in Sorex alpinus, 58, 56 and 54, respec-
tıvely, were invarıably connected with different numbers of large biarmed autosomes (note
that the X was also large and biarmed), one pair in the 58-type, two pairs in the 56-type
and three pairs in the 54-type. It may be noted that Zıma and KrAr (1990) found 56
chromosomes and that “the female karyotype contained three pairs of large biarmed
chromosomes“, of which one was obviously the X-chromosome, in Sorex alpinus from
Czechoslovakia. Thus, it seems highly plausible that the difference in chromosome
number within the species is due to Robertsonian translocations.

G-banding analysis confirmed this. Of the three large biarmed pairs of autosomes the
pair m2 was present in all anımals. The pair mi was present in 56- and 54-animals and
formed by the acrocentric pairs t9 and t11 in 58-animals. The pair sm1 was present only in
54-anımals and was formed by the acrocentric pairs t4 and t20 in the other animals.

164 E. Danneld

id Am... { } i i Ta
st2 st3 ti =»
van An dar gen
t3 t4 t5 t6
sı0 va gE' we
en t8 t9 t10
as. gan Ai 388
t11 t12 t13 +14
8 26% MR 4er
t15 t16 t17 t18
ser Pen Ah si
t19 t20 t>1 t20
ss a «- Aa» ah i : 1 a 2:5
t23 t24 x Y

Fig. 2. G-banded chromosomes of Sorex alpinus. Three different specimens of each chromosome are
shown. For the acrocentric chromosomes involved in Robertsonian fusion, the fusion is also shown

Chromosome polymorphism in Sorex alpinus 165

m] st3
tl t6
17

.

t13 t18
B
t19 t24

3A

Fig. 3. G-banding pattern of the chromosomes of Sorex alpinus

It is generally agreed that ın mammals, centric fusion is amuch more common process
than centric fission. This idea should lead to the hypothesis that the 58-karyotype ıs the
most primitive character state, and that a fusion between pair t9 and pair t11 gave rise to the
56-karyotype. Another fusion, this time between pair t4 and pair t20 led to the 54-
karyotype.

The material examined in this study ıs probably too small in number to determine
whether the varıation in chromosomal number is geographically correlated (as seems
probable). However, at least the 54-karyotype seems to have a specific geographical
distribution. It was the most frequently encountered, eight anımals, all captured in an area
south of Lake Geneva (chiefly Les Houches in French Savoıa and Val d’Illiez in canton
Valais) where all anımals had the 54-karyotype.

The 58-karyotype was encountered ın anımal east of Lake Geneva, from Haslıtal ın

166 E. Dannelid

Table 2. Length of chromosome arms canton Bern in the north to Pont-de-Nant
(inmean percentageoflengthoffemalehaploidset) ın canton Vaud in the south- and east to
Bärfel, Oberwald in canton Valais. These
Shorter Longer Total localities were all situated north of the

m Em) Rhone Valley. The sample consisted, how-
ever, of only four individuals.

The 56-karyotype, finally, presents
some problems. Of the three individuals,
two were captured in the Swiss Jura, north
of Lake Geneva, while the third was from
Val d’Herens in canton Valais, which is
south of the area with 58-anımals. Finally,
it exists ın Czechoslovakıa (ZımA and KrAL
1990). It may be noted that the material
analysed by Zıma and KrAr (1990) origi-
nated from two geographically isolated dis-
tribution areas, in the Carpathians in
Slovakia and in the Jeseniky Mountains in
Czechia.

Previous results from MEYLAN (pers.
comm.) seem to indicate that neither the
distribution of the 54-karyotype nor that of
the 58-karyotype is geographically
homogeneous.

However, the data in this study suggest
that the distribution of the three different
karyotypes ıs geographically correlated,
and assuming this to be true, two alterna-
tive hypotheses may be formed.

The first hypothesis ıs that the 58-kary-
otype is the most ancient and vıa centric
fusions has given rise to the 56- and 54-
karyotypes, as suggested above. In this case
the original 58-karyotype has been replaced
by the 56-karyotype south of the Rhone
and further westward (and northeastward).
The 54-karyotype then arose from the 56-
karyotype in the area south of Lake Geneva, thus separating the 56-populations of the Jura
from those of the Alps.

The second hypothesis takes into account the large and nonhomogeneous distribution
of the 56-karyotype. However, genetically ıt seems to be constant. By studyıng banded
material from western Czechia (Sumava mountains) it was possible to confırm that the
fusion of t9 and t11 into ml is the same there as in the Jura and in Valais south of the Rhone
Valley. This hypothesis should thus be that the 56-karyotype is the most ancient, and has
given rise to the 54-karyotype (vıa centric fusion) and to the 58-karyotype (via centric
fission).

It would be interesting to compare these karyological data with differences in morphol-
ogy and electrophoresis from the different areas. Unfortunately, very few publications
have been concerned with Sorex alpınus. CATZEFLIS et al. (1982) found electrophoretic
polymorphism in this species (6 locı out of 17 studied); all their individuals were, however,
from one localıty (Pont-de-Nant in canton Vaud). Moreover, CATZEFLIS (1984) again
reported polymorphism in Sorex alpinus; anımals from Pont-de-Nant (which should have

Chromosome polymorphism in Sorex alpınus 167

the 58-karyotype) had 5 polymorph locı (out of 35 studied), while in anımals from the
Swiss Jura (which should have the 56-karyotype) only one polymorph locus (not poly-
morph in Pont-de-Nant) was found.

Other polymorphies

Zıma and KrAr (1990) recognized two pairs of subtelocentric chromosomes ın this species.
Actually, there are probably more than two pairs of subtelocentrics (excluding the tiny
st3). In some Giemsa-stained preparations up to six could be distinguished. In some
metaphases almost all autosomes appeared to be biarmed, and Zıma recorded the same
condition for the Czechoslovakian material (ZımaA, pers. comm.). However, only pairs stl
and st2 (in size order corresponding closely to the two pairs described by Zıma and KrAL
(1990) and thus probably identical with these) were easily recognized, moreover, these two
pairs were the only ones possible to identify as subtelocentrics also in the G-banding. The
other chromosomes sometimes appearing as subtelocentrics quite as often were acrocen-
tric. This non-constant condition might be due to centromeric shift, or to different
interpretation because the smaller arms are sometimes exceedingly difficult to see, espe-
cially if the chromosome is in a contracted state. A possible third subtelocentric pair might
be the pair here termed t8.

To ascertain whether the distribution of the 58-, the 56-, and the 54-karyotypes is
geographically correlated and in that case to determine the distributions greater detail,
further studies on the karyology on this species are needed, also from other parts of its
distributional area.

Acknowledgements

The author wishes to express his gratitude to AnnE-MARIE MEHMETI for help with the preparation of
the chromosomes. I am also grateful to Prof. PETER VOGEL, HARALD BRÜNNER and LucA FUMAGALLI
for bringing me shrews when my own trapping success was declining. I am indebted to Prof. JAcQuzs
HaAusser, Lausanne, Prof. EDwARD FARMER, Lausanne, Prof. Bo FERNHOLM, Stockholm, Prof.
KARL FREDGA, Uppsala and Dr. Jan Zıma, Brno for critical comments on the manuscript. Finally, I
wish to thank the Swedish Institute, Wallenbergsstiftelsens jubileumsfond, A. F. Regnells Zoologiska
gavomedel, Helge Ax:son Johnsons stiftelse, Hierta-Retzius fond för vetenskapling forskning and
Stiftelsen Wenner-Gren Center for the financial supply that made this work possible.

Zusammenfassung

Chromosomenpolymorphismus von Sorex alpinus (Mammalıa, Soricidae) in den Westalpen und im
Schweizer Jura

Der Karyotyp von Sorex alpinus wurde für 15 Tiere anhand der Chromosomenmorphologie und des
G-Bandenmusters analysiert. Alle drei bisher bekannten Chromosomenzahlen 2n = 58, 56 und 54
wurden für die Schweiz nachgewiesen. Ebenso konnte gezeigt werden, daß der Unterschied in der
Chromosomenzahl durch einen Robertsonischen Polymorphismus bedingt wird.

References

CATZEFLIS, F. (1984): Systematique biochimique, taxonomie et phylogenie des musaraignes d’Europe
(Soricıdae, Mammalıa). Thesis, Univ. Lausanne.

CATZEFLIS, F.; GRAF, J. D.; HAusser, J.; VoGEL, P. (1982): Comparison biochimique des Musaraig-
nes du genre Sorex en Europe occidentale (Soricidae, Mammalia). Z. zool. Syst. Evolut.-forsch.
20, 223-233.

HUTTERER, R. (1982): Biologische und morphologische Beobachtungen an Alpenspitzmäusen (Sorex
alpinus). Bonner Zool. Beitr. 33, 3-18.

LEvan, A.; FREDGA, K.; SANDBERG, A. (1964): Nomenclature for centromeric position on
chromosomes. Hereditas 52, 201-220.

Meyran, A. (1964): Le polymorphisme chromosomique de Sorex araneus L. (Mamm.-Insectivora).
Revue suisse Zool. 71, 903-983.

168 E. Danneld

Meyran, A. (1966): Donnees nouvelles sur les chromosomes des Insectivores europeens (Mamm.).
Revue suisse Zool. 73, 548-558.

REUMER, J. W. F.; MEyrLan, A. (1986): New developments ın vertebrate cytotaxonomy IX Chromo-
some numbers in the order Insectivora (Mammalia). Genetica 70, 119-151.

SEABRIGHT, M. (1971): A rapid banding technique for human chromosomes. Lancet 2, 971-972.

SPITZENBERGER, F. (1990): Sorex alpinus Schinz, 1837 - Alpenspitzmaus. In: Handbuch der Säugetiere
Europas 3:1. Ed. by J. NIETHAMMER and F. Krapp. Wiesbaden, Aula-Verlag. Pp. 295-312.

Zıma, J.; KrAı, B. (1984): Karyotypes of European mammals I. Acta Sc. Nat. Brno 18, 1-51.

Zıma, J.; KrAr, B. (1990): The karyotype of the Alpine shrew (Sorex alpinus). Biologia (Bratislava).
45, 465-469. (In Czech, Engl. summary).

Author’s address: ERLAND DANNELID, Department of Zoology, Stockholm University, S-106 91
Stockholm, Sweden

Z. Säugetierkunde 59 (1994) 169-173
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Placental scar counts in the Red fox (Vulpes vulpes L.) revisited
By E. R. Linpström

Grimsö Wildlife Research Station, Department of Wildlife Ecology, Swedish University
of Agricultural Sciences, Sweden

Receipt of Ms. 8.7. 1993
Acceptance of Ms. 28. 2. 1994

Abstract

Placental scars used for analyses of reproduction in the red fox (Vulpes vulpes L.) may fade throughout
the season after birth. I report a method to calculate whelping frequency and mean litter sıze from
placental scar analyses of 608 vixens shot during autumn and winter in three areas of Sweden. The
material allowed 6-9 consecutive periods of calculation within the year in each area. By successively
including scars of lighter shades in the counts, I established isopleths of constant whelping frequen-
cies. The isopleth yielding the best estimate of mean litter size as derived from analyses of vixens ın late
pregnancy was accepted for calculations in each area. Although this method appeared to be adequate
for each area, there were inter area differences. Therefore, reference material is needed for each
specific area before this method can be used.

Introduction

Placental scars provide convenient measures of litter size and whelping frequency (pro-
portion of females that give birth) in the red fox (Vulpes vulpes L.), and the method has
been applied extensively (e.g. SHELDON 1949; RıcHarps and HıneE 1953; LAynE and
McKeon 1956; McIntosH 1963; WANDELER 1968; FAIRLEY 1970; EnGLunD 1970;
WANDELER et al. 1974; Ryan 1976; STORM et al. 1976; ULsrıcH 1977; Pırs and MARTIN
1978; Kos and Hrwson 1980; Artoıs et al. 1982; ALLEN 1983; HarRrıs and SMITH 1987;
LinpströMm 1988, 1989, 1992; AnsoRGE 1990). However, different shades of scars may
originate at resorption sites, be persisting from old pregnancies or from cubs born the
previous spring (EnGLunD 1970). Hence, EncLunD (1970) based hıs calculations on the
darkest scars only. LinpströMm (1981) showed that the shade of a scar may fade throughout
the year following birth. Accordingly, I modified the method to account also for the
fadıng, and in such a way that the scars of 4 vixens with known minimum litter sizes were
correctly interpreted (LINDSTRÖM 1981). Here, I report a new attempt to determine which
scars should be counted at different intervals after birth, including new material and with a
new approach.

Material and methods

Placental scars were counted in adult (=1 year) vixens that showed no sıgn of fresh ovulation (an
indication of a new pregnancy). The carcasses were collected from hunters ın three parts of Sweden:
Västerbotten (approximately 65°N, 20°E, n = 203, 1982-1990), Bergslagen (approximately 60°N,
15°E, n = 284, 1975-1990), and Smäland (approximately 57°N, 15°E, n = 121, 1983-1990). The
hunting season, ı.e. the season of collection, lasted from early August to mid-March/mid-April
(depending on area). Adult age was indicated by epiphyseal closure of the tibia. Placental scars were
classified by shade of darkness from 1 (hardly visible) to 6 (completely black).

An additional 30 pregnant vixens with fetuses of a crown-rump length >12 mm (i.e. at least
halfway through pregnancy, see e.g. EnGLUND 1970) were used for calculating mean litter size. For
comparison of mean litter sızes, I used 95 % confidence intervals.

The exact date of death of the fox was not known in all cases. Hence, the week numbers were used

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5903-0169 $ 02.50/0

170 E. R. Lindström

as the time base; a year encompasses weeks 1-52 except every 5th or 6th year, which also includes
week no.53.

Mean litter sıze and whelping frequency were calculated for scars of shade 6, shade 6+5, shade
6+5+4, etc. until scars of all shades were included. This was done for consecutive periods throughout
the season of collection in each area. At least 20 individuals were desired for each period, but
especially ın early autumn and late winter, this number was not possible to attain. The shades that had
to be included during the different periods to obtain isopleths of constant whelping frequencies (40,
50, 60 and 70 %) throughout the season were noted, and the resulting mean litter sıze was calculated
for each isopleth. This was then compared with the mean litter size as calculated from pregnant vixens
and the isopleth giving the best fit was chosen for each area. To make a final adjustment I also checked
the chosen isopleths for the best estimates of the mean litter size during each consecutive period.

Results and discussion
Mean litter size in late pregnancy

Mean embryonic litter size did not differ statistically among areas (means + 95%
confidence intervals: Västerbotten 5.7 # 1.2, n = 9; Bergslagen 5.1 + 0.89, n = 15; and
Smäland 4.8 # 1.6, n = 6). Thus, I pooled the materials and calculated an overall average of
5.2 # 0.58 cubs. The confidence interval thus obtained covered the average of 4.8 cubs per
litter (n = 489) noted by EnsLunD (1970) in his total material of pregnant vixens from four
areas of Sweden.

In five comparisons, embryo counts overestimated mean litter size as calculated from
cubs at dens by an average of 18.5% [range 0-63%; data from McIntosH (1963);
WANDELER etal. (1974); STORM et al. (1976); PITZsCHkE (1972) combined with STUBBE and
StusBE (1977); Pırs and Marrın (1978)]. Appling this figure to EnGLunn’s (1970) data
would yield 4.1 cubs at dens. For different reasons, embryo counts are likely to
overestimate the number born, whereas counts of cubs at dens are subject to the opposite. I
have used a tentative estimate of 4.5 cubs born per litter in all three areas.

Selecting isopleths

The results did not vary much when scars of all shades were included in the calculations
(Tab.1). However, the darkest scars were primarıly found early in the season. Thus,
although dark scars faded, no scars seemed to disappear completely within one season.

Isopleths of constant whelping frequencies could be established by successively includ-
ing scars ot lighter shades (Tab. 1), and the corresponding mean litter sizes were calculated
(Tab. 2). In Västerbotten the best estimate of the mean 4.5 cubs born was attained by the
60 % ısopleth, whereas the 70 % isopleths yıelded the best fits in Bergslagen and Smäland.

Adjustments to obtain the best fit of mean litter size in single time periods yıelded the
finally accepted trajectories (series) of successive shades to be included in the counts:
(shades counted early ın autumn; subsequently included shade/week when first included)
Västerbotten 6-4; 3/46; 2/5, Bergslagen 6-4; 3/41; 2/52, Smäland 6-3; 2/1.

All previously counted scars were included in the new counts, but 22% of the vixens
with accepted scars had an increased number of scars as compared with the old method.
Half of these vixens were previously considered as having had no cubs. The number of
scars now counted in the uteri of the four vixens wıth known minimum (!) lıtter sızes
(LinDstRÖöM 1981) overestimated these by 1-3 cubs.

The series of shades to be successively considered in the scar counts presented above
should not be applied in any other areas. However, the method of calculating these
provides a possibility to analyze fading of the scars whenever a material large enough is
available, and also to correct for fading if an independent estimate of mean litter size can be
obtained. There is no limitation concerning species as long as the method of placental scar

Al

Placental scar counts in the Red fox (Vulpes vulpes L.) revisited

c6/lr

SL/L'v

BE /L’o

8 /0°E

127

02/9’

OL/L'E

0E/0’E

ol

07

Or

Le

er ei Iıı 08.16

06/3°%

08B/EV

08/0 E

ol

LLIE'S | 68/5

Liv |S8/L'v

0S/72 | EIU/r'E

ez/s'l | LEIL

sl/eis | SWRl

0/7 r 01

9 LG

08/0°9 | S8/S'S

02/85 | 8/09

geimw'e | Le/BEe

slie'a | 6L/0°E

8

0/7 v /0'9
Or Or

0g Le

12 19

395 EIEP y>e9 JO do au} Ye PO9YEIIPUT 9ZIS apdures "sıegq TEIIOA Aq p>3esrpur SPOLOT

6/17

98/17

er/ier

6l/B’e

Le

18/09

v8/lv

09/6

L2/6 |

8/03

S/st

BE

L6/5°S

8/25

ge/6E

AA?

v/0%

0%

es

8/7’

LLIS'v

SriL’e

ce/g'c

6 /0'°S

OR

4

€e8/0

LS/S'S

931’

Se

98/5°5

98/6°%

Ss/r's

OL/ZL

OR

(TR

[X4

6L/L'S

SLIL'S

t9/0’P

gerg’e

rLis’e

LIE

8°

9l/e'g |SL/L'9

gYE'S |EL/S'S

Lr/l’e | 89/8

63/92 | Br/L’e

st/oL |BZ/L'e

07 sl/io’e

Zt or

g8/7'S

08/1’

Bricht

0e/L’E

01/0'S

e 00€

or

16/89 | 96/0'5

GL/8'S | 96/9°p

r2IS’e |64/1’v

Be/g’o | vS/L’E

6l/s’e | 12/9'S

07 el/0's

LS 124

08/E'S

89/19

09 L'v

9g/E'd

sli0’c

r /01

ER

r9/1'9

v6/E's

95/8€

LE/B'S

98/19

c8/2'S

LLUIG'Y

6S/C4

0s/g’E

gl/o’e

[4

68/8'9

€8/7'9

85/1L'9

cvle's

92/09

8/0

sl

88/19

B8/E'S

GLIE'S

L2/6’P

ey/e'r

Be/l’e

124

yeoMm

9-1}

9:3

3

9

9-9

9 opeys

91

9:7

9 epeys

9

9-9

9 epeys

(5) puejews pue ‘q) us3egsdisg “(e) usgoqssIse‘ UI (EI-TE Jaquınu Y99M) UONJ9]J09 Jo uoseas 3yJ Inoy3noıyy pue

apeys 197431] AJ3urses1aur Jo sıeds [eyuasegd wos} payenajes se (Afpamaadsaı “yses Jo I1y3rı pue I75]) sspewaz IurdjsyMm Jussıad pue aZıs 19391] ueaW '/ 27991

172 E. R. Lindström

Table 2. Mean litter sizes as calculated from the isopleths of 40, 50, 60, and 70% whelping
frequencies in Västerbotten, Bergslagen, and Smäland

Whelping frequency Mean litter size

Västerbotten Bergslagen Smäland

counts ıs valıd. The only assumption needed is that the uteri collected during each
subperiod provide an unbiased sample from the population of females alıve previous

spring.
Acknowledgements

Special thanks go to YLvA LINDROOTH and CHRISTINA LINDSTRÖM for doing the necropsies and the
accompanying analyses. Critique and support were offered by Jon Swenson. He also corrected my
English language. CHRISTINA LINDSTRÖM, EGon NiLsson, JAN and BIRGIT ÄBERG organızed the
collection of carcasses. I am also ın debt to all hunters who provided the raw material of the study.
ANDREAS SEILER helped me with the Zusammenfassung. The study was an integrated part of the
environmental monitoring program (PMK) of the Swedish Environmental Protection Agency.

Zusammenfassung

Zählungen von Implantationsnarben beim Rotfuchs (Vulpes vulpes L.), eine Revision

Die Wurfgröße und Wurfhäufigkeit von Säugern mit einer Placenta zonarıa können über Implanta-
tionsnarben im Endometrium ermittelt werden. Beim Rotfuchs (Vulpes vulpes L.) verblassen die
Narben jedoch nach erfolgter Geburt. Hier präsentiere ıch eine revidierte Methode zur Feststellung,
welche Färbungsgrade berücksichtigt werden sollten, um eine korrekte Schätzung der Reproduktion
zu verschiedenen Zeitpunkten zu erhalten. Diese Methode basiert auf sukzessiven Berechnungen von
Narbenanzahlen unterschiedlicher Intensität. Die Narben, welche die besten Übereinstimmungen mit
den beobachteten Durchschnittswurfgrößen ergaben, wurden für jährliche Reproduktionsanalysen
bei schwedischen Füchsen verwendet.

Literature

Auen, $. H. (1983): Comparison of red fox litter sizes determined from counts of embryos and
placental scars. J. Wildl. Manage. 47, 860-862.

AnsoRGE, H. (1990): Daten zur Fortpflanzungsbiologie und Reproduktionsstrategie des Rotfuches,
Vulpes vulpes, in Oberlausitz. Säugetierkdl. Inf. 3, 185-199.

ARTO1s, M.; AUBERT, M. F. A.; G£RARD, Y. (1982): Reproduction du renard roux (Vxlpes vulpes) en
France. Rythme saisonnier et fecondite des femelles. Acta (Ecologica, (Ecol. Applic. 3, 205-216.

EnGLun, J. (1970): Some aspects of reproduction and mortality rates in Swedish foxes (Vulpes
vulpes) 1961-63 and 1966-69. Viltrevy (Swedish Wildlife) 8, 1-82.

FAIRLEY, J. S. (1970): The food, reproduction, form, growth and development of the fox Vulpes vulpes
(L.) in north-east Ireland. Proc. R. Irish Acad. 69 sec. B, 5, 103-137.

HarRıs, $.; SMITH, G. C. (1987): Demography of two urban fox (Vulpes vulpes) populations. J. Appl.
Ecol. 24, 75-86.

Koıs, H. H.; Hewson, R. (1980): A study of fox populations in Scotland from 1971 to 1976. J. Appl.
BEco177 19:

LAyne, J. N.; McKeon, W. H. (1956): Some aspects of red fox and grey fox reproduction in New
York. N. Y. Fısh and Game ]. 3, 44-74.

Linpström, E. (1981): Reliability of placental scar counts in the Red fox (Vulpes vulpes L.) with
special reference to fadıng of the scars. Mam. Rev. 11, 137-149.

— (1988): Reproductive effort in red foxes (Vulpes vulpes L.) and future supply of a fluctuating prey.
Oikos 52, 115-119.

— (1989): Food limitation and social regulation in a red fox (Vulpes vulpes L.) population. Holarctic
Ecol. 12, 70-79.

— (1992): Diet and demographics of the red fox (Vulpes vulpes) in relation to population density. —

Placental scar counts in the Red fox (Vulpes vulpes L.) revisited 173

The sarcoptic mange event ın Scandinavia. In: Wildlife 2001: Populations. Ed by D. R. McCur-
LOUGH and R. H. BARRETT, London, New York: Elsevier Applied Science. Pp. 922-931.

McInTosH, D.L. (1963): Reproduction and growth of the fox in the Canberra distriet. C. $. I. R. ©.
Wildl. Res. 8, 123-141.

Pırs, C. M.; MarTın, M. A. (1978): Population dynamics, predator-prey relationships and manage-
ment of the red fox in Wisconsin. Wis. Dep. Nat. Res., Tech. Bull. 105, 1-56.

PITZscHkE, H. (1972): Untersuchungen über die Fuchspopulation - ein Beitrag zur Erforschung von
Grundlagen für eine wirksame Tollwutbekämpfung. Mh. Vet.-Med. 27, 926-932.

RıcHaRrDs, $. H.; Hıne, R. L. (1953): Wisconsin fox populations. Wis. Conserv. Dep. Tech. Bull. 6,
1-78.

Ryan, G. E. (1976): Observations on the reproduction and age structure of the fox, Vulpes vulpes L.,
in New South Wales. Aust. Wildl. Res. 3, 11-20.

SHELDON, W. G. (1949): Reproductive behaviour of foxes in New York state. J. Mammalogy 30,
236-246.

STORM, G. L.; ANDREwSs, R. D.; PhHıLLıps, R. L.; Bishop, R. A.; SINIFF, D. B.; TESTER, J. R. (1976):
Morphology, reproduction, dispersal and mortality of mid-western red fox populations. Wildl.
Monogr. 49, 1-82.

STUBBE, M.; STUBBE, W. (1977): Zur Populationsbiologie des Rotfuchses Vulpes vulpes (L.) -— I.
Hercynia N. F. (Leipzig) 14, 160-177.

ULBRICH, F. (1977): Weitere Angaben zur Fortpftlanzungsbiologie des Rotfuchses (Vulpes vulpes L.)
Beitr. Jagd- und Wildforsch. 10, 322-326.

WANDELER, A. (1968): Einige Daten über den Bernischen Fuchsbestand. Rev. Suisse de Zool. 75,
19711073.

WANDELER, A.; MÜLLER, J.; WACHENDÖRFER, G.; SCHALE, W.; FÖRSTER, U.; STECK, F. (1974):
Rabies in wild carnıvores ın central Europe. III Ecology and biology of the red fox in relation to
control operations. Zbl. Vet. Med. B 21, 765-773.

Author’saddress: Pu. D. Erık R. Linpstköm, Grimsö Wildlife Research Station, S-730 91
Riddarhyttan, Sweden

Z. Säugetierkunde 59 (1994) 174-180
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Zur Reproduktionsleistung des Alpensteinbockes (Caprai. ibexL.)
in der Freilandkolonie Albris (Graubünden, Schweiz)

Von M. GIACoMETTI und P. RaTTI

Forschungsinstitut für Wildtierkunde und Ökologie der Veterinärmedizinischen Universität Wien,
Österreich, und Jagd- und Fischereünspektorat des Kantons Graubünden, Schweiz

Eingang des Ms. 13.12.1993
Annahme des Ms. 22.4. 1994

Abstract

On the reproductive performance of the free-ranging alpine ibex population (Capra i. ibex L.) at Albris
(Grisons, Switzerland)

In order to evaluate the reproductive performance of alpine ibex (Capra ı. ıbex L.), the uteri and

ovaries of 80 females were investigated. The anımals came from the colony at Albris (Grisons,

Switzerland) and samples were collected during the months December to June in 1989-1990 and

19901991.

The females conceived between December 1 and January 21. The earliest age of females to conceive
was 2 years with a conception rate of 0.13. Conception rate in 3 year old females was 0.56 and of
4 year old females 1.00. Gravidity rate amounted to 0.88. For 70 females older than 2 years 48 viable
embryos could be determined (fecundity rate = 0.69). For the age class 4-13 years fecundity rate
equalled 0.83. At 14 years of age a probably age dependent decrease of fecundity was observed
(fecundity rate of females 14-16 years old = 0.40). Among 59 mature females 48 viable embryos were
found (natality 0.81) and fertility was 0.25. Twins were observed in only 1’ out of 52 cases investigated.
The reproductive performance detected in this study clearly surpasses the one previously described for
this population, which is approaching the carrying capacity of its habitat.

Einleitung

Eine wesentliche Voraussetzung zum Verständnis der Populationsdynamik einer Wildwie-
derkäuerart ist die Kenntnis der Fertilität, d. h. der ın einem Jahr erzeugten Nachkommen-
zahl einer Population (vgl. SCHWERDTFEGER 1978). Die Beurteilung der Fortpflanzungs-
leistung ist aber auch für die Bewirtschaftung freilebender Wildtiere, zum Beispiel von
Alpensteinbockpopulationen, von besonderer Bedeutung. In der Schweiz, im Fürstentum
Liechtenstein, in Österreich, in Deutschland und in Slowenien (GIACoMETTI 1991) sowie
in Südtirol werden zahlreiche Steinbockkolonien zur Regulierung der Bestände jagdlich
genutzt.

Die beim Alpensteinbock diesbezüglich vorliegenden Angaben beziehen sich aus-
schließlich auf die Beurteilung der Anzahl der im Anschluß an die Setzperiode zählbaren
Kitze (vgl. CoUTURIER 1962; NIEVERGELT 1966; Rartı 1981; PERACINO und BAssano
1990). Als Methode diente dabei die Direktbeobachtung. Bei dieser Methode wird aber die
peri- und postnatale Kitzmortalıtät, die hinsichtlich der Gesamtmortalität eine wesentliche
Rolle spielt, nicht berücksichtigt (vgl. HEPTNER et al. 1966; SCHRÖDER 1971; MITCHELL et
al. 1977; SALZMANN 1977; ELLENBERG 1978; HOFMANN 1990; BALLARD et al. 1991; KurT
1991). Es muß somit davon ausgegangen werden, daß die Fortpflanzungsleistung des
Alpensteinbockes ın freier Wildbahn größer ist als bisher angenommen.

Um diese Hypothese zu prüfen, wurden in der vorliegenden Arbeit die Ergebnisse einer
zweijährigen Untersuchung über die Fortpflanzungsleistung des Alpensteinbockes der
Freilandkolonie Albris im Kanton Graubünden (Schweiz) ausgewertet; Teilaspekte wur-
den an anderer Stelle publiziert (TATARUCH et al. 1991; Weıss et al. 1993).

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5903-0174 $ 02.50/0

Zur Reproduktionsleistung des Alpensteinbockes (Capra i. ibex L.) 175

Material und Methode

Von 80 Geifßen wurden Uteri und Ovarien untersucht. Die Geißen wurden in den Monaten Dezember
bis Juni der Jahre 1989-1990 und 1990-1991 von Wildhütern und Jagdaufsehern im Rahmen eines
Sonderabschusses erlegt. Die Tiere stammen von der im Südosten des Kantons Graubünden liegenden
Steinbockkolonie Albrıs. Der dortige Steinwildlebensraum liegt zwischen 1600 und 3200 m Seehöhe.
Mit einem gezählten Frühjahrsbestand von 1667 Tieren im Jahre 1990 ist die Kolonie Albrıs die
zahlenmäßig stärkste des Kantons. Im Wintereinstand beträgt die Dichte derzeit etwa 50, jene im
Sommereinstand etwa 15 Steinböcke pro 100 ha.

Unmittelbar nach dem Erlegen der Geißen wurden Ovarıen und Uteri in Polyethylensäcke
verpackt und bis zur Analyse bei -18°C aufbewahrt. Die Untersuchung erstreckte sich auf die
Anzahl, das Geschlecht, das Gewicht und die Entwicklungsfähigkeit der Keimlinge sowie die Anzahl
und den Maximaldurchmesser der Corpora lutea. Zur Zählung der Corpora lutea und zur Messung
ihres Maximaldurchmessers wurden die Ovarien im frisch aufgetauten Zustand in Scheiben von 0,5 bis
1 mm Dicke geschnitten. Dann wurde eine Scheibe nach der anderen aufgeklappt, hierbei die
vorhandenen Gelbkörper gezählt und deren Durchmesser auf einen Millimeter genau gemessen. Das
Keimlingsalter wurde nach Huscer und Wınpas (1951) berechnet. Dabei wurde eine Trächtigkeits-
dauer von 167 Tagen (STÜwE und GroDInskY 1987) und eine tierartspezifische Fetus-Wachstumskon-
stante a von 0,103 angenommen. Das nach RArTTı und HABERMEHL (1977) aufgrund von Merkmalen
am Gehörn bestimmte Alter der Geiffen wurde in ganzen Zahlen angegeben, wobei die Periode der
vorangegangenen Brunft, unabhängig vom Erlegungsdatum, als Stichzeitpunkt angenommen wurde.
Zur Feststellung, ob eine Trächtigkeit vorlag oder nicht, wurden nur jene Geißen berücksichtigt, die
nach dem 1. Februar erlegt wurden, und diejenigen, die nachweislich trächtig waren (Keimling
vorhanden). Der früheste Zeitpunkt war in diesem Zusammenhang der 11. Januar.

Ergebnisse
Gelbkörperrate und Gelbkörperdurchmesser

Die Gelbkörperrate (Anzahl der Corpora lutea periodica und graviditatis bezogen auf die
Anzahl reproduktiver Geißen) betrug 1,05 (n=55) (Geißen, die bereits gesetzt hatten oder
bei welchen eine pathologische Gravidität festgestellt wurde, blieben unberücksichtigt).
Bei trächtigen Steingeißen war ın den Monaten Februar bis Juni die Zahl der Corpora lutea
graviditatis mit jener der Keimlinge ıdentisch (n=34). Der Maximaldurchmesser der
Corpora lutea graviditatis beträgt durchschnittlich 14,8 mm (n=35, s= 2,2); er nımmt bei
fortschreitender Trächtigkeitsdauer an Größe zu (r?=0,655, p< 0.001).

Konzeptionszeitpunkt

Die Konzeption erfolgte in der Zeit zwischen dem 1.Dezember und dem 21. Januar
(n=37) (vgl. Abb.1); 86,5% der Geißen konzipierte zwischen dem 6. und dem
29. Dezember, somit innerhalb einer Zeitspanne von 23 Tagen. Der mittlere Zeitpunkt, an
welchem die Geifen konzipierten, war der 19. Dezember.

Geschlechtsreife und Konzeptionsraten

Die Geißen der Kolonie Albris konzipierten frühestens im Alter von 2 Jahren. Die
Konzeptionsrate der 2jährigen Geißen (Anzahl der entwicklungsfähigen und abgestorbe-
nen Keimlinge bei Geißen einer bestimmten Altersstufe oder -klasse bezogen auf die
Anzahl Geißen derselben Altersstufe oder -klasse) betrug 0,13 (n=8). Die Konzeptions-
rate der 3jährigen Geißen lag bei 0,56 (n=9), jene der 4jährigen bei 1,00 (n=9).

Trächtigkeitsrate

Von allen geschlechtsreifen Geißen (n=59) waren 52 trächtig, was eine Trächtigkeitsrate
von 0,88 ergibt.

M. Giacometti und P. Ratti

N
N
oN

6) (0%) > O1

—

Anzahl der Konzeptionen

12345678 910111213114 15 16 17 18 1920 2122232425 262728293031 12345678 gi 1213114 15 16 17 18 1920 21
Dezember Januar

Abb. 1. Konzeptionszeitpunkt bei den Steingeißen der Kolonie Albris (n = 37)

Pränatale Mortalität

In fünf von 52 Fällen wurde eine pathologische Gravidität festgestellt (2 Fälle im Resorb-
tionsstadium, 2 Frühaborte, 1 Mumifikation). Die pränatale Mortalität beträgt demnach
9,6%. Aufgrund ihres Verhaltens intra vitam und des Sektionsbefundes können die
Geißsen als gesund beurteilt werden. Bei einer GeifS wurde jedoch serologisch ein Titer von
1:100 gegen Leptospira bratislava (Geiß Nr. 579), bei der Geiß Nr. 593 ein Titer von 1:16

gegen Chlamydıa psittaci festgestellt. Mißgebildete

et d icht festgestellt.
Ausgewählte Fekunditätsraten beim GIEDHWALLLENANIE estzeste

Alpensteinbock der Kolonie Albris

Fekunditätsraten
Alterskl Fekunditäts- : : 5
m i Ber y Ausgewählte Fekunditätsraten (im Sinne von ASDELL
————————— nn 1946; ANDREWARTHA et al. 1954; NIEVERGELT 1966a)
0-1 sind aus der Tabelle und in Abb. 2 ersichtlich (bei den

Fekunditätsraten wırd, im Gegensatz zu den Kon-
zeptionsraten, nur die Anzahl entwicklungsfähiger
Keimlinge berücksichtigt). Die für die Berechnung
der Fertilität herangezogene und deshalb besonders
bedeutsame Fekunditätsrate der 2jährigen und älteren
Geißsen beträgt 0,69 (n=70).

Potentielle Natalität

Bei den 59 reproduktiven Geißen konnten insgesamt 48 sich bildende, entwicklungsfähige
Keimlinge nachgewiesen werden. Das ergibt eine potentielle Natalität (im Sinne von
SCHWERDTFEGER 1978) von 0,81. In Abb. 3 sind die Gelbkörperrate, die Trächtigkeitsrate
und die potentielle Natalität dargestellt. Die potentielle Natalität beträgt 77,1% der
Gelbkörperrate.

Fertilität

Als Fertilität wird nach SCHWERDTFEGER (1978) die Zahl der Nachkommen bezogen auf
1 Individuum einer Population (Tiere aller Geschlechts- und Altersklassen) verstanden.

Zur Reproduktionsleistung des Alpensteinbockes (Capra ı. ibex L.) 11777

Ausgehend von der oben angeführten Fe-

kunditätsrate der 2jährigen und älteren 0,9
Geißen (0,69) und von den ın der Kolonie 0,8
Albris im Frühjahr 1990 und 1991 gezähl- ._ 07
ten Steinböcken (3211 Tiere, davon 1185 > 06
Geißen von 3 und mehr Jahren) kann eine 2
Fertilität von 0,25 berechnet werden. zZ 0,5
or!
UL ’
Satzgröße und Geschlechterverhältnis 0,3
0,2
Einzig bei einer 5jährigen Geiß waren 01
Zwillinge nachzuweisen. Alle anderen
Geifßen waren unipar, die Häufigkeit der 01°» SIE EEE

Zwillingsträchtigkeiten betrug 1,9%.
Das Geschlechterverhältnis (Weibchen zu
Männchen) der ungeborenen Kitze betrug Abb. 2. Fekunditätsraten von 5 ausgewählten Al-
1:1,21 (n=31). tersklassen

Altersklasse [Jahre]

Diskussion

Bisher liegen keine Angaben zu Gelbkörperraten und zum Gelbkörperdurchmesser beim
Alpensteinbock vor. Bei anderen Caprinae (z.B. VALENTINCIC et al. 1974; BAUER 1972;
Kıra et al. 1987) und bei Cerviden (z.B. VALEnTIn@ıC 1958; WANDELER 1975; BucHLi
1979; SCHWARTZ et al. 1993) sind hingegen Zählungen von Corpora lutea zur Bestimmung
der Ovulationsrate durchgeführt worden. Die Tatsache, daß die potentielle Natalität
lediglich 77,1% der Gelbkörperrate beträgt, zeigt, daß auch beim Alpensteinbock die
während der Brunftzeit ermittelte Gelbkörperrate sıch nicht zur Beurteilung der Natalität
eignet (vgl. HorMmAnNn 1990). Makroskopisch lassen sich Corpora lutea periodica und
Corpora lutea graviditatis nicht sicher unterscheiden (SALZMAnN 1977). Beim Serau
(Capricornis crispus) war eine sichere Unterscheidung ım brunftnahen Zeitraum sogar
durch histologische Untersuchung nicht möglich (Kıra et al. 1987). Der von uns ermittelte
durchschnittliche Maximaldurchmesser der Corpora lutea liegt über jenem, der bei Gem-
sen (Rupicapra rupicapra) gemessen
wurde (11 mm, VALENTINCIC et al. 12
1974; 11,5 mm, SALZMANN 1977). KITA gl eis
et al. (1987) fanden ım Verlauf der 10
Trächtigkeit eine Größenabnahme, ı
VALENTINCIC et al. (1974) konnten keı- 7)
ne Größenänderung feststellen. 9,8
Im Freiland findet die Brunft meist
im Dezember bis Anfang Januar statt 0,86
(RAaucH 1937; COUTURIER 1962; AscH-
BACHER 1978; RATTI 1986; |NNIEVERGELT
und ZınGG 1986). Unsere Daten bestäti-
gen diese Angaben. Als späteste Kon-
zeptionstermine werden 19.Februar 0,2
(COUTURIER 1962) und 28. Februar
(STÜWE und GROoDINSkY 1987) angege- 0
ben. Der Befund bei einer am 21. Au- Gelbkörper- Trächtigkeits- potentielle
gust 1989 erlegten, trächtigen 12jährigen DD AD Natallızı
Steingeiß der Kolonie Albris (die Frucht Abb. 3. Drei Fortpflanzungsraten bei den ge-
war ohne pathologische Veränderun- schlechtsreifen Steingeißen der Kolonie Abris

0,4

178 M. Giacometti und P. Rattı

gen) läßt vermuten, daß bei freilebenden Alpensteinböcken Konzeptionen in seltenen
Fällen auch noch im Monat März möglıch sind.

Steingeißen setzen ın freier Wildbahn erstmals mit 3 bis 5 Jahren, in der Regel mit
4 Jahren (RaTTı 1981). Diese Angaben werden durch unsere Untersuchungen bestätigt. In
Gehegen gehaltene Steingeißen können hingegen bereits im zweiten Lebensjahr
geschlechtsreif sein (COUTURIER 1962; NNIEVERGELT 1966b; STÜWE und GRoDINSKY 1987).
Die Ergebnisse von NIEVERGELT (1966a), wonach viele Geißen der Kolonie Albris erst im
6. Altersjahr zu setzen beginnen, können nicht bestätigt werden. Die Verzögerung des
durchschnittlichen Eintrittes der Geschlechtsreife bei Erreichen der Lebensraumkapazität
dürfte beim freilebenden Alpensteinbock nicht wesentlich mehr als 1 Jahr betragen.

Bei Rindern (Bos primigenins f. taurus) und den kleinen Hauswiederkäuern wird die
Absterberate in der frühen Embryonalphase, der Periode der höchsten Gefährdung, mit 20
bis 50% angegeben (WALSER 1990). Bei Rehen (Capreolus capreolus) wurden Raten von
6,6 % (BoRG 1970) und 3,2 % (ELLENBERG 1978) angegeben. Die von uns gefundene Rate
scheint, nachdem sie sich auf Postimplantationsverluste bezieht, relativ hoch zu sein,
obwohl noch nicht von einem gehäuften Auftreten gesprochen werden kann.

Die Fekunditätsraten der 3- bis 13jährıgen bzw. der 1- bis 15jährigen Geißen betragen
im Wildpark Peter und Paul ın St. Gallen 0,99 bzw. 0,78 (STÜwE und GroDinsky (1987),
25% bzw. 24% mehr als die entsprechenden für die Kolonie Albrıs gefundenen Raten.
Die höhere Reproduktionsleistung in Wildgehegen kann insbesondere mit dem häufigeren
Vorkommen von Zwillingsgeburten (NIEVERGELT 1966a) und mit dem früheren Eintritt
der Geschlechtsreife erklärt werden.

NIEVERGELT (1966a) gibt für die Kolonie Albris eine Nachwuchsrate von 0,44 # 0.17
an, PERACINO und Bassano (1990) für die Steinbockpopulation des Parco Nazionale Gran
Paradiso eine Natalıtät von 0,39-0,50. Beide Raten sind deutlich niedriger als die in dieser
Arbeit gefundene potentielle Natalıtät. Die Anzahl der tatsächlich vorhandenen Kitze
kann in typischen Steinbocklebensräumen in den Monaten Juni oder Juli nicht immer mit
ausreichender Genauigkeit erfaßt werden (RaucH 1937). Zudem muß davon ausgegangen
werden, daß auch in freier Wildbahn ein gewisser Anteil der gesetzten Kitze ın den
allerersten Wochen nach der Geburt eingeht. Dieser Anteil liegt bei verschiedenen
Wildwiederkäuerarten zwischen 17 und 61% (Kurt 1968; STAInEs 1970; ELLENBERG
1978; BALLARD et al. 1991; WHITTEN et al. 1992). Die reale Natalıtät dürfte beim
freilebenden Alpensteinbock am besten durch die potentielle Natalıtät als höchstmögliche
Zahl charakterisiert werden (vgl. KıRKPATRICK 1980). Zur Vermeidung einer unkorrekten
Interpretation von Angaben zur Fortpflanzungsleistung bei freilebenden Alpensteinbock-
populationen wird angeregt, die Begriffe Natalität und Nachwuchsrate ausschließlich bei
Gebärmutteruntersuchungen ın der Postimplantationsphase zu verwenden. Für die
Bezeichnung der Anzahl der im Anschluß an die Setzperiode durch Direktbeobachtung
gezählten Kitze bezogen auf die Gesamtzahl fortpflanzungsfähiger bzw. adulter Geißen
sollte der Begriff Kitzrate verwendet werden, wobei stets der Beurteilungszeitpunkt und
die bei den Geißen gewählte Altersklasse mitangegeben werden müßten.

In freier Wildbahn setzen Geißen jährlich 1 Kitz, sehr selten 2 (CoUTURIER 1962; RATTI
1981), eine Aussage, die durch unsere Untersuchung bestätigt wird. HEPTNER et al. (1966)
fanden beim sibirischen Steinbock (Capra i. sıbirica) in einer Population in Zentraltien-
schan und im Pamir 2 Zwillingsträchtigkeiten unter 56. Daraus resultiert eine Rate von
3,6 %, die mit jener in dieser Arbeit gefundenen vergleichbar ist. In Wildgehegen kann die
Häufigkeit von Zwillingsgeburten 29% erreichen (vgl. NIEVERGELT 1966a). Wie ım
Wildpark Peter und Paul (Srüwe und Gropinsky 1987) war ein leichter Überhang der
männlichen Kitze festzustellen. Allerdings ist die Größe der Stichprobe für eine schlüssige
Aussage noch zu gering.

Zur Reproduktionsleistung des Alpensteinbockes (Capra ı. ıbex L.) 179

Danksagung

Ein besonderer Dank gilt den Wildhütern D. Gopti, J. SCHANIEL und A. PLozza, dem Jagdaufseher
E. EGGENBERGER und lic. phil. IIH. Jenny für Organisation und Durchführung der Probenentnahme
sowie für die Angaben über die Steinbockkolonie Albris. Bestens gedankt wird Dr. F. ToLarı für die
serologischen Untersuchungen, Dipl.-Ing. F. VöLk und Dr. F. REıMOsErR für die konstruktiven
Gespräche, Prof. Dr. K. ZEROBIN für die kritische Durchsicht des Manuskriptes sowie A. KÖRBER für
die Gestaltung der Abbildungen. Der Aufenthalt des Erstautors am Forschungsinstitut für Wildtier-
kunde und Okologie der Veterinärmedizinischen Universität Wien wurde durch ein Stipendium des
Schweizerischen Nationalfonds zur Förderung der wissenschaftlichen Forschung ermöglicht. Dem
Leiter des Instituts, ©. Univ. Prof. Dr. K. ONDERSCHEKRA, wird für die breite Unterstützung bestens
gedankt.

Zusammenfassung

Zur Beurteilung der Reproduktionsleistung des Alpensteinbockes (Capra 1. ibex L.) wurden Uteri und
Ovarıien von 80 Geifen untersucht. Die Tiere stammten aus der Kolonie Albris (Graubünden,
Schweiz) und wurden in den Monaten Dezember bis Juni der Jahre 1989-1990 und 1990-1991 erlegt.

Die Geifen konzipierten zwischen dem 1. Dezember und dem 21. Januar und wurden frühestens
mit 2 Jahren geschlechtsreif. Die Konzeptionsrate der 2jährigen Geißen betrug 0,13, diejenige der
3jährigen 0,56, jene der 4jährigen Geißen 1. Die Trächtigkeitsrate lag bei 0,88. Bei 70 zweijährigen
und älteren Geifen konnten 48 entwicklungsfähige Keimlinge nachgewiesen werden (Fekunditätsrate
0,69). Für die Altersklasse der 4- bis 13jährıgen Geißen betrug die Fekunditätsrate 0,83. Mit 14 Jahren
konnte ein wahrscheinlich altersbedingtes Nachlassen der Fortpflanzungsleistung festgestellt werden.
Bei 59 geschlechtsreifen Geifen wurden 48 entwicklungsfähige Keimlinge gefunden, was eine
potentielle Natalität von 0,81 ergibt. Die Fertilität betrug 0,25. Zwillingsträchtigkeiten wurden
lediglich in einem unter 52 Fällen nachgewiesen (1,9 %).

Die in dieser Arbeit festgestellte Fortpflanzungsleistung übersteigt die bisher für diese Alpenstein-
bockpopulation angenommene deutlich.

Literatur

ANDREWARTHA, H. G.; BIRcH, L. C. (1954): The distribution and abundance of anımals. Chicago:

University of Chicago Press.

ÄSCHBACHER, A. (1978): Das Brunftverhalten des Alpensteinbockes. Erlenbach-Zürich: Eugen
Rentsch Verlag.

AsDELL, $. A. (1946): Pattern of mammalıan reproduction. Ithaca, New York: Comstock Publ.

BALLARD, W. B.; WHITMaNn, ]. S.; REED, D. J. (1991): Population dynamics of moose in southcentral
Alaska. Wildl. Monogr. 114, 1-49.

BAUER, ]. J. (1982): Untersuchung zur Dynamik von stabilen und kolonisierenden Gemspopulationen
(Rupicapra rupicapra L.) ın Neuseeland. Diss. Forstzool. Univ. Freiburg/Breisgau.

Borg, K. (1970): On mortality and reproduction of roe deer in Sweden during the period 1948-1949.
Viltrevy 7, 121-149.

Buchui, C. (1979): Zur Populationsdynamik, Kondition und Konstitution des Rothirsches (Cervus
elaphus L.) in und um den Schweizerischen Nationalpark. Diss. Phil. Fak. II Univ. Zürich.

COUTURIER, M. A. ]J. (1962): Le Bouquetin des Alpes. Edit. par l’auteur. Grenoble: Imprimerie
Allier.

ELLENBERG, H. (1978): Zur Populationsökologie des Rehes (Capreolus capreolus L., Cervidae) in
Mitteleuropa. Spixiana, Suppl. 2, 1-199.

GIACOMETTI, M. (1991): Beitrag zur Ansiedlungsdynamik und aktuellen Verbreitung des Alpenstein-
bockes (Capra ı. ibex L.) ım Alpenraum. Z. Jagdwiss. 37, 157-173.

HEPTNER, V. G.; Nasımovic, A. A.; BAnNIKov, A. G. (1966): Die Säugetiere der Sowjetunion, Band
I: Paarhufer und Unpaarhufer. Jena: VEB G. Fischer Verlag.

Hormann, R. R. (1990): Reproduktion und Zuwachsrate in Beziehung zur Ernährungshege. In:
Fütterung und Äsungsverbesserung für Reh- und Rotwild. Hrsg. Wildforschungsstelle des Landes
Baden-Württemberg, Aulendorf, Bd. 1, 54-64.

HuccGeT, A. S. G.; Wıppas, W. F. (1951): The relationship between mammalıan foetal weight and
conception age. J. Physiol. 114, 306-317.

KIRKPATRICK, R. L. (1980): Physiological indices in wildlife management. In: Wildlife techniques
manual, 4th. Ed. by S. D. ScHEmITz. Washington, D.C.: The Wildl. Soc. Washington. Pp.
33-12.

Kıra, I.; SuGIMURA, M.; Suzukı, Y.; TıBa, T.; Mıura, $. (1987): Reproduction of female Japanese
serow based on the morphology of ovaries and fetuses. In: The biology and management of
Capricornis and related mountain antelopes. Ed. by H. Soma. Kent, North Ryde, New York:
Croom Helm. Pp. 321-331.

180 M. Giacometti und P. Ratti

Kurt, F. (1968): Zusammenhänge zwischen Verhalten und Fortpflanzungsleistung beim Reh
(Capreolus capreolus L.). Z. Jagdwiss. 14, 97-106.

— (1991): Das Reh in der Kulturlandschaft. Hamburg, Berlin: Paul Parey.

MITCHELL, B.; STAINEs, B. W.; WELcH, D. (1977): Ecology of red deer. Cambridge: Graphic Art.

NIEVERGELT, B. (19662): Der Alpensteinbock (Capra ıbex L.) ın seinem Lebensraum. Hamburg,
Berlin: Verlag Paul Parey.

— (1966b): Unterschiede in der Setzzeit beim Alpensteinbock (Capra ıbex L.). Rev. Suisse Zool. 73,
446454.

NIEVERGELT, B.; ZınGG, R. (1986): Capra ibex L. - Steinbock. In: Handbuch der Säugetiere Europas.
Hrsg. von J. NIETHAMMER und F. Krapp. Wiesbaden: Aula-Verlag. Bd. 2/II, 384-404.

PERACINO, V.; Bassano, B. (1990): Andamento della popolazione di Stambecco nel Parco Nazionale
Gran Paradiso: annı 1986-1989. 3. Notiziario Gruppo Stambecco Europa. Torino: Collana
Scientifica Parco Nazionale Gran Paradiso.

RarTı, P. (1981): Zur Hege des Steinwildes im Kanton Graubünden. Z. Jagdwiss. 27, 41-57.

— (1986): Das Steinwild. In: Leitfaden für Bündner Jäger. Hrsg. P. RAaTTı. Disentis/Muster: Stampa
Romontscha, Condrau SA. Kap. 2, 21-31.

Rattı, P.; HABERMEHL, K.-H. (1977): Untersuchungen zur Altersschätzung und Altersbestimmung
beim Alpensteinbock (Capra 1. ıbex.) ım Kanton Graubünden. Z. Jagdwiss. 23, 188-214.

Rauch, A. (1937): Der Steinbock wieder in den Alpen. Zürich: Orell Füsslı Verlag.

SALZMANN, H. C. (1977): Untersuchungen zur Fortpflanzungsbiologie der Gemsen im Schweizeri-
schen Jura. Z. Säugetierkunde 42, 180-189.

SCHRÖDER, W. (1971): Untersuchungen zur Okologie des Gamswildes. Z. Jagdwiss. 17, 113-168 und
197-235.

SCHARTZ, C. C.; HUNDERTMARK, K. J. (1993): Reproductive characteristics of alaskan moose.
J. Wildl. Manage. 57, 454-468.

SCHWERDTFEGER, F. (1978): Lehrbuch der Tierökologie. Hamburg, Berlin: Verlag Paul Parey.

STAINES, B. W. (1970): The management und dispersion of a red deer population in Glen Dye,
Kincardineshire. Diss. Univ. Aberdeen.

STÜwE, M.; GRoDINsKY, C. (1987): Reproductive biology of captive alpine ibex (Capra 1. ıbex L.).
Zoo Biology 6, 331-339.

TATARUCH, F.; STEINECK, T.; KLANSEK, E.; VAVRA, 1.; RATTI, P.; GIACOMETTI, M. (1991): Untersu-
chungen an Steinwild aus Graubünden (Schweiz). I. Analysen der Nahrungszusammensetzung,
der Aktivität der Schilddrüsen und Nebennieren sowie der Reproduktion. Wien. Tierärztl.
Mschr. 78, 351-356.

VALENTINCIC, S. (1958): Beitrag zur Kenntnis der Reproduktionserscheinungen beim Rotwild. Z.
Jagdwiss. 4, 105-130.

VALENTINCIG, S.; BAVDEK, $.; KuseJ, M. (1974): Gravidität der Gamsgaisen in den Julischen Alpen. Z.
Jagdwiss. 20, 50-53.

WALSER, K. (1990): Pränatale Entwicklungsstörungen. In: Neugeborenen- und Säuglingskunde der
Tiere. Hrsg. K. WALSER und H. BOTSTEDT. Stuttgart: Enke-Verlag.

WANDELER, A. I. (1975): Die Fortpflanzungsleistung des Rehs (Capreolus capreolus L.) ım Berner
Mittelland. Jb. Naturhist. Museum Bern 5, 245-301.

Weıss, R.; GIACOMETTI, M.; GEYER, H. (1993): Histologische Untersuchungen zur Entwicklung der
Milchdrüse bei trächtigen Alpensteingeißen (Capra ı. ıbex. L.). Z. Säugetierkunde 58, 144-154.

WHITTEN, K. R.; MAUER, F. ]J.; Harrıs, R. B. (1992): Productivity and early calf survival ın the
porcupine carıbou herd. J. Wildl. Manage. 56, 201-212.

Anschriften der Verfasser: Dr. Marco GIACOMETTI, Forschungsinstitut für Wildtierkunde und
Okologie der Veterinärmedizinischen Universität Wien, Savoyenstr. 1,
A-1160 Wien; Dr. PEIDER RATTI, Jagd- und Fischereiinspektorat Grau-
bünden, Loestr., CH-7000 Chur

Z. Säugetierkunde 59 (1994) 181-188
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

Ouü se situe la limite nord de repartition geEographique de
Myotis blythii(Chiroptera: Vespertilionidae) en Europe centrale?

Par R. ArLETTAZ, A. BECK, R. GÜTTINGER, MIRIAM LUTZ, M. Rueoı et P. ZıncG

Institut de Zoologie et d’Ecologie Animale, Universite de Lausanne, Suisse et Ethologie und Wild-
forschung, Zoologisches Institut, Universität Zürich, Suisse

Reception du Ms. 15. 11. 1193
Acceptation du Ms. 05. 05. 1994

Abstract

Where is the northern border of the distribution range of Myotis blythii (Chiroptera: Vespertilionidae)
in Middle Europe?

At least four mouse-eared bat colonies from the Swiss Rhine Valley visited in 1993 were mixed nursery
roosts of M. myotis and M. blythıi and not pure populations of M. myotis as previously believed. The
recent discovery of M. blythiüi in that area does not result from a recent colonization process, but from
the misidentification of these two cryptic species up to now. The northernmost mixed nursery found
in Switzerland is located just a few dozen kilometers from Germany where M. blythü has apparently
never been recorded. It is suggested to check the identity of further nursery roosts in Central Europe
in order to avoid autoecological research on populations which may actually not be conspecific.

Introduction

Sı le Grand Murin Myotis myotis est une espece repandue sur l’ensemble du territoire
helvetique, des dizaines de colonies Etant connues dans le Moyen-Pays (STUTZ et HAFFNER
1991), la repartition du Petit Murin Myotis blythii semblait jusqu’ici limitee au sud de la
Suisse ou l’espece cohabite dans ses gites avec le Grand Murin (MoRrETTI etal. 1993). Seules
quatre colonies abrıtant M. blythii Etaient jusqu’icı connues en Suisse, trois en Valais
(Alpes centrales du sud-ouest de la Suisse) et une au Tessin (versant sud des Alpes). Par
ailleurs, une seule mention existait A cette date en Suisse en dehors du Tessin et du Valais:
un ındıvıidu mäle collecte par Bovey (1954) dans les souterrains du chäteau de Chillon/
Montreux (Musee zoologique Lausanne No 1556), et attribue a M. myotis est en fait un
M. blythı (Rueoı 1987; MORETTI et al. 1993). En septembre 1993, dans le cadre des etudes
de sa these de doctorat, R. ARLETTAZ visitait les colonies de Murins connues dans les
Grisons, en vue d’effectuer des mensurations de la morphologie externe d’une population
allopatrique de Grand Murin M. myotis. A notre plus grande surprise, ıl s’est avere que les
quatre gites visits abritaient tous une importante proportion de M. blythü inumement
associes aux M. myotis. Ces faıts nous ont incites A visiter d’autres colonies, notamment
dans le nord-est de la Suisse, afın de preciser la repartition geographique du Petit Murin en
Suisse.

Materiel et methodes

Colonies examinees

Selon les travaux de Lutz (1986 et comm. pers.), cing colonies de reproduction de Grand Murin
existent dans les Grisons; elles sont toutes quatre situees dans la vallee du Rhin, soit de l’amont vers
l’aval: a Surrein, a Trun (colonie tres proche de la pr&cedente, habitee probablement par les m&mes
individus) et Laax (Surselva), a Pratval dans le Domleschg, enfin a Fläsch aux confins du Liechtenstein
et du Vorarlberg autrichien. Les troıs colonies de reproduction les plus accessibles (la colonie de
Pratval ne permet pas de capture directe des anımaux) ainsi qu’un gite occup& soit par des mäles

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5903-0181 $ 02.50/0

182 R. Arlettaz, A. Beck, R. Güttinger, Miriam Lutz, M. Ruedi et P. Zingg

solitaires soit par des couples durant la periode d’accouplement ont £t& vistes les 9 et 10 septembre
1993. Les 15 et 16 septembre, nous avons visite l’eglise d’Eichberg, qui est la colonie la plus nordique
connue dans la vallee du Rhin helvetique, ainsi que les colonies d’Oberglatt/Flawil (en bordure du
Plateau et des pre£alpes saint-galloises et appenzelloises), de Lipperswil (Plateau thurgovien ä proximite
du lac de Constance) et de Veltheim (pied sud du Jura argovien). Nous disposons par ailleurs
d’informations recentes au sujet de la colonie de Meiringen, Oberland bernois (visitee en 1992) ainsi
que sur celles d’Eysins, dans le bassın lemanique, au pied du Jura vaudois (1992) et de Roche (vallee
vaudoise du Rhöne, visitee en 1991).

Identification

L’identification des especes a Et& Etablie sur la base des criteres de la morphologie exterieure proposes
par ARLETTAZ et al. (1991): 1. precence d’une tache blanche sur la tete des Petits Murins qui permet
d’effectuer un premier reperage visuel dans les essaims; 2. mesure de la longueur de l’avant-bras et de
celle de l’oreille, qui permettent, lorsqu’on les associe dans une fonction discriminante, de separer
correctement les deux especes. Au total, nous avons capture et mesure 102 individus differents
(Tableau). Ce lot ne represente pas un Echantillonnage aleatoire, l’accent ayant et€ mis sur la capture
des Petits Murins en priorite, du moins la ou ls existaient (29 M. blythii vs. 73 M. myotis au total). La
visite des colonies a eu lieu apres une p£riode fraiche et maussade, et la dislocation postreproductrice
etait deja bien entame&e, preuve en est le nombre de jeunes de l’annee captures au seın des colonies de
parturition (71.6%, soit 68 immatures sur 95 individus d’äge connu, gite d’accouplement non
compris). Il faut par ailleurs savoir que la mise-bas est plus tardıve chez M. blythii, ce qui a pour
corollaire une dispersion plus tardıve des femelles adultes et des immatures chez cette espece, en tout
cas en Suisse (ARLETTAZ, inedit). En consequence, il est probable que les estimations des proportions
des deux especes au sein des colonies mixtes (cf. ci-dessous) sont bıaisees en faveur des Petits Murins.

Resultats

La colonie de Surrein comportait un essaim principal totalisant enyiron 250 chauves-souris
dont 20 a 30 % £Etaient de la petite espece. L’eglise de Laax comptait 100 a 150 individus
dont un quart A un tiers de M. blythii. Le gite d’accouplement de Bonaduz totalisait 7
couples ainsı que deux individus isoles; deux des couples appartenaient a la petite espece M.
blythii, alors que cing £taient des couples de M. myotis. L’eglise de Fläsch qui rassemble
plus de 1200 tetes en periode d’elevage ne contenait plus qu’environ 140 individus disperses
en quatre essaims distincts lors de notre visite. Trois essaims Etaient composes en majorite
(plus de 95% des individus) de M. myotis tandıs que le quatri&me essaim ne contenait
pratiquement que la Petite espece (soit une total estime a 100 M. myotis et 40 M. blythn).
Les Murins de l’eglise de Eichberg (50-100 anımaux lors de notre visite) ne pendent pas
librement dans la sous-pente, mais gitent dans l’entretoit, ce qui a rendu impossible
l’estimation des proportions des deux especes. Les chauves-souris de Oberglatt/Flawil
(environ 50 individus lors de notre visite) se retirent €galement dans les fissures entre
charpente et couverture du clocher ou des combles; 32 individus captures £taient tous des
M. myotis. A Lipperswil, ou ıl restait encore 50-60 individus, et ä Veltheim (1400 individus
en p£riode de reproduction, soit la plus grande colonie de Murins connue en Suisse; 150
individus lors de la visite), les chauves-souris pendaient librement sous les lambris. Ces
deux dernieres colonies n’abritaient aucun M. blythüi. Enfin, les colonies de Meiringen et
d’Eysins sont egalement ä consıiderer comme des colonies pures de Grand Murin. La carte
(Fig.) presente la localisation des colonıes mixtes et pures visitees dans le cadre de cette
etude.

Discussion

La difficulte d’identifier ces deux especes jumelles de Murins sur la base de leur morpholo-
gie externe explique que la presence de colonies mixtes ait passe inapercue dans le nord-est
de la Suisse. A cet Egard, ıl faut rappeler que ce n’est que recemment que la distinction
genetique entre M. myotis et M. blythü a te Etablie (Ruenr et al. 1990) et que c’est par le

La limite nord de repartition de Myotis blythu en Europe centrale 183

Localite, sexe äge, longueur de l’avant-bras, longueur de l’oreille, score discriminant (M. myotis:
z>0;M. blythii: z < 0) et appartenance specifique des Murins captures dans les colonies du nord-
est de la Suisse dans le cadre de la presente etude

La methode d’identification au moyen des scores discriminants est decrite par ARLETTAZ et al. (1991)

Localıite Avant-bras Longueur Score dis- Espece
(mm) oreille (mm) criminant (z)

61.0
60.9

Surrein GR . blythii
myotis
blythii
blythii
myotis
blythii
blythü
blythüi
blythü
blythii
blythu
myotis
myotis
myotis
myotıis
blythıi
myotıis
blythü
blythu
myotis
myotis
myotıs
blythüi
myotis
myotis
myotis
blythui
blythii
blythü
blythii
blythi
blythu
blythui
blythii
blythü
blythii
myotis
myotis
myotis
myotis
myotis
myotıis
myotis
myotis
myotis
myotis
myotis
myotis
myotis
myotis
myotis
myotis
. myotis

Laax GR

Bonaduz GR

Fläsch GR

Oberglatt SG

SSSSSSSISSISISISISSISISSIISSISIISIERIIIRRI IIIRI IIIIRIIERIIIREN

W
Ww
m
m
W
m
Ww
Ww
m
m
Ww
m
W
W
W
m
m
W
Ww
m
m
Ww
W
Ww
m
m
m
m
W
Ww
m
Ww
m
m
m
m
W
m
Ww
m
Ww
m
W
Ww
m
Ww
Ww
Ww
W
W
Ww
W
Ww

184 R. Arlettaz, A. Beck, R. Güttinger, Miriam Lutz, M. Ruedi et P. Zingg
Tableau (suite)

Localite 3 Avant-bras Longueur Score dis- Espece
(mm) oreille (mm) _criminant (z)

Oberglatt SG 58.0 26.9 3.803

SI 25.8 BIRD

58.6 26.2 2.877

335, 253 1.309

58.2 25.6 1.984

22 4.521

26.8 3.987

26.6 3.498

2.245

4.086

4925

1.952

3-47

220

1.853

2450
2.613
—2.255

0.720

myotis
myotis
myotis
myotis
myotis
myotıs
myotis
myotıis
myotis
myotis
myotis
myotis
myotis
myotis
myotis
blythu
blythui
blythii
myotis
bhythui
blythu
myotis
blythu
myotıs
myotıs
myotis
myotis

Eichberg SG

myotis
myotis
myotis
myotis
myotis
myotis
myotıis
myotis

Lipperswil TG

Veltheim AG myotis
myotis
myotis
myotis
myotis
myotis
myotis
myotis
myotis
myotis
myotis
myotis
myotis
. myotis

Meiringen BE

Ww
m
m
m
m
Ww
Ww
Ww
Ww
Ww
Ww
m
mn
m
m
m
m
Ww
m
Ww
Ww
m
m
Ww
Ww
Ww
mn
w
Ww
m
w
Ww
Ww
W
Ww
m
w
Ww
Ww
m
Ww
Ww
Ww
Ww
Ww
Ww
Ww
Ww
Ww

SSSSSSSSSSSISSSSSISISSSISS SISIISSIISISS SSIIIIIIIISSISISS

AG = Argovie ; BE = Berne; GR = Grisons; SG = St-Gall; TG = Thurgovie, soit les differents
cantons suisses oU se trouvent les colonies de parturition (les sites sans reproduction n’ont pas Ete
cartographies) etudiees dans le cadre de la presente £tude.

185

La limite nord de repartition de Myotis blythü en Europe centrale

neojgqeı np »puodg] 1TOA ‘sajentur sonne sop nod !sıepey = Sy pneA=AA -UISS9], = ]] :S9TUOJOI Sa] JU9ANOM 95 no sassıns
suorues SIUSIZFIP so] Juanbıpur sopentur so "uoneurmyuo9 sToFsIN0I ITe1sLıgW Inb szzodw "yy op sınd »9soddns aruoJo9 sun anbıpur uone3o11sgut,p
urod 97 'assıng U9 g4d7q "W >p vonngtnsıp ap arrung e] % 91puodsa1109 apquias ınb urdje frsseur np pxou 2uo1F 97 onbrpur 995113 opueq ET 'assıng
U9 SO9SU9991 JUJUNWUIIH (ALS) Zg1d7g "w 32 sol "ap sorxıun 19 (110U) sırodw ap sand s9TUOJ09 sap saa1feso] 19 anbryde13093 uonnsedoy ‘7 ‘17

VITVL

x Y E
\ I I
\ z \ j h Ba gyr ES

f \ N a \ ) 3
® r Sn y Sn uoleg Sana ET:
iR — DIET ( f 9 (E 2 3 Cr one Tal \ )
D 7 EM N N f Rz a8 :
j in 7 \ f ” } Ä N) L SA SIOJEN a 2 Sy9od Be ___ \ AL
= m \ Re % - Da > = s d > 7)
y wu 5 | au = / = \ “ 8 ————— N Suıskg3z
\ = \ NS 6, ] / \ — j r

f =, © Ya \ / 3g uedunıeN _ | / | l IE 8
ash SE | 49 uieung { \ EN ( > nm icar
) 7; _ SZ > { ) j - IP / / N 4
Dee ee: Su

[ HN ——& a
N E N S / HH xee] ,
vn — (
e—J > ) |
& 3 ch = \

In

NISLSNILH931

BO Uyoseı4 \

HOISHH3LSO > A \

? N N dat $ 7 4 u E
( \ SS IS e— N <a
> N \ N m a9
2 DS Yelßlsgo IN \ En N £ J N
7 & x Le S TS a.)

N
GG Hy wıeynen. nf 5

|
S a an > S rn ] 5 —. = eG
®:: msieddi1 Free —_ x
DE
Zu FJONVHJ
m

ANVAHOSLNIA De

186 R. Arlettaz, A. Beck, R. Güttinger, Miriam Lutz, M. Ruedi et P. Zingg

bıais de cette approche que ARLETTAZ et al. (1991) ont Et€E en mesure de mettre au point une
methode de determination sur le terrain plus rigoureuse que ce ne fut le cas prec&demment.
Notons toutefois que la simple r&colte et mensuration des cränes des cadavres d’adultes ou
subadultes qui finissent fatalement par se retrouver sous les essaims auraient permis de
deceler la presence de la petite espece, les differences cränıennes interspecifiques &tant
decrites de longue date en Europe centrale (e.g. STRELKOV 1972). En Suisse orientale,
l’accent particulier qui a Et€ mis sur la protection de ces colonies, et qui s’accompagnait
d’une quası absence de capture, a reporte la decouverte de ces deux especes cryptiques ä
une date bien tardıve. Sur la soixantaine de colonies recensees en Suisse orientale et
attribuees par STUTZ et HAFFNER (1991) A la seule espece M. myotis, quatre au moins se sont
averees des colonies mixtes lors de nos visites. Et il est probable que trois autres colonies au
moıns (Pratval/Grisons, Triesen/Liechtenstein, Gams/St-Gall) sont aussi habitees par M.
blythii, puisque situees entre et ä peu de distance des colonies mixtes precitees. Une
proportion non negligeable des colonies de Murins recensees en Suisse orientale seraient
donc des colonies mixtes. On notera egalement que toutes les colonies mixtes aujourd’hui
connues en Suisse sont sıtuees dans le massif alpin. Dans ce contexte, on peut s’interroger
sur le statut exact des colonies de Suisse centrale, notamment celles existant dans les vall&es
soumises a un regime de foehn; A cet Egard, un nouveau contröle de la colonie de Meiringen
se justifierait, seuls 7 indıvidus y ayant ete captures en 1992 (Tableau). Enfin, qu’en est-ıl des
regions limitrophes (Baviere du sud) sıtu&es A quelques dizaines de kilometres seulement de
la colonie mixte la plus nordique de Suisse? D’autant plus que SPITZENBERGER (1988)
mentionne la presence d’une colonie de M. blythü dans la vallee de !’Inn (Tyrol autrichien), ä
une latitude comparable a la colonie helvetique mixte la plus nordique. Le bassin de
Rosenheim, situe dans le prolongement naturel de la vallee de l’Inn et aux portes des Alpes
est-ıl vraiment habıte par la seule espece M. myotis? Les etudes intensives effectuees au cours
de la derniere decennie dans cette region concernent-elles vraiment des colonies pures de
Grand Murin? Il est urgent que les chiropterologistes d’Europe centrale, notamment de
France septentrionale, de Suisse et d’Allemagne, effectuent un nouveau contröle de leurs
colonies afın qu’une reponse definitive soit apportee A cette question cruciale. Sinon, on peut
redouter que certains travaux de recherche soient A considerer, au moins partiellement,
comme caducs, A l’exemple des etudes de ACKERMANN (1984) et de GRAF et al. (1992) qui
portaient sur la biologie et le regime alimentaire de M. myotis en Suisse orientale. Ces deux
chercheurs on &tudie sans le savoir des colonies mixtes et non des colonies pures de Grand
Murin. Une immigration recente de M. blythüi dans cette region est A Ecarter, les habitats de
chasse (milieux herbaces; voir ARLETTAZ et al. 1993) dont l’espece depend ayant plutöt subi
une diminution ou une p&joration (engraissement) qu’une extension au cours des dernieres
annees. ACKERMANN (1984) qui a justement travaille A Eichberg, dans la vallee du Rhin,
decrit le regime alimentaire, les routes de vol, les rythmes d’activite, et accessoirement les
zones de chasse de sa colonie supposee pure. Or, il est fort probable que ses observations
concernent les deux especes. Parmi les 14 gites etudies par GRAF et al. (1992) dans le cadre
d’une comparaison geographique du regime alimentaire de M. myotis en Suisse orientale,
trois au moins se sont averdes des colonies mixtes. Ainsi, la proportion inferieure des
Carabidae et la part superieure des Acrididae en region alpine, pour ne citer que ces deux
groupes particulierement &vidents, n’a vraisemblablement pas pour origine principale la
difference dans l’offre du milieu mais dans le fait que l’on a Etudi& deux expeces aux regimes
alimentaires bien tranches (ArLETTAZ et al. 1993) au lieu d’une seule. En effet, ARLETTAZ et
al. (1993) ont recemment montre que, dans les Alpes valaisannes (sud-ouest de la Suisse), M.
myotis capture surtout des Carabes tandis que M. blythii prefere les Orthopteres appartenant
a la famille Tettigoniidae. D’autres travaux en cours sur l’Ecologie de ces deux especes en
conditions de sympatrie et d’allopatrie, devraient permettre de mieux cerner les exigences de
M. blythii du point de vue habitat et regime alimentaire, et partant d’expliquer sa distribution
geographique en Europe.

La limite nord de repartition de Myotıs blythü en Europe centrale 187

Remerciements

Nous remercions les responsables des edifices publics et sacres qui nous ont permis d’acceder aux
differentes colonies visitees, en particulier M. W. D. BURKHARD.

Resume

Une visite effectu&e en 1993 dans les colonies de Vespertilions murins de la vallee helvetique du Rhin
(cantons des Grisons et de Saint-Gall) ont montr& que quatre nurseries au moins y abritaient les deux
especes jumelles M. myotis et M. blythuü et non des populations pures de M. myotis comme on le
pensait jusqu’alors. Ceci porte A huit le total des nurseries mixtes qui ont Et& localisees a ce jour en
Suisse. La decouverte recente du Petit Murin dans les cantons des Grisons et de Saint-Gall ne resulte
pas d’une colonisation recente, mais d’une non reconnaissance de cette espece cryptique. La colonie
suisse la plus nordique est situee au sud du lac de Constance (Bodensee) et a quelques dizaines de
kilometres seulement de la Baviere (Allemagne) ou l’espece n’a jamais ete sıgnalee. Les auteurs
insistent sur la necessite de proceder a une nouvelle identification des Murins des colonies connues en
Europe centrale, notamment dans la moiti€ nord de la France, le sud de l’Allemagne et l’Autriche, afın
de mieux cerner l’aire de repartition geographique du Petit Murin et d’eviter des Etudes auto£cologi-
ques sur des populations qui seraient en realıte mixtes.

Zusammenfassung

Wo liegt die nördliche Verbreitungsgrenze von Myotıs blythü (Chiroptera: Vespertilionidae) in
Mitteleuropa?

Eine 1993 erfolgte Kontrolle von Wochenstuben-Quartieren ım schweizerischen Rheintal (Kantone
Graubünden und St. Gallen) hat ergeben, daß mindestens vier der bisher in der Region gefundenen
Mausohr-Wochenstuben sowohl Myotis myotis wie Myotis blythii beherbergen. Bis zum heutigen Tag
sind damit insgesamt acht Mischkolonien der beiden Arten in der Schweiz nachgewiesen. Mischkolo-
nien waren bisher nur aus dem Wallis (3) und dem Tessin (1) bekannt. Die erst jetzt bekannt
gewordenen Vorkommen von M. blythü weisen auf ein bisheriges Übersehen dieser Fledermausart hin
und weniger auf eine erst kürzlich erfolgte Besiedlung neuer Gebiete. Die nördlichste Kolonie mit M.
blythü ın der Schweiz liegt wenige Kilometer südlich des Bodensees und damit nur wenige Dutzend
Kilometer entfernt von Bayern (Deutschland), wo die Art bisher nicht nachgewiesen ist. Aufgrund
der vorliegenden Befunde erachten es die Autoren der vorliegenden Arbeit als dringend nötig,
sämtliche Mausohrkolonien ın Mitteleuropa von neuem auf ihre Artzugehörigkeit zu überprüfen, vor
allem in der nördlichen Hälfte Frankreichs sowie in Süd-Deutschland und Österreich. Dadurch
könnte die geographische Verbreitung von M. blythir präziser erfaßt und somit künftig verhindert
werden, dafs autökologische Untersuchungen irrtümlicherweise an Kolonien durchgeführt werden,
welche tatsächlich gemischte Kolonien von M. myotis und M. blythü sind.

Bibliographie

ACKERMANN, G. (1984): Diät, Aktivitätsmuster und Jagdgebiete des Großen Mausohrs Myotis myotis
(Borkhausen, 1797). Diplomarbeit Universität Zürich.

ARLETTAZ, R.; RuEDI, M; Hausser, J. (1991): Field morphological identification of Myotıs myotis and
Myotis blythü (Chiroptera, Vespertilionidae): a multivariate approach. Myotis 29, 7-16.

— — — (1993): Ecologie trophique de deux especes jumelles et sympatriques de chauves-souris:
Myotis myotis et Myotis blythü (Chiroptera: Vespertilionidae); premiers resultats. Mammalıa 57,
519-531.

Bovzy, R. (1954): Observations sur les Chiropteres du canton de Vaud et des regions voisines.
Bulletin de la Societe vaudoise des Sciences naturelles 66, 1-18.

GRAF, M.; Sturz, H. P. B.; ZıswiLer, V. (1992): Regionale und saisonale Unterschiede in der
Nahrungszusammensetzung des Großen Mausohrs Myotis myotis (Chiroptera, Vespertilionidae)
in der Schweiz. Z. Säugetierkunde 57, 193-200.

LuTZ, M.; ZAHNER, M.; StuTz, H.P. (1986): Die gebäudebewohnenden Fledermausarten des Kantons
Graubünden. Jber. Natf. Ges. Graubünden 103, 91-140.

MOoRETTI, M.; ARLETTAZ, R.; MADDALENA, T. (1993): Decouverte d’une colonie mixte de parturition
de Myotis myotis et Myotis blythii au Tessin (Sud de la Suisse) et cartographie sommaire de la
presence de M. blythü en Suisse. Le Rhinolophe 9, 59-62.

Ruepı, M. (1987): Statut specifique de deux chauves-souris jumelles Myotis myotis (Bork.) et Myotis
blythü (Tomes): une approche morphologique, caryologique et biochimique. Travail de diplöme,
Universite de Lausanne.

188 R. Arlettaz, A. Beck, R. Güttinger, Miriam Lutz, M. Ruedi et P. Zingg

Ruepı, M.; ARLETTAZ, R.; MADDALENA, T. (1990): Distinction morphologique et biochimique de
deux especes jumelles de chauves-souris: Myotis myotis (Bork.) et Myotis blythı (Tomes) (Mam-
malia; Vespertilionidae). Mammalıa 54, 415-429.

SPITZENBERGER, F. (1988): Großes und Kleines Mausohr, Myotis myotis Borkhausen, 1797, und
Myotis blythüi Tomes, 1857 (Mammalıa, Chiroptera) in Österreich. Mitt. Abt. Zool. Landesmus.
Joanneum, Mammalıa austriaca 154, 21-68.

STRELKOV, P. (1972): Myotis blythi (Tomes, 1857): Distribution, geographical varıiability and
differences from Myotıs myotis (Borkhausen, 1797). Acta Theriologica 17, 355-380. _

Sturz, H. P. B.; HArrner, M. (1991): Wochenstubenkolonien des Großen Mausohrs. Ein Überblick
über die Arbeiten der Quartierbetreuer zum Schutze der Wochenstubenquartiere des Großen
Mausohrs. Koordinationsstelle Ost für Fledermausschutz (KOF), Zürich. Zoologisches Museum
der Universität Zürich.

ZAHNER, M.; Lutz, M. (1986): Die gebäudebewohnenden Fledermausarten des Kantons Graubün-
den. Jber. Natf. Ges. Graubünden 103, 91-140.

Adresses des auteurs: RarHAEL ARLETTAZ, Institut de Zoologie et d’Ecologie Anımale, Bätiment de
Biologie, Universit€ de Lausanne, CH-1015 Lausanne; ANDRES BECK, Zweiern
19, CH-5443 Niederrhordorf; RENE GÜTTINGER, Ethologie und Wild-
forschung, Zoologisches Institut, Universität Zürich, Irchel 2, CH-8057
Zürich; Mırıam LuTz, Encarden 51, CH-7152 Sagogn; Dr. MAnUEL RUuEDI,
Institut de Zoologie et d’Ecologie Anımale, Bätiment de Biologie, Universite de
Lausanne, CH-1015 Lausanne; Dr. PETER ZınGc, Riedmattenweg 19, CH-
3700 Spiezwiler, Suisse

Z. Säugetierkunde 59 (1994) 189-191
© 1994 Verlag Paul Parey, Hamburg und Berlin
ISSN 0044-3468

NZISSEIN SCHRIAEIDETETIERIURZMTEREFEITLUNG

Feeding habits of the Stone marten Martes foina and
environmental factors in western France

By TH. LoDeE

Laboratoire d’Ethologie, Universite Rennes 1, Rennes, France

Receipt of Ms. 8.9. 1993
Acceptance of Ms. 12. 1. 1994

The stone marten Martes foına Erxleben, 1777 has been portrayed as an omnivorous feeder
(WAECHTER 1975; DELIBES 1978; AMORES 1980; CHOTOLCHU et al. 1980; CLEMENT and
SAINT-GIRONS 1982; HoLısovA and ORBTEL 1982; KALPERS 1983; RAsMUSsENn and MAD-
SEN 1985; SKIRNISSON 1986; TESTER 1986; ROMANOWSKI and LesınskI 1991); moreover, it
has been assumed that food availabilities affect diet composition. Although some previous
studies have assessed seasonal dietary patterns, only very few have attempted to evaluate ın
parallel resource availabilities (LoDE 1991). The aim of the present investigation is to relate
abiotic factors and the main resource availabilities to a detailed study of the diet of stone
martens.

The study was performed ın farmlands ın the Briere regional Parc (47° 26’ N, 2° 14’ W)
and near marshes of the Lake Grand-Lieu (47°01’ N, 1°46’ W). The clımate ıs mıld and
humid, mean temperatures ranging between 21.5°C ın August (1990) (Grand-Lieu) and
2.9°C in February (1991) (Briere), precipitation reached 851 mm per year over 140 to 153
rainy days.

In Briere, 663 scats from stone martens were collected monthly from February 1990 to
November 1991, and 486 scats from July 1989 to December 1990 in Grand-Lieu, in barns
used as resting places. Remains were assigned to the most specific taxon possible by
examining the external characteristics and by microscopic observation of medullary hair
structure and cross-section, compared with our collection and an atlas (Dav 1966;
CHALINE et al. 1974; DEBROT et al. 1982). Diet composition was estimated as frequency of
occurrences. The trophic niche breadth was calculated using the Shannon index h’ =
-2 P; log; P; where P; was the frequency of a tood category (mammals, birds, insects, fruit
and other). Availability of Microtus arvalıs was made by traplıne success every second
month. At each sıte, a line of 32 live-traps was set for 3 nıghts through two meadows. The
relative abundance of rodents was expressed as the number of individuals captured per
trapnight. The abundance of insects (mainly Orthoptera and Coleoptera) was ascertained
by counting the number of individuals captured monthly in visual traps (LEBERRE 1969) in
3 habitats (wood, hedge, meadow). The fruit availability was estimated by ascrıbing a
monthly rank order according to the ripeness and productivity of fruit in 3 control trees
each for blackberry bush (Rubus sp.), sloe (Prunus spinosa), elder (Sambucus nigra) and
apple tree (Malus sylvestris).

Dietary variations showed a clear seasonality (Briere, x’ = 159.4, df 9, p<0.001;
Grand-Lieu x? = 160.3, df 9, p< 0.001). Mammals formed the bulk of the diet reaching
75 % of the winter diet ın Briere and 84% ın Grand-Lieu but represented only 40 % and
31 % of the summer diet (Tab. 1). Birds were mainly consumed in spring and insects were
especially found ın summer. Fruit was of major importance during summer and autumn.

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5903-0189 $ 02.50/0

190 Th. Lode

Table 1. Seasonal proportions of different food categories in Stone marten diet at two sites in
western France

Briöre Grand-Lieu

Spring Summer Autumn Winter Spring Summer Autumn Winter

Insectivores

Microtus

Other mammals

Total mammals

Birds

Insects

Fruit ö | 3 : 5 :

Others 157 4.1

n prey = 180 ZI

N’= 1.630 12952 1.463 1.157 1.480 1.924 1.684 0.882

The niche breadth varıed throughout the year with a minimum in winter and a maximum in
summer. Annual diet was significantly different between the two sites (x? = 24.22, df 4,
p<0.001). In summer, birds represented a less important food category in Grand-Lieu
than in Briere, while the proportion of insects and fruit was larger in Grand-Lieu (x? =
14.9, df 3, p<0.002). During autumn, more insects and less fruit were eaten in Grand-
Lieu than in Briere (x? = 11.86, df 3, p< 0.008). Due to the predominance of mammal
prey, no significant difference appeared either in winter (x? = 7.7, p> 0.05) or in spring (x?
= 0.74, p>0.8).

Photoperiod and mean temperatures were negatively correlated with the monthly
frequency of mammals in the diet (Spearman rank correlation, df 20 in Briere, df 16 in
Grand-Lieu, Briere r, = -0.699, p<0.001 and r, = -0.845, p<0.001, Grand-Lieu r, =
-0.798, p<0.001 and r, = -0.909, p<0.001), whereas the number of rainy days (r, =
0.448, p<.0.05) ın Briere and the number of rainy days and precipitations in Grand-Lieu
were positively related (r, = 0.579, p<0.02, r, = 0.501, p<0.05). Close negative
relationships were obtained between monthly frequency of Microtus arvalis and mean
temperatures (Briere r, = -0.709, p<0.001, Grand-Lieu r, = -0.862, p<0.0Ol) or
photoperiod (Briere r, = 0.750, p<0.001, Grand-Lieu r, = -0.876, p<0.001). On the
other hand, precipitation and number of rainy days were positively correlated with dietary
frequency of voles (Briere r, = 0.494, r, = 0.451, p<0.05, Grand-Lieu r, = 0.634, r, =
0.605, p<0.01). A few insectivores were consumed during the coldest months. Other
mammals eaten included long-tailed field mice (Apodemus sylvaticus), mice (Mus domes-
ticus), brown rats (Rattus norvegicus), bank voles (Clethrionomys glareolus), water voles
(Arvicola sapidus) and rabbits (Oryctolagus cuniculus). Birds were mainly passeriforms.
Coleoptera and orthoptera were the most common insects found and showed correlations
with photoperiod and temperatures (Briere r, = 0.874, r, = 0.851, p<0.001, Grand-Lieu r,
= 0.815, r, = 0.934, p< 0.001). Other invertebrates, mainly earthworms, were eaten in
winter and spring. The frequency of fruit (Rubus sp., Prunus sp., Sambucus niger, Malus
sylvestris, Crataegus oxyacantha, Rosa canina) correlated wıth temperatures (Briere r, =
0.544, p<0.02, Grand-Lieu r, = 0.806, p< 0.001) and in Grand-Lieu to photoperiod (r, =
0.815, p<0.001), precipitations (r, = 0.575, p<0.02) and rainy days (r, = -0.623,
p<0.0l).

Availability of Microtus arvalıs increased from spring to autumn (Tab. 2) but monthly
variation of the trap-night index did not correlate with vole occurrence in the marten diet.
Insect abundance showed an increase in summer, significantly correlated with occurrence
of this category (Briere r, = 0.777, p< 0.005 df = 12, Grand-Lieu r, = 0.947, p< 0.001 di =
16). Fruit availability was greatest from July to October, associated with the fruit
frequency in the diet (Briere r, = 0.894, Grand-Lieu r, = 0.848, p< 0.001).

Feeding habits of the Stone marten and environmental factors in western France 9

Table 2. Mean seasonal variations in availability index of three main food resources at two sites in
western France

Briere Grand-Lieu

Spring Summer Autumn Winter Spring Summer Autumn Winter

Microtus 52 312 52.0 41.6 10.4 93.2 Sl 26.0
Fruit 0.0 16.3 2246 . 0.0 19.2 31:8 0.0
Insects 1197 67.0 36.0 5 Le) 67.2 9087, 0.0

Characteristics of feeding habits of Martes foina in western France particularly illustrated
the trophic opportunism of this species. The diet showed a strong seasonality with a
winter/spring diet based on Microtus alternating wıth a summer/autumn exploitation of
insects and fruit. Not surprisingly, the diversity index reached ıts lowest level ın winter and
increased in summer, according to the increase of food availabilities. Close relationships
found between abiotic variables, food availabilities and dietary variations emphasized the
decisive influence of environmental factors on feeding ecology of the stone marten.

References

AMoRESs, F. (1980): Feeding habits of the Stone marten, Martes foina (Erxl., 1777) ın south western
Spain. Säugetierkdl. Mitt. 28, 316-322.

CHALINE, ]J.; Baupvin, H.; JAMMOT, D.; SAINT-GIRoONS, M. C. (1974): Les proies des rapaces. Paris:
Doin.

CHOTOLCHU, N.; STUBBE, M.; Dawaa, N. (1980): Der Steinmarder Martes foina (Erxleben, 1777), ın
der Mongolei. Acta Theriol. 25, 105-114.

CLEMENT, R.; SAINT-GIRONS, M. C. (1982): Note sur les mammiferes de France. XVIII. Le regime de
la fouine, Martes foina (Erxleben, 1777) dans l’agglomeration nantaise et en milieu rural.
Mammalıa 46, 550-553.

Day, M. G. (1966): Identification of hair and feather remains in the gut and faeces of stoats and
weasels. J. Zool. (London) 148, 210-217.

DEBROT, $.; Fivaz, G.; MERMOD, C.; WEBER, J. M. (1982): Atlas des poils de mammiferes d’Europe.
Neuchatel: Institut Zool.

Derises, M. (1978): Feeding habits of the stone marten Martes foına (Erxleben, 1777), ın Northern
Burgos, Spain. Z. Säugetierkunde 48, 282-288.

Horısova, V.; ORBTEL, R. (1982): Scat analytıcal data of the diet of urban stone martens, Martes foina
(Mustelidae, Mammalia). Folia Zool. 31, 21-30.

KALPpErs, J. (1983): Contribution & l’etude Eco-Ethologique de la fouine (Martes foina): strategies
d’utilisation du domaine vital et des ressources alimentaires. I. Introduction generale et analyse du
regime alimentaire. Cahiers Ethol. Appl. 3, 145-163.

LEBERRE, J. R. (1969): Les methodes de piegeage des invertebres. In: Problemes d’ecologie, l’Echantil-
lonnage des peuplements des milieux terrestres. Ed. by M. LAMOTTE and F. BOURLIERE. Paris:
Mason. Pp. 54-96.

Lone, T. (1991): Exploitation des milieux et organisation de l’espace chez deux mustelides europeens,
la fouine Martes foina et le putois Mustela putorius. Vie et Milieu 41, 29-38.

Rasmussen, A. M.; Mansen, A. B. (1985): The diet of the Stone marten Martes foina in Denmark.
Natura Jutlandica 21, 141-144.

ROMANOWSKT, J.; LESINSKI, G. (1991): A note on the diet of stone marten in southeastern Romania.
Acta Theriol. 36, 201-204.

SKIRNISSON, K. (1986): Untersuchungen zum Raum-Zeit-System freilebender Steinmarder (Martes
foina Erxleben, 1777). Beitr. Wildbiol. 6, 1-200.

TESTER, U. (1986): Vergleichende Nahrungsuntersuchung beim Steinmarder Martes foina (Erxleben,
1777) ın großstädtischem und ländlichem Habitat. Säugetierkdl. Mitt. 33, 37-52.

WAECHTER, A. (1975): Ecologie de la Fouine en Alsace. Rev. Ecol. (Terre Vie) 29, 399-457.

Author’s address: Dr. THIERRY LoDE, Laboratoire d’ethologie, Universite de Rennes I, F-35042
Rennes, France

MESTEINEUNIGENEDEREGESEBESCLENEHT

Tagung der Tierschutzkommission der DGS am Institut für Haustierkunde
der Christian-Albrechts-Universität in Kiel

Am 24. Aprıl 1994 fand in Kiel das erste Treffen der neugegründeten Tierschutzkommis-
sıon statt. Wichtiger Top dieser Sitzung war die Erarbeitung des Entwurfs einer Evalu-
ierung der Eingriffe an Tieren in bezug auf ihren Schweregrad. Dieser Entwurf soll
insbesondere für diejenigen DGS-Mitglieder eine Hilfestellung sein, die im Rahmen ihrer
Feldforschung etwa Wildtiere fangen müssen (zum Zwecke der Markierung etc.).

Ein Konzept ın Form eines Merkblattes soll unter Berücksichtigung des momentanen
Standes der Tierschutzgesetzgebung Richtlinien bezüglich der Einordnung verschiedener
Forschungsvorhaben geben (anmeldeptlichtiger, genehmigungspflichtiger Tierversuch
bzw. kein Tierversuch) und auf der Mitgliederversammlung in Wien 1994 zur Diskussion
gestellt werden.

Tierschutzkommission der DGS:

Dr. DorIT FEDDERSEN-PETERSEN, Institut für Haustierkunde, Christian-Albrechts-Uni-
versität, Olshausenstr. 40, D-24118 Kiel, Tel. (0431) 880-4506, Fax (04 31) 880-13 89
(Vorsitz).

Prof. Dr. CHRISTIAN WELKER, Gesamthochschule Kassel, Zoologie und vergl. Anatomie,
Heinrich-Plett-Str. 40, D-34132 Kassel, Tel. (0561) 804 46.04.

Dr. AnDREAs HAEMISscH, Medizinische Hochschule Hannover, Zentrales Tierlabor,
D-30623 Hannover, Tel. (0511) 5323747, Fax (05 11) 53237 10.

Dr. DIETER ZscHEILE, Waldschulenweg 1, D-19061 Schwerin, Tel. (0385) 213301, Fax
(03 85) 2133.00.

Tagungsankündigungen

1. Fifth International Conference “Rodents and Spatium”, Biodiversity and Adaptatıon.
20.-24. März 1995 in Rabat, Marokko, Institut Agronomique et Veterinaire Hassan II.
Auskünfte: Dr. ABELKADER ZAIME, Convenor, Conference Rodents and Space, Depar-
tement de Physiologie Anımale et Therapeutique, I.A.V. Hassan II, B.P. 6445 Rabat-
Instituts, MA-10101 Rabat, Morocco. Fax (212-7) 777119 oder 775838.

2. VIII Curso Intensivo: 17. Oktober-22. Dezember 1994: Master Internacional en enfer-
medades Parasitarias Tropicales.

Programas Erasmus y Comett de la UE

Auskünfte: Universitat de Valencia, Departamento de Parasıtologia, Fakultad de
Farmacıia. Av. Vicent Andres Estelles s/n, E-46100 Burjassot, Valencia, Espana, UE.
Fax 34 63 86 47 69.

Erscheinungsweise und Bezugspreis 1994: 6 Hefte bilden einen Band. Jahresabonnement Inland:
DM 378,- zuzüglich DM 13,80 Versandkosten; Jahresabonnement Österreich: 65 2949,— zuzüg-
lich öS 164,- Versandkosten; Jahresabonnement Schweiz: sfr 364,— zuzüglich sfr 21,— Versand-
kosten; Jahresabonnement EG-Binnenmarkt-Länder mit USt-ID-Nr.: DM 353,27 zuzüglich
DM 19,63 Versandkosten; Jahresabonnement EG-Binnenmarkt-Länder ohne USt-ID-Nr. und
Drittländer: DM 378,- zuzüglich DM 21,- Versandkosten. Das Abonnement wird zum Jahres-
anfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens
am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhand-
lung oder bei der Verlagsbuchhandlung Paul Parey GmbH & Co. KG, Spitalerstraße 12,
D-20095 Hamburg, Bundesrepublik Deutschland, bestellt werden. Die Mitglieder der „Deut-
schen Gesellschaft für Säugetierkunde“ erhalten die Zeitschrift unberechnet im Rahmen des
Mitgliedsbeitrages.

Z. Säugetierkunde 59 (1994) 3, 129-192

(4), 193-256, August 1994 ISSN 0044-3468 C 21274 F

ITSCHRIFT FÜR
SAÄUGETIERKUNDE

INTERNATIONAL JOURNAL
OF MAMMALIAN BIOLOGY

|

Organ der Deutschen Gesellschaft für Säugetierkunde

|

'Canova, L.; Maistrello, Lara; Emiliani, D.: Comparative ecology of the Wood mouse Apodemus sylvaticus in two
differing habitats. — Vergleichende ökologische Untersuchungen an Waldmäusen, Apodemus sylvaticus, in zwei

unterschiedlichen Lebensräumen 193
Stubbe, Annegret; Wiegand, Sabine: Ontogenesis of pelage and the course of moulting in Microtus brandti (Radde,

1861). -— Ontogenese des Haarkleides und Fellwechselverlauf von Microtus brandti (Radde, 1861) 199
Brooks, J. E.; Ahmad, E.; Hussain, I.: Reproductive biology and population structure of Rattus rattus in Rawalpindi,

Pakistan. — Reproduktionsbiologie und Populationsstruktur von Rattus rattus in Rawalpindi, Pakistan 209
Christian, S. F.: Dispersal and other inter-group movements in badgers, Meles meles. — Ausbreitung und andere

Ortsbewegungen zwischen Gruppen von Dachsen, Meles meles 218

'Löpez-Luna, Pilar; Arevalo, F.; Burgos, M. J.; Del Hoyo, N.: Lipid deposits in pregnant and non-pregnant bats
(Pipistrellus pipistrellus). — Fettspeicher in trächtigen und nicht trächtigen Fledermäusen (Pipistrellus pipistrellus) 224

Ruckstuhl, K.; Ingold, P.: On the suckling behaviour of Alpine chamois Rupicapra rupicapra rupicapra. — Zum

Saugverhalten bei der Alpengemse Rupicapra rupicapra rupicapra 230
Vassart, M.; Granjon, L.; Greth, A.; Catzeflis, F. M.: Genetic relationships of some Gazella species: an allozyme

survey. — Genetische Verwandtschaft einiger Gazella-Arten: Eine Allozym-Untersuchung 236
Plug, Ina: Springbok, Antidorcas marsupialis (Zimmerman, 1780) from the past. — Springböcke, Antidorcas

marsupialis (Zimmerman, 1780), aus vergangenen Zeiten 246

Wissenschaftliche Kurzmitteilung

Mustrangi, Meika A.: Marmosops scapulatus Burmeister, 1856, and the brown mutation in didelphids (Marsupialia).
—- Marmosops scapulatus Burmeister, 1856, und die braune Mutation bei Didelphiden (Marsupialia) 252

Buchbesprechungen 255

ZRIITHSNAT
wi NOV “Ay,

MAY I 7 1995

OnAKILO

Verlag Paul Parey Hamburg

IIERAUISGEBIERGZEDIENORS

P. J. H. van BreEE, Amsterdam — W. FIEDLER, Wien — H. Frick, München - G. B.

HARTL, Wien -— W. HERRE, Kiel - R. HUTTERER, Bonn - H.-G. Krös, Berlin - H.-].

Kuhn, Göttingen -— E. Kurzer, Tübingen -— W. MAIER, Tübingen — J. NIETHAMMER,

Bonn - ©. Anne E. Rasa, Bonn - H. ReıcHstEin, Kiel -— M. Rönrs, Hannover —

H. SCHLIEMAnN, Hamburg — D. STARCK, Frankfurt a. M. - E. THEnıus, Wien - P. Vo-
GEL, Lausanne - H. Wınkıng, Lübeck

SCHRIETLETTUNGY/EDTEORTAT- OFEIGE

D. Kruska, Kiel - P. LANGER, Gießen

This journal is covered by Biosciences Information Service of Biological Abstracts, and by Current Con-
tents (Series Agriculture, Biology, and Environmental Sciences) of Institute for Scientific Information

Die Zeitschrift für Säugetierkunde veröffentlicht Originalarbeiten und wissenschaftliche Kurzmittei-
lungen aus dem Gesamtgebiet der Säugetierkunde, Besprechungen der wichtigsten internationalen
Literatur sowie die Bekanntmachungen der Deutschen Gesellschaft für Säugetierkunde. Verantwort-
licher Schriftleiter im Sinne des Hamburgischen Pressegesetzes ist Prof. Dr. Dieter Kruska.

Zusätzlich erscheint einmal im Jahr ein Heft mit den Abstracts der Vorträge, die auf der jeweiligen
Hauptversammlung der Deutschen Gesellschaft für Säugetierkunde gehalten werden. Sie werden als
Supplement dem betreffenden Jahrgang der Zeitschrift zugeordnet. Verantwortlich für ihren Inhalt
sind ausschließlich die Autoren der Abstracts.

Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z. Hd. Prof. Dr. Dieter
Kruska, Institut für Haustierkunde, Biologiezentrum der Christian-Albrechts-Universität, Am
Botanischen Garten 9, D-24118 Kiel, Bundesrepublik Deutschland. Für die Publikation vorgese-
hene Manuskripte sollen gemäß den „Redaktionellen Richtlinien“ abgefaßt werden. In ihnen
finden sich weitere Hinweise zur Annahme von Manuskripten, Bedingungen für die Veröffent-
lichung und die Drucklegung, ferner Richtlinien für die Abfassung eines Abstracts und eine

Korrekturzeichentabelle. Die Richtlinien sind auf Anfrage bei der Schriftleitung und dem Verlag
erhältlich.

Sonderdrucke: Anstelle einer Unkostenvergütung erhalten die Verfasser von Originalbeiträgen und
Wissenschaftlichen Kurzmitteilungen 50 unberechnete Sonderdrucke. Mehrbedarf steht gegen
Berechnung zur Verfügung, jedoch muß die Bestellung spätestens mit der Rücksendung der
Korrekturfahnen erfolgen.

Vorbehalt aller Rechte: Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich
geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks,
des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der
Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei-
ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch ım Einzelfall grund-
sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines
Beitrages ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber-
rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 ın der Fassung vom
24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen
den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sınd
mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen.

Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of
the fırst page of an article in this journal indicates the copyright owner’s consent that copies of the article may
be made for personal or internal use, or for the personal or internaluse of specific clients. This consent ıs given
on the condition, however, that the copier pay the stated percopy fee through the Copyright Clearance
Center, Inc., 21 Congress Street, Salem, MA 01970, USA, for copying beyond that permitted by Sections 107
or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying
for general distribution, for advertising or promotional purposes, for creating new collective, or for resale.
For copying from back volumes of this journal see “Permissions to Photo-Copy: Publisher’s Fee List” of the
SEC,

Fortsetzung 3. Umschlagseite

© 1994 Paul Parey. Verlag: Paul Parey GmbH & Co. KG, Hamburg. Anschrift: Spitalerstr. 12, D-20095 Hamburg,
Bundesrepublik Deutschland.
Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., Heide/Holst.

Z. Säugetierkunde 59 (1994) 193-198
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Comparative ecology of the Wood mouse Apodemus sylvaticus
in two differing habitats

By L. CanovA, LARA MAISTRELLO, and D. EMILIANI

Dipartimento di Biologia Animale, Universita di Pavia, Pavia and Dipartimento di Biologia
Evolutiva, Universita dı Ferrara, Ferrara, Italia

Receipt of Ms. 27.7. 1993
Acceptance of Ms. 28. 10. 1993

Abstract

Densities, variations in body mass, sex ratios, breeding activities and size of home-range were studied
in wood mouse populations living in a woodland and areed bed. In the reed bed population: 1. spring
densities were lower, 2. body mass differed, and was lower in early autumn, and 3. home ranges were
larger than in the woodland population.

It is suggested that differences in food availability and quality strongly influenced the behaviour of
the two populations; the main effect of these differences was that reed bed mice entered winter with a
lower body mass and suffered higher winter mortality. The larger size of the home range in the reed
bed suggests that mice living in a poor-food habitat will enhance their survival by patrolling a wider
area than in a rich habıtat.

Introduction

Habitat features such as quality and structure, spatial heterogeneity and some related
environmental parameters (availability of food and shelter and of suitable nest sites) may
influence certain aspects of intraspecific rodent behaviour (NIETHAMMER 1978; HESTBECK
1982; WIGER 1982; GURNELL 1985; GoORMAN and AHmaD 1993). Populations living in
differing habitats may be characterized by demographic parameters such as fecundity and
mortality (KrROHNnE 1980), spacing behaviour in females (voles: BujaLskA and JANION
1981) and juvenile dispersal between high- and low-quality habitats (voles: GLiwıcz 1989).
It is also suspected that habitat heterogeneity may partially explain the observed asyn-
chronous fluctuation in neighbouring populations of noncyclic voles (ALIBHAT and Gipps
1985).

The aım of this study is to investigate wood mouse populations lıving in two differing
habıtats that are characterized by different productivity and structure: a mesıc woodland

and a reed bed.

Material and methods

The study was performed at the “Punte Alberete” and the “Pineta di S. Vitale” Nature Reserves,
located along the eastern Adriatic coast of Italy. “Punte Alberete” is a marshland covered with
extensive reed beds Phragmites sp. and Carex sp. The “Pineta di S. Vitale” biotope is a mixed spruce
and deciduous woodland. The general features of the two areas, which we shall call “reed bed” and
“woodland”, differed strongly: arboreal cover was obviously high in the woodland, whereas in the
reed bed the ground cover increased continuously from May to October in relation to reed growth.
The reed bed was constantly characterızed by wet soils, while in the woodland the water table
decreased from May to an August minimum; the water table may sometimes rise in response to
meteorological factors such as storms or rain, but usually lowers rapidly.

The study was carried out from May to October 1992. A grid of 52 Sherman traps (trapped area =
1 ha) was located ın each study area and one trapping session was carried out for five days every
month at each study site. The traps were baited with sunflower seeds and checked at dusk and dawn;
after capture mice were weighed (g), sexed and marked by toe clipping (Twıcg 1978). Each individual
was aged as adult, subadult or juvenile according to fur colour and reproductive condition (GURNELL

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5904-0193 $ 02.50/0

194 L. Canova, Lara Maistrello and D. Emiliani

and FLOWERDEwW 1990). Reproductive conditions in males were detected by the degree of testes
activity (1.e. their position in the scrotum); reproductive status for female was described as pregnant,
lactating, pregnant and lactating, reproductively active (perforate vagina) or inactive (imperforate
vagına).

Population abundance and home range size were estimated by the Minimum Number Alive and
Minımum Area Methods (GuURNELL and FLOWERDEW 1990).

Results
Population abundance and sex ratio

Seventy nine individuals were caught 172 times in the reed bed and 91 individuals were
caught 187 times ın woodland; the trapping effort was equal to 3120 trap checks per
habıtat.

Densities increased in early autumn in both habitats (Fig. 1). Throughout the study
period the woodland population density was consistently higher than that of the reed bed
(Mann-Whitney, U, z = 2.22, P = 0.026); however, the differences were only very marked
in May and June. After June, the woodland population decreased to a minimum, while the
reed bed population remained at the density attained in the late spring.

40

Mm REEDBED
e \WOODLAND

30

20

MINIMUM NUMBER ALIVE/ha

May June July August September October

Fig. 1. Population densities in the two habitats

The adult sex ratio was biassed towards males in nearly all the months and ın both
habitats (Fig. 2); a slight prevalence of females was observed during June in the reed bed
and in May in the woodland; only in July did the sex ratio approach parity.

Body mass variation and breeding

During the study period, variation in body mass was observed between the two popula-
tions. The weight of wood mice living in the reed bed increased rapidly from May to June
and then decreased constantly up to October, whereas the woodland mice gained weight
more slowly and reached their highest weight in August. The maximum weight of
woodland mice was lower than that of reed bed mice and weight differences were always
significant except during the July trappıng session (Fig. 3).

Comparative ecology of the Wood mouse in two dıiffering habitats 195

= REED BED
e WOODLAND

SEX RATIO (male:female)

M J J A S O

Fig. 2. Male:female sex-ratio, expressed as the logarıthm of the ratio. Positive values show a male
biassed ratio

40
SIREEDIBEB
39 © WOODLAND
= |
=EE ! 8
®) 3 ‘
AT 20 |
S
=
<
LI] 10
=
0} -
M J J A S (6)
Fig. 3. Average weight (+/- SE) variations during the study period; differences were tested by Mann-
Whitney U test. (* = P<0.05, ** = P<0.001, *** = P<0.001, ns = not significant)

At the beginning of the study nearly all the males were in sexual activity in both habıtats
(Fig.4); however, in October nearly all of them entered a non-reproductive phase
(abdominal testes) ın the reed bed, whereas more than half the woodland males were still in
a reproductive condition (scrotal testes).

A sımilar pattern was observed for females (Fig. 4): in the reed bed the proportion of
females in a reproductive condition (perforate vagina, lactating or pregnant) was high from
May to September, while in October nearly all the females were found to be ın a non-
reproductive condition (imperforate vagina); only one female showed the final phases of
lactation activity. In the woodland, on the other hand, nearly all the females were
imperforate in May and June; in October, however, ten out of fourteen females captured
were pregnant or lactating.

Home-range size

Relative areas of home ranges by sex and habıitat are given in the table. Because of the lack
of recapture data for each trapping session we considered the average value for the home

196 L. Canova, Lara Maistrello and D. Emiliani

REED BED

oO

PROPORTION OF TOTAL CAPTURES

MALES FEMALES
NON-BREEDING ADULT EJ] BREEDING ADULT BE JUVENILE

Fig. 4. Proportion of breeding and non-breeding males and females during the study period, expressed
as the proportion of monthly captures

range throughout the study period. The average home-range size of reed bed mice was
much higher than ın woodland mice (male, Mann U, z = -4.31, P<0.001; female, Mann
U, z = -3.52, P<0.01; from data in the table). In the reed bed the home range was
sıgnificantly larger for males than for females (Mann U, z = -2.58, P< 0.01), whereas the
home range of females was slightly larger than that of males in woodland. This latter
difference, however, was not statistically significant (Mann U, z = -0.07, P = 0.93).

Home range sizes (m? + SD) in the woodland and the reed bed
Differences are tested by Mann-Whitney U test

Reed bed Woodland

Males Females Males Females
(8070 + 478) (6670 + 978) (4080 + 401) (4120 + 401)

Discussion

The population trend in the two habitats follows approximately the same pattern and fits
with current information concerning the reproductive biology of temperate woodland
rodents (FLOWERDEW 1985). The autumn increases may be explained by recruitment of
juveniles or subadults to the adult population. A male-biased sex ratio ın both habıtats ıs a
common finding in capture-recapture studies of wood mice (FLOWERDEW 1985; HALLE
1993), probably reflecting the polygynous social system of this rodent (WoLTon and
FLOWERDEW 1985). Several studies have shown that male home ranges overlapped those ot
more than one female (BRown 1969; WoLron and FLOWERDEW 1985). In this way males
improve their chance of encountering sexually active females (BRown 1966). Moreover, ın

Comparative ecology of the Wood mouse in two differing habitats 197.

covering a wider area they should encounter more traps compared to females and thıs
should explain their prevalence in capture-recapture data sets.

The data from our study areas fit, at least for the reed bed, current information about
sex-related differences in wood mouse home-range size.

The temporal pattern of body mass variation seems to fit the current picture of weight
variation in wood mice living in temperate areas (TAnton 1969; GRODSZINSKY 1985).
According to GRODSZINSKY (1985), seasonal changes in body weight are related to the
dynamics of water, fat and protein content; and even though water loss and protein
transformation are more involved in seasonal weight decreases than fat, the individuals that
lost most weight were subject to higher winter mortality (Gropszınsky 1985). The
outcome of our study may be, in conclusion, summed up as follows: in the reed bed - a
habitat that may offer a lower availability of winter food since trees (and abundant seed
crops) are not present - the spring density was lower, weight increased earlier, reproduc-
tive activity stopped earlier in both sexes, and male and female home-range sizes were
larger than in the woodland population. In the woodland, probably as a response to better
food availability, wood mice went into winter with a higher weight, stopped their
reproductive activity later, maintained smaller home ranges, and reached the following
spring with a higher density than the reed bed population. These last data may reflect a
differential winter mortality between populations, since the October densities were not
significantly different.

The differences in home-range size support the hypothesis that food availabilıty
differed between the habitats. According to a recent study (GoRMAN and AHMmAD 1993),
differences ın the home-range size of mice living in contrasting habiıtats are related, at least
in part, to differences ın food quality, availabıliıty and dispersion. Differences in food
availability and quality may be then an important limiting factor for the reed bed wood
mice; their main effect ıs that the mice will enter winter with a lower body weight and will
suffer higher winter mortality. Their larger home-range size suggests that mice living in a
poor-food habitat will enhance their survival by patrolling a wider area than in a rıch

habıtat.
Acknowledgements

Thanks are due to Dr. S. VıstoLı (Assessorato Ambiente; Comune di Ravenna) and to C. LAZZARI
(Cooperativa ARCA) for permitting access to the Nature Reserve. The study was carried out with
personal funds.

Zusammenfassung

Vergleichende ökologische Untersuchungen an Waldmäusen, Apodemus sylvaticus, in zwei
unterschiedlichen Lebensräumen

Einige ökologische Parameter der Waldmaus wurden in zwei unterschiedlichen Lebensräumen
ermittelt: in einem Nadelmischwald und in einem Marschgebiet mit Phragmites- und Carex-Bewuchs.
Die Marschpopulation zeichnete sich gegenüber der Waldpopulation durch folgende Unterschiede
aus: 1. Die Populationsdichte war ım Frühjahr deutlich niedriger, 2. die Körpergewichte waren im
frühen Herbst geringer, 3. die mittleren individuellen Aktionsräume waren größer. Daraus wird
geschlossen, daß den Marschwaldmäusen weniger Nahrung bei gleichzeitig geringerer Qualität zur
Verfügung stand. Die Waldmäuse in der Marsch gehen mit einem geringeren Körpergewicht in den
Winter, sie unterliegen gleichzeitig einer höheren Wintersterblichkeit, was eine geringere Frühjahrs-
Populationsdichte zur Folge hat. Die größeren individuellen Aktionsräume in der Marsch werden mit
knapperem Nahrungsangebot in Zusammenhang gebracht.

References

ALIBHAI, $S. K.; Gipps, J. H. W. (1985): The population dynamics of bank voles. Symp. Zool. Soc.
London 55, 277-305.

Brown, L. E. (1966): Home range and movement of small mammal. Symp. Zool. Soc. London 18,
111-142.

198 L. Canova, Lara Maistrello and D. Emiliani

Brown, L. E. (1969): Field experiments on the movements of Apodemus sylvaticus L. using trapping
and tracking techniques. Oecologia 2, 198-222.

BuJALSKA, G.; JAnIoN, S. M. (1981): Bank vole response to an increase in environmental capacity.
Bull. Acad. Pol. Sci. 26, 129-133.

FLOWERDEW, J. R. (1985): The population dynamics of wood mice and yellow-necked mice. Symp.
Zool. Soc. London 55, 315-332.

Griwicz, J. (1989): Individual and populations of the bank vole in optimal, suboptimal and insular
habitat. J. Anım. Ecol. 52, 237-247.

GORMAN, M. L.; AHmap, Z. A. B. (1993): A comparative study of the ecology of wood mice
Apodemus sylvaticus in two contrasting habitats - Deciduous Woodland and Marıtime sand dunes.
J. Zool. London 229, 385-396.

GRODZINSKI, W. (1985): Ecological energetic of bank vole and wood mice. Symp. Zool. Soc. Lond.
55, 169-188.

GURNELL, J. (1985): Woodland rodent communities. Symp. Zool. Soc. London 55, 377402.

GURNELL, J.; FLOWERDEW, J. R. (1990): Live trapping small mammals. A practical guide. Occ. Publ.
Mammal Soc. London 3, 1-39.

HALLE, S. (1993): Wood mice Apodemus sylvaticus as pioneers of recolonization in a reclaimed area.
Oecologia 94, 120-127.

HESTBECK, J. B. (1982): Population regulation of cyclic mammals: the social fence hypothesis. Oikos
39, 157-163. E

KROHNE, D. T. (1980): Intraspecific litter size varıation in Microtus calıfornicus. Variation among
populations. Evolution 34, 1174-1182.

NIETHAMMER, J. (1978): Apodemus sylvaticus (Linneus, 1758) -— Waldmaus. In: Handbuch der
Säugetiere Europas. Ed. by J. NIETHAMMER and F. Krapp. Wiesbaden: Akademische Verlagsges.
Vol.1, 337-358.

Tanton, M. T. (1969): The estimation of biology of populations of the bank vole Clethrionomys
glareolus and wood mouse Apodemus sylvaticus. J. Anım. Ecol. 191, 413—418.

Twıcc, G. 1. (1978): Marking mammals by tissue removal. In: Animal marking: recognition marking
of anımals ın research. Ed. by B. STONEHOUSE. London: The MacMillan Press. Pp. 109-118.
WIGER, R. (1982): Roles of self-regulatory mechanisms in cyclic populations of Clethrionomys with

special reference to ©. glareolus: a hypothesis. Oikos 38, 60-71.

WOLTOoN, R. J.; FLOWERDEW, J. R. (1985): Spatial distribution and movements of wood mouse, yellow

necked mouse and bank voles. Symp. zool. Soc. London 55, 249-275.

Authors’ addresses: Luca Canova, Dipartimento dı Biologia Anımale, Universitä dı Pavıa, I-27100
Pavia; LARA MAISTRELLO, Dipartimento dı Biologia Anımale, Universita di
Ferrara, I-44100, Ferrara; DAVIDE EMILIANI, cooperativa Arca, 1-44030 Ravenna,
Italia

Z. Säugetierkunde 59 (1994) 199-208
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Ontogenesis of pelage and the course of moulting in
Microtus brandti (Radde, 1861)

By ANNEGRET STUBBE and SABINE WIEGAND

Institute of Zoology, Martin-Luther-University, Halle/Saale, Germany

Receipt of Ms. 29. 1. 1993
Acceptance of Ms. 24. 1. 1994

Abstract

Two methods are used in the investigation of the development of fur and moulting in small mammals.
The most common way is to observe the pigmentation of the skin. We have used a more recent
method, staining the whole anımal with standard hair colours and observing changes ın pelage
patterns. It is thus possible to investigate ontogenetic changes ın voles, and also the influences of
different conditions of photoperiod and temperature on moulting. Contrary to the information in the
literature, we found that newborns of M. brandti are pilose. During ontogenesis the following moults
take place: I. Moult into the second immature pelage, II. moult into the first mature pelage, II.
seasonal moultings (spring and autumn). Moults I. and II. occur, depending on the age of the anımal.

Like other Arvicolidae M. brandti shows a sublateral course of moulting. The moult of the ventral
body sıde is completed before the dorsal side. Deviations from this scheme, so-called “moulting
variants”, are possible.

Within one population of M. brandti the spring generation passes through five moults and the
autumn generation through only four in the first year of life. Because of the shortened duration of
moulting during low temperatures the first moults of the autumn generation are faster than those of
spring-born anımals.

These results represent a preliminary attempt to clarıfy conflicting observations on moulting from
field investigations of small mammals, involving, in most cases, anımals of different age classes and
generations from one population.

Introduction

Investigations concerned with moulting are rare. At the beginning of this century, studies
of mammalian pelt were undertaken mostly because of the commercial use of hunted game
furs. Initial results were obtained from mammals that changed their fur colour. Later, the
connection between moult and the state of pigmentation of the inner surface of skin was
discovered. Continuous observations of hair changes in living anımals are still rare, but
they still offer the best opportunity for investigating the postnatal ontogenesis of moulting
under different environmental conditions. We examined the moulting processes in the vole
Microtus brandti, because of their ease in keeping, their highly developed social behaviour
and our rather good knowledge of this species. The aim of the present study was to
establısh a general model of the moulting processes in small rodents, especially Ar-
vicolidae.

Material and methods

Species and maintenance conditions (MC)

The vole Microtus brandti ıs a very social anımal, living in large colonies and often exhibiting cyclic
fluctuations and gradiations in its abundance. The species occurs throughout the steppes of Central
Asıa and has extended its distributional area eastwards in recent years. The vole shows a high
reproductive potential and, as a diurnally active herbivore, is regarded as the most important
competitor of pasture anımals.

In our laboratory, voles were bred in a HAN-rotation system for rigorous outbreeding (Rapp
1982) and were kept under standard environmental conditions (21 # 2°C, 50-60 % humidity, L:D =

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5904-0199 $ 02.50/0

200 Annegret Stubbe and Sabine Wiegand

14:10, Lon 4.00 a.m., light intensity = 300-400 Lux). After weaning at an age of 21 days, the anımals
were kept in sibling groups. For these investigations voles were maintained under different conditions
(MC):

MC 1: 21 # 2°C, L:D = 14:10 (L:D, T = const.)

MC 2: 20 + 2°C, L:D adapted to the natural photoperiod (L:D-, T = const.)

MC 3: T and L:D adapted to the natural changing conditions (L: D, T-)
MCA4: 20 # 2°C, LL (light throughout the day).

Feeding was the same for all animals: standard pellets and water ad lıb., plus fresh food (apples,
carrots, cabbage, Taraxacum) three times each week.

Staining and pelt investigations

To investigate moulting we used a treatment modified from MiıLiTzEr (1989). Animals were
anesthetized by using a 1.43 % solution of sodiumhexobarbital and then stained with standard hair
colouring (LONDANCOLOR, black). Following completition of one moult, the voles were re-stained.
Changes in pelage patterns were recorded weekly. For each MC, twenty animals were investigated
over a timeperiod of one year. We distinguished between a spring generation (animals born in spring)
and an autumn generation (animals born in late summer or autumn). To complete these studies, we
observed the skin pigmentation of dead animals from the laboratory, and also from those caught
between 1988 and 1990 in their natural environment near Ulan-Bator (Mongolıa).

Results
Genesis of neonatal pelage

The following results are based on observations of 22 anımals born in three litters. Voles
are born with eyes and ears closed, and only the vibrissae are visible without the
microscope; they seem to be naked. Contrary to the opinion in the relevant literature, the
whole body excluding the soles ıs covered with short, fine haırs after birth (Fig. 1, 10 mın.
after birth, umbilical cord not bitten through at this time).

Table 1 gives an overview of the genesis of neonatal pelage.

Fig. 1. Skin surface of a newborn Microtus brandti, 10 min. post partum

Postnatal ontogeny of fur development and moulting

After the first juvenile pelage, voles moult into a second juvenile coat and then into the
mature pelage, which may be a summer or a winter fur, depending on date of birth and
season. Spring-born animals pass through five moults, because the first mature coat ıs a
summer fur. They change into a winter pelage in autumn, whereas the autumn generation

Ontogenesis of pelage and the course of moulting in Microtus brandti 201
Table 1. Genesis of hair coat

Day ot life Appearance

whole animal covered with downy hair, especially dorsal side and on the back of the

head

DS down to the top of the feet darker, VS more light coloured, vibrissae growing

forefeet pilose

hairs begin to grow in length (2 mm)

hind feet pilose, body hairs 3 mm long

vibrissae 10 mm, hairs on DS 4 mm, with some dark pile hairs in between, VS hairs
3 mm long

portion of pile hairs growing, eyes start to open (up to 11"” day of life)

vibrissae 13 mm, DS with underwool
eyes and ears open, VS with good visible underwool, DS with longer pile hairs
VS hairs 5 mm long, vibrissae 15 mm long, pile haırs rising above the coat

development of first juvenile pelage (development of new hairs) is complete, after 15"
day no skin pigmentation was found, hairs only growing in length up to 23'“ day

vıbrissae 20 mm
vıbrissae 30 mm

DS and VS hair length 5 mm, woolly hairs and guard haırs are fully grown, so pile haırs
are no longer above the rest of the coat

VS = ventral side, DS = dorsal side of anımal.

Table 2. Frequencies of moulting variants

Variants Number Frequency

(%)

Dorsal A 51.64
Dorsal B 4.31
Dorsal C 17.97
Dorsalg;rr 26.08

total 100.00

Ventral A 13.45
Ventral B 54.82
Ventralg;s 3173

total 100.00

moults into a winter coat with the first adult fur in the first year of life. All anımals living
under changing environmental conditions, and born in spring develop a second winter fur.
Thus, the autumn generation undergoes only four moults in the first year.

For better understanding we defined the different courses of moulting as “varıants”.
Excluding the moultings of senescence, Microtus brandti generally shows a synchronous
growing of hairs. To allow systematic observation of moulting we have regarded the dorsal
and ventral body surfaces as independent, although they are both one entity in moulting.
The moult generally begins ventrally and advances over the flanks to the back; therefore it
is completed ventrally earlier than dorsally. This course is defined as sublateral moulting
by Kryırzov (1964). Figure 2 shows the variants of ventral moulting (A - beginning
laterally and forming a band, B - beginning cranially and at the axillae; both ending at the
feet and tail). Figure 3 explains the dorsal variants; all three start at the shoulders and

202 Annegret Stubbe and Sabine Wiegand

FIT
KRFEr,

Fig. 2. Moulting variants of the ventral side, black: old coat, white: new pelage

haunches and end at the tail. Dorsally as well as ventrally the paws moult least or are
excluded from the moult. The tail may also show irregularities, such that we found a
“raccoon-like” marking, although the tail is always completely moulted. These described

moulting variants were observed in all on-
Table 3. Combinations of variants and their togenetic hair changes (excluding senes-

frequencies cence) and, under each maintenance condi-

tion. They seem to be independent of abio-

Variants Number Frequency tic factors or birth date, suggesting they
Dorsal/ventral (n) may be endogenous. Table 2 shows the
recorded dorsal and ventral body side var-

An | iants of moulting and their frequencies. The
B/A variants Dorsalg;sr and Ventralg;r comprise
B/B patterns, which cannot be categorized

C/A
C/B

within the described variants above, and
include the scattered moultings of senes-
cence. Table 3 explains the frequencies of
the observed variant combinations. Through its life time one animal can change from one
moult variant combination to another. We observed that, in most cases, the first moults
show other combinations than the later ones. The following changes occurred:

AA — AB,BB, CA, CB
AB — AA,BB, CA, CB
BB — AA, AB

CA—> CB

CB — AA,AB, CA

Moultings of senescence (scattered moultings)

With increasing age ordered moulting begins to change to scattered patterns. Anımals of
higher age no longer show regular moults (Fig. 4). Normally, such a scattered moulting
pattern was recorded after the first year of life, but ıt can also occur earlier; thus the first
irregular moults can be observed in anımals which are sıx or seven months old. These
diffuse hair changes are included in the seasonal moults and are not separate entities.

Influence of age on the moulting process

Under defined maintenance conditions (MC) the start and the end of the moults from first
to second juvenile pelage and from the second to the mature coat occur very close together
for all anımals. These dates seem to be more strongly correlated with age than the
subsequent (seasonal) moultings. Figure 5 shows the duration of the spring (S) and autumn

Ontogenesis of pelage and the course of moulting in Microtus brandti 203

ERERNT,

Fig. 3. Moulting variants of the dorsal side, black: old coat, white: new pelage

+%%
FEN

Fig. 4. Examples of scattered moulting; above dorsal side, below ventral side

(A) moults of anımals under different maintenance conditions (MC 1, 2, 3). The first hair
changes begin between the 28°" and 37" day, and end between the 37"® and 71'® day of life.
The average duration is twenty days (average between 31'" to 50'* day of life, dark bars
within the white and structured ın figure 5). Comparing the average durations of juvenile
moults within the spring or autumn generation, they are nearly equal under all mainte-
nance conditions. The spring generation simply takes longer to moult into the second
juvenile coat than the autumn-born anımals. Moulting into the mature fur requires the
longest period (35 days for the autumn generation, 50 days for the spring generation). It
starts between the 51° and 111° day of life, and finishes between the 139" and 215" day
(average: start 81 to 120%, end 122” to 158* day). Again, the autumn generation moults
more rapidly than spring-born anımals. The anımals in MC 3 (LD, T-) are the first to
start and finish the hair changes influenced by temperature and photoperiod, whereas the
anımals of MC 1 (L:D, T = const.), without any “zeitgeber”, show the latest dates and the
largest varıation in moulting times. Therefore, the moult from first into second juvenile fur
seems to be very closely connected with a defined age, whereas moulting into the mature
coat ıs intluenced by day length and temperature as are, to a greater extent, the subsequent
seasonal moults.

Table 4 shows the results of varıous authors for the commencement and the end of the
first and second moults during postnatal ontogenesis in Arvicolidae and Muridae (after
BüHnLow 1970, completed).

204 Annegret Stubbe and Sabine Wiegand

Table 4. Commencement and end of moulting into the second juvenile and into the first adult
coats of some Arvicolidae and Muridae

Authors

Species

BECKER
BoRUM
COLLINS
Dry

EcKE and
Kinney
FRANK and
ZIMMERMANN
FULLAGAR
FULLAGAR
KÄSTLE

(1952)

Rattus norvegicus

Mus musculus
Peromyscus maniculatus
Mus musculus

Microtus cahfornicus

Microtus oeconomus
Apodemus sylvaticus
Apodemus flavicollis

Micromys minutus

First into second
juvenile coat

Age (days)
Start End

Second juvenile into

first mature coat

Age (days)

KEMPER Psendomys
novaehollandiae
LANGENSTEIN-

IssEL Pitymys subterraneus
LinzEy and
LinzEy
MAZAK
McMants and
ZURICH
MILITZER
STEIN
SYKORA
VIITALA
STUBBE and
WIEGAND

(1967)
(1962)

Ochrotomys nuttalli
Clethrionomys glareolus

Meriones unguiculatus
Mesocricetus auratus
Microtus arvalıs
Microtus arvalıs
Clethrionomys rufocanus

(1972)
(1987)
(1960)
(1959)
(1981)

(this study) Microtus brandti

MC
ZZ G GGG GG RG GGG GDGLYDLLLCETIEZ, 1
2 ZI
S 2
3
CZ 1
A RE EÄTTELTT N 2
3
20 40 60 80 100 120 140 160 180 200 220 age(d)

MC 1 (L:D,T = constant )
RI MC2 (L:D = adapted to the natural photoperiod, T = constant )
LI MC3 (L:D,T = adapted to the natural changing conditions )

Fig. 5. Start and end of age-dependent juvenile moults; S: spring generation, A: autumn generation

Ontogenesis of pelage and the course of moulting in Microtus brandti 205

Comparative studies on pelts

In addition to the observations on living animals, we have also investigated pelts of dead
animals (N = 1158, MC 1 -.n, = 728, MC 2 - n, = 330, pelts of anımals caught in Mongolıa
- nz = 100).

The results of both methods agree completely, but with stained living anımals it is
possible to follow all moults through the life of one animal and, in the end, we required
about 7000 voles less for the present study by staining the anımals as we had used only the
old method to observe skin pigmentation. Figure 6, for example, shows some typical
patterns of juvenile moulting of killed animals. Hair replacements are visible ın the
epidermis by the pigment production of melanocytes, which were accumulated before the
hair follicles maturate (RypeEr 1973). The colour of the pigmentation ıs therefore extremely
dark in the areas of active hair growth and becomes lighter as the hairs change and growth
ceases (Vıro and KoskeLA 1978). Figure 6 shows the beginning (left), middle and end
phases (right) of the second juvenile moult. The left picture conforms with variant A2, the
middle with A3 and the right wıth A5 from figure 3.

Fig. 6. Typical skin patterns of juvenile moultings; left: start, right: endphase of moult

Discussion

According to many mammalogists, neonates of small rodents are born with a nude body
surface except for the vibrissae, 1 to 2 mm in length (Torpr 1935; NIETHAMMER and
Krapp 1978, 1982). BARE (1981), Sykora (1959) and STEIN (1960) reported on Microtus
arvalıs that the first hair tips appear on the second day of life. For Pitymys subterraneus
LANGENSTEIN-IssEL (1950) stated that, on the day of bırth, 0.5 mm long, light hair tips can
be seen under a magnifier. SCHRÖPFER (1977) found the fırst hair fuzz on the dorsal side on
the third day of life and sinus hairs one day earlier. FRANK and ZIMMERMANN (1956)
observed the first dark hairs in Microtus oeconomus on the second day of life. CoLLıns
(1923) also determined this date for the genus Peromyscus. In Microtus agrestis the hair
colouration is visible on the fourth day of life (NIETHAMMER and Krarp 1982). For
Micromys minutus opinions vary: SLEPsOV (1947) observed the fırst hairs on the fourth day
and KästLe (1953) on the second day of life.

We have studied newborn Microtus brandti 10 minutes after birth, and by use of a
binocular microscope observed sparse, small hairs over the entire body surface, as well as
vibrissae. This differs from the general opinion; only Kourıst (1957) also described

206 Annegret Stubbe and Sabine Wiegand

colourless hairs over the entire body a few hours after birth in Cricetus cricetus. It is our
opinion that the type of hair development noted in Microtus brandti is also common in
other Arvicolidae; we also found it ın Altıcola semicanus, the high mountain vole.

The genesis of neonatal pelage seems to be comparatively uniform; we only observed
individual variations of 1 or 2 days. The representation given may be a general guide to the
genesis of first juvenile pelage, not solely in Microtus brandti. The following authors gave
the same developmental states: LANGENSTEIN-IssEL (1950) delimited the end of develop-
ment of the neonatal coat of Pitymys subterraneus between 12 and 14 days; Bake (1981),
FRANK and ZIMMERMANN (1956), STEIN (1960) and Sykora (1959) cited the 14"? day for
Microtus arvalis and KÄstLe (1953) the 15"? for Micromys minutus.

Hairs have a limited life span. After the duration of the growth process, they die and are
rubbed off during moulting. ToLpT (1935) distinguished between the periodic or seasonal
moults which occur in wıld mammals of cold and temperate zones, and the continuous
(partial) moults of domesticated anımals and humans. According to ToLpr (1935) animals
of tropical and arctic zones, or adapted to aquatic habitats, show only one moult per year.
A companion paper to thıs publication (STUBBE and WIEGAND 1994) will deal with seasonal
moults in Microtus brandtı.

Concerning the course of moulting, synchronous and asynchronous hair growth has
been described. In Microtus brandti we found synchronized hair growth. KryLTzov (1964)
noted sublateral moulting, with only minimal variations, in 18 species of Arvicolidae,
LANGENSTEIN-IssEL (1950) for Pitymys subterraneus, and EckE and Kınney (1956) for
Microtus cahifornicus. In evolutionary terms all these are comparatively “young” species.
Primitive genera change their hair coat according to other patterns; Clethrionomys, ın
particular, shows the so-called cephalo-sacrale type. In Microtus brandti the sublateral
type of moulting was corroborated and described from the investigation of stained lıving
anımals. Ä

The scattered moulting of senescence has been observed by varıous authors: BAKE
(1981) described this for 9- to 15-month-old Microtus arvalıs, CoLLıns (1918) — Peromys-
cus spec.; BECKER (1952) — Rattus norvegicus, EsPANA et al. (1985) - Mus spretus; BÜHLOw
(1970) - Arvicola terrestris; and KryLtTzov (1964) for some Microtus species. OLIVEIRA et
al. (1992) found different pelages in sexually active males and females of Marmosa incana,
but did not note any influence of seasons. ROwSEMITT et al. (1975) found a relationship
between diffuse moulting and reproductive activities, but we have the impression that
gravidity simply accelerates normally occuring moults. Generally, ageing processes were
considered a primary premise of irregular moults. This ıs also our opinion, derived from
observations on anımals kept under MC 4 (T = const., LL), which age more rapidly than
other voles because of light stress, and already show scattered moulting patterns within
5 months.

There is general agreement with regard to the induction of juvenile moults. Our
observations show that the changes from first to second juvenile, and from there to the fırst
mature coat, are not dependent on season, but we did find a close correlation between age
and the start of these moults.

From a comparison of Microtus species a high correspondence of dates is evident, and
the results from Microtus brandti also coincide closely. This high conformity, and the
small possibility of the influence of exogenous factors, indicate the probability of an age-
dependent course in juvenile moults. These results agree with those of LANGENSTEIN-ISSEL
(1950), Steın (1960) and Vırrara (1981). Vırrara (1981) refered to the influence of
exogenous factors (temperature) on the duration of moults, and from keeping anımals
under different temperature regimes in the laboratory. Animals under colder conditions
changed their hair coat faster than anımals in warmer environments.

In our study, we found moult duration to be dependent on birth date (spring or
autumn) and on abiotic factors (day length and temperature). The influence of abiotic

Ontogenesis of pelage and the course of moulting in Microtus brandti 207

factors on juvenile moults, however, is not as strong as in the subsequent adult moults. The
seasonal dependence of mature hair changes will be discussed ın the companıon paper
(STUBBE and WIEGAND 1994).

Acknowledgements

This publication is no. 225 of the “Results of Mongolian-German Biological Expeditions since 1962”.
We thank Dr. H. GriFFITHs from the Department of Genetics of the University of Leeds for
linguistice corrections.

Zusammenfassung

Ontogenese des Haarkleides und Fellwechselverlauf von Microtus brandti (Radde, 1861)

Neben dem üblichen Verfahren, den Fellwechsel bei Kleinsäugern anhand der Pigmentierung der
Hautunterseite abgebalgter Felle zu beurteilen, wurde mit Hilfe der Methode des Einfärbens der Tiere
mit einem herkömmlichen Haarfärbemittel und der wöchentlichen Aufzeichnung der entstehenden
Fellmuster eine Beobachtung des Mausergeschehens am lebenden Tier möglich. Somit konnte nicht
nur der Mauserverlauf während der Ontogenese, sondern auch der Fellwechsel bei Einfluß der
Photoperiodik und von Temperaturänderungen an ausgewählten Tieren über längere Zeiträume
beobachtet werden.

Entgegen der in der Literatur vertretenen Auffassung wurde erstmals eine Behaarung neugebore-
ner Tiere nachgewiesen. Innerhalb der Ontogenese treten folgende Härungen auf: I. Fellwechsel in
das 2. Jugendkleid, II. Fellwechsel ın das 1. Alterskleid, III. saisonale Haarwechsel (Frühjahr und
Herbst). I. und II. verlaufen altersabhängig.

Insgesamt zeigt Microtus brandti wie auch andere Arvicolidae einen sublateralen Mauserverlauf.
Der Fellwechsel der Ventralseite ist eher abgeschlossen als der der Dorsalseite. Abweichungen von
diesem Schema, sogenannte „Mauservarianten“, sind möglich. Innerhalb einer Population durchläuft
die Frühjahrsgeneration fünf, die Herbstgeneration nur vier Haarwechsel im ersten Lebensjahr.

Die vorliegende Studie möchte zur Klärung der sich häufig widersprechenden Beobachtungen des
Fellwechsel bei im Freiland gefangenen Kleinsäugern beitragen, die durch den gleichzeitigen Fang von
Tieren verschiedener Altersklassen und Generationen entstehen können.

References

BAkE, U. (1981): Beitrag zum Aufbau des Haarkleides und zum Haarwechsel der Feldmaus Microtus
arvalıs (Pallas, 1779). Säugetierkdl. Inf. 5, 3-49.

BECKER, K. (1952): Haarwechselstudien an Wanderratten (Rattus norvegicus Erxl.). Biol. Zbl. 71,
626-640.

Borum, K. (1954): Hair pattern and hair succession in the albino mouse. Acta pathol. microbiol.
scand. 34, 521.

Bünrow, E. (1970): Untersuchungen über den Haarwechsel bei Schermäusen, Arvicola terrestris (L.,
1758). Zool. Anz. 184, 18-32.

Corıns, H. H. (1918): Studies of the moult and of artıfıcıally induced regeneration of pelage ın
Peromyscus. J. Exp. Zool. 27, 73-99.

CorLins, H.H. (1923): Studies of the pelage phases and of the nature of the color varıations in mice of
the genus Peromyscus. J. Exp. Zool. 38, 45-107.

Dry, F. W. (1926): The coat of the mouse. J. Genet. 16, 287-340 .

EckE, D.; Kınney, A. (1956): Aging meadow mice, Microtus calıfornicus, by observation of moult
progression. J. Mammalogy 37, 249-254.

EspaNA, M.; PALOMO, M. ]J.; ZAMORANDO, E. (1985): Moulting and hair covering of Mus spretus from
southern Spain. Spixiana 8, 1-16.

FRANK, F.; ZIMMERMANN, K. (1956): Zur Biologie der Nordischen Wühlmaus (Microtus oeconomus
stimmingi Nehring). Z. Säugetierkunde 21, 58-83.

FULLAGAR, P. ]J. (1967): Moult in field mice and the varıation in the chest markings of Apodemus
sylvaticus (Linne, 1758) and Apodemus flavicollis (Melchior, 1854). Säugetierkundl. Mitt. 15,
138-149.

KÄsTLeE, W. (1953): Die Jugendentwicklung der Zwergmaus (Micromys minutus soricinus Her., 1789).
Säugetierkundl. Mitt. 1, 49-59.

KEMPER, C. M. (1976): Maturational and seasonal moult in the New Holland Mouse, Pseudomys
novaehollandiae. Aust. Zool. 19, 9-17.

Kourıist, W. (1957): Das Haarkleid des Hamsters (Cricetus cricetus cricetus Linne, 1758). Wiss. Z.
Univ. Halle, Math.-Nat. VI/3, 413-438.

Kryırrzov, A. I. (1964): Moult topography of Microtinae, other Rodents and Lagomorphs. Z.
Säugetierkunde 29, 1-17.

208 Annegret Stubbe and Sabine Wiegand

LANGENSTEIN-IssEL, B. (1950): Biologische und ökologische Untersuchungen über die Kurzohrmaus
(Pitymys subterraneus DeSelys-Longchamps). Z. Pflanzenbau und Pflanzenschutz 1, 145-183.
Lınzey, D. W.; Limzey, A. V. (1967): Maturational and seasonal molts in the Golden mouse,

Ochrotomys nutallı. J. Mammalogy 48, 236-241.

Mazax, V. (1962): Zur Kenntnis der postnatalen Entwicklung der Rötelmaus, Clethrionomys
glareolus Schreber, 1780. Acta soc. zool. Bohemoslov. 26, 77-104.

McManus, ]J. J.; Zurich, W. M. (1972): Growth, pelage development and maturational moults of
Mongolian gerbil (Meriones unguiculatus). Amer. Midl. Nat. 87, 264-271.

MILITZER, K. (1987): Die Ontogenese der Haarwachstumszyklen beim Goldhamster (Mesocricetus
auratus W.) - makroskopische und histometrische Befunde an 2 Stämmen. Z. Versuchstierk. 29,
181-192.

NIETHAMMER, J.; KrApp, F. (Eds.) (1978): Handbuch der Säugetiere Europas. Vol. 1: Rodentia I.
Wiesbaden: Akad. Verlagsges.

— — (Eds.) (1982): Handbuch der Säugetiere Europas. Vol. 2/I: Rodentia II. Wiesbaden: Akad.
Verlagsges.

OLIVIERA, J. A. De; Lorını, M. L.; Persson, V. G. (1992): Pelage varıation in Marmosa incana
(Didelphidae, Marsupialia) with notes on taxonomy. Z. Säugetierkunde 57, 129-136.

Rapp, K. G. (1982): HAN-Rotation - a new system for rigorous outbreeding. Z. Versuchstierk. 14,
133-142.

RowseEMITT, C.; Kunz, T. H.; Tamarın, R. H. (1975): The timing and patterns of molt in Microtus
breweri. Occ. Papers Mus. Nat. Hist. Univ. Kansas 34, 1-11.

Ryper, M. (1973): Hair. Studies in Biology 41. London: Arnold.

SCHRÖPFFER, R. (1977): Die postnatale Entwicklung der Kleinwühlmaus, Pitymys subterraneus De
Selys-Longchamps, 1836 (Rodentia, Cricetidae). Bonn. zool. Beitr. 28, 249-268.

SLEPTSOv, M. (1947): On the biology of Micromys minutus ussuricus. Ham. Contr. Faune et Flore de
P’USSR, N.S., Sect. Zool. 8, 69-100.

STEIN, G.H. W. (1960): Zum Haarwechsel der Feldmaus (Microtus arvalıs Pallas, 1779) und weiterer
Muroidea. Acta theriol. 3, 2744.

STUBBE, A.; WIEGAND, $. (1994): Intluence of photoperiod and temperature on moulting processes in
Microtus brandti (Radde, 1861). Z. Säugetierkunde (im Druck).

SYKORA, J. (1959): Die postnatale Entwicklung der Feldmaus. Der postnatale Prozess der Wachstums-
veränderungen bei 1 bis 21 Tage alten Tieren. In: Hrabos polni Microtus arvalıs. Ed. by ].
KRrATOcHVIL, F. BALAT, C. FoLk, I. GRULICH, J. HavLin, V. HoLısovA, K. Hupec, J. PELIKAN,
B. Rosıckv, I. Sykora, Z. SEBEK und M. ZAPLETAL. Praha: Naklad. CSAV, Pp. 53-60, 322.

TOEDEHK (955): Aufbau und natürliche Färbung des Haarkleides der Wildsäugetiere. Leipzig:
Dtsch. Ges. Klein- und Pelztierzucht.

VIITALA, J. (1981): Hair growth patterns in the vole Clethrionomys rufocanus (Sund.). Biol. Res. Rep.
Jyväskylä 7, 3-17.

Vıro, P.; KoskeLa, P. (1978): Moult topography, moulting and structure of the fur in the Harvest
mouse. Acta theriol. 23, 503-517.

Authors’ address: Dr. ANNEGRET STUBBE and Dipl.-Biol. SABINE WIEGAND, Institut für Zoologie,
Martin-Luther-Universität Halle-Wittenberg, Domplatz 4, D-06099 Halle/Saale,
FRG

Z. Säugetierkunde 59 (1994) 209-217
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Reproductive biology and population structure of Rattus rattus
in Rawalpindi, Pakistan

By J. E. Brooks, E. AHMAD, and I. Hussaın

Denver Wildlife Research Center, Denver, Colorado, USA and Vertebrate Pest Control Project,
National Agricultural Research Centre, Islamabad, Pakistan

Receipt of Ms. 4. 11. 1993
Acceptance of Ms. 24. 3. 1994

Abstract

Studied the breeding biology and population structure of roof rats (Rattus rattus) in the wholesale
grain and commodities market in Rawalpindi, Pakistan. Rats were trapped from the grain market
monthly for 14 months. We necropsied 2327 rats, comprising 1175 males and 1152 females, essentially
a 1:1 sex ratio. Males were found fertile in every month, with no significant seasonal differences;
females were pregnant in every month, and the adjusted frequency of pregnancy averaged 39.8 %.
Litter size during the last quarter of pregnancy averaged 6.1 + 1.7 Standard Deviation (SD). Based
upon primiparous and multiparous females, we calculated that an average adult female had 1.8
pregnancies. Average production of young was 10.9/female (6.1x 1.8). Immature anımals (weaned,
but not sexually mature) constituted 13.1 % of the total collection, and recruitment was continuous
during the study, indicating high mortality among nestling and weanling rats. Numerous wounds and
scars on adults of both sexes indicated a high degree of social strife and aggression among the rats.

Introduction

The rat commonly found ın grain storage facılities, wholesale commodities markets, cities,
towns, and farm houses in Pakistan ıs the roof rat (Rattus rattus). TABER et al. (1967) refer
to the subspecies in the Indus Plains and upland areas in northern Pakistan as R. rattus
rufescens (Gray), a brownish-gray-backed form with a venter either creamy white or light
gray with a rufous tint. The typical mammary formula ıs 2 + 3 = 10, with an occasıonal
pairing of the postaxial mammae to givea3 + 3 = 12 formula.

Previous studies dealing with the reproductive biology and population dynamics of this
subspecies in similar habitats gave somewhat contradictory findings. BEG et al. (1983) ın
Faisalabad, Pakistan, found that trapping success was lowest in winter and peaked in
summer: pregnancy occurred only in 10 months of the year and averaged 45 % during the
study; the proportion of young was highest in the winter and averaged 31% of the
collections during the year. Rana et al. (1983) in Jodhpur, India, found that temales were
pregnant throughout the year; the annual frequency of pregnancy was 25.5%; the
proportion of young averaged 54% and was lowest from January to April. Additional
information on the reproductive biology and population dynamics of this subspecies is
needed to use control methods with proper timing. Thus, we carried out a 14-month study
of the roof rat population in the wholesale grain market ın Rawalpındi, Pakistan.

Material and methods

Description of the grain shops

The Rawalpindi market consists of several hundred dealers occupying ground floor, small (25 m?-
45 m? floorspace) shops, 40-50 years old, with capacities for 500-800 bags of commodities (50-80
metric tons). None of the shops are ratproof. The main commodity is rice; other foods available are

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5904-0209 $ 02.50/0

210 J. E. Brooks, E. Ahmad and I. Hussain

lentils, grams, sorghums, wheat flour, and groundnuts. The amounts of grains sold annually vary
from 1,000 bags to 60,000 bags (BRooxs et al. 1987).

Trapping

Commercial wire-mesh live traps (41x14x14 cm) and Sherman galvanized steel traps (25x 7.5 x
8 cm) were used. The smaller traps were more suitable for smaller rats. We baited the traps with fresh
seasonal vegetables or fruit, which the rats readily accepted, and trapped for 4 consecutive nights each
month from April 1987 through May 1988 by setting ten wire-mesh traps and five Sherman traps in
each of eight shops. Each month we selected eight different shops, four on each side of the roadway,
to minimize ınvasıon from surrounding shops and to avoid the effect of removal trapping on
population sıze. The Sherman traps were not available during the first 2 months of the study.

Necropsy procedure

After using chloroform to kill rats in the laboratory, we necropsied 2327 rats. Slight discrepancies in
totals resulted when all data were not taken on some rats. We weighed and measured each anımal
before necropsy and recorded reproductive data: for females, this included the condition of the vaginal
orifice (perforate or not), condition of the uterus (nulliparous, pregnant, placental scars), condition of
the ovaries (corpora lutea visible or not), number of embryos, their crown-rump length, and whether
any were resorbing. For males, we recorded the position (abdominal or scrotal), length, and weight of
testis, and whether the tubules of the cauda epididymis were visible or not. We recorded the number
of scars, wounds, fractures, and missing limbs on the trapped rats.

We classified anımals as sexually mature by calculating the 50% points at which females showed
visible corpora lutea and males showed visible tubules in the cauda epididymis (Davıs 1964). Animals
meeting or exceeding these 50 % points in body weight (BW) and in head and body length (HBL)
were adults; all others were immatures.

Using HArRIsoN’s (1952) methods to compare rat populations, we calculated the “embryo rate”
(the crude pregnancy rate X the unadjusted embryo number = the number of embryos per 100 females
of all sizes) and the “reproduction rate” (the number of embryos per 100 head ot population, males
and females). We believe these rates to be more indicative of a population’s reproductive potential than
the reproductive rate as defined in SourtHwick’s (1966) procedure in which he used the yearly
production per female, calculated from the average litter size and the incidence of pregnancy (yearly
pregnancies per female). The incidence of pregnancy ıs an artıfıcıal figure indicating the number of
times a female rat could be pregnant during a year; actually very few female rats live a year, so the
incidence of pregnancy gives an overestimation of potential production. We used 16 days as the period
of visible pregnancy in the roof rat (SOUTHWICK 1966).

Population estimates

Monthly population estimates were calculated from the linear regression of cumulative captures on
daily captures (BLOWER et al. 1981). Because captures were taken from a different set of eight grain
shops each month, we cannot generalize much about seasonal population changes. We compared the
number of animals captured with population estimates to see it there was a correlation.

Results
Characteristics of the sample

The proportion of males to females was 50.5:49.5, which does not difter significantly from
a 1:1 sex ratio. Males grew to a greater body weight than did females, but dıd not exceed
females in head and body length (Tab. 1). There was no significant difference in mean body
weight at equivalent HBL’s between sexes below 160 mm; above that point, males were
significantly heavier (P = 0.01). Males predominated in the largest size classes. The largest
male weighed 273.5 g and measured 214 mm. The heaviest female weighed 246.8 g and
measured 199 mm; however, the longest female measured 215 mm. The mean weight of
pregnant females was 163.2 g + 29.5, ranging from 84.4 to 246.8 9.

Reproductive biology and population structure of Rattus rattus in Rawalpindi 211

Table 1. Head and body length (HBL), body weight (BW), and sex ratio of Rattus rattus from
Rawalpindi, Pakistan

(Mean + SD)

Head and body Males Females
length sıze Number of Body weight Number of Body weight
classes (mm) animals (8) anımals (g)

ker helanltelke he helkelk- keller

Se
+
te
at
ze
2
ste
==
==
as
de
ef
Se

210-219
Totals
Means, BW
Means, HBL

Reproduction

We defined mature females as those that weighed >80 g or were >139 mm in HBL;
mature males weighed >96 g or were >152 mm in HBL. We observed fertile males in
every month of the year. Male fertility begins when testis weight reaches 0.7-0.8 g and a
length of 16-17 mm. Many males with testis smaller than these nevertheless were observed
to be scrotal. There was little seasonal fluctuation in mean testis weights when adjusted for
mean body weights (Fig. 1).

The adjusted frequence of pregnancy, as judged by visible pregnancy ın all adult females
(=80 g BW), ranged from 30% in April to 58% in July (Fig. 2). The frequence of

Testis wt (mg/100 g body wt)
1

HESSSAEBBEBRE?

0.4

0.2 ———

AM SEI RERESEEO7 INT DJ EM ‚AM

Fıg. 1. Mean testis weights (+ SD) in male R. rattus adjusted for monthly body weights

DD J. E. Brooks, E. Ahmad and I. Hussain

Percent of female adults visibly pregnant

100

0 |
Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May
1987 1988

Fig. 2. Percentage of visible pregnancies monthly in adult female R. rattus

pregnancy for the entire study was 39.8%. The smallest pregnant female was 145 mm
HBL, but most pregnancies occurred in females = 150 mm HBL and > 100 g body weight
(abn)!

Primigravid females constituted 53.7% of all pregnancies (Tab. 3). The 1:0.78 ratio
between females pregnant for the first time and those bearing their second litter indicates
that, on the average, primigravid females have a 78% chance of living long enough to have
a second pregnancy.

Litter size

The number of embryos per female ranged from 1 to 16, and 80 (3.0 %) of the total of 2682
embryos were resorbing. We examined litter size by embryo size (CR = crown-rump
length), roughly corresponding to the second, third, and fourth quarters of pregnancy, and

Table 2. Relationship of HBL and BW to visible pregnancy in female R. rattus from Rawalpindi

Body size Number Pregnant Body weight Number Pregnant
class (mm) examined Number class (g) examined Number

<140° i <80*°
140-149 . 80- 99
150-159 ; 100-119
160-169 ; 120-139
170-179 : 140-159
180-189 , 160-179
190-199 s 180-199
200-209 : 200-219
210-219 I 220-239

Total rats
> 140 mm 1039
=808

Total/percent
rats pregnant

? Immatures.

Reproductive biology and population structure of Rattus rattus in Rawalpindi 218

Table 3. Corpora lutea and parity in R. rattus

Corpora Parous
HBL sıze Number lutea Nulli- Primi- Multi- Non-gravid with scars
class (mm) examined visible parous gravid gravid 1 set 2 sets 3 sets

80- 89
90= 99
100-109
110-119
120-129
130-139
140-149
150-159
160-169
170-179
180-189
190-199
200-209
210-219

Totals 1126

ooVuıNsSs*-+ooo00o000000

N
ON

found that embryo counts in the second quarter averaged 6.90/female and decreased to
6.09/female in the last quarter. This decrease indicated an intrauterine loss of 0.81 embryo,
a mortality of 11.7 % of embryos from all causes. Litter size changed with increasing body
sıze of the female from 5.7 # 1.8 (in females 150-159 mm HBL) to 7.4 + 2.0 (in females
190-199 mm HBL).

Population and age structure

Sex ratios from immature and adult rats were basıcally alıke, although males predominated
among the immature rats (142 males to 123 females), but the difference was not significant
(X? = 1.36, P = 0.20). Data for April and May 1987 were not used because the smaller traps
suitable for immature rats were not available then.

The recruitment of young anımals into the population occurred throughout the year
(Tab. 4). We found that recruitment of immatures each month showed a significant
correlation with the frequence of pregnan-
cy ın the preceding month (r = 0.742, t =

Table 4. Recruitment of young R. rattus
3.5, P = 0.005).

Month Immatures Adults % immatures
Population estimates
Jun.

Except for April 1987, the captures per trap Jul.
night varıed only between 0.24 and 0.41 Aug.

(Tab. 5). The capture rate and numbers of S>

animals removed were virtually the same in Nov

June 1987 and May 1988. This is indicative Dec.

of stable populations; most differences in Jan.
population estimates occurred because of ee
trappıng vagaries, (e.g., more rats captured Apr.

on the second or third night, or disturban- May

ces ın shops that may have affected cap- oals 244

tures). There was a fair correlation between
the number of anımals captured and the

April and May 1987 captures are deleted
because small live traps for immature rats
estimated population (power regression, were not available in those months.

correlation coefficient = 0.64).

214 J. E. Brooks, E. Ahmad and I. Hussain

Table 5. Roof rat captures, trap success, and estimated populations from Rawalpindi

y Total Trap Captures/ Estimated
3 captures niıghts trap night population

Apr. 87
May
Jun.
Jul.
Aug.”
Sep.
Oct.
Nov.
Dec.
Jan. 88
Feb.
Mar.
Apr.
May
Totals

* 7 shops only.

Table 6. The frequence of scars, wounds, and Intraspecific strife

missing body parts in R. rattus
Sehr The most frequently observed evidence of

fighting among the rats was that parts of

Disabilities Females Males > Be :
tails were missing (Tab. 6). Wounds on tails
Part of tail gone were seen next in frequence. In aggressive
Wounds on tail encounters among rats, the winner quite
Fracture of tail often bites at the lower back or tail of the
Swollen tail
Word 0x dasesss am hariı loser (personal observation). Fractures and
Wounds or swollen legs swellings of tails were seen less often. Many
Fracture of hind foot females were wounded and had missing
Missing hind leg body parts (18% of the total female sam-
Totals ple). This is indicative of extreme strite

since aggressive exchanges are usually con-
fined to those between males (BARNETT
1963). Altogether, anımals bearing evidence of aggressive exchanges were about 23% of
the total collection.

The incidence of injuries increased sharply for rats of both sexes >150 mm in body
size. This could be due to cumulative encounters.

Discussion

We compared our findings with those of two similar studies cıted earlıier. The sızes of rats
at maturıty depend on the average body sızes of rats ın each population. The average sızes
of rats differ in the three populations: Faisalabad females averaged 87 g and males averaged
94 g; Rawalpindi females averaged 138 g and males averaged 147 g. Roof rats from Jodhpur
were the smallest, with females averaging 77 g and males 76 g. The average for Jodhpur is
low because many immatures were trapped; however, maximum body weights recorded
were not notable. Rats from Faisalabad and Rawalpindi were larger, but not necessarıly
older, than rats from Jodhpur.

Males from Rawalpindi showed little seasonal change in testis weights and were fertile

Reproductive biology and population structure of Rattus rattus in Rawalpindi 215

in all months. Sexually active males were seen in all months in Jodhpur, but the proportion
of fertile males dropped to low levels ın April, July, and September. No data on seasonal
fertility in males rats were reported from Faisalabad.

Pregnant females occurred in every month in Rawalpindi and Jodhpur. In Faisalabad,
pregnant rats were not seen in November or December. The frequency of pregnancy and
ovarıan weights varıed seasonally in Jodhpur; it was lowest in June and October and
highest in July and December. The annual frequency of pregnancy was highest in
Faisalabad (46 %), next highest in Rawalpındi (39.8 %), and lowest in Jodhpur (25.5 %).

The comparative litter sizes were similar. In Jodhpur, Faisalabad, and Rawalpindi, they
ranged from 6.60 to 6.19. These crude litter sizes, however, are not adjusted to account for
the number of embryos actually seen at birth. Often, litter size decreases as pregnancy
progresses. Counts taken in the last trimester of visible pregnancy approximate the
numbers seen at bırth.

We calculated the “embryo rate per 100 of all temales” and from this derived the
“reproductive rate” (HARRISON 1952); this is the number of embryos per 100 head of
population (Tab. 7). This rate is highest in rats from Faisalabad and Rawalpindi. This
reproductive rate in stable rat populations is sometimes interpreted as a measure of the
death rate, since births should equal deaths to maintain stability. However, high mortality
in weanlıngs and juveniles can ınvalidate this assumption. In that case, the entry of young
breeders (recruitment) into the population should be a better measure of the death rate.

Table 7. Comparative reproductive data and calculated parameters for R. rattus from several
Asian urban localities

Crude Unadjusted
Sex pregnancy embryo Embryo Reproduction
City ratio” rate number rate rate“

Jodhpur 121 168 92
Rawalpindi 100 Ä 232 116
Faisalabad 117 223 120

2 Females/100 males. - P Number of embryos for all females in population regardless of size. -
“ Number of embryos for 100 head of population, males and females.

The percentage of immature rats in the several populations was 13% in Rawalpindi,
31 % ın Faisalabad, and 54% in Jodhpur. We assume that roof rat mortality (in adults)'is
least ın Rawalpindi and greatest in Jodhpur. The reasons for the high recruitment of young
rats in Jodhpur are not given by the authors, but may be due, in part, to a relatively harsh
environment. In Rawalpindi, the main cause of rat mortality is year-round intraspecific
strife.

Despite a high reproductive rate in Rawalpindi roof rats, the low proportion ot
immatures entering the population indicates high mortality among nestling and weanling
rats. This could be due to desertion of litters before weaning (few lactating females were
seen) or predation by adult rats.

The differing population strategies are exemplified by the Rawalpindi and Jodhpur rat
populations. Whereas Rawalpindi relies upon a high reproductive rate to offset high
mortality in immatures to maintain the population. Jodhpur rats rely upon a high
recruitment of immatures into the population to offset the apparent high mortality in adult
rats. Pregnancies at Jodhpur remain at a rather lower level than at either Rawalpindi or
Faisalabad.

The implications for management of Rawalpindi rat populations are not promising.
Sınce juvenile mortality is already high, efforts to increase adult mortality with poisons

216 J: E. Brooks, E. Ahmad and I. Hussain

would tend to decrease the pressure on the juveniles. The birth rate would be adjusted by
the rats following population reduction to compensate for mortality, and both frequence
of pregnancy and litter size would be expected to increase.

More effective control measures would be those that change the environment. Rat-
proofing of grain shops, though difficult, would bring about population decreases. There
are numerous holes in walls between adjoining shops that could be filled to stop rat
movements. Grain shop sanitation is another measure that would be beneficial. This
includes keeping grain spillage swept up and damaged bags repaired. If these measures
were instituted along with poisoning and trapping, then population control would become
evident — but it would have to be a continuous effort.

Acknowledgements

We appreciate the help MoHAMMAD IryAas, Yousar KHAn and MOHAMMAD Nawaz provided in
capturing the rats and assisting in the necropsies. We thank LıAQAT Auı for entering all the necropsy
and morphomerric data into the dBASE III computer program. RiCHARD A. DOLBEER and DoNALD ].
Erias, Denver Wildlife Research Center, critically reviewed the manuscript and provided helpful
suggestions.

This study was supported with funds provided by the U. S. Agency for International Develop-
ment under the project “Food Security Management, Vertebrate Pest Control, PASA IPK-0491-P-IF-
5017-05”. We also thank the Pakistan Agricultural Research Council and the National Agricultural
Research Centre, Islamabad, for providing office and laboratory space for the project.

Zusammenfassung

Reproduktionsbiologie und Populationsstruktur von Rattus rattus in Rawalpindi, Pakistan

Die Fortpflanzungsbiologie und Populationsstruktur von Rattus rattus wurden in einem Großhan-
delsmarkt für Getreide und andere Gebrauchsgüter in Rawalpindi, Pakistan, untersucht. Innerhalb
von 14 Monaten wurden die Ratten in monatlichen Abständen auf dem Getreidemarkt gefangen. Wir
sezierten und untersuchten insgesamt 2327 Ratten, 1175 männliche und 1152 weibliche. Es wurde
festgestellt, daß die männlichen Ratten jeden Monat reproduktionsfähig waren ohne bedeutsame,
saisonbedingte Unterschiede; weibliche Ratten waren jeden Monat trächtig, und die Trächtigkeitsfre-
quenz betrug durchschnittlich 39,8 %. Die jeweilige Embryonenzahl während des letzten Viertels der
Trächtigkeit betrug durchschnittlich 6,1 # 1,7 (SD). Auf der Basıs unserer Studie von Ratten, die zum
ersten Mal trugen, und von Ratten, die schon vorher trächtig waren, kalkulierten wir, daß eine
geschlechtsreife weibliche Ratte im Durchschnitt 1,8 Trächtigkeiten hat. Die durchschnittliche
Produktion von jungen Ratten war 10,9 pro Weibchen (6.1x 1.8). Unreife Tiere (unabhängig, aber
nicht sexuell reif) dagegen machten 13,1 % der Gesamtstichprobe aus. Nachwuchs war fortlaufend
während der Studie vorhanden und wies darauf hin, daß beträchtliche Todesfälle bei Nestlingen und
sexuell unreifen Tieren vorkamen. Viele Wunden und Narben an erwachsenen Tieren beider
Geschlechter wiesen darauf hin, daß ein hohes Maß von Streit und Aggression zwischen den gesellig
lebenden Ratten herrschte.

References

BARNETT, $. A. (1963): The rat. A study ın behavior. Chicago: Aldine Publ. Co.

BEc, M. A.; MusHTaQ-UL-Hasan, M.; Kausar, $.; KHan, A. A. (1983): Some demographic and
reproductive parameters of house rat population i in Faisalabad, Pakistan. Pakistan J. Zool. 15,

BLOWER, L. G.; Cook, L. M.; BısHor, J. A. (1981): Estimating the size of anımal populations.
London: George Allen and Unwin Ltd.

Brooks, J. E.; AnmaAD, E.; Hussarn, I. (1987): Rat population and stored food losses at a Pakistan
grain market. Vertebrate Pest Control Project, National Agricultural Research Centre, Islamabad,
Pakistan: Technical Report No. 12 (unpubl.).

Davıs, D. E. (1964): Manual for analysis of rodent populations. Penn. State Univ., University Park,
Penn.

HARRISION, J. L. (1952): Reproduction in rats of the subgenus Rattus. Proc. Zool. Soc. Lond. 121,
674-694.

Rana, B. D.; Apvanı, R.; Son, B. K. (1983): Reproductive biology of Rattus rattus rufescens in the
Indian desert. Z. Angew. Zool. 70, 207-216.

Reproductive biology and population structure of Rattus rattus in Rawalpindi 217

SOUTHWICK, C. H. (1966): Reproduction, mortality and growth of murid rodent populations. In:
Indian Rodent Symposium, 1966, Calcutta. Calcutta: Johns Hopkins University Centre for
Medical Research and Training.

TABER, R. D.; SHER1, A. N.; AnmaAD, M. S. (1967): Mammals of the Lyallpur Region, West Pakistan.
J. Mammalogy 48, 392-407.

Authors’ addresses: Joe BRooxs, USDA/APHIS/ADC/Denver Wildlife Research Center, P. ©. Box
25266, Denver, Colorado 80225-0266, USA; EjJaz AHMAD, Department of
Fishery and Wildlife Biology, Colorado State University, 135 Wagar Building,
Fort Collins, Colorado 80523, USA; and IrtIıkHArR HussAaın, Vertebrate Pest

Control Laboratory, National Agrıcultural Research Centre, Park Road,
Islamabad, Pakistan

Z. Säugetierkunde 59 (1994) 218-223
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Dispersal and other inter-group movements in badgers,
Meles meles

By S. F. CHRISTIAN

School of Biological Sciences, University of Sussex, Brighton, UK

Receipt of Ms. 3. 12. 1993
Acceptance of Ms. 24. 3. 1994

Abstract

Investigated dispersal and other movements of badgers between social groups in a high density badger
population in Sussex, southern England, during the period November 1988 to October 1993.
Fourteen badgers (10 female, 4 male) were radio-collared, tracked and observed during nocturnal
ranging. Daily sleeping locations of collared badgers were also recorded for a 2-year period. In
addition bait-marking was carried out twice annually to determine territory boundaries. Twenty-eight
movements between groups were recorded, including 7 permanent group changes, one temporary
group change and 20 visits of shorter duration. All of the adult badgers that permanently changed
groups were mature females. An adult male changed groups temporarily and another was killed
3.5 km from where he was trapped. No aggression from territorial residents towards immigrants was
observed and 3 females bred successfully following their group change.

Introduction

Badgers are unusual among the mustelids in that they typically live in mixed-sex social
groups, sharing a communal range and one or more setts (Kruuk 1978; PoweELL 1979;
CHRISTIAN 1993). The social groups scent-mark and actively defend territories that are
usually contiguous, mutually-exclusive and remarkably stable over time (Kruuk 1978;
Kruuk and ParısH 1982). Kruuk (1978, 1989) reportet that badgers are highly aggressive
towards their neighbours and NEaAL (1986 p. 151) stated that “a resident boar will fight any
intruder which hasn’t the correct communal scent of his social group’. Serious territorial
fights have been reported (Kruuk 1978; NEAL 1986; CHRISTIAN 1993) and intraspecitic
aggression is thought to be a significant cause of badger mortality (GALLAGHER and
Neıson 1977; ANDERSON and TREWHELLA 1985). Thus, there is a high cost to leaving the
group territory and, in populations where the social structure is stable, evidence from lıve-
trapping suggests that dispersal of juveniles from their natal group is rare (KRuUUK and
PArısH 1982; CHEESEMAN et al. 1987, 1988; Evans et al. 1989). However, little has been
reported about the movements of individual badgers between social groups.

The aım of this study is to report observations of social group changes and other
intergroup movements by radio-collared individuals in a high-density rural population of
badgers with a stable territorial structure.

Material and methods

The observations were made as part of a long-term, intensive radio-tracking study of badger
behaviour and ecology in Sussex, southern England over the period November 1988-October 1993.
The estimated population density was 16.7 adults per km? (for further details see CHrısTIan 1993). A
total of 14 badgers (4 males, 10 females) from six adjacent social groups were radio-collared, tracked
and observed during the night for an average of 6 hrs per badger per night during the period 1988 to
1992. Over the period October 1989 to October 1991 the setts used by radio-collared badgers as
underground sleeping locations were also recorded daily. In addition, bait-markıng (Kruuk 1978) was

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5904-0218 $ 02.50/0

Dispersal and other inter-group movements in badgers, Meles meles 219

carried out twice each year in Aprıl and November, and the results were used to determine the

boundaries of the 16 group territories in the study area.

Four categories of movement between social groups were distinguished:

1. Nocturnal visits. These included brief visits to the main setts of other social groups during
nocturnal ranging. They lasted from a few minutes to one hour, during which the visitor might
investigate the sett from the outside, or enter it and remain underground for some time.

2. Diurnal visits. These were visits in which individuals slept for a day or two in another social
group’s main sett but subsequently returned to their own territory.

3. Temporary group changes. These were longer-term movements in which badgers moved from
their original social group to a new one, but returned to their original group after a period of
several months.

4. Permanent group changes. These were movements in which the badgers moved from their original
social group to a new social group, where they remained for the rest of the study period.

Results
Permanent and temporary group changes

Twenty-eight movements between socıal groups were observed during the study period:
eight were classed as Permanent and Temporary group changes and 20 as other move-
ments. Seven individuals made Permanent group changes and one individual made a
Temporary group change (Fig. 1). All of the six adult badgers that changed groups
permanently were mature females in their second year or older and two of them were
known to have bred successfully prior to moving. The seventh badger to change social
groups permanently was a young cub which accompanied its mother to her new territory.
The badger which made a temporary group change was an adult male that moved to a
neighbouring territory for a 3-month period.

All of these badgers were in
good general body condition,
showed no signs of injury or ill
health and were not observed to be
socially peripheral or subject to
persecution in their original
groups. In each case of Permanent
group change the social group that
was joined was well-established
and contained several adults of
either sex, occupying a stable,
well-defined territory. Two of the
groups to which females moved
contained at least two resident fe-
males which were known to have
bred successfully in the year pre-
ceding the movement and went on
to raise cubs in the year following
the movement. Three of the sıx
females that moved permanently
bred successfully in their new so-
cıal group within a year of moving.

In no case was aggression from
residents towards immigrants or

SE Fig. 1. Permanent (black arrows) and Temporary group
from immigrants towards residents changes (white arrows) by badgers, 1988-1993. Bold lines

observed. No injuries caused by denote territory boundaries and dots numbered 1-16 de-
intra-specific fighting were re- note main setts

2209 S. F. Christian

corded during tracking or re-
capture on any of the indi-
viduals that changed groups
and there was no evidence of
the immigrants avoiding the
residents at the new sett or
during nocturnal ranging.
The movement of the
Group 2 badgers to Group 8
was ımmediate and unex-
pected: the badgers concerned
(two females, one accom-
panıed by a cub, and an adult
male) began foraging near to a
neighbouring main sett one
night and by the following
day had taken up residence
there. In all other cases the
movement from one territory
to another was a more gradual
process, the badgers con-
cerned moving into outlying
setts in their own territories,
from these into outlying setts
Fig.2. Home range of female 15Fl, during November in the new territory and only
1988-October 1990. The study area was divided into I-ha grid then into the new main sett.
squares (not shown), and open circles denote squares in which One case of a permanent
15F1 foraged. Bold lines denote territory boundaries and dots group change is worth report-
numbered 1-16 denote main setts

10) 500m

ing ın more detail because the
badger concerned was excep-
tional in the extent to which it ranged within other groups’ territory. The badger in
question (an adult female) was trapped and collared at the main sett of Territory 1 in
November 1988, and continued to inhabıt that sett until February 1989, when she began to
spend her days in outlying setts near the territory boundary. In early March 1989 she was
found during the day at the main sett in Territory 15, and thereafter she continued to sleep
at that sett (over 1.5 km from her original sett). On the night following her move, during a
5-h period, she made an extensive tour of the study area, moving through 10 different
territories, passing close to the residents of several of them, and entering four main setts
(Setts 1, 3, 8, and 10). Similar excursions were seen on subsequent nights and until October
1990 she was frequently located (often accompanied by a cub) foraging ın other territories:
particularly Territory 1 (her original territory) whose outliers she sometimes entered
during the night and with whose residents she frequently associated; Territory 3 where she
was recaptured in March 1990 during a nocturnal visit; and Territory 10 where she foraged
with members of Groups 1 and 10 (Fig. 2): Despite this extensive extra-territorial ranging
she was never observed in aggressive interactions with members of other social groups.

Other movements between social groups

Two individuals (a male and a female) made Diurnal visits to other main setts on four
occasions, ın January, July, August and September respectively, on each occasion spending
the day underground with the residents. Fifteen Nocturnal visits (10 by females, 5 by
males) were also recorded (Fig. 3). These usually lasted 5-20 minutes, during which time

Dispersal and other inter-group movements in badgers, Meles meles 221

the visitor sniffed around the
sett entrances and sometimes
entered the sett. On one occa-
sion a female spent an hour
underground at a neighbour-
ing sett with several of its resı-
dents, having previously
spent more than two hours
peacefully feeding with them
in a nearby maize-patch. On
two occasions a male visiting a
neighbouring sett was fought
by an occupant and chased
away, and on two occasions a
male visiting a neighbouring
sett was observed to copulate
with a resident female.

Finally one other badger
was known to have moved a
long way outside of its origi-
nal territory. An adult male
badger was captured whilst at
Sett 6 in April 1993 but was
never recorded in the study
area again. In October 1993 Fig. 3. Nocturnal (black arrows) and Diurnal visits (white
he was found dead at the side arrows) by badgers, 1988-1993. Bold lines denote territory
of a road 3.5 km from Sett 6. boundaries and dots numbered 1-16 denote main setts

(0) 500m

Discussion

Several studies suggest that the movement ot adult badgers between groups is rare. KRUUK
(1978) reported that movements between territories in Wytham Woods were infrequent
and ın Speyside (Kruuk 1989) he found that of 60 badgers caught in consecutive years (37
females, 23 males) only sıx (all male, all adult) emigrated from their original territories. He
concluded that male badgers remain in their natal territories for as long as it takes to move
to a vacancy in a neighbouring territory, or until evicted by a relative or a usurper, whereas
females remain permanently in their natal territories.

In other populations, however, both sexes may change social groups. CHEESEMAN et al.
(1988) reported dispersal and movement patterns in badger populations in rural Glouces-
tershire and suburban Bristol. In the lower-density Bristol population the socıal structure
appeared to be fluid and inter-group movements were relatively common. Movements
were less frequent in the high-density Gloucestershire population, although their incidence
increased as a result of disturbance, involving the complete removal of 11 social groups for
reasons ot disease control. In both populations, more males than females changed groups.
Most of these movements involved sexually mature individuals and constituted movements
to neighbouring territories. In a low-density rural Irısh population subject to frequent
human persecution, SLEEMAN (1992) reported that long-distance movements between
social groups by individuals of both sexes may be relatively common. Similarly, ROPER
and Lürs (1993) have reported an increase ın extra-territorial ranging and nocturnal visits
to a main sett following the sudden death ot all of ıts male occupants. Thus, overall the
evidence points to the ıdea that in undisturbed high-density rural populations, where all

222 S. F. Christian

the available habiıtat ıs divided up in a well-defined and stable territorial system, move-
ments from one social group to another are uncommon and largely involve the occasional
movement of individuals, usually males, to neighbouring social groups. In areas where the
removal of social groups, persecution or some other factor causes territorial disruption, or
severely depresses the population, extra-territorial movements and group changes are
probably much more frequent.

My results are consistent with these findings in two respects, in that all of the
movements involved adults (with the exception of an accompanying cub) and all of the
permanent group changes involved movement to an adjacent territory. However, the
observed movements were relatively frequent (7 of 14 radio-collared badgers changed
groups during a 3-year period and visits to other setts were relatively common) and
permanent group changes exclusively involved mature females. Females also visited other
main setts more frequently than males.

My observations of group changes are notable in several other respects. Firstly,
dispersers were never observed suffering aggression from members of their own social
group prior to changing social groups. This suggests that the individuals did not disperse
because they were induced to do so by other group members. Secondly, dispersers were
never observed fighting with members of their new social group. Female immigrants
showed no signs of persecution but went on to breed sımultaneously with resident females
in the new social group. This suggests that badgers may be more free to move social groups
than previously thought and suggests females may not be in direct competition wıth each
other for opportunities to breed as has been reported in other populations (Kruuk 1989;
WOODROFFE and MAcDonALD 1993). Thirdly, the unchallenged movements through
other territories suggest that under certain circumstances some individuals, or categories of
individual, may be free to range and forage within other territories and to visit other setts.
Finally, the movement of all of the known members of Group 2 to join Group 8 “en
masse” (2 adult females, a dependent cub and an adult male) is the first reported group
movement of its kind. It is important to note that the Group 2 badgers did not move to a
previously unoccupied territory, nor did they take over a previously occupied territory
and expel its occupants. Rather, the immigrants joined the existing group members in their
main sett, associating frequently and harmoniously with them, and both immigrant and
resıdent females went on to successfully rear cubs following the group change.

Visits to other setts by radıo-collared badgers of both sexes were relatively common and
were characterised by amicable association between visitors and residents. Visitors some-
times entered the new sett and even spent the next day underground there with residents.
Visiting males twice achieved successful matings with resident females. On two other
occasions, however, male visitors were attacked and chased away by a resident male.

These radio-tracking observations indicate that, even within a stable and undisturbed
high-density rural population, interaction and movements between neighbouring social
groups may be more common than has previously been estimated on the basis of recaptures
at main setts. They also suggest that territoriality may be more flexible (seasonal or
context-related), or more specific (only concerning specific individuals or categories of
individual, e.g. adult males) than has been previously supposed.

Acknowledgements

I would like to thank Davıp Fee for his company and assistance with fieldwork. I am also grateful to
Tım Ropek for allowing me to use his unpublished observations and for helpful comments on drafts of
this manuscript. My research was supported by an SERC studentship.

Dispersal and other inter-group movements in badgers, Meles meles 222)

Zusammenfassung

Ausbreitung und andere Ortsbewegungen zwischen Gruppen von Dachsen, Meles meles

Untersucht wurden Ausbreitung und andere Ortsbewegungen zwischen sozialen Gruppen einer
Dachs-Population mit hoher Dichte in Sussex, Südengland, von November 1988 bis Oktober 1993.
Vierzehn Dachse (10 weiblich, 4 männlich) wurden mit Radio-Transmittern markiert und während
ihrer nächtlichen Wanderungen verfolgt und beobachtet. Über einen Zeitraum von 2 Jahren wurden
die Schlafplätze täglich registriert. Außerdem wurde zur Feststellung der Territoriengrenzen zweimal
im Jahr das Futter mit farbigen Plastikquadern markiert. Es wurden 28 Bewegungen zwischen
Gruppen beobachtet, darunter 7 dauerhafte Gruppenwechsel, ein vorübergehender Wechsel und 20
von kürzerer Dauer. Alle adulten Dachse, die dauerhaft die Gruppe wechselten, waren geschlechts-
reife Weibchen. Ein adulter männlicher Dachs wechselte die Gruppe nur vorübergehend, ein anderer
wurde 3,5 km von der Fangstelle entfernt getötet. Territorienbesitzer zeigten keine Aggression
gegenüber Einwanderern. Drei weibliche Dachse zogen nach Gruppenwechsel erfolgreich Nach-
wuchs auf.

References

ANDERSON, R. M.; TREWHELLA, W. (1985): Population dynamics of the badger (Meles meles) and the
epidemiology of bovine tuberculosis (Mycobacterium bovis). Phil. Trans. Roy. Soc., B, 310,
327-381.

CHEESEMAN, C. L.; CRESSWELL, W. J.; HARRIS, $.; MALLINson, P. J. (1988): Comparison of dispersal
and other movements in two badger (Meles meles) populations. Mammal Rev. 18, 51-59.

CHEESEMAN, C. L.; WILESMITH, J. W.; Ryan, J.; Marrınson, P. J. (1987): Badger population
dynamics in a high-density area. Symp. Zool. Soc. Lond. 58, 279-294.

CHRISTIAN, $. F. (1993): The behavioural ecology of the Eurasian badger, Meles meles: space use,
territoriality and social behaviour. D. Phil. thesis, University of Sussex.

Evans, P. G. H.; MAcDonaLD, D. W.; CHEESEMAN, C. L. (1989): Social structure of the Eurasıan
badger (Meles meles): genetic evidence. J. Zool., Lond. 218, 587-595.

GALLAGHER, J.; NELSON, J. (1979): Causes of ill-health and natural death in badgers in Gloucester-
shire. Vet. Rec. 105, 546-551.

Kruuk, H. (1978): Spatial organisation and social behaviour of the European badger, Meles meles.
]J. Zool., Lond. 184, 1-19.

— (1989): The social badger. Oxford: Oxford University Press.

Kruuk, H.; ParısH, T. (1982): Factors affecting population density, group size and territory size of
the European badger, Meles meles. J. Zool., Lond. 196, 31-39.

NEAL, E. G. (1986): The natural history of badgers. London: Croom Helm.

PowELı, R. A. (1979): Mustelid spacing patterns: variations on a theme by Mustela. Z. Tierpsychol.
50, 153-165.

SLEEMAN, D. P. (1992): Long-distance movements in an Irish badger population. In: Wildlife
telemetry: remote monitoring and tracking of animals. Ed. by I. G. PrıEDE and $. M. Swırr.
Chichester: Ellis Harwood.

ROoPER, T. J.; Lürs, P. (1993): Disruption of territorial behaviour in badgers Meles meles. Z.
Säugetierkunde 58, 252-255.

WOODROFFEE, R.; MAacponaLDd, D. W. (1993): Female/female competition in European badgers
(Meles meles): effects on breeding success. J. Anım. Ecol. (in press).

Author’s address: S. F. CHrısTIan, School of Biological Sciences, University of Sussex, Brighton
BN1 9QG, UK

Z. Säugetierkunde 59 (1994) 224-229
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Lipid deposits in pregnant and non-pregnant bats
(Pipistrellus pipistrellus)

By PıLar LöPpEz-Luna, F. AREVALO, M. J. Burcos, and N. DEL Hoyo

Departamentos de Fisiologia y Farmacologia, Biologta Animal y Bioguimica y Biologia Molecular,
Universidad de Alcala, Alcalä de Henares, Madrid, Spain

Receipt of Ms. 15. 12. 1993
Acceptance of Ms. 1. 3. 1994

Abstract

Studied the white and brown adipose tissues of female bats (Pipistrellus pipistrellus) in order to
ascertain differences in relative lipid, fatty acıd and phospholipid contents that might be related to
pregnancy. Relative lipids were higher in the white adıpose tissue of pregnant bats than in that of non-
pregnant bats, and the same was true for phospholipids in the brown adipose tissue. Pregnancy in P.
pipistrellus could be responsible for the changes in the length and unsaturation of some fatty acids in
both kinds of adıpose tissue.

Introduction

Reproduction is accompanied by an increase in energy demands (TowsEnD and CaLLow
1981). The total energy investment of pregnancy involves many components such as
production ot fetal, uterine, placental, and mammary tissue, as well as the production and
increased maintenance costs associated with the new tissues. It has been suggested that
suppressing metabolic energy expenditure increases energetic efficiency during pregnancy
(SCHNEIDER and WADE 1987). Although in many mammals the energy thus conserved is
first stored as white adıpose tissue and later mobilized to meet the energy demands of
lactation (NAISMITH et al. 1982; SADLEIR 1984), this reserve is apparently not used by small
insectivorous bats (RAcEY and SPEAKMAN 1987). However, fat storage may be important in
bats because of their unavoidably high activity levels and obvious tlıght costs (GITTLEMAN
and THomPpson 1988).

Maternal body fat accumulation is one of the most striking features of gestation in both
women (HyTTten et al. 1966; Hyrten and LertcH 1971) and experimental anımals
(BEAToN et al. 1954; LopEz-Luna et al. 1986; HERRERA et al. 1988). Fat storage may also
be important in pregnant bats since, when fat ıs unavailable, fetal growth rates and
pregnancy rates often decline (RAcey 1973; KurTA 1986).

The main purpose of this study is to learn whether pregnant females of Pipistrellus
pipistrellus show differences in body weight, total lipids, fatty acıds and phospholipids in
the white and brown adipose tissues compared with non-pregnant female individuals.

Material and methods

Animals

The active bats used in this study were pregnant and non-pregnant Pipistrellus pipistrellus temales
captured alive in eastern Madrid, between June 28 and July 31 and between March and September,
respectively. The pregnant bats were gathered from their maternity roosts (an inhabited building)
between 19:00 and 22:00 h.

Bats were placed in a wet sack for transportation to the laboratory and weighed and killed by
exsanguination under diethyl ether anaesthesia 2-3 hours after collection. The abdomen was immedi-
ately opened and conceptus (fetus and placenta) were delivered by hysterectomy and weighed.

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5904-0224 $ 02.50/0

Lipid deposits in pregnant and non-pregnant bats 225

The subeutaneous white and interscapular brown adipose tissues were quickly removed, stripped
of connective tissue, excised, weighed and immediately frozen in liquid nitrogen. They were then kept
frozen at -30°C until lipid analysis. Standard fatty acıds were purchased from Carlo Erba (Milan,
Italy) and Alltech Assocıates (Deertield, IL, USA).

Lipid analysis

Lipids were extracted from frozen tissues and purified with chloroform:methanol (2:1), according to
the method of Foıch et al. (1957). After evaporatıon of the solvent, the lipids were stored in a N,
atmosphere at -30 °C.

The method of Rouser et al. (1966) was used for quantitative determinations of total phospholipid
from both tissues. Purified total lipids from both tissues were directly treated with BF/CH3OH to
obtain the fatty acid methyl esters as previously described by PuLivo et al. (1986). A Pye Unicamp
Philips gas chromatograph equipped with a flame ionızation detector was used to assay the products ın
a5mx2mm glass column packed with 20 % Sılar 10 C on 80/100 Chromosorb WHP (Alltech
Associates); the flow rate of the nitrogen carrier gas was 32 ml/mın.

Statistical treatment of the results

Absolute and relative values are expressed as mean + SE. The statistical significance of differences
between groups was determined by Student’s t-test and differences were considered significant when
p<0.05.

Consent for capturing the anımals was given by the Consejeria de Medio Ambiente (Comunidad
de Madrid) in accordance with the Bern Agreement (1979) ratified by the Spanish Government in
1986.

Results

As shown in table 1 the conceptus-free body weight of pregnant bats was significantly
higher (p<0.05) than the body weight of non-pregnant bats. The pregnant bats studied
had each one fetus which weighed between 0.76 and 1.12 g. However, the subcutaneous

Table 1. Absolute body weights and relative tissue weights, relative lipid content and relative
phospholipids in white and brown adipose tissues in non-pregnant and pregnant females of
P. pipistrellus

The results are given as mean + SE. Comparisons of pregnant vs non-pregnant

Non-pregnant pregnant
(n = 28) (n = 15)

Body weight
(8)

White adıpose tissue
weight
(mg wet tissue/g body wt)
lipids
(g lipid/g wet tissue)
Phospholipids
(umol/g wet tissue)

Brown adipose tissue
weight
(mg wet tissue/g body wt)
lipıds
(g lipid/g wet tissue)
Phospholipids
(umol/g wet tissue)

” p = 0.05, Ei p < 0.01, aa P < 0.001.

226 Pılar Lopez-Luna, F. Arevalo, M. J. Burgos and N. del Hoyo

%

100

PREGNANT

3 NON-PREGNANT

80
60
40
20 “x

kakk j a

12:0 14:0 16:0 16:1 18:0 18:1 18:2 20:0 20:1 20:4 SAT UNS

Fig. 1. Fatty acıd composition of subcutaneous white adıpose tissue in P. pipistrellus. All assays were
performed in duplicate. The values given as % total fatty acıds are means + SE. Asterisks show
statistical comparisons between pregnant and non-pregnant bats **p<0.01, ***p< 0.001

white adıpose tissue (WAT) weight relative to body weight does not show significant
differences between pregnant and non-pregnant anımals. The relative lipid content
(g. lipıd/g wet tissue) was higher in pregnant than in non-pregnant bats (p< 0.01). Inter-
scapular brown adipose tissue (BAT) weight and relative lipıd content was not modified
during pregnancy. Relative phospholipids from WAT were unmodified during pregnancy
although BAT relative phospholipids increased significantly (p < 0.001) during pregnancy
(aba)

Distribution of fatty acıds in the total lipids from WAT tissue for non-pregnant and
pregnant bats is shown in figure 1. The proportion of myristic acıd (14:0) and palmitic acıd
(16:0) inthe WAT of pregnant bats ıs higher (p < 0.001 and p< 0.01 respectively) than that
observed in non-pregnant bats. However, lauryl acid (12:0) was only present in WAT
from pregnant bats and arachidonic acıd (20:4) only appeared in non-pregnant bats.

Pregnancy did not significantly affect the percentages of unsaturated fatty acıd ın WAT,
with the exception of oleic acid (18:1), which decreased (p< 0.001) in pregnant bats.

The fatty acid in total lipids of BAT (Fig. 2) shows an increase in oleic acıd (18:1)
monounsaturated fatty acid and a decrease in linoleic acid (18:2) polyunsaturated fatty acıd
that were both statistically significant (p< 0.001). The fatty acıd composition of the total
lipids in WAT and BAT (Figs. 1, 2) shows few differences between pregnant and non-
pregnant bats. Unsaturated fatty acıds are the most abundant (71-78 %) in both kinds of
tissues and in both pregnant and non-pregnant bats. However, the level of saturated fatty
acids increased during pregnancy from 21 to 27% (p<0.001) in WAT, but not in BAT,
where the level remained constant (21 %).

Discussion
Although copulation by temperate-zone insectivorous bats normally occurs in autumn,

fertilization does not occur until early spring when the bats arouse from hibernation and
enter maternity colonies (Racey 1982). The length of gestation in heterothermic bats may

Lipid deposits in pregnant and non-pregnant bats 227,

%

100

PREGNANT
J NON-PREGNANT

80

[use] EEE Mn |
20:1 20:2 20:4 SAT UNS

|
|
el

14:0 16:0 16:1 18:0 18

Fig. 2. Fatty acid composition of interscapular brown adipose tissue in P. pipistrellus. The same
symbols as in figure 1 were used

vary with the external conditions of food supply and temperature. The pregnant P. pıpi-
strellus bats studied here carrıed only one pup, and, according to date of capture (last week
of June or first of July), sıze of the fetuses, and the results reported by RAaceEy and SwIFT
(1981), gestation was probably far advanced.

The increase of body weight in these ?. pıpistrellus at late gestation is not due only to the
presence of the conceptus since the excess weight persists after calculating the conceptus
free maternal body weight. Such weight increase in gestating females has also been
observed in other bat species (STEBBINGS 1976; STUDIER and O’FARRELL 1976; SPEAKMAN
and RAacey 1987). The mass increase in P. pipistrellus ıs partially explained by the
enlargement of the mammary glands.

In our study the weight of subcutaneous WAT ın pregnant bats showed no significant
difference compared with that in non-pregnant individuals. However, although increased
fat during pregnancy has been detected in M. Ilucifugus, P. auritus (SPEAKMAN and RACEY
1987) and M. grisescens (KRULIN and SEALANDER 1972), ıt cannot be completely confirmed
since STUDIER and O’FARRELL (1976) found no evidence of fat increase during pregnancy
in M. lucifugus and M. thysanodes.

Our results for WAT weight and lıpid content suggest that P. pipistrellus responds to
the increased energy requirements of pregnancy by increasing the caloric content of the fat
rather than by increasing the amount of fat stores, as PısroLE (1989) has also suggested in
pregnant Eptesicus fuscns.

The main alteration ın the pattern of total lipid fatty acıd composition in WAT of
P. pipistrellus caused by pregnancy could indicate increased lipogenesis in this tissue during
gestation as also occurs in pregnant rats (PaLacın et al. 1991).

Brown adipose tissue, tradıtionally associated with thermogenesis, plays an important
role in maıntaining the energy balance in small mammals (ROTHwELL and Stock 1983;
Hımms-HAGEN 1983; TRAYHURN 1984). It has been previously shown that there ıs an
increase of BAT during late pregnancy in the rat (Acıus and WırLLıamson 1980; LoPEZ
Luna et al. 1991). The present results show that BAT tissue weight and total lipids remain
unmodified in P. pipistrellus during pregnancy.

Sımilarly, the fatty acıd profile of BAT from pregnant bats was similar to non-pregnant
ones; however the levels of the 18:2 and 20:4 fatty acıds were decreased, which could be

228 Pılar Lopez-Luna, F. Arevalo, M. J. Burgos and N. del Hoyo

related to their importance in the prostaglandin production involved in the normal
parturition process (ARAHUETES et al. 1982).

Although BAT wheight, lipid content, and fatty acıd do not differ between pregnant
and non-pregnant bats, the phospholipid levels are higher in pregnant than in non-
pregnant bats. The scarce bibliography does not help explaining the rise we found in BAT
phospholipids from pregnant bats.

Cold and diet have also been reported to modify BAT phospholipid composition and/
or content (GIRADIER 1983; IDE and SuUGANO 1988; AREVALO et al. 1990). MAsoro (1968)
suggests that muscle phospholipids may serve as an energy source, but we cannot affirm
whether the phospholipids of brown fat are an energy source or not.

Several factors, like quantity and nature of food, temperature, or physiological state,
are known to affect the fat composition of mammals (PEARcE 1983). Present findings show
that ın P. pipistrellus gestation may produce changes in the length and/or unsaturation of
some of the fatty acid chains as well as in their fat composition. In this sense LOcCkwooD et
al. (1970) have shown that sex hormones are involved in controlling the specific activities of
lipogenic enzymes in both male and female rats. In fact, and as PısToLE (1989) suggests,
bats alter their fat composition in response to a particular energy need such as hibernation,
pregnancy or lactation.

Acknowledgements

We wish to thank Dr. PEREZ-SUAREZ for his field assistance and CAROL F. WARREN of the translation
service at the Instituto de Ciencias de la Educaciön (ICE) of the University of Alcalä de Henares. This
investigation was supported in part by the Comisiön Asesora de Investigaciön Cientifica y Tecnica,
Project: PB85--0242, Spain.

Zusammenfassung

Fettspeicher in trachtigen und nicht trächtigen Fledermäusen (Pıpistrellus pipistrellus)

Weißes und braunes Fettgewebe von graviden und nicht graviden Fledermaus-Weibchen (Pıpistrellus
pipistrellus) wurde analysiert, um trächtigkeitsbedingte Änderungen im Gesamtgehalt von Lipiden,
Fettsäuren und Phospholipiden festzustellen. Der Gesamtanteil an Lipiden war in weißfem Fettgewebe
bei trächtigen höher als bei nicht trächtigen Fledermäusen. Das gleiche traf für die Phospholipide ım
braunen Fettgewebe zu. Die Trächtigkeit bewirkt bei P. pipistrellus in beiden Typen von Fettgeweben
Änderungen in der Länge der Fettsäuren und deren Anteil an ungesättigten Bindungen.

References

Acıus, L.; WıLLıamson, D. H. (1980): Lipogenesis in interscapular brown adipose tissue of virgin,
pregnant and lactating rats. The effects of intragastric feeding. Biochem. J. 190, 477480.

ARAHUETES, R. M.; FRAILE, A.; SUAREZ, A. (1982): Lipid metabolism in pregnancy. Perfusion ot liver
in pregnant rats. Rev. Esp. Fisol. 38, 201-208.

ARE£VALO, F.; BurGos, M. J.; DEL Hovo, N.; L6Pez-Luna, P. (1990): Seasonal variations in the lipid
composition of white and brown tissues in the bat Pipistrellus pipistrellus. Comp. Biochem.
Physiol. 95, 535-539. |

BEAToN, G. H.; BEARE, J.; Ryv, M. H.; McHewrry, E. W. (1954): Protein metabolism in the pregnant
rat. J. Nutr. 54, 291-313.

FoLchH, J.; LEEs, M.; STANLEY, G.H. S. (1957): A simple method for the isolation and purification of
total lipids from animal tissues. J. Biol. Chem. 226, 497-509.

GITTLEMAN, J. L.; THOMPson, $. D. (1988): Energy allocation in mammalian reproduction. Amer.
Zool. 28, 863-875.

GIRARDIER, L. (1983): Brown fat. An energy dissipating tissue. In: Mammalıan thermogenesis Ed. by
L. GIRARDIER and M. ]J. STock. London: Chapman and Hall. Pp. 50-98.

HERRERA, E.; LASUNCION, M. A.; GÖMEZ-CORONADO, D.; ARANDA, P.; LÖPEZ-Luna, P. (1988):
Role of lipoprotein lipase actıvity on lipoprotein metabolism and the fate of circulating trigly-
cerides in pregnancy. Am. J. Obstet. Gynecol. 158, 1575-1583.

Hımms-HaGen, J. (1983): Brown adıpose tissue thermogenesis in obese anımals. Nutr. Rev. 41,
261-267.

Hyrren, R. E.; LertcH, I. (1971): The physiology of human pregnancy. 2nd ed. Oxtord: Blackwell.

Lipid deposits in pregnant and non-pregnant bats 22

Hyrren, R. E.; THomPpson, A. M.; TAGGART, N. (1966): Total body water ın normal pregnancy.
Obstet. Gynecol. Br. Commonw. 73, 553-561.

Ipe, T.; Sucano, M. (1988): Effects of dietary fat types on the thermogenesis of brown adıpocites
isolated from rat. Agrıc. Biol. Chem. 52, 511-518.

KruLın, G. $.; SEALANDER, J. A. (1972): Annual cycle of the gray bat, Myotis grisescens. Comp.
Biochem. Physiol. 42 A, 537-549.

Kurra, A. (1986): Factors affecting the resting and postflight body temperature of little brown bats,
Myotis lucıfigus. Physiol. Zool. 59, 429-438.

LockwooD, E. A.; Baırey, E.; Taytor, C. B. (1970): Factors involved in changes in hepatic
lipogenesis during development of the rat. Biochem. J. 118, 155-162.

LöPpez-Luna, P.; MuNoz, T.; HERRERA, E. (1986): Body fat in pregnant rats at mid and late
gestation. Life Science 39, 1389-1393.

LöPEz-Luna, P.; MAIER, I.; HERRERA, E. (1991): Carcass and tissue fat content in the pregnant rat.
Biology Neonate 60, 29-38.

Masoro, E. J. (1968): Physiological chemistry of lipıds in mammals. Philadelphia: Saunders.

NAISMITH, D. J.; RICHARDSON, D. P.; PRITCHARD, A. E. (1982): The utilization of protein and energy
during lactation in the rat, with particular regard to the use of fat accumulated in pregnancy. Br. ].
Nutr. 48, 433441.

Paracin, M.; Lasuncıön, M. A.; Asuncıon, M.; HERRERA, E. (1991): Circulating metabolite
utilization by periuterine adıpose tissue “in sıtu” in the pregnant rat. Metabolism 40, 534-539.
PEARCE, ]J. (1983): Fatty acıd synthesis ın lıver and adıpose tissue. Symposium on comparative aspects

of fatty acid metabolism. Proc. Nutr. Soc. 42, 263-271.

PıstoLe, D. H. (1989): Sexual differences in the annual lipid cycle of the big brown bat Eptesicns
fuscus. Can. J. Zool. 67, 1891-1894.

PuLıoo, J. A.; per Hoyo, N.; PEREZ-ALBARSANZ, M. A. (1986): Composition and fatty acıd content
of rat ventral prostate phospholipids. Biochim. Biophys. Acta 879, 51-55.

Racer, P. A. (1973): Environmental factors affecting the length of gestation in bats. J. Reproduct.
Fertil. Suppl. 19, 175-189.

— (1982): Ecology of bat reproduction. In: Ecology of bats. Ed. by H. T. Kunz. New York: Plenum
Press Rp. 57-102.

Racev, P. A.; SPEAKMan, J. R. (1987): The energy cost of pregnancy and lactation in heterothermic
bats. In: Reproductive energetics in mammals. Ed. by A. S. I. Lonpon and P. A. Racey, New
York: Oxford University Press. P. 371.

Racey, P. A.; Swırt, $S. M. (1981): Variations ın gestations length in a colony of pipistrellus bats
(Pipistrellus pipistrellus) from year to year. J. Reprod. Fertil. 61, 123-129.

ROTHWELL, N. J.; STOCK, M. J. (1983): Diet-induced thermogenesis. In: Mammalıan thermogenesis.
Ed. by L. GIRADIER and M. J. STock. London: Chapman and Hall. Pp. 208-233.

ROUSER, G.; SIAKOTOS, A. N.; FLEISCHER, $. (1966): Quantitative analysis of phospholipids by thin
layer chromatography and phosphorus analysıs of spots. Lipids 1, 85-87.

SADLEIR, R. M. F. S. (1984): Ecological consequences of lactation. Acta Zool. Fennica 171, 179-182.

SCHNEIDER, J. S.; WADE, G. N. (1987): Body composition, food intake, and brown fat thermogenesis
in pregnant Djungarıan hamsters. Am. ]J. Physiol. 253, R314-R 320.

STEBBINGS, R. E. (1976): Studies on the population ecology of British bats. Ph. D. Thesis, Univ. East
Anglıa.

SPEAKMAN, J. R.; Racey, P. A. (1987): The energetics of pregnancy and lactation in the brown long-
eared bat Plecotus auritus. In: Recent advances in the study of bats. Ed. by M. B. FEnton,
P. Racey, and J. M. V. Rayner. Cambridge: Cambridge University Press. Pp. 367-394.

STUDIER, E. H.; O’FARRELL, M. ]. (1976): Biology of Myotis thysanodes an M. Incifugus (Chiroptera:
Vespertilionidae). III. Metabolism heart rat and general energetics. Comp. Biochem. Physiol.
54 A, 423432.

TowsenD, C. R.; Carrow, P. (1981): Physiological Ecology: An evolutionary approch to resource
use. Oxford: Blackwell.

TRAYHURN, P. (1984): The development of obesity in anımals: the role of genetic susceptibility. Clin.
Endocrinol. Metab. 13, 451-474.

Authors’ addresses: Prof. P. Lörez Luna, Departamento de Fisiologia y Farmacologia; Dr.
F. AREvALO, Departamento de Biologia Animal; Dr. N. per Hoyvo, Depar-
tamento de Bioquimica y Biologia Molecular; M. J. BurGos, Departamento de
Fisiologia y Farmacologia, Universidad de Alcalä, E-28871 Alcalä de Henares,
Madrid, Spain

Z. Säugetierkunde 59 (1994) 230-235
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

On the suckling behaviour of Alpine chamois
Rupicapra rupicapra rupicapra

By K. RuckstuHL and P. InGoLD

Zoology Department, University of Berne, Switzerland

Receipt of Ms. 24. 9. 1993
Acceptance of Ms. 9. 12. 1993
Abstract

The suckling behaviour of 9 mother-kid pairs of alpıne chamois (Rupicapra rupicapra rupicapra) was
studied from May to October 1991. Duration of suckling bouts decreased with age of the young, from
a mean of 48 sec in the first month, to 22 sec in October, whereas the time between suckling increased
from 25 min to over 160 min. Suckling success (number of successful suckling attempts over all
attempts) of the young decreased from the first month up to the time when the young were weaned.
Suckling was always terminated by the mother (except 2 times). After the third month of life, mothers
only allowed suckling by their kids from the side. Each suckling attempt from behind was rebuffed.

Introduction

Lactation is of fundamental importance for young mammals, for growth and building up
body reserves (CLUTTON-BRocK 1982). While the young attempt to obtaın as much
parental care (e.g. milk) as possible, mothers are selective in maximising the difference
between the benefit and cost of parental care. If they invest too much in their current
offspring, ıt may decrease their chances of survival and the number of future offspring
(especially when the females are young). This asymmetry of benefits and costs between
mothers and young gives rise to a parent-offspring conflict over the amount and termina-
tion of parental investment (TRIvErs 1985). Studies on ungulates indicate that the young
attempt to suckle as long as possible, while mothers increasingly refuse to suckle them
(e.g. BERGER 1979; CLUTTON-BRocK 1982). Moreover, it could be of some importance
from which position (i.e. from behind or from the sıde) a kıd attempts to suckle.

In chamois, no detailed study is yet available describing the suckling behaviour of
females and kids for the entire lactation period.

The aim of the present study was to investigate the suckling behaviour of mothers and
kids over the entire suckling period: the duration and frequency of suckling bouts and the
suckling success rate of the kıds. Furthermore, it was of interest to determine who
terminates such suckling bouts and ıf mother and kid agree or ditfer ın their preference ot
suckling positions.

Material and methods

Study area and animals

Nine individually marked female chamois and their kids were observed on 70 days (617 hours)
between May and October 1991, in the region of Augstmatthorn, Bernese Oberland, Switzerland.
The study site lies within the borders of an area where hunting is prohibited. Chamois were captured
in a wooden trap (with salt as bait) and individually marked with yellow-coloured plastic stripes glued
around their horns. Females with their young spent most of their time on grassy slopes at an altıtude
between 1400 and 2137 m a.s.l. with only little tree cover, where they could be easily observed
(InGoLD and MARBACHER 1991).

Time of birth was estimated to be mid-way between the last observation, that a female was seen
without a kid and the first day when she was observed with it. Kids were not marked, but as female

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5904-0230 $ 02.50/0

On the suckling behaviour of Alpine chamoıs 231

chamois only suckle their own kids (KRÄMER 1969), ıt was possible to determine which kid belonged
to each marked female.

The sex of the kıd was determined through the kid’s position while urinating (Tab. 1).

Data on suckling behaviour were collected from an observation point where most of the slope used
by the chamois during daytime was visible. Observations were made with a spotting scope (30x 60)
and binoculars (10x 40).

Table 1. Age of female Alpine chamois in years, sex and date of birth of young and observation
time of each mother-kid pair (in days and hours visible), at the Augstmatthorn, Switzerland, 1991

Age of female Date of birth Sex of kid Days of observation

08.-20. May
21.-26. May
08.-20 May
20.-22. May
08.-20. May
08.-20. May
03.-12. June
08.-20. May
08.-20. May

BSH nn

5
5
5
7
8
8
8
1
3

REN

Data collection

Each mother-kid pair was observed for 9 to 17 days, depending on their presence and visibility. If
several marked females were visible at the beginning of data collection, the female with the least
observation days or hours was chosen as focal anımal. Data on suckling behaviour were collected with
the focal anımal-continued sampling method. Durations of suckling bouts for females that were not
focal animals were collected ad libitum (ALtmann 1974).

Duration of suckling bout was measured as the time from the first contact of the kid with the
udder, until the kid itself or the mother terminated the suckling. Suckling bouts were timed to the
nearest second.

During a suckling bout the kid was either in contact or pulled at the udder. Suckling attempts were
considered successful, when they lasted more than 5 seconds. Attempts where no contact was made
with the udder, or bouts where the contact was shorter than 5 seconds, were treated as unsuccessful
suckling attempts. A kid’s suckling attempt was rejected, when the female did not allow it to suckle or
even approach the udder. Time between suckling bouts was measured in minutes from the end of the
last successful suckling bout until the commencement of the next bout.

Suckling success was calculated from the number of successful suckles divided by the number of all
suckling attempts.

The position of the kid relative to the mother during suckling was recorded after each suckling
bout or attempt.

Statistical analysis

Mean duration of suckling bouts and time between suckling were calculated separately for each
mother-kid pair for the first six months of life of the young.

Differences in duration of suckling bout and time between suckling (dependent variables) were
tested with a 2-way ANOVA (Zar 1984) according to individual variation and age of the kids. Age of
the kids and individual differences were treated as independent variables.

Suckling success: the mean suckling success was calculated (for each kid) for each month of life and
tested with a Spearman rank correlation coefficient.

Results

The duration of suckling bout decreased with age of the young (Fig. 1; 2-way ANOVA,;
F = 63.77, df=1, p<0.001). Differences in suckling duration between individuals had no
influence on the decrease in suckling duration over the months observed (F = 1.01, df= 8,
p = 0.44).

22 K. Ruckstuhl and P. Ingold

Although suckling was observed occasıionally in October, it did not occur in November
and after the rut (4 days of observation).

The time between suckling bouts increased during the first 5 months of the kids’ life
(Fig.2; 2-way ANOVA; F = 33.24, df = 1, p<0.001). Individual differences had no
influence on the average increase in time between suckling bouts (F = 0.99, df = 8,
p= 0.47).

Kids either suckled from the side (their body antiparallel to their mothers’) or from
behind (Tab. 2). They butted the udder 3 to 5 times before holding their heads still and

+ ae

T T T T

1 T
1 2 3 4 5 6 months
n=123 n=101 n=77 n=82 n=31 n=33

Fig. 1. Duration of suckling bouts in the first six months of life for Alpine chamois at the Augst-
matthorn, Switzerland, 1991. (Mean, minima, maxıma, 1° +3rd quartile). Data of 9 mother-kid pairs

minutes

350 7
300
250 |

200

150 7

100

50 7

(0) zZ T
1 2 3 4 5 months

n=54 n=26 n=39 n=28 n=5
Fig. 2. Time between suckling bouts in the first 5 months of life of Alpine chamois at the

Augstmatthorn, Switzerland, 1991. (Mean, minıma, maxıma, 1” are quartile). Data of 9 mother-kid
pairs

On the suckling behaviour of Alpine chamois 238
Table 2. Number of successful or unsuccessful suckling attempts from the “side” or “from behind”
position

Data from Alpine chamois at the Augstmatthorn, Switzerland, 1991. Data of 9 mother-kid pairs

Age in months

Position
Motallattempts 169° 77.

% successful A
% unsuccesful 29 49

b = from behind; s = from the side.

suckling. During the first months 75 % of the successful bouts were from the side. In the
second and third months, 88% and 97 % of all successful sucklings were from the side.
After the third month of life, suckling attempts from behind were always rejected by the
mother.

Suckling success in general decreased with the age of the kids (Fig. 3; Spearman Rank
correlation coefficient; rs = 3.64 N = 40, p<.0.05).

Suckling was always terminated by the mother (n = 474) except in May, when 2 kids
interrupted the suckling themselves (n=2 out of 163 suckling bouts; 9 individuals were
observed for a total of 40 hours).

In May, kids sometimes were observed attempting to suckle from mothers other than
their own, but each of these attempts was rebufted by the females, either by walking away
or by butting the kid away.

le

0,8 -

0,6 |
IL

0,4

0. 2

T T T
1 2 3 5 6 months
n=20 n=18 n=25 n=15 n=9 n=7

Fıg. 3. Suckling success (number of successful suckling attempts over all attempts) of chamois kids
during the first 6 months of life at the Augstmatthorn, Switzerland, 1991. (Mean, minima, maxima,
1°° +3” quartile), n = total number of suckling attempts. Data of 9 mother-kid pairs

234 K. Ruckstuhl and P. Ingold
Discussion

The decrease in duration of the suckling bout, frequency and success is very similar to
those described in other studies on ungulates (red deer: CLuUTTONn-Brock 1982; bighorn
sheep: BERGER 1979; FEsTA-BIANCHET 1988; bison: GREEN 1986). As the young grow
older, they appear to receive less milk and are admitted to the udder less often. In spite of
being rebufted the kids try to gain access, and after the 4th month of life the total number
of suckling attempts decreases. Weaning in chamois ıs a gradual process, with little conflict
between mother and young, and aggressive behaviour by the young was never observed, as
in monkeys (GoOMENDIO 1991; GOODALL 1990; TRIVERS 1974, 1985) towards the mother
when refused to suckle. Kids simply attempt to suckle again, after being rebuffed, or wait
for another opportunity. The mother decided how long and when she wanted to suckle her
kid. The probability of a kıd suckling successfully not only depended on the frequency and
duration of suckling bouts but also on the kid’s position towards the udder during suckling
attempts. Approaches from the side were more successful. After the third month of life, all
suckling attempts from behind were rejected by the mother. It probably was easier for the
mother to keep on walking when the kid attempted to suckle from behind. When it tried to
suckle from the side, the female had to lift her leg and step over her young. Perhaps the
mother also had better control (visible and olfactory) over the kiıd who wanted to suckle,
when it approached her from the sıde. Mother and kid therefore not only differed in their
interest in the duration and frequency of suckling bouts but also in their preference of the
suckling position.

Acknowledgements

We thank the Federal Game Department for supporting the study financially and the Game
Department of the Canton of Berne for its cooperation and P. NEUHAUS, M. GÜNTERT and D. SEP
for comments on earlier drafts of this paper.

Zusammenfassung

Zum Saugverhalten bei der Alpengemse Rupicapra rupicapra rupicapra

In einem Gebiet im Berner Oberland, Schweiz, wurde von Mai bis Oktober 1991 das Saugverhalten
von Gemsen (Rupicapra rupicapra rupicapra) untersucht. Während die durchschnittliche Saugdauer
bei den Kitzen über die Monate immer kürzer wurde, nahm die Zeit zwischen dem Saugen zu. Kitze
wurden immer seltener und in größeren Abständen gesäugt. Außer zweimal im Mai wurden alle
Saugakte von der Geiß abgebrochen. Saugversuche der Kitze blieben immer öfter erfolglos. Ab dem
dritten Lebensmonat der Kitze lehnte die Geiß alle Saugversuche des Kitzes von hinten ab.

References

ALTMANN, J. (1974): Observational study of behaviour: sampling methods. Behaviour 49, 227-267.

BERGER, J. (1979): Weaning conflict in desert and mountain bighorn sheep (Ovis canadensis): an
ecological interpretation. Z. Tierpsychol. 50, 188-200.

CLUTTON-Brock, T. H.; Guiness, F. E.; Arson, S. D. (1982): Red Deer. Chicago: University of
Chicago Press.

CZAKERT, H. (1985): Beiträge zur Verhaltensökologie des Gamswildes (R. rupicapra L.) ım FUST-
Projekt Achenkirch. PhD thesis, Universität für Bodenkultur Wien.

FestA-BIAncHET, M. (1988): Nursing behaviour of bighorn sheep: correlates of ewe age, parasitism,
lamb age, birthdate and sex. Anim. Behav. 36, 1445-1454.

GooDALL, J. (1990): Through a window. Thirty years with the Chimpanzees of Gombe. London:
Weidenfeld and Nicolson.

GOMENDIO, M. (1991): Parent/offspring conflict and maternal investment in rhesus macaques. Anım.
Behav. 42, 993-1005.

GREEN, W. C. H. (1986): Age-related differences in nursing behaviour among American bison cows
(Bison bison). J. Mammalogy 67, 739-741.

InsoLp, P.; MARBACHER, H. (1991): Dominance relationship and competition for resources among
chamois Rupicapra rupicapra rupicapra ın female social groups. Z. Säugetierkunde 56, 88-93.

On the suckling behaviour of Alpine chamois 255

KRÄMER, A. (1969): Soziale Organisation und Sozialverhalten einer Gemspopulation (Rupicapra rup.

L.) der Alpen. Z. Tierpsychol. 26, 889-964.
TRIVERS, R. L. (1974): Parent-Offspring Conflict. Amer. Zool. 14, 249-264.
TRIvERs, R. L. (1985): Social Evolution. Menlo Park, California; the Benjamin/Cimmings Publishing

Comp.
Zar, J. H. (1984): Biostatistical Analysıs. 2nd ed. Engelwood Cliffs, N. J.: Prentice-Hall.

Authors’ address: Lic. phil. nat. K. RuckstunL and Prof. Dr. P. Incorp, Zoology Department,
Ethology and Nature Conservation, University of Berne, Länggaßsstr. 27, CH-
3012 Berne, Switzerland

Z. Säugetierkunde 59 (1994) 236-245
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Genetic relationships of some Gazella species:
an allozyme survey

By M. VassarT, L. GRANJON, A. GRETH, and F. M. CATZEFLIS

Ecole Nationale Veterinaire de Toulouse, Toulouse, France; Museum National d’Histoire Naturelle,
Paris, France; National Wildhfe Research Center, Taif, Saudi Arabia; and Institut des Sciences de
l’Evolution, Universite de Montpellier II, Montpellier, France

Receipt of Ms. 22. 11. 1993
Acceptance of Ms. 17. 1. 1994

Abstract

An allozyme comparison of eight taxa in the genus Gazella (Bovidae: Artiodactyla) was conducted to
clarıfy the systematic relationships of endangered gazelles currently bred in Saudi Arabia for
reintroduction. Electrophoretic varıation at 16 genetic locı suggested that several similar taxa of
Arabian gazelles, namely G. gazella gazella, G. gazella erlangerı, G. gazella farasanı, and G. gazella
cora, belong to the same species of G. gazella sensu lato. Four other species proved to have diverged
genetically: G. thomsoni and G. dorcas, which cluster together, G. rufıfrons, and G. subgutturosa. The
subgenus Trachelocele, in which the latter species has been placed according to morphological
characteristics, is not supported. Polymorphism and heterozygosity values found in Gazella were
generally sımilar to average values reported for mammals. The results are discussed in terms of the
strategy to follow a conservation program that take genetic data into account.

Introduction

There ıs urgent need for conservation action concerning gazelles (RyDEr 1987), both in
captive and wild populations. Of the 12 species of Gazella (CoRBET and Hırr 1980), nıne
are considered vulnerable or endangered (IUCN 1988), mainly due to overhunting and
habitat destruction. In Saudi Arabia, where at least three “good biological species” (MAYR
1963) are found: G. saudiya, G. subgutturosa and G. gazella (see Fig. 1), the situation is of
particular concern (THOULESss et al. 1991). Over the last few years, tremendous efforts have
been undertaken towards the conservation of gazelles in this country (ABU-ZINADA et al.
1989). It ıs widely accepted that species conservation must be based on proper systematics
of the endangered taxa (RyDER 1986; GRoVvEs 1988).

Morphological characters such as size and shape of horns and skull have been previ-
ously used to establish systematics in Gazella (Groves 1969, 1983; LAnGE 1972; GROVES
and Lay 1985; ALapos 1986/1987). Color and coat patterns have also been used, but these
characters may vary due to environmental conditions (HARRISON and BATESs 1991; GROVES
and Lay 1985). Cytotaxonomy seems to be particularly informative, and can be helpful ın
characterizing some species, as the Indian-gazelle, G. bennetti (FuURLEY et al. 1988). Up to
now, only limited data on the genetic diversity in the genus Gazella are available based on
protein electrophoresis (TEMPLETON et al. 1987; GRANJoN et al. 1991). This technique has
proven to be useful ın a number of studies both for captıve breeding management purposes
(see WAYNE et al. 1986) and for systematic and phylogenetic studies (see BuTH 1984 for a
review). Advocating a phylogenetic basis for taxonomy, including subspecies groupings
(see CRACRAFT 1989 for a discussion of the phylogenetic species concept), the present
study shows the results of allozyme varıation in eight taxa of gazelles and proposes some
hypotheses on the phylogeny and conservation of this group.

In particular the following questions were posed: 1. Among the three species present ın

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5904-0236 $ 02.50/0

Genetic relationships of some Gazella species: an allozyme survey 297,

N FARASAN ISLANDS

=
iM

G.thomsoni

G.dorcas isabella

G.subgutturosa marica

G.gazella gazella

G. gazella cora

G.9.erlangeri
(Me)

| G. gazella muscatensis Bere il

Fig. 1. Distribution of eight taxa of Gazella in the Arabian peninsula and in East Africa. On Farasan
Islands lives G. g. farasani. Areas of sympatry for G. g. gazella and G. g. cora, G. g. cora and G. g.
erlangeri are temptative. Data from Grovzs (1985), Dorst and DAnDELOT (1972), HARRISON and
Bates (1991) and Kınaswoop and KumAmMOTOo (1988)

Saudı Arabia, G. subgutturosa has been considered very divergent morphologically from
all other members of the genus, and placed alone in the subgenus Trachelocele (ELLERMAN
and MORRISOoN-ScoTT 1951). Is this morphological dıvergence accompanied by significant
molecular genetic divergence? 2. GRovEs (1983) proposed that the insular gazelles living on
the Farasan Islands ın the Red Sea should be treated as a separate species, namely G.
arabica. However, GRoVvESs (1989) pointed out that G. arabica resembles G. gazella more
closely than any other species. Based on skull measurements, THOULEsSS and Ar Baskı
(1991) proposed to consider gazelles from the Farasan ıslands as a subspecies of G. gazella,
named G. g. farasani. Are electrophoretic data able to help ditferentiate between these two
hypotheses? 3. At least five subspecies have been attrıbuted to Gazella gazella sensu lato:
G. g. muscatensis, G. g. gazella, G. g. cora, G. g. farasanı and G. g. erlangeri (GROVvES
1989; THouLess and Ar Basrı 1991; GrovEs et al. 1994) based on morphological
characteristics. Do some fixed allelic differences characterize any of these taxa, hence
supporting the concept of separate gene pools in G. gazella sensu lato? 4. The taxonomic
position of G. rufifrons ıs not clear. GRovEs (1975, 1985, 1988) described the Red-fronted
Gazelle G. rufifrons with 7 subspecies: rufıfrons, laevipes, kanurı, tilonura, albonotata,
nasalıs and thomsoni. The last three subspecies have usually been grouped in a different
species: G. thomsoni (DorRst and DANDELOT 1972; CoRBET and Hırr 1980; Nowak and
Parapıso 1983). Is G. thomsoni a subspecies of G. rufifrons?

To address these questions, we have analysed by means of protein electrophoresis,
samples of gazelle species from the Arabian peninsula, and 3 African species.

238 M. Vassart, L. Granjon, A. Greth and F. M. Catzefls

Material and methods

Origin of the animals

Gazella thomsoni albonotata (n = 8): eight anımals from private collections in Saudi Arabia, all
originating from Sudan. This taxon is called G. rufifrons albonotata by Grovss (1985, 1988).

Gazella dorcas (n = 25): 16 individuals from different private collections in Saudi Arabia, 5 from
Taif breeding center (National Wildlife Research Center, Saudi Arabia), and 4 from Thumama
breeding center (King Khaled Wildlife Research Center, Saudi Arabia). The probable origin for all is
Sudan. These anımals would be representative of the taxon G. dorcas isabella mentioned by GrovEs
(1969) and Aranos (1986/1987).

Gazella subgutturosa (n = 30): 30 anımals from Thumama center. They represent the subspecies G.
s. marica (NADER 1989). This species ıs endangered (lucn 1988) and limited to a very small geographic
area in the Arabian peninsula, but is still common ın Mongolia and in some parts of Russia (GRovES
1988).

Gazella gazella gazella (n = 16): 16 individuals from the Thumama center. This gazelle is locally
abundant in Northern Israel.

Gazella gazella cora (n = 7): 5 anımals from Taif center and 2 from a private Saudi collection; they
may have originated in Saudi Arabia. G. g. cora is rare and endangered in the mountains east of the
Red Sea (Grovzs 1988). i

Gazella gazella farasanı (n = 5). 4 individuals from a private collection on Farasan Kebir island
(Red Sea) and 1 wild gazelle from Farasan Kebir island. This taxon, distributed on Farasan islands,
was previously called G. arabica by Grovss (1985) but is more probably a subspecies of G. gazella
(THourzss and Ar Basrı 1991). Its population status has been recently examined by FLAMAND et al.
(1988), who found that these gazelles were still present ın fairly large numbers.

Gazella gazella erlangeri (n = 15): 6 anımals from Taif center, some of them thought to have been
caught in the South of Saudi Arabia, 5 anımals held in private collections in Saudi Arabia, 4 individuals
from a pet shop in Djeddah. For the latter samples the origin was reported to be the region of Aden
(Yemen). This subspecies has recently been described by GrovsSs et al. (1994) after morphological and
skull measurement comparisons wıth G. g. cora and G. g. muscatensıis..

Gazella rufifrons (n = 1): 1 specimen from a private collection in Saudi Arabia. The origin of this
sample is unknown. This gazelle could belong to the subspecies kanuri (synonym = centralis) or
laevipes (synonym = haslerı) which are very sımilar.

Electrophoresis

Horizontal starch gel electrophoresis was performed on blood extracts collected by jugular puncture.
An isotonic saline solution was added to the samples, and the plasma was separated from the red blood
cells by centrifugation. Samples were duplicated for a reference collection and stored at -30°C until
electrophoresis was performed.

Electrophoresis, staining of the proteins, and scoring the results were conducted according to
PASTEUR et al. (1988). Many allozyme systems were assayed under different electrophoretic condi-
tions, but only 16 locı that gave consistent results were retained (Tab. 1). Statistical and phylogenetic
treatments were performed without taking hemoglobin into account, as the A and B subunits were not
separated before running the gels. This prevented us from scoring the different alleles. Nevertheless,
haemoglobin pattern appears very useful to distinguish G. subgutturosa as will be discussed further.

For each taxon, the percentage of polymorphic locı (P 95 %, i.e. a locus ıs considered polymorphic
when the frequency of the most common allele is not higher than 0.95), mean number of alleles per
locus (A), and mean heterozygosity (H) were calculated. Genetic distances between samples were
estimated using formulae from NEI, “unbiased minimum distance” (1978) and Rocers (1972). The
Arabian oryx (Oryx leucoryx) was used as an outgroup to root the tree for a phenetic analysıs, with
only 14 locı, because MPl and ACP could not be scored in this species. Allelic frequencies for Arabian
oryx were published by Vassarr et al. (1991). Phenetic analysis was performed via the Distance
Wagner method (in BIOSYS-1, SvOFFORD and SELANDER 1989), using the RoGERs distance (1972) as
modified by WRIGHT (1978).

For cladistic analysis, allelic frequencies were coded as „locus-as-character“ (see BurH 1984).
Alleles with low frequencies were kept even though this might have introduced a bias for small sample
sizes. Data for oryx were not used here, as the information they bring after coding was too weak.
Cladistic analysis was performed using PAUP 3.0 (SwoFFoRD 1990). Unordered option was used, and
the consensus tree obtained with branch-and bound search after 1000 bootstrap replications was
finally retained (threshold of 95 %).

Genetic relationships of some Gazella species: an allozyme survey 239
Table 1. Enzymes surveyed and electrophoretic buffers used

Enzymes Tissue Locus Buffer (pH)

GOT: Aspartate aminotransferase (EC 2.6.1.1) RBC
ACP: Acıd phosphatase (EC 3.1.3.2) RBC
DIA: Diaphorase (EC 1.6.4.3) RBC
ES 10-14: Esterase 10 and 14 (EC 3.1.1.X) RBC
GLO: Glyoxalase (EC 4.4.1.5) RBC
GPI: Glucose phophate isomerase (EC 5.3.1.9) RBC
LDH: Lactate dehydrogenase (EC 1.1.1.27) RBC
MDRH: Malate dehydrogenase (EC 1.1.1.37) RBC
MOD: Malic enzyme (EC 1.1.1.40) RBC
MPI: Mannose phosphate isomerase (EC 5.3.1.8) RBC
NP: Nucleoside phosphorylase (EC 2.4.2.1) RBC
IPO: Superoxide dismutase (EC 1.15.1.1) RBC
ALB: Albumin Serum
ESEIenEsterasenl (EE3:11.29) Serum
TRF: Transferrin Serum

TME 6.9/6.9
TC 6.4/6.0
TC 6.4/6.0
TME 6.9/6.9
TBE 8.6/8.6
TC 6.4/6.0
TC 6.4/6.0
TC 6.4/6.0
TC 6.4/6.0
TC 6.4/6.0
TME 6.9/6.9
TC 6.4/6.0
LiOH 8.3/8.1
LiOH 8.3/8.1
LiOH 8.3/8.1

zz ss ze ee zT

TME = Tris Maleate, TC = Tris Citrate, TBE = Tris Borate EDTA, LiOH = Lithium Hydroxyde.

Results

Among the 8 Gazella taxa, 8 locı were found to be polymorphic; mean heterozygosities
(H) varıed from O to 0.085, mean numbers of alleles per locus (A) from 1.0 to 1.31, and
percentages of polymorphic locıi (P) from 0 to 18.7 (Tab. 2). Of the eight polymorphic locı,
only four exhibited significant varıability: ES14, MOD, NP, and TRF (Tab. 2). NP with
three alleles was polymorphic in five of seven taxa. ROGERS (and NET) genetic distances
(Tab. 3) varıed from 0.012 (0.001) to 0.304 (0.322) among taxa. IPO showed a fixed allelıc
difference between G. thomsoni, G. dorcas and G. rufifrons, on the one hand, and the five
other taxa, on the other. GOT, DIA and GPI displayed minor varıation, with a rare (less
than 5 % frequency) allele in only one taxon each.

According to phenetic analysis using the Wagner method (Fig. 2), G. subgutturosa
clusters together with the 4 samples of G. gazella sensu lato, and appears more similar to
G. g. erlangeri mainly due to sımilar allelic frequencies at the NP locus. When hemoglobin
data are considered, all G. subgutturosa samples had a unique pattern which distinguishes
that species from all others (Fig. 4). On the phenogram, G. rufıfrons lies in an intermediate
position between G. dorcas/G. thomsoni and the cluster of G. gazella/G. subguttnrosa.

For cladistic analysıs, allelic frequencies were coded according to the qualitative method
described in BurH (1984) (Tab. 4). The unrooted cladogram obtained after parsımony
analysıs (Fig. 3) confirms to a large extent the picture obtained with the phenetic analysıs.
Most of the samples in the G. gazella/G. subgutturosa group are characterized by minor
autapomorphies (corresponding to low frequency alleles). G. rufifrons and the pair G.
dorcas/G. thomsoni are clearly distinguishable from each other and from the G. gazella/G.
subgutturosa group. Two synapomorphies separate G. rufifrons from each of the other two
clusters.

Discussion
These results raise two points that require further discussion: the systematic ımplications

of these allozyme data, and their implications for the conservation biology of the Arabian
gazelles. It should be kept in mind that all the samples used here came from captive

M. Vassart, L. Granjon, A. Greth and F. M. Catzeflis

240

00°I
00°1

00°I

00°1

00°I

00°1
00° 1

00°1

xKı09n2] ©)

00°1

00°1

00°1

00°1

00°1

00°I

00°I

(SI =N)
1193uvJ49 3 °9

90] tydıowdjod jo uonıodord = %,66 A !sN90] 19d sajajje Fo ssquınu ueau = Y ‘A11so3Akzo1Iay urauı = H ‘azıs apdwes = N

0
901
0’E

00°I

00° I
09°0
0r'0

00° I

00°I

00°I

00° I

00°I

(S=N)
1upspirf[ 23'9

ao
90° 1
80

00°1

00° I

60

ZOO

00°1

00°I

00'1

00° I

Z=N)
v109 3°

00°I

00°1

00°I

L60

00

(II =N)
vpjozvd 8 °9

Sc
6l I

IF“
c0’0
L6'0
00°1
O1’0
06°0
00°1
soo
60
00°I
00°I

00°I

(0E=N)
psouniIndqns ‘9

00° 1

00°I

00°1

00°1

00°1

00° 1

00°I

(I=N)
suo4fılnı ‘9

71]
LEI

1474
r0’0
960
00°I
OL’O
06°0
890
ceo
00°I
r9°0
959
860
co’o
00°I

(G-= Ni)
sy940p ‘DI

ejj>zen) jo sajpdures g ur 1907] afgerreA 107 sarmuanba.ay Sıpayjy 'Z 279P1L

L’8l
Sal

gg
940
cl’d
cl’o

00° 1
00°1

390
20

00° I

E90

LEO

00°I

00°I

Ol
001
06

Ocl
001
DE

ora1
001
09

Ol
001
06

07a
001
08

sol
001
06

ora1
001
08

ra
001
08

(8=N)

zuoswog? 'D

[21V

%S6d
V
%H

JUL

Odl

dN

AON

IdD

v1 SA

vIa

LOD

sn907T

Genetic relationships of some Gazella species: an allozyme survey

populations, often derived from a few individuals, and
therefore have probably been subject to sampling ef-
fect and subsequent inbreeding and genetic drift.
Thus, they may not reflect allele frequencies of the
parental populations and the results inferred from
these samples should be taken with caution. Unfortu-
nately, it is difficult to correct for these biases because
the captive history of these samples is not well
documented.

Our allozyme analysis does not support G. subgut-
turosa as an isolated taxon in the genus Gazella, or the
concept of the subgenus Trachelocele for this species.
This is in contradiction to the relationships suggested
by ELLERMAN and MOoRRISoN-ScoTT (1951), HAL-
TENORTH (1963), and GRovESs (1969). Rather, G. sub-
gutturosa appears to be related to Gazella gazella
sensu lato. Among the protein systems considered
here, hemoglobin appears to represent the only dis-
criminative one between these two taxa (see Fig. 4).

The clustering of G. thomsoni and G. dorcas ın the
cladistic analysis (2 synapomorphies represented by
ES14 and MOD) is surprising, since this was never
suggested by studies of comparative morphology.
From our data it ıs difficult to support the view of
Grovss (1975, 1985, 1988) or LanGE (1972), who
described G. thomsoni as a subspecies of G. rufıfrons.
However, chromosome numbers (58) also appear
identical in G. thomsoni and G. rufıfrons (VASSART,
unpubl. results). Considering allelic frequency (see
ES14) G. rufifrons could belong to the G. thomsoni
group, but with a single specimen for G. rufıfrons, thıs
conclusion would be speculative.

There are no standard levels of genetic divergence
associated with subspecies or species rank (see exam-
ples in LinneLL and Cross 1991 and references ın
CRACRAFT 1989). However, our results suggest that
several Arabian gazelles do belong to Gazella gazella
sensu lato, namely, G. g. erlangeri, G. g. farasanı, G.
g. gazella, and G. g. cora. The small genetic distances
between them are not larger than those found between
geographic samples in other ungulates (see, for in-
stance, HARTL and REIMOSER 1988; HarTL et al.
1990). The position of G. g. cora well inside the G.
gazella group does not agree with the hypothesis of G.
g. cora being the Arabian representative of G. dorcas
(Grovss 1988). This has been confirmed by cytogene-
tie results: G. g. cora, G. g. gazella, G. g. erlangeri and
G. g. farasani have 35 chromosomes in males and 34 in
females (VAssarT et al. 1993) whereas G. dorcas have
31 chromosomes in males and 30 in females.

Also, and contrary to the hypothesis of GrovEs
(1985) and Groves and Lay (1985), G. arabica (our G.

Table 3. Rogers (above diagonal) and Nei’s (below diagonal) genetic distances between 8 taxa of Gazella, and Oryx leucoryx

G. g. cora G. rufıfrons O. leucoryx

G. thomsoni

ES
N)
80
S

—
65}
&%

&

G. dorcas G.g. gazella G.g. farasanı

G. subgutturosa

G. subgutturosa

G. g. gazella

G. g. farasani

G. g. erlangeri
G. thomsoni
G. g. cora

G. rufifrons
O. lencoryx

241

242 M. Vassart, L. Granjon, A. Greth and F. M. Catzeflıis

mm amt üpppepappeetmt m
0.00 0.08 0.15 0.23 0.30 0.38 0.45
G. subgutturosa

G. g. erlangeri

G. g. gazella
G.g. cora
G. g. farasani
G. rufifrons

G. thomsoni

G. dorcas
Oryx leucoryx

Fıg. 2. Dendrogram derived from the WAGNER procedure on the matrix of Rogers genetic distances

G. dorcas
G. subgutturosa

DIA-2
G. rufifrons GPI-2

NP-2
2 G. 9. gazella
GOT
G. g. erlangeri
NP-4
TRF-4 ES14-1 MOD-1 NP-3 =
. cora . q. farasani

G. thomsoni

Fig. 3. Cladogram obtained with “locus as character”” coding using branch-and-bound search after
1000 bootstrap replications. Synapomorphies are underlined, other characters are autapomorphies.
The arrow points to the root leading to the outgroup (see Fig. 2)

Table 4. Qualitative coding of allelic presence in sample data set from table 2
See text for explanations

=)
®)
4
=

G. thomsoni
. dorcas

. rufifrons

. subgutturosa

. 8. cora
. g. farasanı
. 8. erlangeri

ma N mm
ma m N m
Mama ll (yJNM
mm DD
Mean Mm
RW DW WeamND —
Mark | NDMNDM
NDDNDNDHN WW]

G
G
G
G. g. gazella
G
G
G

g. farasani) should not be treated as a different species, but as a morphological and/or
geographical race or subspecies of G. gazella: G. g. farasanı as proposed by THouLEss and
Ar Basrı (1991).

As stated by Furrey et al. (1988: page 48), “no common agreement has yet been
reached on the number of genuine species within [Gazella]. . ..”, because of the consider-
able morphological variation shown by some species and the possibility of morphological

Genetic relationships of some Gazella species: an allozyme survey 243

&

GRGRGERGTGIT.

Fig. 4. Hemoglobin gel with TBE 8.6/8.6 buffer: R; G. subgutturosa, G; G. gazella, T; G. thomsoni.
G. dorcas has the same patterns as G. gazella

convergence between taxa as illustrated by the cladistic analysıs of skull and jaw characters
of Aranos (1986/1987). For this reason, we feel that a non-morphological approach, such
as allozyme electrophoresis, should be used for reconstructing the systematics and
evolutionary relationships of the gazelles. Examination of other species in the genus is
needed, as well as the scoring of a greater number of locı.

With respect to conservation issues, GROVES (1989) stated that „for conservation
purposes it is quite clear that each of the Arabian forms [G. subgutturosa marıca, G. gazella
cora and G. dorcas saudiya] represents a unique gene pool.” Our data on the former two
species support this view. But, as far as local forms are concerned, we have to deal with the
“dilemma of subspecies”” (RyDEr 1986) when considering G. g. gazella, G. g. cora, G. g.
farasani, and G. g. erlangeri. On the basis of our results, they are genetically very similar,
however, from a conservation biologist’s point of view, they might represent “unique gene
pools,”” each one of which adapted to a particular local environment.

Among the different guidelines entering into consideration for selecting wild anımals to
be used for breeding purposes, genetic parameters such as mean heterozygosity and
percentage of polymorphic locı should be taken into account. The data derived from our
samples of Gazella (see Tab. 2) are similar to those found in natural populations of
artıiodactyls (see Baccus et al. 1983; Vassarrt et al. 1991, for reviews). The mean
heterozygosity for 184 species of mammals is 4.1% (+ 3.5 SD) (NEvo etal. 1984), a value
similar to the one observed for the three Gazella samples represented by more than 10
individuals in our study. On the other hand, the absence of genetic varıability (despite the
different orıgins of the samples) for G. g. erlangeri ıs potentially problematic. This gazelle
is only known from captive individuals and there is no protected area on its supposed range
(southwest of the Arabian peninsula). It is possible that the lack of polymorphism and
heterozygosity resulted from the population bottleneck experienced by this subspecies.
This could lead to inbreeding problems (O’Brızn et al. 1983).

These results have to be confirmed by other molecular techniques such as mitochondrial
DNA sequencing. This phylogenetic tool could be useful to differentiate all the different
species of gazelles even ıf they are of recent origin, but could be of limited value at the
subspecific level (CRonın 1992).

Acknowledgements

This work was carried out under the patronage of HRH Prince SauD Aı Faısar and Dr. ABUZINADA,
Secretary of the National Commission for Wildlife Conservation and Development of the Kingdom of
Saudi Arabia. The technical help of J.-F. AGn&se, G. BERREBI and L. DuranD during the establish-
ment of the laboratory experiments has been invaluable. We wish to express our thanks to two
anonymous reviewers and to M. A. CRonIn who provided valuable comments on the manuscript. We
thank J. F. Voısın for help in translating the summary into German. This is contribution 94-014 of the
Institut des Sciences de l’Evolution (Montpellier, France).

244 M. Vassart, L. Granjon, A. Greth and F. M. Catzeflıs

Zusammenfassung

Genetische Verwandtschaft einiger Gazella-Arten: Eine Allozym-Untersuchung

Acht Arten der Gattung Gazella (Bovidae: Artiodactyla) wurden mit einer Allozymanalyse vergli-
chen, um die genetische Verwandtschaft von Gazellen zu überprüfen, die derzeit in Saudi-Arabien zur
Wiedereinbürgerung gezüchtet werden. Die elektrophoretische Untersuchung von 16 polymorphen
Loci deutet darauf hin, daß mehrere ähnliche Taxa, nämlich G. gazella gazella, G. gazella erlangeri,
G. gazella farasani und G. gazella cora zur gleichen Art G. gazella sensu lato gehören. Vier andere
Arten sind davon deutlich genetisch verschieden: G. thomsoni und G. dorcas, die genetisch ähnlich
sind, sowie G. rufifrons und G. subgutturosa. Die Untergattung Trachelocele, der die letzte Art
aufgrund morphologischer Merkmale zugeordnet worden ist, konnte aufgrund der Allozymvergleiche
nicht bestätigt werden. Die Polymorphismus- und Heterozygotie-Grade der Gattung Gazella waren
insgesamt den bei anderen Säugern gefundenen Werten ähnlich. Die Ergebnisse werden im Hinblick
darauf diskutiert, welche Schlußfolgerungen sich aus den genetischen Untersuchungen für ein
angestrebtes Schutzprogramm ergeben.

References

INTERNATIONAL UNION FOR CONSERVATION OF NATURE (1988): Red list of threathened animals.
Gland, Switzerland: TUCN. '

Asu-ZınADaA, A. H.; Gorıup, P. D.; NADER, 1. A. (1989): Wildlife conservation and development in
Saudi Arabia. Riyadh: NCWCD.

Aranos, C.L. (1986/1987): A cladistic approach to the taxonomy of the Dorcas gazelles. Isr. J. Zool.
34, 3349.

Baccus, R.; Ryman, N.; SMITH, M. H.; REUTERWALL, C.; CAMERON, D. (1983): Genetic varıability
and differentiation of large grazing mammals. J. Mammalogy 64 , 109-120.

BuTH, D. G. (1984): The application of electrophoretic data ın systematic studies. Ann. Rev. Ecol.
Syst. 15, 501-522.

CoRBET, G. B.; Hırr, J. E, (1980): A world list of Mammalıan Species. London: British Museum;
Comstock Publishing Ass. Cornell University Press.

CRACRAFT, J. (1989): Speciation and ıts ontology: the empirical consequences of alternative species
concepts for understanding patterns and processes of ditfentiation. In: Speciation and its conse-
quences. Ed. by D. OTTE and J. A. EnDLER. Sunderland, Massachusetts: Sinauer Ass. Pp. 28-59.

Cronin, M. A. (1992): Intraspecific varıation in mitochondrial DNA of north American cervids. ].
Mammalogy 73, 70-82.

Dossr, J.; DANDELOT, P. (1972): A field guide of larger mammals of Africa. London: Collins.

EFFRoN, M.; BoGART, H.; KuMAMOTO, A. T.; BENIRSCHKE, K. (1967): Chromosome studies ın the
mammalıan subfamily Antilopinae. Genetica 46, 419-444.

ELLERMAN, ]J. R.; MORRISON-SCOTT, T. C. S. (1951): Checklist of Palearctic and Indian mammals 1758
to 1946. London: British Museum (Natural History).

FLAMAND, J. R. B.; THouLEss, C. R.; Tarwany, H.; AsmoDe£, ]J. F. (1988): Status of the gazelles of the
Farasan Inslands, Saudi Arabia. Mammalia 52, 608-610.

FuRLEyY, C. W.; TıcHy, H.; UERPMANN, H. P. (1988): Systematics and chromosomes of the Indian
gazelle Gazella bennetti (Sykes, 1831). Z. Säugetierkunde 53, 48-54.

GRANJON, L.; VAsSART, M.; GRETH, A. (1991): Genetic study of Sand gazelles from Saudi Arabia;
chromosomal and isozymic data. Z. Säugetierkunde 56, 169-176.

Grovzs, C. P. (1969): On the smaller gazelles of the genus Gazella Blainville, 1816. Z. Säugetier-
kunde 34, 38-60.

— (1975): Notes on the gazelles. Gazella rufifrons and the zoogeography of Central African Bovidae.
Z. Säugetierkunde 40, 308-319.

— (1983): Notes on the gazelles. IV. The Arabian gazelles collected by Hemprich and Ehrenberg. Z.
Säugetierkunde 48, 371-381.

— (1985): An introduction to the gazelles. Chinkara 1, 4-16.

— (1988): A catalogue of the genus Gazella. In: Conservation and biology of desert antelopes. Ed. by
A. Dixon and D. Jones. London: C. Helm. Pp. 193-198.

— (1989): The gazelles of the Arabian peninsula. In: Wildlife conservation and development ın Saudi
Arabia. Ed. by A. H. Asu-Zınapa, P. D. Gorıup, and I. A. Naper. Riyadh: NCWCD. Pp.
237-248.

Grovss, C. P.; Lay, D. M. (1985): A new species of the genus Gazella (Mammalia: Artiodactyla:
Bovidae) from the Arabian peninsula. Mammalia 49, 27-36.

GROVES, C. P.; THOULESS, C.; VAssarT, M. (1994): Subspecies of Gazella gazella ın Saudi Arabia.
Fauna of Saudi Arabia (in prep.)

HALTENORTH, T. (1963): Klassifikation der Säugetiere: Artiodactyla. Hdb. Zool. 32, 1-167.

Harrıs, H.; Hopkınson, D. A. (1976): Handbook of enzyme electrophoresis in human genetics.
Amsterdam: North-Holland.

Genetic relationships of some Gazella species: an allozyme survey 245

Harrıson, D. L.; Bates, P. J. J. (1991): The mammals of Arabia. Sevenooks, England: Harrison
Zool. Museum Publ.

HarTL, G. B.; REIMOSER, F. (1988): Biochemical variation in roe deer (Capreolus capreolus L.): are r-
strategists among deer genetically less variable than K-strategists? Heredity 60, 221-227.

HarTL, G. B.; Wırrıng, R.; Lang, G.; Krein, F.; KÖLLeEr, J. (1990): Genetic varıability and
differentiation in red deer (Cervus elaphus L.) of Central Europe. Genet. Sel. Evol. 22, 289-306.

Kıngswoon, S. C.; KuMAMOTO, A. T. (1988): Research and management of Arabian Sand gazelle in
the USA. In: Conservation and biology of desert antelopes. Ed. by A. Dixon and D. Jones.
London: C. Helm. Pp. 212-226.

LAnGe, J. (1972): Studien an Gazellenschädeln. Ein Beitrag zur Systematik der kleinen Gazellen,
Gazella (De Blainville, 1816). Säugetierkdl. Mitt. 20, 193-249.

Linneıt, J. C. D.; Cross, T. F. (1991): The biochemical systematics of red and sıka deer (genus
Cervus) ın Ireland. Heredita 115, 267-273.

Mayr, E. (1963): Animal species and evolution. Cambridge: Harvard Univ. Press.

NADER, I. A. (1989): Rare and endangered mammals of Saudi Arabia. In: Wildlife conservation und
development in Saudi Arabia. Ed. by A. H. Asu-Zınapa, P. D. Gorıup, and I. A. NADER.
Riyadh: NCWCD. Pp. 220-233.

Neı, M. (1978): Estimation of average heterozygosity and genetic distance from a small number of
individuals. Genetics, 89, 583-590.

NEvo, E.; BEILEs, A.; BEN-SHLOMO, R. (1984): The evolutionary significance of genetic diversity:
Ecological, demographic, and life history correlates. In: Evolutionary dynamics of genetic
diversity. Ed. by G. S. Manı. New York: Springer-Verlag. Pp. 13-213.

Nowak, R. M.; PARADISO, J. L. (1983): Walker’s mammals of the world. London: Johns Hopkins
Univ. Press.

O’BRIEN, $. J.; GOLDMAN, D.; MERRIL, C. R.; BusH, M. (1983): The cheetah is depauperate in genetic
varıatıon. Science 221, 459-462.

PASTEUR, N.; PASTEUR, G.; BONHOMME, F.; CATALAN, J.; BRITTON-DAVIDIAN, J. (1988): Practical
isozyme genetics. Shister, G. B: Ellıs Horwood Ltd.

ROoGeERs, J. S. (1972): Measures of genetic similarıty and genetic distance. Studies in Genet. VII, Univ.
Texas Publ. 7213, 145-153.

RyDeEr, ©. A. (1986): Species conservation and systematics: the dilemna of subspecies. Trends in
Ecology and Evolution 1, 9-10.

— (1987): Conservation action for gazelles: an urgent need. Trends in Ecology and Evolution 2,
143-144.

SWOFFORD, D. L. (1990): PAUP: Phylogenetic analysıs using parsımony, Version 3.0. Champaign,
Illinois: Computer program distributed by the Illinois Natural History Survey.

SWOFFORD, D. L.; SELANDER, R. B. (1981): BIOSYS-1: A FORTRAN program for the comprehen-
sıve analysıs of electrophoretic data in population genetics and systematics. J. Heredity 72,
281-283.

TEMPLETON, A. R.; Davıs, S. K.; ReaD, B. (1987): Genetic varıability in a captive herd of speke’s
gazelle (Gazella spekei). Zoo Biology 6, 305-313.

THouLzss, C. R.; Ar Baskı, K. (1991): Taxonomic status of the Farasan Island gazelle. J. Zool.
(London) 223, 151-159.

THOULEss, C. R.; GRAINGER, J. G.; SHOBRAK, M.; Hasısı, K. (1991): Conservation status of gazelles
in Saudi Arabia. Biological Conservation 58, 85-98.

VAsSART, M.; GRANJON, L.; GRETH, A. (1991): Genetic varıability in the Arabian Oryx. Zoo Biology
10, 399-408.

VASSART, M.; GRETH, A.; CrıBIu, E. P. (1993): Cytogenetic survey of gazelles from Middle East. X.
European Collogium on Cytogenetics of Domestic Anımals, Utrecht. Pp. 253-256.

WAYNE, R. K.; FORMaNn, L.; NEwMmaAnN, A. K.; SIMoNson, J. M.; O’BrIEnN, S. J. (1986): Genetic
monitors of zoo populations: morphological and electrophoretic assays. Zoo Biology 5, 215-232.

WRIGHT, $. (1978): Evolution and the genetics of populations. Vol 4: Variability within and among
natural populations. Chicago, Illinois: University of Chicago Press.

Authors’ addresses: Marc VaAssArT, Ecole Nationale Veterinaire de Toulouse, 23 chemin des Capel-
les, F-31076 Toulouse, France; LAURENT GRANJON, Museum National d’His-
toire Naturelle, 55 rue Buffon, F-75005 Paris, France; ARNAUD GRETH, National
Wildlife Research Center, P. ©. Box 1086, Taif, Saudi Arabia; Francoıs M.
CATZEFLIS, Universite de Montpellier II, Institut des Sciences de l’Evolution,
URA 327 CNRS, F-34095 Montpellier, France

Z. Säugetierkunde 59 (1994) 246-251
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Springbok, Antidorcas marsupialis (Zimmerman, 1780)
from the past

By Ina Pruc

Department of Archaeozoology, Transvaal Museum, Pretoria, South Africa

Receipt of Ms. 29. 11. 1993
Acceptance of Ms. 17. 6. 1994

Abstract

Analysed 6000 springbok, Antidorcas marsupialis (Zimmerman, 1780), bone fragments from an
archaeological site in South Africa. This enabled deductions on herd composition and seasonality of
springbok from the period before settled farmıng and large scale hunting disrupted herd mobility and
reduced numbers. Techniques of osteomorphology and osteometry were used to compare the
archaeological bone wıth modern skeletal material. The results show that the excavated remains are
predominantly from foetal, neonate and adult individuals. On average the anımals from the
archaeological samples were somewhat smaller than modern springbok females, suggesting that most
of the animals in the archaeological samples were also females. The age profiles and the large numbers
of foetal and neonate individuals indicate that the assemblage was taken from lambing herds.
Climatological and environmental conditions of the region indicate that these herds visited the area
during spring and early summer on a seasonal basis. They were not part of the occasional mass
movements, or smaller mixed herds, as such herds consist of anımals of both sexes and all ages.

Introduction

Archaeozoological and palaeontological studies provide information on the anatomy of
wild anımals from the past, but most samples are too small to allow deductions on herd
composition, behaviour and average sizes of the anımals. The large sample of springbok,
Antidorcas marsupialis (Zimmerman, 1780), bones from an archaeological site provided an
unique opportunity to determine ıf such deductions can be made on larger samples, and to
compare the results with those of modern springbok studies and historical observations.

The distribution of springbok in South Africa ıs mainly restricted to game reserves and
farms where herd mobility and herd size are limited. Little is known about structure and
behaviour of free roaming herds in the past. Until 1950, in the Kalaharı Gemsbok Park and
Botswana, springbok sporadically congregated in their thousands and mega-migrations
occurred (CONWRIGHT-SCHREINER 1925; CHILD and LE RıcHE 1969; BIGALKE 1972).
Springbok herds of the Kalaharı Gemsbok Park, consist of individuals of both sexes and all
age groups, with the exception of neonates. These herds are small in winter and larger in
summer when the grazing is good. At the start of the lambing season females congregate
and male animals of six months and older are driven off to form bachelor herds (SKINNER
and SMITHERS 1990). A few territorial males remain with the females. The lambing season
may vary from region to region, but most lambs are born between September, at the
beginning of the rainy season, and January, with the highest peak in October. Females of a
particular herd usually drop their lambs within one or two months (van ZyL and SKINNER
1970; SKINNER et al. 1974; SKINNER et al. 1977).

Material and methods

The springbok bones, other faunal remains and artefacts of hunter-gatherer communities were
retrieved from Abbot’s Cave (ABB) in the Seacow Valley, Karoo, South Africa, (31°27'S, 24°39' E)

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5904-0246 $ 02.50/0

Springbok, Antidorcas marsupialis from the past 247

(Fig.). Most of the deposit dates to between 1270 and 1682 AD, predating European contact and the
establishment of demarcated farms (Sampson and VoGEL 1989; Sampson et al. 1989; Pruc 1993). The
samples therefore reflect the past fauna endemic to the region.

Early historical records of the 18th century only make passing reference to the wild anımals in the
valley. However, in his diary, JACOB GORDON (RAPER and BOUCHER 1988) mentions that small herds
of springbok remained in the valley for most of the year, but that larger herds (not mega-herds) moved
in from September to November, migrating southwards. He observed that most lambs were dropped
in August and September (RAPER and BouCHER 1988). Mating must therefore have taken place at the
end of the rainy season in late summer and autumn, when ewes would have been in optimum
condition.

The heavily fragmented ABB macrofaunal sample consists of over 142000 fragments of which
6.6% could be identified, representing 46 different species. Of the 9415 identified fragments 4586
(49 %) are from springbok and 1418 (15 %) from probably springbok. Determinations were made by
using the comparative skeletal collections of the Transvaal Museum and the osteomorphological
criteria for fragments of springbok (PruG and PETERS 1991) and for modern springbok skeletons
(PETERS and BrInk 1992).

Measurements were taken with dial calipers to the nearest 0.1 mm. Only bones from fully adult
animals were measured according to procedures defined by von DEN DRIESCH (1976) and PETERS
(1988). Means and standard deviations were calculated. Only 337 post-cranıal springbok fragments
were measurable, including complete as well as partially complete bones, mostly carpals, tarsals and
phalanges.

The measurements of modern springbok adults of known sex (PETERS and BrınK 1992) were
compared to those of the archaeological sample. The table lists the measurements of the archaeological
post-cranial elements that number five or more, as well as the relevant measurements of modern
springbok.

No detail information exists on the ages of springbok in relation to skeletal development. Based on

*= Pretoria

Map of southern Africa showing the position of Abbot’s Cave (ABB). N = Namibia; B = Botswana;
Z = Zimbabwe; M = Mozambique; RSA = Republic of South Africa; L = Lesotho; $ = Swaziland;
KPG = Kalaharı Gemsbok Park

248 Ina Plug

Table. The measurements of postcranial skeletal elements of archaeological and modern
springbok where sample size is five or more

The measurements of the modern springbok are taken from PETERS and Brınk (1992)

Archaeological sample Modern females Modern males

Os carpi radiale
26 197. 122
26 13.4 1.00
26 99 0.64

Os carpi intermedium
25 19.0 132
25 127, 12.09

Os carpi ulnare
26 18.4 1.06

Os carpale II + III
25 14.6 0.96
25 1539 0.96

Os carpale IV
25 10.4 0.73
25 10.9 0.89

Os metacarpale III + IV
26 DD
26 18.0

Patella
26 28.5
26 25.1

Os malleolare
»3 15.4

Talus
29 30.5
2,9 28.7
29 17.6
29 18.9

Os centroquartale
27, 2
220 26.1

Os tarsale II + III
24 176%

Os metatarsale III + IV
26 23.4 1.43
26 18.5 1.63

Phalanx proximalıs pedis
25 43.4 2.89
223 111162)
25 10.1
25 8.8

Phalanx media manus
23 2555
23 10.1
23 8.6
23 Ze

Springbok, Antidorcas marsupialis from the past 249

Table (continned)

Archaeological sample Modern females Modern males
n X

Phalanx media pedis

GL
BP
Bd
SD

Phalanx distalis
DES i 5 30.2
Ld & e 25.4
Hp 17.5
BFp : 8.4

n = number; x = mean; s = standard.

tooth eruption, tooth wear and epiphyseal fusion the following relative age categories, with brief
descriptions, are distinguished (Pruc 1988, 1993):
Foetal/neonate: deciduous teeth unerupted, unfused proximal radıus, os centroquartale and metapo-
dials (longitudinally, proximally and distally), bones spongy;
Neonate/juvenile: deciduous teeth erupting/just in wear, M 1 unerupted/erupting, proxımal radıus, os
centroquartale and metapodıals longitudinally still unfused or beginning to fuse;
Juvenile: deciduous teeth and M1 in wear, M2 erupting, metapodials longitudinally fused, most
epiphyses unfused, articulation surfaces well defined;
Sub-adult: M2 in wear, M3 erupting, heavy wear on deciduous teeth, most epiphyses beginning to
fuse;
Young adult: M3 in wear, most deciduous teeth replaced, most epiphyses fused;
Adult: all permanent teeth present and ın wear, heavy wear on MI, all epiphyses strongly fused;
Mature: heavy wear on all teeth, no central ıslands on M1 and disappearing on M2 and M3;
Aged: no central ıslands on M3, M I and M2 worn down to roots, ossification of muscle attachments
and cartılage.

Pelvis fragments were used to determine the sex of adult individuals. No other bones were well
enough preserved to allow sexing.

Results

The springbok bone sample contains an unprecedented amount, almost 15 %, of foetal/
neonate bones and an additional 2% neonate/juvenile bones. This ıs the first time that so
many bones from such young (nondomestic) anımals have been found in a South African
archaeologıcal site. Foetal anımals are seldom represented and neonates only occasionally,
but juveniles are usually well-represented in most assemblages. As foetal and neonate
bones are extremely fragile, they may even be underrepresented in the ABB sample. The
other age categories represented are: 1% sub-adult, less than 1% young adult, 75 % adult,
4% mature, and 2% aged. There were no juveniles present in the sample.

The measurements ot the archaeological sample fall mainly within the range of those of
modern females (Tab.). Some of the archaeological bone, phalanges in particular, are
somewhat smaller on average, than modern females, but the differences are small and not
consistent. Of the 40 pairs of mean values, the mean of the archaeological measurements is
smaller than the modern mean of females in 19 cases, larger in 18 cases and equal in three.
Compared to modern males, the archaeological sample is smaller in 36 cases, larger in one
and equal in three.

The standard deviations of the measurements of the archaeological sample were also
compared with those of the modern sample. This shows that in 26 and 30 of the
measurements the archaeological sample has a smaller standard deviation and ın 14 and 10a
larger standard deviation than the modern female and male samples respectively. This

250 Ina Plug

indicates that the anımals from the ABB herds were slightly more homogenous in size than
those of the modern sample. This result is not unexpected as the modern sample consists of
animals that were obtained from different geographical regions.

Osteomorphologically ıt was not possible to distinguish with confidence between the
sexes, as attempts to determine sex were hampered by the fragmented nature of the
assemblage. Subsequently only eight females and two males were identified on pelvis
fragments, but as male pelves are more robust than those of females and are less susceptible
to natural attrition, females are probably underrepresented. The possibility that there is
overlap between young males and large females should be considered. The young adult
category is poorly represented ruling out the possibility that young males (or young
females) influenced the sample significantly. It can therefore be argued that, combined
with the high incidence of foetal/neonates, the majority of the adult springbok were
females.

Discussion

Archaeological faunal samples have inherent limitations for research, related to fragmenta-
tion, preservation and attrıtion. Nevertheless, the results show that even samples as heavily
fragmented as the ABB assemblage, have research value.

The age structure of the springbok from ABB suggests that the majority of the animals
did not come from mixed herds or migrating mega-herds. Foetal/neonate animals are not
part of the former and can be expected in small numbers only, in the latter, whereas the
juvenile, sub-adult and young adult categories should be better represented in both herd
types.

The results indicate that females were more frequently hunted than males, while the
large foetal/neonate component shows that they were culled from female lambing herds.
The hunters of ABB appear to have been famıliar with the migratory and breeding cycle of
the springbok and deliberately preyed on pregnant and lactating females as these would
have had a relatively high body fat content.

The ABB deposits give no indication of major climate changes within the past 1000
years (Bousman 1991). It would therefore be reasonable to assume that the prıme mating
season for springbok was in late summer and autumn when grazing in this region is at its
best after the summer rains. As a result the majority of foetal and neonate springbok from
ABB would have been born in August-September, as has been observed in the 18th
century.

The age categories of the springbok from ABB indicate that the herds dıd not remain in
the vicinity long after lambing. Juvenile, sub-adult and young adult anımals are either
absent or poorly represented, suggesting that the new crop of lambs matured elsewhere.
The area near ABB would have been attractive to springbok in August-September, as it has
a small seasonal wetland that could have supported some good grazing towards the end of
winter.

In summary, the springbok living in the Seacow Valley before settled farming, usually
dropped their lambs in early spring (August to September). They congregated in female
herds for the occasion, but did not remain in the area. The herds seem to have had some
mobility, probably on regional scale, to optimize grazing opportunities.

Acknowledgements

I am grateful to Prof. C. G. Sampson of Southern Methodist University in Dallas and the South
African Museum in Cape Town. My thanks also go to the Foundation for Research Development for
their financial support.

Springbok, Antidorcas marsupialis from the past 251

Zusammenfassung

Springböcke, Antidorcas marsupialis (Zimmerman, 1780), aus vergangenen Zeiten

Die Entdeckung einiger tausend Springbockknochen (Antidorcas marsnpialis Zimmerman, 1780) von
einer archäologischen Fundstelle in der Karoo, Südafrika (13.-17. Jahrhundert), bot die Gelegenheit,
Populationsstruktur, Körpergröße und saisonale Fluktuationen der Art zu untersuchen, bevor
Farmgründungen und starke Bejagung das Wanderverhalten störten und die Bestände reduzierten.
Die ausgegrabenen Knochen stammen überwiegend von ungeborenen und ganz jungen Individuen
und von ausgewachsenen Weibchen. Osteometrische Vergleiche mit rezenten Springbockweibchen
zeigten, daß die damaligen Tiere im Durchschnitt etwas kleiner waren. Das Vorkommen zahlreicher
Feten und Jungtiere macht es wahrscheinlich, daß die Herden das Gebiet und die Fundstelle im
Frühjahr und zum Sommeranfang aufsuchten und hier dann gejagt wurden.

References

BIGALKE, R. C. (1972): Observations on the behaviour and feeding habits of the springbok Antidorcas
marsupialis. Zool. Afr. 7, 333-359.

Bousman, C. B. (1991): Holocene paleoecology and the Later Stone Age hunter-gatherer adaptations
in the South African interior plateau. Ph. D. diss. thesis, Southern Methodist Univ., Dallas.

CHıLD, G.; LE RıcHe, J. D. (1969): Recent springbok treks (mass movements) in South-Western
Botswana. Mammalia 33, 499-504.

CONWRIGHT-SCHREINER, $. C. (1925): The migratory springbok of South Africa. London: Fisher
Unwin Ltd.

PETERS, J. (1988): Osteomorphological features of the appendicular skeleton of African buffalo,
Syncerus caffer (Sparrman, 1779) and cattle, Bos primigenins f. taurus Bojanus, 1827. Z. Säugertier-
kunde 53, 108-123.

PETERS, J.; Brink, J. S. (1992): Comparative postcranial osteomorphology and osteometry of
springbok, Antidorcas marsupialis (Zimmerman, 1780) and grey rhebok, Pelea capreolus (Foster,
1790). Navors. nas. Mus. Bloemfontein 8, 161-207.

Prug, I. (1988): Hunters and herders: an archaeozoological study of some prehistoric communities in
the Kruger National Park. D. Phil. diss. thesis, Univ. Pretoria.

— (1993): The macrofaunal remains of wild anımals from Aboot’s Cave and Lame Sheep Shelter.
Koedoe 36, 15-26.

Prug, 1.; PETERS, J. (1991): Osteomorphological differences in the appendicular skeleton of Antidor-
cas marsupialıs (Zimmerman, 1780) and Antidorcas bondi (Cooke and Wells, 1951) (Mammalia:
Bovidae) with notes on the osteometry of Antidorcas bondi. Ann.Transvaal Mus. 35, 253-264.

RAPER, P. E.; BOUCHER, M.M. (eds.) (1988): Robert Jacob Gordon Cape travels, 1771 to 1786. Vol.
1. Johannesburg: Brenthurst Press.

SAMPSON, C. G.; HART, T. J. G.; WALLSMITH, D. L.; Bracc, J. D. (1989). The ceramic sequence in the
upper Seacow Valley: problems and implications. S. Afr. archaeol. Bull. 44, 3-16.

SAMPSON, C. G.; VOGEL, J. C. (1989). A painted pebble and associated C-14 date from the Upper
Karoo. Pictogram 2, 1-3.

SKINNER, J. D.; NEL, J. A. J.; MirLar, R. P. (1977): Evolution of time parturition and differing litter
sızes as an adaptation to changes in the environmental conditions. In: Reproduction and evolution.
Ed. by J. H. Carasy and C. H. TynDaLe-Biscoe. Canberra: Australian Acad. Sci. Pp. 39-44.

SKINNER, J. D.; SMITHERS, R. N. (1990): The mammals of the southern African subregion. Pretoria:
Univ. Pretorıa.

SKINNER, J. D.; van Zyr, J. H. M.; OaTtes, L. G. (1974): The effect of season on the breeding cycle of
plains antelope of the western Transvaal Highveld. J. sth. Afr. Wildl. Mgmt Ass. 4, 15-23.

Van Zyı, J. H. M.; SKINNER, J. D. (1970): Growth and development of the springbok foetus. Afr.
wild Life 24, 308-316.

VON DEN DkısschH, A. (1976): A guide to the measurements of anımal bones from archaeological sites.
Peabody Museum Bulletin 1: Harvard Univ.

Author’s address: Ina Pruc, Department of Archaeozoology, Transvaal Museum, P. ©. Box 413,
Pretoria 0001, South Africa

Z. Säugetierkunde 59 (1994) 252-254
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

WISSENSCHAFTLICHE KURZMITTEILUNG

Marmosops scapulatus Burmeister, 1856, and the brown
mutation in didelphids (Marsupialia)

By MEIKA A. MUSTRANGI

Museum of Vertebrate Zoology, University of Calıfornia at Berkeley, Berkeley, USA

Receipt of Ms. 23. 4. 1993
Acceptance of Ms. 15. 11. 1993

A female mouse opossum of the genus Marmosops Matschie, 1916, showing an unusually
pale coat color was collected in the Atlantic Coastal Forest ın Brazil in July of 1992 (see
figure). The site is the Intervales biological reserve in the State of Säo Paulo, 24° 20’ S, 48°
25’ W. The anımal was captured in a Sherman trap tied to a tree limb at approximately
1.5 m from the ground ın primary forest habitat. It was immediately recognized for its
unusual coloration since the only species of Marmosops known to occur in all southeastern
Brazil, M. incanus, shows an ınvarıably dark gray-brown coloration above and creamy
white below (Emmons 1990). The pale female ıs light cinnamon-brown dorsally, corre-
sponding to a score ot 7.5 YR 4/4 in the Munsell system of color (MunseLL CoLoR
Company 1976). The underparts are of a pinkish, creamy white. The dorsal coloration of
M. incanus (as measured from the other individuals collected in Intervales) corresponds to
a score of 10 YR 2/1. The pale individual was prepared as skin, skull, and partial skeleton
and will be deposited at the Museu de Zoologıa da Universidade de Säo Paulo. A sample of
liver tissue was preserved in ethanol for molecular analyses. Female mouse opossums are
generally much smaller than males, but old females may reach the size of males (TATE
1933). The pale female was a very old individual (age class 7 of TRıBE 1990), and equalled ın
size the adult males of M. incanus collected at that localıty.

Not surprisingly, new species of mammals are still being described in South America
(e.g. the marmoset, Callıthrix manes, of the Amazon basın by MITTERMEIER et al. 1992;
the lion tamarın, Leontopithecus caissara, of the Atlantic Forest by Lorını and PERSSON
1990; three new species of gracile mouse opossums, genus Gracilianus, from the forested
Andean slopes by HERSHKoVvITZ 1992). To evaluate the possibility that the pale female
represented a new species of Marmosops, I sequenced the first 380 base pairs of the
mitochondrial cytochrome b gene for 4 of the individuals collected at that locality. I also
looked for differences ın cranıal characters. No significant differences were found among
the several skulls, apart from size-related ones. The comparison of the cyt. b sequences
revealed the presence of two haplotypes, differing by only one silent, third position
transition. The pale female and an individual of normal coloration had one haplotype, wıth
the other 2 individuals showing the other. Sequence variation (calculated as the number of
varying sites divided by the total number of sites compared), therefore, amounts to 0.3
percent, well within the range of intrapopulation varıatıon for this gene observed ın other
species (SMITH and PATToN 1991).

DNA sequencing was a quick way to test the hypothesis that the pale female
represented a different, so far undescribed, species of Marmosops. Since it is not a new
species, ıts unusual coat color might be an age-related phenomenon, or might represent a
rare mutation. The first alternative seems unlikely since no other individual with the same

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5904-0252 $ 02.50/0

Marmosops scapulatus and the brown mutation in didelphids (Marsupialia) 2:55

contrasting coloration has ever been col-
lected in that area, despite recent exten-
sive collecting efforts (DE Vıvo, pers.
comm.).

HARTMAN (1921, 1922) reported on a
brown or cinnamon variety of the Vir-
ginia opossum (Didelphis virginiana). In
those individuals, the underfur was of a
“uniform and delicate light brown” in-
stead of black. Harrtman (1921, 1922)
associated this color morph with the so-
called brown mutation in house mice.
Currently, this recessive allele is known
to occur in several orders of mammals
(SEARLE 1968). The predominant effect of
the b allele is to change the nature of
eumelanın from a black to a brown pig-
ment, and thus the coat color from gray
to cınnamon brown (SILvErs 1979). As
expected in a brown individual, the pig-
ment granules in the pale M. incanus
female are of a brownish color, instead of
the black color displayed by the typically
colored M. incanus individuals.

In a recent study, DE OLiVvEIRA et al.
(1992) reported on the results of their
analysıs of pelage varıation in thıs same
species. The authors concluded that the
different pelage types of M. incanus rep-
resent different ontogenetic and sexually
dimorphic states, and not seasonal (sum- Marmosops incanus Lund, 1840, collected in Inter-
mer and winter) pelage types as ıt was vales, Brazil: an adult male (left) and the brown
previously believed (e.g. Emmons 1990). female (right)

The authors agree wıth Emmons (1990) in
synonymizing M. scapulatus Burmeister, 1856, to M. incanus, believing that it had been
described as a different species on the basıs of the misinterpreted pelage varıation.

Besides the pelage type, however, the M. scapulatus male individual ıs also distinct from
other M. incanus individuals by its general coloration, originally described by BURMEISTER
as being composed of haırs “at base slate gray, then pale yellow-red, and last cınnamon”
(TATE 1933). It seems reasonable to argue that the M. scapulatus individual and the pale
female described here represent two independent manifestations of the same rare, recessive
allele, separated in time by at least 136 years.

Interestingly enough, a female Philander of a similar, pale coloration was recently
collected in the Atlantic Forest by P. HERSHKOVITZ. The female Philander later gave birth
to several young of normal, dark coloration. Pending on a closer study of this specimen, it
raises to three the didelphid genera known to present the brown mutation, making it more
common and widespread in didelphid marsupials than presently known.

Acknowledgements

I would lıke to thank James L. PATTON, WILLIAM Z. LIDICKER, SERGIO F. DOS REıs, KEvin Burns,
and ALBERT D. DITCHFIELD for critically reviewing and helping improve this manuscript. This study
was funded by a graduate studies’ fellowship from FAPESP (Brazil).

254 Meika A. Mustrangi

References

MunseLL CoLOR CoMPANY (1976): Munsell book of color: glossy finish collection. Baltimore:
Munsell Color.

DE OLivEIra, J. A.; Lorını, M. L.; Persson, V. G. (1992): Pelage varıation in Marmosa incana
(Didelphidae, Marsupialia) wıth notes on taxonomy. Z. Säugetierkunde 57, 129-136.

Emmons, L. H. (1990): Neotropical rainforest mammals. A field guide. Chicago: University of
Chicago Press.

HaRrTMan, C. (1921): The Virginia opossum. Photographs of two new varieties and the two normal
phases of this anımal. J. Heredity 12, 471-473.

— (1922): A brown mutation in the opossum (Didelphis virginiana) with remarks upon the gray and
the black phases in this species. J. Mammalogy 3, 146-149.

HERSHKOVITZ, P. (1992): The South American gracile mouse opossums, genus Gracilianus Gardner
and Creighton, 1989 (Marmosidae, Marsupialia): a taxonomic review with notes on general
morphology and relationships. Fieldiana Zoology n.s. 70, 1-56.

LorintI, M. L.; PErsson, V. G. (1990): Nova especie de Leontopithecus Lesson, 1840, do sul do Brasil
(Primates, Callithrichidae). Bol. Mus. Nac. Rio de Janeiro Zool. n.s. 338, 1-14.

MITTERMEIER, R. A.; SCHWARZ, M.; Ayres, J. M. (1992): A new species of marmoset, genus Callthrix
Erxleben, 1777 (Callithrichidae, Primates) from the Rio Maues region, State of Amazonas, Central
Brazilian Amazonıa. Goeldiana Zoology 14, 1-17.

SEARLE, A. G. (1968): Comparative genetics of coat colour in mammals. New York: Academic Press.

SILVERS, W. K. (1979): The coat colors of mice: a model for mammalian gene action and interaction.
New York: Springer Verlag.

SMITH, M. F.; PATToNn, J. L. (1991): Variation in mitochondrial cytochrome b sequence in natural
populations of South American akodontine rodents (Muridae: Sigmodontinae). Mol. Biol. Evol. 8,
85-103.

TATE, G. H. H. (1933): A systematic revision of the marsupial genus Marmosa. Bull. Amer. Mus.
Nat. Hist. 66, 1-250.

TRIBE, C. J. (1990): Dental age classes in Marmosa incana and other didelphoids. J. Mammalogy 71,
566-569.

Author’s address: MEıKA A. MusTRanGı, Museum of Vertebrate Zoology, 1120 Life Sciences Build-
ing, University of Calıfornia at Berkeley, Berkeley, CA 94720-0001, USA

BIOI@EIBESPRE@ENUNGEN

NIETHAMMER, ]J.; Krarpp, F. (Hrsg.): Handbuch der Säugetiere Europas. Bd. 3/1:
Insektenfresser, Primaten. Wiesbaden: Aula-Verlag 1990. 524 S., 141 Abb., 133 Tab.,
328,- DM. ISBN 3-89104-027-X

Band 3/1 des Handbuches der Säugetiere Europas behandelt die Insectivoren-Familien der Igel, der
Maulwürfe und der Spitzmäuse sowie den Berberaffen, den einzigen hier in Frage kommenden
Primaten. Der Umfang von 524 Seiten machte es erforderlich, den ursprünglich für diesen Band mit
eingeplanten Hasenartigen einen weiteren Band zu widmen. Neben den Herausgebern, von denen
besonders J. NIETHAMMER für zahlreiche Kapitel als Autor tätig war, haben mitgewirkt: M. GENOUD,
J. Hausser, H. Horz, R. HUTTERER, E.-A. JUCKWER, T. MADDALENA, H. PIEPER, F. SPITZENBERGER,
D. STARCK, S. SULKAVA, P. VLAsAK und P. VocEL.

Auch bei diesem Band wurde die bewährte Gliederung beibehalten. Nach einer kurz gehaltenen
Einführung folgt der Hauptteil mit je einer knappen Vorstellung der Ordnungen Insectivora und
Primates, an die sich die zu ihnen gehörigen Taxa anschließen. Hierbei wird für Familien und
Gattungen ein einheitliches Gliederungsschema verwendet (Diagnose; Angaben zur Verbreitung, zu
Umfang und Gliederung der Familien bzw. der Gattungen, zur Paläontologie; besondere Merkmale);
Gattungs- und Artenschlüssel beschließen die Familien- und Gattungsabschnitte und leiten zu den
Kapiteln der einzelnen Arten (26 Insectivoren- und eine Primatenart) über. Hier findet sich jeweils
wieder eine Diagnose (u.a. Karyotyp), auf die Ausführungen zur Taxonomie, zur Morphologie
(„Beschreibung“, die sich nicht auf Skelettmerkmale beschränkt), zur Verbreitung, zur Merkmals-
variation, zur Paläontologie, zur Okologie, zur Jugendentwicklung und zum Verhalten folgen. Für
nahezu jede der Arten und ebenso für die höheren Taxa gibt es eine eigene, zum Teil sehr
umfangreiche Bibliographie (für Sorex araneus mehr als zehn Seiten); am Ende des Bandes folgt
darüber hinaus noch ein „Allgemeines Literaturverzeichnis“. Eine Fülle von Abbildungen (Zeichnun-
gen von Schädeln, Zähnen, postcranialer Skeletteile sowie anderer Merkmale der Formen; Dia-
gramme; Verbreitungskarten) und eine große Anzahl von Tabellen ergänzen die Texte. Unter
letzteren sollen diejenigen mit individuellen Maßen der Arten von unterschiedlichen Fundorten
besonders erwähnt werden; diese zum Konzept des Handbuches gehörenden Tabellen beanspruchen
zwar viel Raum, zeigen aber auch die besondere Mühewaltung der Autoren und wären nicht durch
Hinweise auf Originalliteratur zu ersetzen.

Wissenschaftliche Kompetenz und Ausstattung dieses Bandes sind so hervorragend wie diejenigen
der vorherigen. Die strikte Einhaltung des Gliederungsschemas erleichtert Orientierung und Arbeit in
diesem Werk, dessen große Informationsfülle in knapp gefaßten, gut lesbaren Texten dargeboten
wird. Den Autoren, dem Verlag und im besonderen den Herausgebern ist für diese besonders
wichtige Fortsetzung der Handbuchreihe zu danken, die auf lange Zeit einen festen Platz in den
Bibliotheken nicht nur der europäischen Mammalogen einnehmen wird. H.ScHLIEMAnn, Hamburg

ECKERT, M.; HERTEL, W. (Hrsg.): Praktikum der Tierphysiologie. Jena, Stuttgart:
Gustav Fischer Verlag 1993. 312 S., 138 Abb., 44 Tab., 58,- DM. ISBN 3-334-60438-1

Der vorliegende Band enthält zahlreiche Anleitungen zu experimentellen Arbeiten auf den wichtigsten
Gebieten der Tierphysiologie und Verhaltensbiologie. Die Versuche, die offensichtlich von den
Autoren gründlich ın der Lehre erprobt worden waren, sind so konzipiert, daß sie bei entsprechender
Betreuung auch von unerfahrenen Experimentatoren durchgeführt werden können.

Die Beschreibung der einzelnen Aufgaben ist gut gelungen. Nach einer kurzen Einführung in die
Theorie folgt eine anschauliche Erklärung der Versuchsdurchführung. Zahlreiche Abbildungen der
für den Versuch benötigten Präparate, Skizzen der Versuchsanordnungen sowie grafische Darstellun-
gen repräsentativer Versuchsergebnisse geben eine wertvolle Hilfe bei der Durchführung und Auswer-
tung der Versuche. Zur statistischen Aufarbeitung der Versuchsergebnisse erhalten die Praktikums-
teilnehmer nützliche Hinweise im Abschnitt 1.5, in dem die wichtigsten Methoden der Statistik
erläutert werden.

Die Praktikumsanleitung enthält weiterhin Hinweise zur Zucht und Haltung verschiedener
Versuchstiere, zum Umgang mit gefährlichen Stoffen (Arbeitsschutz) sowie zu den physikalischen
Grundlagen der elektrophysiologischen Meß- und Registriertechnik. In diesem allgemeinen Teil wird
auch auf die gesetzlichen Bestimmungen zur Durchführung genehmigungspflichtiger bzw. anzeige-
pflichtiger Tierversuche hingewiesen (Tierschutzgesetz, Naturschutzgesetz). Zu diesem Abschnitt
wäre allerdings zu bemerken, daß erstens bei der Genehmigung von Tierversuchen zu Ausbildungs-
zwecken zur Zeit sehr restriktiv verfahren wird, und daß zweitens auch genehmigte bzw. angezeigte
Versuche an Wirbeltieren von den Studierenden häufig abgelehnt werden. Erfahrungsgemäß werden
dagegen Versuche an Evertebraten von den Studierenden im allgemeinen akzeptiert. Diese Schwierig-
keiten werden jedoch in der vorliegenden Praktikumsanleitung berücksichtigt; es werden in nahezu

256 Buchbesprechungen

allen Abschnitten genügend Versuche an Wirbellosen angeboten, so daß auf Experimente an Vertebra-
ten weitgehend verzichtet werden kann.

Das „Praktikum der Tierphysiologie“ stellt eine gut abgestimmte Ergänzung des von PENZLIN
verfaßsten Lehrbuches der Tierphysiologie dar. Die vorliegende Praktikumsanleitung kann deshalb
Lehrenden und Lernenden gleichermaßen empfohlen werden. W. WÜNNENBERG, Kiel

TAMmARIn, R. H.; OsTFELD, R. $S.; Puch, $S. R.; BuJAaLskA, G. (eds.): Social Systems and
Population Cycles in Voles. ALS Advances ın Life Sciences. Basel, Switzerland: Birkhäu-
ser Verlag 1990. 229 pp. 62,-sFr. ISBN 3-7643-2437-6 or 0-8176-2437-6

The themes presented by the two editors at the two International Theriological Congresses in
Edmonton (Social system in Microtus) and in Rome (the relationships between social systems and
population dynamics in Microtine rodents) led to the edition of the present volume. In addition to 4
overview articles, there are 10 papers dealing with Microtus-species, 5 with Clethrionomys-species and
one paper on Arvicola terrestris. The critical remarks and the emphasis put on the flexibility of the
social organızation of Arvicolidae in the overview articles deserve special attention. Of the Microtus-
species, M. agrestis ıs the only European species mentioned. Best represented of the genus Clet-
hrionomys ıs C. glareolus. Both species were investigated in northern European habıtats.

In the species studied, the relationships between the mating system, crowding, natural history,
territoriality, predation by specialists and the impact of the changing in density and population
dynamics are discussed.

It was not possible to agree upon a common concept regarding the relationship between certain
social systems and population dynamics. However, numerous ideas about different methods and
results, which could enhance a fruitful discussion, are presented. This is the reason why this collection
of papers is a rich source of information especially for population biologic-ethological seminars, as
this reviewer has already experienced ın a eco-theriological seminar. R. SCHRÖPFER, Osnabrück

SCHALLER, O. (ed.): Illustrated Veterinary Anatomical Nomenclature. In cooperation
with G. M. CoNSTANTINESCU, R. E. HABEL, W. ©. Sack, P. SIMOENs, N. R. DE Vos.
Stuttgart: Ferdinand Enke Verlag 1992. VIII, 614 pp., 280 plates including 1316 illustr.,
148,- DM. ISBN 3-432-99591-1

This book is a most effective medium for all scientists to obtain correct anatomical nomenclature as
approved by the “International Anatomical Nomenclature Committee”. The lists published as
“Nomina Anatomica Veterinaria” confront the reader only with terms, but do not give illustrations of
the listed structures. This volume, however, which was edited by an Austrian veterinary anatomist
with the cooperation of two Belgian and three US-American specialists, helps the reader by
illustrating the relevant anatomical structures and including them in their topographical surrounding.
Reference is made to seven domestic mammalıan species, namely, cat, dog, pig, ox, sheep, goat, and
horse, but not to birds.

On the right hand pages very clear and informative line drawings are given and the structures
within them are labelled with numbers. On the opposite left hand pages the scientific names and very
often their Anglisized versions, are listed according to the above-mentioned numbers, followed by
very short explanations in English. The book is subdivided according to eleven subheadings: Regions
of the body, osteology, arthrology, myology, splanchnology, angiology, nervous system, sense
organs, common integument, parts of the body, and general terms.

Reference to this volume can be made in three ditferent ways: A scientific name can be brought
into its topographical context with the help of an alphabetical index (39 pages) at the end of the book.
Alternatively, a user who has seen the structure during dissections, can refer to the illustrations of the
respective body system, which is indicated at the page heading and will then be able to identify the
name of the structure. The third alternative includes consulting the tables of names and obtaining
information concerning the topographical situation of the structure.

The spelling in ambiguous cases ıs given in both the American and British forms (e.g., esophagus
[oesophagus]). Terms of orientation are used that make sense in quadrupedal mammals, but are
generally not used by specialists on human gross anatomy (V. cava cranialis instead of V. cava
superior).

This book is a very systematic illustrative dictionary of anatomical terms that is easy to handle and
very well produced. Hopefully, not only veterinarians, but also mammalogists in general will make
use of this publication, which presents a sound and widely applicable nomenclature of the gross
anatomy of terrestrial mammals. P. LAnGeER, Gießen

Erscheinungsweise und Bezugspreis 1994: 6 Hefte bilden einen Band. Jahresabonnement Inland:
DM 378,- zuzüglich DM 13,80 Versandkosten; Jahresabonnement Österreich: 65 2949,- zuzüg-
lich öS 164,— Versandkosten; Jahresabonnement Schweiz: sfr 364,—- zuzüglich sfr 21,- Versand-
kosten; Jahresabonnement EG-Binnenmarkt-Länder mit USt-ID-Nr.: DM 353,27 zuzüglich
DM 19,63 Versandkosten; Jahresabonnement EG-Binnenmarkt-Länder ohne USt-ID-Nr. und
Drittländer: DM 378,- zuzüglich DM 21,- Versandkosten. Das Abonnement wird zum Jahres-
anfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens
am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhand-
lung oder bei der Verlagsbuchhandlung Paul Parey GmbH & Co. KG, Spitalerstraße 12,
D-20095 Hamburg, Bundesrepublik Deutschland, bestellt werden. Die Mitglieder der „Deut-
schen Gesellschaft für Säugetierkunde“ erhalten die Zeitschrift unberechnet im Rahmen des

Mitgliedsbeitrages.
Z. Säugetierkunde 59 (1994) 4, 193-256

4 59 (5), 257-320, Oktober 1994 ISSN 0044-3468 "@2127AF

%ITSCHRIFT FÜR
SAUGETIERKUNDE

NTERNATIONAL JOURNAL
DFMAMMALIAN BIOLOGY

drgan der Deutschen Gesellschaft für Säugetierkunde

alko, Elisabeth K. V.; Handley, C. O., Jr.: Evolution, biogeography, and description of a new species of Fruit-eating
bat, genus Artibeus Leach (1821), from Panama. — Evolution, Biogeographie und Beschreibung einer neuen,

| fruchtfressenden Fledermausart der Gattung Artibeus Leach (1821) aus Panama 257,

ste B., J.; Ibanez, C.: Contribution to the knowledge of the bat fauna of Bioko island, Equatorial Guinea (Central

Africa). —- Beitrag zur Kenntnis der Chiropterenfauna der Insel Bioko, Aquatorial-Guinea 274

orre, I.; Tella, J. L.: Distribution of the Cabrera water shrew (Neomys anomalus) in Northeastern Spain. —

_ Verbreitung der Sumpfspitzmaus (Neomys anomalus) im Nordosten Spaniens 282
atson, J. O.; Blood, B. R.: A report on the distribution of small mammals from Namibia. — Verbreitungsmuster von
Kleinsäugern in Namibia 289

potorno, A. E.; Sufan-Catalan, J.,; Walker, Laura I.: Cytogenetic diversity and evolution of Andean species of
Eligmodontia (Rodentia, Muridae). — Cytogenetische Vielfalt und Evolution von Eligmodontia-Arten in den Anden
(Rodentia, Muridae) 299
tubbe, Annegret; Wiegand, Sabine: Influence of photoperiod and temperature on moulting processes in Microtus

brandti (Radde, 1861). -— Der Einfluß von Photoperiode und Temperatur auf Fellwechselprozesse bei Microtus
\ brandti (Radde, 1861) 309

issenschaftliche Kurzmitteilung

chreiber, A.; Dmoch, R.: Chromosomes of two rare species of neotropical mammals: Southern pudu (Pudu puda)
and Bush dog (Speothos venaticus). - Karyotypen von zwei seltenen südamerikanischen Säugerarten: Südpudu
(Pudu puda) und Waldhund (Speothos venaticus) Sl

(erlag Paul Parey Hamburg

HIERAUSGEBERYEDIIEORS

P. J. H. van BrEE, Amsterdam -— W. FIEDLER, Wien -— H. Frick, München - G. B.

HarTL, Wien -—- W. HERRE, Kiel - R. HUTTERER, Bonn - H.-G. Krös, Berlin - H.-].

Kunn, Göttingen — E. KuLZEr, Tübingen -— W. MAIER, Tübingen — J. NIETHAMMER,

Bonn - ©. Anne E. Rasa, Bonn - H. Reıcastein, Kiel -— M. Rönrs, Hannover —

H. SCHLIEMANnN, Hamburg — D. STARcK, Frankfurt a. M. - E. THEnıus, Wien - P. Vo-
GEL, Lausanne -— H. Wınkıng, Lübeck

SC HIRTEN EI TUNG/EDTIORLAIIFOFEIGE

D. Kruska, Kiel - P. LANGER, Gießen

Thıs journal is covered by Biosciences Information Service of Biological Abstracts, and by Current Con-
tents (Series Agriculture, Biology, and Environmental Sciences) of Institute for Scientific Information

Die Zeitschrift für Säugetierkunde veröffentlicht Originalarbeiten und wissenschaftliche Kurzmittei-
lungen aus dem Gesamtgebiet der Säugetierkunde, Besprechungen der wichtigsten internationalen
Literatur sowie die Bekanntmachungen der Deutschen Gesellschaft für Säugetierkunde. Verantwort-
licher Schriftleiter im Sinne des Hamburgischen Pressegesetzes ist Prof. Dr. Dieter Kruska.

Zusätzlich erscheint einmal im Jahr ein Heft mit den Abstracts der Vorträge, die auf der jeweiligen
Hauptversammlung der Deutschen Gesellschaft für Säugetierkunde gehalten werden. Sie werden als
Supplement dem betreffenden Jahrgang der Zeitschrift zugeordnet. Verantwortlich für ihren Inhalt
sind ausschließlich die Autoren der Abstracts.

Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z. Hd. Prof. Dr. Dieter
Kruska, Institut für Haustierkunde, Biologiezentrum der Christian-Albrechts-Universität, Am
Botanischen Garten 9, D-24118 Kiel, Bundesrepublik Deutschland. Für die Publikation vorgese-
hene Manuskripte sollen gemäß den „Redaktionellen Richtlinien“ abgefaßt werden. In ihnen
finden sich weitere Hinweise zur Annahme von Manuskripten, Bedingungen für die Veröffent-
lichuhg und die Drucklegung, ferner Richtlinien für die Abfassung eines Abstracts und eine
Korrekturzeichentabelle. Die Richtlinien sind auf Anfrage bei der Schriftleitung und dem Verlag
erhältlich.

Sonderdrucke: Anstelle einer Unkostenvergütung erhalten die Verfasser von Originalbeiträgen und
Wissenschaftlichen Kurzmitteilungen 50 unberechnete Sonderdrucke. Mehrbedarf steht gegen
Berechnung zur Verfügung, jedoch muß die Bestellung spätestens mit der Rücksendung der
Korrekturfahnen erfolgen.

Vorbehalt aller Rechte: Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich
geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks,
des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der
Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei-
ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ıst auch im Einzelfall grund-
sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines
Beitrages ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber-
rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 ın der Fassung vom
24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen
den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind
mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen.

Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of
the fırst page of an article in this journal indicates the copyright owner’s consent that copies of the article may
be made for personal or internal use, or for the personal or internaluse of specific clients. This consent is given
on the condition, however, that the copier pay the stated percopy fee through the Copyright Clearance
Center, Inc., 21 Congress Street, Salem, MA 01970, USA, for copying beyond that permitted by Sections 107

or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying
for general distribution, for advertising or promotional purposes, for creating new collective, or for resale.
For copying from back volumes of this journal see “Permissions to Photo-Copy: Publisher’s Fee List” of the
Eee:

Fortsetzung 3. Umschlagseite

© 1994 Paul Parey. Verlag: Paul Parey GmbH & Co. KG, Hamburg. Anschrift: Spitalerstr. 12, D-20095 Hamburg,
Bundesrepublik Deutschland.
Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., Heide/Holst.

Z. Säugetierkunde 59 (1994) 257-273
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Evolution, biogeography, and description of a new species of
Fruit-eating bat, genus Artibeus Leach (1821), from Panamä

By ELISABETH K. V. Karko and C. ©. HANDLEy, Jr.

Division of Mammals, National Museum of Natural History, Smithsonian Institution, Washington,
USA and Smithsonian Tropical Research Institute, Balboa, Republic of Panama

Receipt of Ms. 15. 3. 1993
Acceptance of Ms. 6. 6. 1994

Abstract

We describe and name a new species of Neotropical fruit-eating bat, genus Artibeus. It ıs a local
endemic, found only on Isla Escudo de Veraguas, approximately 18 km off the Caribbean coast of the
province of Bocas del Toro in northwestern Panama. Based on a variety of shared characters we
assume that this new species has evolved from a species also ancestral to Artibeus watsoni, which is
widespread and abundant on the mainland and other islands of Bocas del Toro. The new species is
15% larger in mass than its mainland relative, 10 % larger in body dimensions, and 6 % larger ın
cranıal dimensions. Several discrete morphological characters, particularly ın the dentition, also
distinguish this bat from A. watsoni. We discuss aspects of evolution and biogeography of A.
incomitatus to elucidate factors which might have facilitated its speciation.

Introduction

Laguna de Chiriqui on the Caribbean coast of northwestern Panama (province of Bocas del
Toro, 30 km southeast of the Costa Rican boundary) harbors numerous islands (Fig. 1), all
formed by rising sea level. The islands are small (the largest is only 59 km?), are close to the
mainland, and wıth one exception they lie ın shallow water. Isla Escudo de Veraguas
(hereafter, “Escudo”), is the exception. It ıs surrounded by deeper water and is relatively
remote, approximately 18 km off the coast of the mainland and 50 km from the nearest of
the other Bocas islands.

Most of the islands, except the smallest, are to some degree inhabited by people.
Originally, all of the islands with dry ground were covered with evergreen forest. In recent
centuries, repeated disturbance by humans has caused the extirpation of some of the fauna
and has removed part or all of the old (primary) forest from most of the islands.

There has been surprisingly little interest in biological exploration of these ıslands.
WETMORE in 1958, HANDLEY ın 1960, 1962, and 1963, and Dary and Myers in 1967 made
brief forays to the ıslands. Recently, a group from the Smithsonian Institution initiated
inventories and in-depth studies of the island ecosystems. Its annual expeditions from
1987-93 have touched all of the major islands and some points on the adjacent mainland,
and made collections that include mammals, birds, reptiles, amphibians, and plants.

Like other systems of small ıslands, the Bocas islands have depauperate faunas and
floras. Although most of the islands are very close to the mainland, small in size, and
young in origin, the inventories have revealed relict taxa and morphological differentiation
of taxa between islands and mainland, as well as between the islands themselves (e.g.,
Hanpıey 1959a). This is of special interest, as it raises an essential question: Why is there
large scale response to small scale isolation? Comparative studies of species assemblages
occurring on the mainland and on the islands will reveal part of the answer. Here we
describe a new species of bat, genus Artibeus, which ıs found only on Isla Escudo de
Veraguas, Panama.

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5905-0257 $ 02.50/0

258 Elisabeth K. V. Kalko and C. O. Handley, Jr.

Boca del Drago

B Bocas del Toro
EL,

BASTIMENTOS
„a PANAMA
L „ ZAPATILLA CAYS

PoPA J°
a 5 AGUA

VALIENTE ESCUDO DE VERAGUAS

&

LAGUNA DE CHIRIQUI

Eagelog 8° 15 82°00 81°45 81° 30°

Fig. 1. Archipelago of Laguna de Chiriqui, Bocas del Toro, Panamä

Material and methods

Study area

Escudo is a small (4.3 km’), triangular, heavily forested island in the Carribean, 17.6 km off the
mainland Peninsula Valiente. From a distance Escudo appears to be flat but, except for red mangrove
(Rhizophora) swamps at eastern and western extremes, ıts surface is washboard-like. Its topography
consists of a symmetrical series of low (<50 m), steep-sided, flat-topped, parallel ridges separated by
small, narrow swamps.

Escudo is the most remote of the Bocas ıslands. MERRILL VARN and CHARLES HANDLEY (unpubl.
studies) have found it to be the oldest, formed by rising sea level. Varn and HAnDLEY correlated
ocean floor topography around Escudo with dates from pollen and coral cores from the western

Carrıbean (e.g., BARTLETT and BARGHOORN 1973). Their studies show that until sea level had risen to
about 29 m below present sea level (about 9000 years before present, B. P.) Escudo was part of the
mainland.

As sea level continued to rise, isolation of Escudo proceeded rapidly. Within 500 years (about 8500
years B.P.), at a sea level 25 m lower than present, the seaway between Escudo and the mainland had
already opened to about 3 km. Within another 700 years (about 7200 years B. P.) sea level had risen to
15 m below present, and the distance between Escudo and the mainland had widened rapıdly to about
12 km. Thereafter, the channel gained width gradually to ıts present 17.5 km.

Weather conditions

Frequent storms, accompanied by high winds and high tides, dominate the weather of Escudo.
Distinct raıny seasons, June-August and November-December, are punctuated by indistinct dry
seasons. Rainfall records have not been kept on Escudo, but because it ıs further away from the high
mountains of the mainland, Escudo probably has less rain annually than the 2900 to 3200 mm
recorded on the islands that lie within Laguna de Chiriqui.

A new species of fruıt-eating bat from Panama 259

Habitats

Evergreen forest covers about 95% of Escudo. Less than 5% of the island is currently cleared for
houses, garden plots, plantains, and coconuts. Much of the forest is old growth, but tall young forest
reveals that the western end of the island was extensively cleared in the recent past. All of the forest has
been much damaged by wind, forming a peculiar natural mosaic of old and young growth.

Collections

HanDLEy and F. M. GREENWELL explored the southwestern sector of Escudo 19-24 March 1962, and
collected among other mammals two species of bats, Glossophaga soricina (a nectar-feeder/insectivore)
and specimens of a small canopy-foraging fruit bat resembling Artibeus watsoni, but larger. Again in
1990 (29 March-2 April), HANDLEy and GREENWELL surveyed the eastern quarter of Escudo. They
mist netted many of the Artibeus watsoni-like bats and to the known chiropteran fauna they added a
short-tailed fruit bat, Carollia brevicauda, which feeds on the fruits of shrubs.

During the last two weeks of March 1991, from a camp with indigenous people on the west shore,
Hanpıe£y and Penny NEıson collected bats in the western third of Escudo. They extended the series
of Artibeus watsoni-like bats and added three more species of bats to the known fauna: Saccopteryx
leptura and Myotis riparius, both insectivorous and feeding on the wing, and another small insectivo-
rous bat, Micronycteris megalotis, which mainly gleans insect prey from surfaces. In April 1993
Kaıko and HANDLEY stayed in a camp of native fishermen on the northeastern shore of Esucodo to
net and photograph the Artibeus watsoni-like bat. To date the native mammalıan fauna of Escudo is
known to include six species of bats, a marsupial, a sloth, and one rodent. In addition there are feral
introduced house rats, house cats, and swine.

Specimens examined

Artibeus (new species), total 61. Panama, Bocas del Toro: Isla Escudo de Veraguas (61).

Artibeus watsonı, total 134. Panama, Bocas del Toro: Cayo Agua (12); Cayo Nancy (14); Cayo
Zapatilla Este (3); Isla Bastimentos, Punta Vieja (9); Isla Colön, La Gruta (7); Isla Popa, 1 km SE
Canal de Isla Deer (11); Isla San Cristöbal, Bocatorito (21); Peninsula Valiente (various localities
around the shores of Bahia Azul) (13); Tierra Oscura, 3.5 km S Cayo Tigre (8). Panama, Chiriqui:
Progresso, 1.6km SW, 12.8km SE, and 24.2km SE (34); Puerto Armuelles, 3.2km SW (2).
Specimens representative of each taxon from each localiıty have been returned to Panama. All
specimens mentioned in the text are ın the National Museum of Natural History (USNM),
Smithsonian Institution, Washington, DC.

Morphometrics

External and cranıal dimensions were measured in millimeters; mass in grams. Total length, ear,
wingspan, and mass were measured on fresh specimens in the field. Forearm, hind foot, tibia, and
calcar were measured on dry skins of prepared specimens. External measurements were taken in the
conventional manner (HANnDLEY 1988). Cranial measurements (Fig.2) were taken as follows (a
redefinition of the measurements first described in HanDLEy [1959b], and subsequently widely used
as a standard for bats):

Greatest length: Distance between the anteriormost point of the premaxillae and the hindmost point

of the skull.
Zygomatic breadth: Greatest breadth between the outer edges of the zygomata.

Postorbital breadth: Least breadth across the constriction of the frontals, posterior to the postorbital
processes or bulges.

Braincase breadth: Greatest breadth of the globular part of the braincase; measured by closing the
calıpers on the outer walls of the braincase and sliding down to the point of abrupt flare to the
squamosal edge.

Braincase depth: Greatest distance between the medio-ventral surface of the basioccipital and the
dorsalmost point of the braincase, the sagittal crest not included.

Maxillary toothrow length: Greatest crown length from the anteriormost edge of the canine to the
posteriormost edge of the last molar in a maxillary toothrow.

Postpalatal length: Distance between the anteriormost point of the mesopterygoid fossa (disregarding
a median projection) and the anteriormost point of the foramen magnum.

Maxillary breadth: Greatest alveolar breadth between the outer edges of the maxillary toothrows.
Canine breadth: Greatest distance between outer sides of upper canines at the alveolı.

260 Elisabeth K. V. Kalko and C. O. Handley, Jr.

— GREATEST LENGSTH —

U
0°
0]
0
%
0
]
°
v
v
0
N
0%
1]
v
]
%
0
„oo ..-...

ve

ons
ZYGOMATIC POSTORBITAL BRAINCASE
BREADTH BREADTH BREADTH

MAXILLARY
TOOTHROW POSTPALATAL
LENGTH LENGTH

|
|

MAXILLARY CANINE

BREADTH BREADTH

BRAINCASE

DEPTH

Fig. 2. Dorsal, ventral, and lateral views of a skull of A. incomitatus, illustrating method of taking
cranial measurements

Only measurements of adults, recognized by closure of epiphyses, are included in our tabulations.
As no sexual dimorphism was found, the data sets of both sexes were pooled. For mass, only males
were used, as mass ıs variable in females.

External and cranıal measurements taken in the laboratory were made with dial calıpers to the
nearest 0.1 mm. Data are presented as mean, + two standard errors. They are rounded to the nearest
1/10 for the mean and to the nearest 1/100 for the standard error. The significance of differences in
measurements between the new species and the related A. watsonı was verified by means of the Mann-
Whitney U-test. Bivariate graphs were used to compare proportions of external and cranial features of
both species. Next, we assessed the morphomentric variability of the specimens by examining them in
multivariate space. Specifically, a Principal Components Analysis (PCA) was run, using the 15
varıiables listed in table 2. The Factors module of the program Systat for Windows (version 5.03;
WILKINSON et al. 1992) was used for the analyses. The PCA (not rotated) was performed on a
correlation matrix calculated with standardized data. To maximize the number of specimens used in
the calculations, missing data were handled ın a pairwise manner. That is, specimens with missing data
were not automatically excluded because of a small amount of missing data during the calculation of
the correlation matrix. For further discussion of the methods see WILKInson et al. (1992), Davıs
(1986), and CHAPMman et al. (1981).

Results

Artibeus incomitatus, new species
Holotype

USNM 579125, adult male (testis 7x5 mm), skin and skull, collected 19 March 1991 by
CHARLES HANDLEY a meter or so above sea level, near West Point, Isla Escudo de

A new species of fruit-eating bat from Panama 261

Veraguas, Bocas del Toro, Panamä, in a mist net in a clean coconut palm plantation.
Original number, COH 17002.

Etymology

Latin, incomitatus, unaccompanied, alone; referring to the ısolation of this bat on Isla
Escudo de Veraguas, where it seems to be the only Artibens and indeed the only
stenodermatine.

Distribution

This bat ıs known only from Isla Escudo de Veraguas, province of Bocas del Toro,
Panama, where it is the most frequently netted bat. It was found in all habitats sampled,
including upland forest, swamp forest, and coconut plantations. Elevational range, near
sea level to 50 m, the highest point on the island.

Diagnosıs

Artibeus incomitatus ıs characterized by large sıze; long, shaggy, bicolored dorsal fur;
sooty dorsum; rather dark underparts; ıll-defined facial stripes; cream-color on edges of
ear, tragus and horseshoe; hairy posterior extremities; robust skull; broad, deeply-arched
rostrum; supraorbital swelling not breaking supraorbital outline; capsulelike swellings on
orbital wall always five, distinct, and subequal; subparallel zygomata; U-shaped anterior
margin of mesopterygoid fossa; ill-defined inner edge of pterygoid fossa; equal or subequal
cusps on I1; notches on hınd edge of P4 few and indistinct; distance between paracone and
protocone greater than distance between paracone and metacone on MI; paraconid cusp of
m2 large, high, and situated on medial anterior edge of tooth; metaconid cusp of m2 inset
from lingual margin; conulid between metaconid and entoconid of m2 large; m3 variably
present or absent.

Description

Size large (forearm x = 42.7 mm, wingspread x = 327.0 mm, male mass x = 13.0 8).
Dorsum varyıng from sooty to brown, average near Burnt Umber (capitalized color terms
are from Rıpaway 1912); underparts paler, near Warm Sepia; facıal stripes usually poorly
defined, whitish, but sometimes well defined, white; face, between stripes, black; mem-
branes, ears, and noseleaf blackish; ear, tragus, and sometimes horseshoe of noseleaf
narrowly edged with cream; lower edge of horseshoe free; wing attached to side of
metatarsus; tıbıa, foot, and interfemoral membrane usually appear hairy.

Skull large (greatest length x = 20.9 mm, maxillary toothrow x = 7.0 mm); rostrum
broad, deep, and arched, with prominent supraorbital swelling which does not break
supraorbital outline; preorbital-supraorbital rim sharp-edged but low, disappearing before
reaching ill-defined postorbital process; frontal wall of orbit rippled with the outlines of
five well-defined, oblong, subequal capsules; postorbital constriction scarcely narrower
than distance between postorbital processes; braincase narrow and deep, with evenly
convex dorsal profile and low, sharp-edged sagittal crest, only vaguely connected
anteriorly to postorbital processes; Jambdoidal crest low and ill-defined in spots; zygomata
weak and subparallel; palate wide, subeircular, with well-marked lateral depressions
between canine and MI; postpalatal extension parallel-sided, with U-shaped posterior
margin; pterygoid fossa with ill-defined inner edge, opening diagonally inward; basial pits
(interauricular depressions) deep and well-defined, with rounded median septum; auditory
bullae small.

Tooth formula I 2/2, C 1/1, P 2/2, M 2/2-3 X 2 = 28-30. I1 with equal or subequal
cusps; notches on hind edge of blade of P4 usually few and indistinct; M1 wide, with both
protocone and hypocone expanded and distance between paracone and protocone consid-

262 Elisabeth K. V. Kalko and C. O. Handley, Jr.

erably greater than distance between paracone and metacone; lingual longitudinal sulcus on
p4 deep, extending to near base of tooth; anterolingual (paraconid) cusp of m2 large and
displaced medially from lingual margin of tooth to form, with the anterolabial (pro-
toconid) cusp, an anterior rım for the tooth (see Fig. 3 for nomenclature of cusps); conulid
between metaconid and entoconid somewhat enlarged; m3 tiny, often lacking from one or

both mandibles.

Measurements of the holotype

Adult male: Total length 59 mm, tail vertebrae 0 mm, hind foot (dry) 11 mm, ear from
notch 17 mm, forearm 42.4 mm, wingspread 325 mm, tibia 15.2 mm, calcar 4.2 mm, mass
12.5 g. Greatest length of skull 20.6 mm, zygomatic breadth 11.7 mm, postorbital breadth
5.1 mm, braincase breadth 8.8 mm, braincase depth 7.7 mm, maxillary toothrow length
7.1 mm, postpalatal length 6.8 mm, maxillary breadth 8.5 mm, canıne breadth 5.5 mm.
See table 1 for a summary of measurements of a series.

Comparisons with related species

Compared with its close relative, A. watsoni of the Bocas islands and mainland, A.
incomitatus averages 15 % greater ın mass, 10% larger in external dimensions, and 6 %
larger in cranıal dimensions. Every measurement except braincase depth is significantly

A

anterior

paraconid(1)

— protoconid(4)
metaconid(2)—

conulid

entoconid(3) — — hypoconid(5)
posterior

lingual labial

Fig. 3. Occlusal view of lower second molar (schematic drawings based on camera lucida sketches). A:
A. incomitatus from Isla Escudo de Veraguas. B: A. watsoni of Peninsula Valiente. Names of cusps are
from Hanprev (1959b); numbers of cusps are from ANDERSEN (1908)

A new species of fruit-eating bat from Panama 263

larger in A. incomitatus (p<0.001, Mann-Whitney U-test). Relative to A. watsoni, the
largest measurements are tibia and calcar; the smallest are hind foot, zygomatic breadth,
braincase breadth, and braincase depth. Bivariate scatter graphs of all individuals measured
illustrate very well the high degree of separation between the two taxa. Figure 4 shows a
pair of external characters; figure 5, cranıal characters.

Further, the Principal Components Analysis (PCA) confirmed the large size difference
between A. incomitatus and A. watsoni. The results of the PCA are summarized in tables 2
and 3, and figures 6 and 7. Fifteen components (PCs) were derived but only the first two
axes had eigenvalues greater than 1; i.e. they represent more variance than any single
varıable.

PC-I accounted for almost 59% of the varıation in the data and ıs a classic sıze axis,
indicated by the high positive loadings of all the varıables (Tab. 2). This is very frequently
the case in morphometric studies on single or closely related taxa (see CHAPMman et al.
1981). The loadings also indicate that of the original varıables, greatest length is the best
indicator of overall size (loading = 0.95) and braincase depth and postpalatal length are the
least effective (both with loadıngs of approx. 0.55). The scores for the specimens (sample
averages in table 3; scores plotted figures 6 and 7) give an indication of the overall size of
the specimens, based on input from all 15 varıables. A specimen with a high positive value
for PC-I is very large compared to those wıth values near zero. Those with high negative
values are the smallest (see CHmapman et al. 1981).

PC-II accounted for just over 8% of the varıance in the data matrix and represents
shape variation in braincase size (positive; braincase breadth and depth and ear versus
selected standard lengths) (negative; total length, hindfoot, postpalatal length, tibia, and
calcar). Not surprisingly, these varıables tested to have the smallest loadings on PC-I. A
specimen with a high positive value on this axis has a relatively large braincase and ear
measurement and is relatively smaller in the length measurements. A high negative value
would indicate the opposite.

In order to interpret these results we examined the scores; that ıs, how the specimens
related to the new axes; and analyzed patterns in them relative to sex, taxon, and sample.
No apparent trends could be discerned related to sex on either axis, suggesting that sexual
dimorphism is not significant in these taxa.

Plotting the scores with the data coded to represent the specimens of each population
(PC-I) indicate clear differences between A. watsoni and A. incomitatus. In PC-I, the
specimens from Isla Escudo all plot positive, some highly positive, with relatively little
overlap with the other samples (Fig. 6). This reflects the large sıze difference described for
A. incomitatus relative to the much smaller A. watson:. Plotting the average scores for each
population shows trends for PC-II which are more subtle but still apparent (Fig. 7).
Ignoring the Isla Escudo specimens and concentrating on A. watsoni alone, we see a
mainland-ıisland trend in the average values for PC-II. The Chriqui specimens not only
tend to be small but they tend to have larger negative values tor PC-Il, indicating that they
have relatively high values for the length measurements and small values for the braincase
and ear measurements. On the other side of the PC-I trend are the larger near-shore
islands of Colön and Bastimentos. Most other ıislands and Peninsula Valiente are inter-
mediate but closer to the larger ıslands. Isla San Cristöbal and Tierra Oscura, on the
adjacent mainland, plot more toward the middle area. This suggests a northeast (island) —
southwest (mainland) cline.

Dorsal coloration is similar ın A. incomitatus and A. watsoni, but underparts are darker,
gray-brown rather than bufty, in A. incomitatus (12 of 13 Peninsula Valiente specimens are
paler than the average Escudo specimen; only 6 of 45 Escudo specimens are as pale as the
average Valiente specimen). Less than half (44%) of Escudo specimens have prominent
well-defined, white facıal stripes, while more than two-thirds (69%) of Valiente A.
watsonı have prominent stripes. A. incomitatus usually (75%) appears to have posterior

WINGSPREAD

340

320

310

300

290

280

264

Total
length

(mm)

Hind foot
(dry)

(mm)

Ear

(mm)

Forearm

(mm)

Tıbia

(mm)

Elisabeth K. V. Kalko and C. O. Handley, Jr.

Calcar

(mm)

Wingspan

(mm)

Artibeus incomitatus, Isla Escudo de Veraguas

59.7#0.61 I =E0H6 17.7+0.14 42.70.30 14.70.15 320E=0H1B SI YZERLSO 13.0#0.34
55-64 10-13 17-19 40.6-44.4 13.6-15.8 4.0-5.9 305-343 10.8-16.0
(60) (49) (60) (48) (48) (48) (57) (50)
Artibeus watsoni, Peninsula Valiente
INA 10.1#0.26 16.6 40.46 38.4+0.8 13.3#0.38 4.2+0.16 298.3 45.44 10:9-=0 72 ||
46-60 9-11 15-18 35.9-40.7 12.4-14.7 3.44.5 282-310 9.5-13.0 |
(13) (13) (13) (13) (13) (13) (13) (19) l
Artibeus watsoni, Cayo Nancy |
54.1#0.96 10.6+0.26 16.1#0.51 38.0+0.58 133-3050 4.40.20 294.6 +6.00 11.320.028 11
52-59 10-11 14-17 35.9-39.2 12.2-14.9 3.8-5.1 277-315 9.5-13.0 1
(14) (14) (14) (14) (14) (14) (14) (10) |
Artibeus watsonı, all islands (except Escudo) and mainland i
54.5 +0.65 10.4+#0.10 16.0#0.25 38.5+0.25 13.0#0.14 4.40.07 296.6+2.10 HAE=027 |
42-60 9-12 12-18 35.9—41.6 11.3-14.9 3.4-5.4 273-317 9.0-14.0

(96) (98) (97) (98) (98) (96) (96) (77)

a j\ I specimen
2‘
4 7‘
)
= A a 2 specimens m* Ü)
_\ a
SEES Artibeus watsonl a Q Q [\ ww‘
4a [\
a
7 $ Artibeus incomliltatus Y\ a '\
2
AA
[\
N‘ a
y\ \#
aa 4 N
4 h
a a
N \
Z\ 0
& Mi
& & 7‘ a & 7‘ |
[o\ &ä \
i
& MM
A & |
fo 23 & AN aAdh
[03
&
&ä
[e\ &
[03 a 68 & [er \
[03 [eY
[03 N
ö &ö BD A‘
& D &
& SEINE ASEEN A
&
AN
N
ä N
& AAA Ah D &
a ö
0%
[0% a & |
[03 [> j
[eY [>
N
[> fay
& fe ARE NEN
[N &
N
&
a
& |
N
[%\
fo)
fe)
|
35 36 SH 38 39 40 41 42 43 44

FOREARM LENGTH \

Fig. 4. Comparison of wingspread and forearm length in 45 specimens of A. incomitatus from Isla
Escudo de Veraguas and 96 specimens of A. watsoni from shores and islands of Laguna de Chiriqui

265

A new species of fruit-eating bat from Panama

incomitatus and A. watsoni

measurment is given as the mean + 2 standard errors, Min/Max and the number of specimens measured

Greatest Zygomatic Postorbital Braincase DBraincase Maxillary Postpalatal Maxillary Canine
length breadth breadth breadth depth toothrow length breadth breadth
(mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm)

|
1 E20,9EE 041022121007 5.0004 9120.05. 77.6=50.06.. 7.0=50.04 6.80.06 8.6#0.6 5.6+0.04
20.0-21.7 11.3-12.8 4.7-5.3 8.8-9.6 7.1-8.1 6.7-7.4 6.3-7.3 8.0-9.2 5.3-5.9
(60) (58) (60) (60) (59) (60) 657) (60) (60)
| II7ES07 7025774800 8.8#0.11 76-20, EEE 652201370013 78.1=E017 75.2200
1 © 19.1-20.6 10.6-12.4 4.4-5.1 8.9-9.2 7.1-8.0 6.2-6.9 6.0-7.0 7.6-8.8 4.9=3.9
(13) (13) (13) (13) (13) (13) (13) (13) (13)
IOSE-OH I ZEN O== 031722 74.7:= 0.06 8.922.098 7775==0,132765=:01092 76.3=80.10778:.0=2041777 75220412
19.1-20.1 11.1124 4.5—4.9 8.4-9.3 7.1-7.8 6.1-6.7 6.1-6.7 7.AZS.A 4.8-5.5
(11) (13) (13) (13) (10) (13) (11) (12) (12)
1976-=0:082=:11%6=2.0.072 74.720.03 828210. 04552272521=0.052,26:5-50.04 2764:=.0:05, 358.1:=20:052755.320.04
18.7-20.6 10.6-12.4 4.3-5.1 8.39.3 7.0-8.2 6.1-6.9 5.9-7.2 7.4-8.8 4.8-3.7
(94) (91) (6) (96) (93) (97) (93) (95) (95)

I specimen

2 specimens

3+ specimens

Artibeus Iincomliltatus

Artibeus watsonl

A
a

[N
A
a
&
&
A

6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 7.0 Voll 0.2

MAXILLARY TOOTHROW
Fıg. 5. Comparison of greatest length of skull and maxillary toothrow length ın 60 specimens of A.
incomitatus from Isla Escudo de Veraguas and 94 specimens of A. watsoni from shores and islands of
Laguna de Chiriqui

266 Elisabeth K. V. Kalko and C. O. Handley, Jr.

Table 2. Variables used in Principal Compo- extremities (tıbıa, foot, and interfemoral
nents Analysis and their loadings for the first membrane) hairy, whereas A. watsoni ap-

two Principal Components pears to have these parts almost naked (12
Eigenvalues and percent variance explained for of 13 specimens from Peninsula Valiente).
first two Principal en are also tabu- Under magnıification all specimens of both

ate

species prove to have hairy posterior ex-
tremities. Hairs are longer and possibly

Loadings denser in A. incomitatus.
rsöilles De mein The skull of A. incomitatus is heavier
Worin 0.6674 0.4435 and more bulky than the skull of A. wat-
a kraydl Koss 0.6288 0.3199 soni. Supraorbital swelling ıs pronounced,
Bar 0.7035 0.2067 but usually not sufficiently in Escudo
Sn neh ae Bi specimens (23 of 30) to break the smooth-
re ach 08525 0.1941 edged supraorbital outline, whereas only 11
Postorbital breadth 0.7235 0.0401 of 30 A. watsoni have the outline unbroken.
Braincase breadth SEI Capsule-like swellings on the frontal wall
en a N > of the orbit always five, distinct, and sub-
xıllar rOW ö . . . . .
Porpäll nah 05485 _0 3666 equal ın A. incomitatus, sometimes only
Maxillary breadth 0.8696 0.1430 four, often indistincet, and always variable
Canine breadth 0.8351 0.0499 in size ın A. watsoni. Zygomata usually
Tibıa 0.8050 0.2376 subparallel (18 of 30) ın Escudo specimens,

Calcar 0.6107 -0.3233

Eigenvalue (Total = 15) 8.8138 1.2193
1 % Varıance explained 58.7589 8.1287

but almost always (26 ot 30) narrow an-
teriorly and swept-back in A. watsoni. An-
terior margin of mesopterygoid fossa al-
most always (26 of 30) U-shaped ın A.
incomitatus, often V-shaped in A. watsoni (17 of 30). Inner edge of pterygoid fosa always
rıdged in A. watsoni, but rıdge almost always (26 of 30) ıll-defined or absent in A.
incomitatus.

Cusps of I1 almost always (39 of 42) equal or subequal in A. incomitatus, while the
outer cusp usually (30 of 44) is larger in A. watsoni. Notches on the hind edge of the blade
of P4 usually few and indistinct in Escudo specimens (29% with one sharp notch, 62 %
with 1-3 obscure notches, and 9% notchless), whereas 56 % of Valiente specimens have
1-3 sharp notches and 44% have 1-3 obscure notches. The paraconıd cusp of m2 has a
medial location on the anterior edge of the tooth in 29 of 35 specimens of A. incomitatus, a

Table 3. Statistics for scores from Principal Components Analysis of samples of Artibeus
incomitatus from Isla Escudo de Veraguas and A. watsoni from ten localities in Bocas del Toro
and Chiriqui, Panama

Localities Abbrev. Factor Score Data
for PD@1 PC-II
Locality Mean/S.E. Mean/S.E.

Isla Escudo
Peninsula Valiente
Isla Colön

Isla Bastıimentos
Cayo Nancy

Cayo Zapatilla Este
Cayo Agua

Isla Popa

Isla San Christöbal
Tierra Oscura
Progresso, Chiriqui

1.31/0.07 —0.43/0.14
—0.47/0.19 0.88/0.19
-0.03/0.62 1217049
—0.54/0.21 1.30/0.35
—0.44/0.18 0.27/0.29
—0.51/0.43 0.79/0.40
-0.78/0.13 0.36/0.26
-0.32/0.09 0.22/0.22
-0.30/0.10 0.08/0.18
-0.56/0.14 —0.21/0.23
-0.98/0.09 -1.00/0.17

OHVReSNZUuNn<H

A new species of fruit-eating bat from Panama 267

Principal Components Analysis
Scores for Specimens

3
a A /\
2 I z /\ z A D
'\ 2 An, B e)
4 Y\ AN
IirhöNl Re er Eeire Bite
(2) 0. r ee a ie) zo
D. N NASR Na. 8 r
IE 57 ad, Ada, n =: ug &
-1 y\ „a: ame] Au) je)
N z 0
2} RL
2 1 1 l 1 8 N
-3 -2 -1 0 1 2 3
PCI
a A. watsomi A. incomitaltus

Fig. 6. Results of Principal Components Analysis. Scores of specimens plotted for PC-I (X-axıs)
versus. PC-II. Labels indicate species of specimen; triangle = Artibeus watsoni; square = Artibeus
incomitatus
Principal Components Analysis
Sample Average Scores

3 i Es
| Islands
B c

il; zV
= A NE
a O2 T E

Mainland

-1 } Q

2, Small g————m—> Large

-3 L l L L L

-3 -2 -1 0 1 2 3

PCI - Size

Fig. 7. Results of Principal Components Analysıs. Scores of specimens plotted for PC-I (X-axis)
versus PC-II. Labels are mean values for populations (see Tab. 3)

lateral location ın 45 of 46 specimens of A. watsoni (Fig. 2). This cusp is large (24 of 35) and
high (18 of 35) in A. incomitatus, but small (42 of 46) and low (34 of 46) ın A. watsoni. The
metaconid cusp of m2 is inset from the lingual margin of the tooth ın 22 of 35 specimens
from Escudo, but ıs on the margın of the tooth in 41 of 46 A. watsonı. The conulid between
the metaconid and entoconid cusps of m2 is large ın 28 of 35 A. incomitatus, but ın only 13
of 46 A. watsoni. Presence or absence of m3 on either mandible varıes among populations
of these bats, from a frequency of zero on Cayo Zapatilla Este and 0.5 on Cayo Nancy to
1.7 on Isla Popa and 2.0 on Isla Colön. The mean for Isla Escudo is 1.2 and the mean for
the other ıslands and mainland of Bocas del Toro collectively is 1.5.

268 Elisabeth K. V. Kalko and C. O. Handley, Jr.

There is another small species of Artibeus (A. phaeotis), on the mainland and on some of
the ıslands of Bocas del Toro. Compared with A. incomitatus, it is smaller, in the size range
of A. watsoni; dorsal fur distinetly tricolor versus bicolor or indistinctly tricolor, shorter
(6 mm versus 8 mm), smooth-Iying rather than shaggy, and brownish rather than sooty;
underparts more buffy; facıal stripes pure white and sharply defined; margins of ears,
tragus, and sometimes noseleaf yellow or orange rather than creamcolor. Rostrum shorter,
shallower, flatter, and less arched; supraorbital not swollen; postorbital constriction
narrower; braincase deeper and wider; zygomata taper (narrow anteriorly, flaring post-
eriorly); palate shorter and broader, subcircular; lateral palatal depressions usually ill-
defined or absent; anterior edge of mesopterygoid fossa V-shaped; inner edge of pterygoid
fossa sharply ridged, constricting roof of mesopterygoid fossa, and causing pterygoid fossa
to open posteriorly rather than inward. Protocone of Mi relatively closer to metacone, so
that protocone, metacone, and paracone usually come close to forming an equilateral
triangle; m2 longer and narrower, but otherwise like m2 of A. watsoni; m3 always absent.

Remarks

To check for the possibility that A. watsoni as well as A. incomitatus might occur on
Escudo, during the 1991 expedition to Escudo HanDLey selectively collected the smallest
Artibeus that were netted. This skewed the A. incomitatus series toward the small side, but
did not turn up any specimens in the size range or with morphological characters of A.
watsoni. Actually, there is surprisingly little variation, considering the size of the series
(159 specimens), among the skulls from Escudo on the one hand and those from all other
islands and mainland on the other (Tab. 1; Figs. 4, 5).

Ectoparasites

Parasitic tlies (Streblidae) collected by HAnDLEY and NELSoN from A. incomitatus on Isla
Escudo prove to represent a well-ditferentiated undescribed species, apparently evolved
from Paratrichobius lowei Wenzel, a parasite of A. watsoni (R. V. PETERSON, unpubl.
results). P. loweı has been reported from A. watsoni taken in eastern Panamä (San Blas and
Darien) (WENZEL et al. 1966) and from “Artibeus cinereus“ taken in Venezuela (Bolivar)
(WENZEL 1976). The Venezuelan hosts have been reidentified as Artibeus glaucus bogoten-
sis, amember of the A. glaucus group, which also includes A. watsoni (HANDLEY 1987).

Discussion
Taxonomic status of A. incomitatus

Insular populations of organısms showing morphological differentiation from mainland
taxa pose taxonomic problems in interpreting the observed varıability. As summarızed by
Wırson (1991), differentiated island populations can depending on the conceptual view-
point of the researcher be regarded either as subspecies of mainland species or as endemic
island species.

Because geographic isolation interrupts or diminishes gene flow between populations, i it
represents one of the most important factors promoting speciation. However, assessing
whether spatial isolation on an island has been sufficient to produce reproductive isolation,
the basic criterion of the biological species concept, remains an imponderable problem
always confounding systematic studies of island organisms. Often this question can be
addressed only indirectly.

Ideally, information about formation of the island, degree and duration of isolation

A new species of fruit-eating bat from Panama 269

from the mainland, detailed studies on morphological and genetic varıability of island and
mainland populations, as well as studies of ecological (e.g., community structure),
behavioral, and physiological characteristics should be available to allow a taxonomist to
judge adequately the significance of observed differences between island and mainland
populations. In reality such information seldom ıs available.

In our study we have described A. incomitatus as a species endemic on Isla Escudo de
Veraguas, Panamä. We compared it with a similar species, A. watsoni, which ıs widespread
on the mainland from southern Mexico to Ecuador (Harr 1981). Based on a set of
morphological characteristics shared by the two taxa we assume that A. incomitatus and A.
watsoni have evolved from a common ancestor. We further assume that the characters
which distinguish A. incomitatus and A. watsoni were variable in the ancestral bat before
Isla Escudo de Verguas was ısolated. After isolation the characters were varıously
accentuated in the evolving A. incomitatus and A. watsonı.

For several reasons we regard A. incomitatus and A. watsoni as species rather than as
subspecies. First, there is the matter of scale, acommonly used criterion for ranking insular
populations. If differentiation of the organısm on an ısland is greater than differentiation
between contiguous populations on the mainland which are supposed to be geographic
variants, then the insular population should be regarded as a species. A. watsoni ıs regarded
as monotypic throughout its range (Harı 1981), but actually it ıs geographically varıable
(Hanpıey 1987; unpubl. results). However, the degree of difference is much greater
between the insular A. incomitatus and the mainland A. watsonı than between any of the
mainland populations of A. watsonı. There ıs no evidence of intergradation between the
taxa, and every external and cranıal dimension we measured, except cranial breadth, is
significantly larger in A. incomitatus.

Principal Components Analysıs confirms the great size difference apparent in the
Escudo specimens. It also indicates an interesting trend, from the mainland across the inner
islands of Laguna de Chiriqui, in both shape and sıze of skull that invites further study,
especially documentation of genetic relationships of the samples. The results could
demonstrate a morpho-genetic clıne, indicating a species in the process of spinning off
from a sister species, or show the progressive fragmentation of a species-range, with the
predictable effect of ısolating varıous populations.

However, molecular analysıs alone, which ıs increasingly used to determine phylogeny
and taxonomic status of organisms, is not necessarıly the solution to the species dilema.
For example, recent studies on Peruvian Sturnira (Phyllostomidae) led to contlicting
evidence in morphological and genetic characters (PACHEco and PATTERSON 1992). Some
of the morphological characters did not show in the genetic analysis and, in reverse, some
molecular differences did not show in morphology.

Second, in addition to statistical differences in external and cranial measurements, A.
incomitatus and A. watsoni difter also in numerous discrete morphological characters in the
dentition, cranıum, and pelage. Among the dental characters that distinguish A. incom-
itatus the one we consider to be the most significant involves the size, shape, and location
of cusps on the second lower molar. The placement of the paraconıid in A. incomitatus on
the anterior margin of the tooth, close to the protoconid, has the effect of giving the tooth a
substantial anterior rım, undeveloped in A. watsoni. In contrast, A. watsoni almost always
has a wide, low gap between the paraconıd and the protoconid. These differences in
dentition must give the teeth of the two species slightly different functional characteristics,
which remain to be explained. The most distinctive cranial feature of the Escudo bat is
suppression of the ridge at the inner edge of the pterygoid fossa. In pelage, A. incomitatus
is characterized by drabness: sooty dorsal coloration, darkening of the underparts, and
obsolescence of facıal stripes, only to a limited extent seen in the mainland bat.

Third, the host-specific parasitic streblid fly of A. incomitatus has differentiated to the
species level from the ancestral fly on A. watsoni. Assuming that the coevolution in host

270 Elisabeth K. V. Kalko and C. O. Handley, Jr.

and parasite proceeds at a sımilar rate this evidence also supports the specific status of A.
incomitatus. |

Fourth, long-term studies of A. watsoni ın Isla Barro Colorado Island (BCI), a field
station of the Smithsonian Tropical Research Institute, Panama, suggest that this bat, and
presumably also A. incomitatus, have very small home ranges, thus further enforcing the
ısolation of mainland and island populations. On this small (1500 ha, 5x 5 km) island in
Gatun Lake (Panama Canal), 40 marked A. watsoni were recaptured up to three times
during a five year interval. Most individuals (36) were recaptured within 1 km of their
mark sites, more than 50% of them (22) at the original mark site. The greatest recorded
movement was only 3 km. The only movement over water probably was no more than
200 m. Furthermore, the wing morphology of A. incomitatus and A. watsoni makes it
highly unlikely that either could commute between the mainland and Escudo. Both have
rather short, broad wings typical of most stenodermatine bats. This wing shape adapts
them for slow, maneuverable flight in and around vegetation but constrains them from
sustained flight over long distances (NORBERG and RAyNER 1987). Moreover, the frequent
storms around Escudo would further limit movement between mainland and island.

Evolution on Escudo

The isolation of Escudo limits immigration and facılitates evolution. Escudo is more
isolated than any other island of Bocas del Toro. On it A. incomitatus has differentiated to
the species level. On the other islands, which are much closer to the mainland, populations
of A. watsoni are barely distinguishable from those of the mainland.

Seven of the nine species of native mammals of Escudo are differentiating from their
mainland counterparts. Artibeus incomitatus, Glossophaga soricina, Micronycteris
megalotis, and Hoplomys gymnurus are larger; Caluromys derbianus, Carollia brevicanda,
and Bradypus varıegatus are smaller. Only the aerial insectivorous bats, Saccopterys leptura
and Myotıs riparins, seem to be morphologically unchanged. Some of the birds of Escudo
clearly are isolated and have undergone morphological differentiation. The hummingbird
(Amaziha handleyi), manakin (Manacus amitinus), and wren (Thryothorus nigricapıllus)
are strikingly larger and more colourful than their mainland congeners, near the species
level of differentiation. The blue tanager (Thranpis episcopis) ıs less differentiated and may
be a later immigrant, or may be evolving more slowly. S. L. Orson (pers. comm.) has
found the rail, kingfisher, pigeon, parrot, flycatchers, and warbler of Escudo to be little ıt
any differentiated. Among the resident land birds, at least the pigeon and parrot may fly
periodically to the mainland.

Escudo has existed at most only about 9000 years, but once established, the channel
separating it from the mainland widened rapıdly and the sedentary nature of the Artibeus
insured its quick isolation from parental populations on the receding mainland. Although
9000 years appears to be little time for speciation there are other examples in mammals.
BERGSTROM and HorFMmAN (1991) have inferred for example that the most recent cycle of
differentiation and speciation in chıpmunks (Tamias) in the montane islands in the Great
Basin and southern Rocky Mountains of the southwestern United States has occurred ın
the 10,000 years since the Pleistocene (see also PATTERSON 1982).

Factors leading to evolution and extinction on Escudo

We assume that all of the mammals of Escudo may be survivors of selective extinction from
a species-rich mainland fauna following the fragmentation of Escudo. The inner islands,
closer to the mainland, harbor more species. With increasing distance from the mainland
the number of species declines. Extinction is rapid and extensive on small ıslands, leading

A new species of fruit-eating bat from Panama ZU

soon to diıminished resource and habıtat diversity and consequently to reduced faunal
diversity.

Compared with their mainland counterparts, island populations of anımals often
undergo significant changes in body size. Smaller anımals tend to increase in size whereas
larger anımals tend to decrease ın size (e.g. Case 1978; HEAnEy 1978; LawLor 1982;
Lomouino 1985). On the Bocas ıslands we have found both phenomena — dwarfism and
gigantism. For example the agouti (Dasyprocta), long-nosed armadillo (Dasypus), and
sloths (Bradypus and Choloepus) on the ıslands are smaller than their mainland congeners
whereas some rodents (e.g., Tylomys) are larger (HANDLEY, unpubl. results). A number of
hypotheses, including resource and habıtat limitation, interspecifice competition, and
predator pressure, have been put forward to explain this pattern (for summaries see as
examples ANGERBJÖRN 1985 and LomoLino 1985).

The factors we regard as most significant in this process on Escudo are reduction ın
habitat diversity and the resulting resource limitation. At first glance resource limitation
would seem to contradict our observation that A. incomitatus ıs sıgnificantly larger, and
has a much higher population density than its mainland counterpart, A. watsoni. However,
whereas A. incomitatus ıs the only stenodermatine on Escudo, on the mainland and on all
other islands of Bocas del Toro, A. watsoni and other small stenodermatines compete for
fruit with larger species, often in graded series of sıze classes. The decrease in abundance
and diversity of fruiting plants on islands ıs a particularly serious problem for frugivorous
bats which live on a very tight energy budget and need a constant supply of fruits
(HANDLEY et al. 1991; MoRRISoN 1980). First observations on Escudo indicate that there is
insufficient food for larger bats. For example, there are few fıgs of any kind on Escudo,
and Ficus insipida seems not to occur there. Throughout Panama thıs species is a major
food source for large numbers of fruit-eating bats ranging in mass from 8-75 8 (e.g.
Bonaccorso 1979; HANDLEY et al. 1991). We assume that release from interspecific
competition with other frugivorous stenodermatines might have freed the evolving A.
incomitatus to occupy a wider nıche and thus to become bigger and more abundant. A
larger body sıze would adapt it to utilize the full (although limited) range of fruit sızes
available on Escudo. A. incomitatus may have already reached its potential maximum body
size. Its sıze ratio of 1.17 to ıts mainland relative, A. watsoni, ıs comparable to means of
ratios of large suits of pairs of ısland and mainland species compiled by LomoLino (1985).

Another factor pressing for larger size in A. incomitatus might be the reduced and
seasonally varıable abundance of fruit on Escudo. LinDstEDT and Boyck (1985) have
argued that seasonality selects for larger body size, which enhances survivorship through
increased fasting endurance. Larger bats tolerate hunger better than smaller bats do.
Although frugivorous neotropical bats typically have little body fat (McNag 1976) we
found substantial amounts of fat in A. incomitatus but little or none in A. watsoni collected
at the same season.

Wırson (1991) found sımilar trends in size among small mammals on the Tres Marias
islands of Mexico. These ıslands are further from the mainland and much larger than
Escudo. As on Escudo, only one stenodermatine is present (A. intermedius koopmanı).
However, Artibens intermedius already is one of the largest stenodermatines, yet the island
bat ıs still larger than mainland A. intermedins. Furthermore, Wırson (1991) tound higher
diversity in insectivorous vespertilionid bats than in the mainly frugivorous and/or
nectarivorous phyllostomids. He postulated that their lower diversity might reflect
unpredictability of supplies of fruit and flowers.

Acknowledgements

More than to any other person we are indebted to LıGıa PAGET of Bocas del Toro for her unfailing
support and friendship. We are grateful to FRancıs M. GREENWELL and PEnny NErLson who helped

272 Elisabeth K. V. Kalko and C. O. Handley, Jr.

to collect and to prepare the specimens of A. incomitatus. The Instituto Nacional de Recursos
Naturales Renovables of Panamä (INRENARE), notably ROBERTO ARRANGO, provided, collecting
and export permits. MERCEDES ARROYO, Isıs IvaAncIc, GLORIA MaAGGIoRTL, and MArfa MoRELLO of
the Smithsonian Tropical Research Institute (STRI) expedited administrative and logistical details. We
are most grateful t0 ARLOINE HooD and KErrRY THOMSON who processed the specimens in the
museum; MERRILL VARN for her reconstructions of prehistoric shorelines; RALPH CHapman for his
invaluable generosity of sharing his expertise in statistics and morphometrics with us; and DARELYN
HaAnDLEY who prepared the table of measurements and edited the manuscript. The Escudo expedi-
tions were financed by grants from the National Science Foundation and the National Museum of
Natural History (NELSON Fund and Reseaerch Opportunities Fund), the Smithsonian Office of
Products Development and Licensing, and a gift from James and PaurA NErson of Hillsboro,
Oregon. E. K. V. Karko was suported by a NATO postdoctoral fellowship. Transporation was
furnished by the U.S. Army (Maj. VERNON J. Tırron), STRI (JOHN CHRrIsTY, EGBERT LEIGH and
Davıp West), United Brands, Changuinola Division (CAMERON FORSYTHE, General Manager), and
Asocıacıön Nacional para la Conservaciön de la Naturaleza, Panama (ANCON).

Zusammenfassung

Evolution, Biogeographie und Beschreibung einer neuen, fruchtfressenden Fledermausart der Gattung
Artibeus Leach (1821) aus Panama

Wir beschreiben und benennen eine neue, fruchtfressende Fledermausart der Gattung Artibeus aus
den Neotropen. Die neue Art ist ein Lokalendemit und kommt nur auf der Insel Escudo de Veraguas,
Panama, vor. Escudo de Veraguas liegt in der Provinz Bocas del Toro in Nordwest-Panamä, ungefähr
18 km von der karibischen Küste entfernt. Aufgrund einer Vielzahl gemeinsamer Merkmale nehmen
wir an, daß sich die neue Art aus einer Artibeus watsoni-nahestehenden Fledermausart entwickelt hat.
A. watsoni ist auf dem angrenzenden Festland und auf den anderen Inseln in der Provinz Bocas del
Toro weitverbreitet und häufig. Im Vergleich zu A. watsoni wiegt die neue Art ca. 15% mehr, die
Körperproportionen sind um ca. 10% und die Schädeldimensionen um ca. 6% größer. Zudem
unterscheidet sich die neue Art ın einer Anzahl diskreter morphologischer Merkmale, insbesondere in
der Bezahnung von A. watsoni. Wir diskutieren Aspekte zur Evolution und Biogeographie von A.
incomitatus und leiten daraus Mechanismen ab, die möglicherweise zur Artbildung dieser neuen
Fledermausart beigetragen haben.

Literature

ANDERSEN, K. (1908): A monograph of the chiropteran genera Uroderma, Enchisthenes, and Artibeus.
Proc. Zool. Soc. London, 204-319.

ANGERBJÖRN, A. (1985): The evolution of body size in mammals on islands: some comments. Am.
Nat. 125, 304-309.

BARLETT, A. $.; BARGHOORN, E. $. (1973): Phytogeographic history of the Isthmus of Panama during
the past 12.000 years (a history of vegetation, climate, and sea-level change). In: Vegetation and
vegetational history of northern Latin America. Ed. by A. GraHam. Amsterdam: Elsevier Scıe.
Publ. Co.

BERGSTROM, B. J.- Horrmann, R. $. (1991): Distribution and diagnosis of three species of chipmunks
(Tamias) in the Front Range of Colorado. The Southwestern Naturalist 36, 14-28.

Bonaccorso, F. ]J. (1979): Foraging and reproductive ecology ın a Panamanıan bat community. Bull.
Florida State Mus. Biol. Scı. 24, 359-408.

Case, T. J. (1987): A general explanation for insular body size trends in terrestrial vertebrates. Ecol.
59, 1-18.

CHAMPMAN, R. E.; GALTON, P. M.; SEPKosk1, ]. J. Jr.; War, W. P. (1981): A morphomertric study of
the cranıum of the pachycephalosaurid dinosaur Stegoceras. J. Paleontology 55, 608-618.

Davıs, J. C. (1986): Statistics and data analysıs in geology. 2nd ed. New York: John Wiley.

Haıı, E. R. (1981): The mammals of North America. 2nd ed. New York: John Wiley.

Hanntey, C. O., Jr. (1959a): A review of the genus Hoplomys (Thick-spined rats), with description of
a new form from Isla Escudo de Veraguas, Panama. Smithsonian Miscell. Coll. 139, 1-10.

— (1959b): A revision of American bats of the genera Euderma and Plecotus. Proc. United States
Nat. Mus. 110, 95-246.

— (1987): New species of mammals from northern South America: Fruit-eating bats, genus Artzbeus
Leach. In: Studies in Neotropical mammology: Essays in honor of Philip Hershkovitz. Ed. by B.
D. PATTERson and R. M. Tımm. Fieldiana: Zoology 39, 163-172.

— (1988): Specimen preparation. In: Behavioral and ecological methods for the study of bats. Ed. by
T. H. Kunz. Washington: Smithsonıian Institution Press. Pp. 437457.

HanDLey, ©. O., Jr.; Wırson, D. E.; GARDNER, A.L. (1991): Demography and natural history of the

A new species of fruit-eating bat from Panama 278

common fruit bat, Artibeus jamaicensis, on Barro Colorado Island, Panama. Smithsonian Con-
tributions to Zoology 511, 147-149.

HEANnNEY, L. R. (1978): Island area and body size of ınsular mammals: evidence from the tri-colored
squirrel (Callosciurus prevosti) of Southeast Asıa. Evolution 32, 29-44.

LAwLoR, T. E. (1982): The evolution of body size in mammals: evidence from insular populations in
Mexico. Am. Nat. 119, 54-72.

LinDSTEDT, $. L.; Bovce, M. S. (1985): Seasonality, fasting endurance, and body size in mammals.
Am. Nat. 125, 873-878.

Lomorıno, M. V. (1985): Body size of mammals on islands: the island rule reexamined. Am. Nat.
125, 873-878.

McNas, B. K. (1976): Seasonal fat reserves of bats in two tropical environments. Ecol. 57, 332-338.

Morrison, D. W. (1980): Efficiency of food utilization by fruit bats. Oecol. 45 (1), 270-273.

NORBERG, U. M.; Rayner, J. M. (1987): Ecological morphology and flight in bats (Mammalıa:
Chiroptera): wing adaptations, flight performance, foraging strategy and echolocation. Phil.
Trans. R. Soc. Lond. B 316, 335-427.

PACHECco, V.; PATTERSoN, B. D. (1992): Systematics and biogeographic analysıs of four species of
Sturnira (Chiroptera: Phyllostomidae), with emphasis on peruvian forms. Mem. del Mus. de Hist.
Nat., U. N.M. S. M. (Lima) 21, 57-81.

PATTERSoN, B. (1982): Pleistocene vicariance, montane islands, and the evolutionary divergence of
some chipmunks (genus Eutamias). J. Mammalogy 63, 387-398.

Rınpcway, R. (1912): Color standards and color nomenclature. Washington: Robert Ridgway.

WENZEL, R. L. (1976): The streblid batflies of Venezuela (Diptera: Streblidae). Brigham Young Univ.
Scı. Bull. Biol. Ser. 20, 1-2, 1-177.

WENZEL, R. L.; TıprTon, V. J.; Kurwkuıcz, A. (1966): The streblid batflies of Panamä (Diptera
Calyptera: Streblidae). In: Ectoparasites of Panama. Ed. by R. L. WENZEL and V. ]J. Tırron.
Chicago: Field Museum Nat.

WILKINSoN, L.; HırL, M.; WELNA, J. P.; BIRKENBEUEL, G. K. (1992) SYSTAT for Windows:
Statistics, version 5 ed. Evanston, IL: Systat Inc.

Wırson, D. E. (1991): Mammals of the Tres Marias Islands. Bull. Am. Mus. Nat. Hist. 206, 214-250.

Authors’ addresses: Dr. ELisABETH K. V. Karko and Dr. CHARLES OÖ. HANDLEY, Jr., Division of
Mammals, National Museum of Natural History, MRC-NHB 108, Smithsoni-
nan Institution, Washington, DC 20560, USA, and Smithsonian Tropical
Research Institute, P. ©. Box 2072, Balboa, Republic of Panama

Z. Säugetierkunde 59 (1994) 274-281
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Contribution to the knowledge of the bat fauna of Bioko island,
Equatorial Guinea (Central Africa)

By J. Juste B. and C. IBANEZ

Estaciön Biologica de Donana, Consejo Superior de Investigaciones Cientificas, Sevilla, Spain

Receipt of Ms. 16.3.1994
Acceptance of Ms. 2.6.1994

Abstract

Although having been long studied, the bat fauna of Bioko island (formerly Fernando Poo, Equatorial
Guinea, Central Africa), ıs still little known. The species Hipposideros commersoni, Glauconycteris
beatrix, Pipistrellus (P.) kuhlii, P. (N.) tenuipinnis, and P. (N.) cf. capensis are reported for the first
time. Furthermore, the species Myonycteris torquata, Taphozous mauritianus, Nycteris arge, Hip-
posideros cyclops, Glauconycteris poensıs, Mops (X.) spurrelli and M. (X.) thersites, previously reported
as doubtful, are confirmed on Bioko. These results increase the bat checklist for Bioko island by 25 %,
and it now includes 26 species.

Introduction

Bioko island (formerly Fernando Poo), ıs situated 32 km off the coast of Cameroon
(3° 48’ -3° 12’ N, 8° 25’ -8° 57' E), ın the middle of the Gulf of Guinea. Since ıt was a
commonly used starting point for many scientific expeditions to the African mainland,
many mammal species, including bats, were first described from Bioko specimens in the
19th century (e.g. Dendrohyrax dorsalis, Colobus satanas, Glanuconycteris poensis,
Rhinolophus landeri, etc.). Nevertheless, our understanding of the bat fauna of Bioko, is
still fragmentary. BasıLıo (1962), in a general view of the fauna of Equatorial Guinea
(a former Spanish colony), gave some data on Bioko’s bats. EISENTRAUT (1964, 1973)
summarized the bat fauna of Bioko after collecting on the island for some months. He
considered that up to 20 species were present (in 1973); although he doubted some (e.g.
Myonycteris torguata or Glauconycteris poensis) and stressed the need to confirm others.
Later on, IBANEZ and VALVERDE (1985) added Eptesicus platyops (as a probable synonym of
E. serotinus) to the bat list of Bioko.

Material and methods

From 1988 to 1991, a systematic sampling was carried out by the senior author throughout the island.
Bats were caught by netting and visiting possible roosting places as part of a wider study of the bat
fauna of the Gulf of Guinea Islands. Collected specimens were deposited in the Estacıön Biolögica de
Donana (EBD) collections and were compared with material from the Museo Nacional de Ciencias
Naturales de Madrid (MNCN). Selected measurements of adult specimens are given in mm, together
with the number of specimens (brackets) and ranges (parentheses). Both sexes were summarized when
no significant dimorphism was found.

Abbreviations used are: FA = forearm length; GSL = greatest skull length; CBL = condylobasal
length; CCL = condylocanine length; ZW = zygomatic width; DCC = distance between canınes (from
outer side) and MW = mastoid width. Species are named according to CoRBET and Hırr (1986),
except for Vespertilionidae, which follow Hırı and Harrıson] (1987), and Molossidae, which follow
Koopman (1993).

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5905-0274 $ 02.50/0

Contribution to the knowledge of the bat fauna of Bioko island 273
Results

As a result of this research, five new species are recorded and another seven confırmed. The
list of bat species known to occur in Bioko comprises now 26 species.

Pteropodidae
Myonycteris torguata (Dobson, 1878)

Material [1]: Eolä river, Patio Vivancos, Baney: 1 subadult 9.

Selected measurements: FA 58.0; GSL 31.0; CBL 29.8; ZW 18.9.

Remarks: The only known specimen was reported by KRUMBIEGEL (1942) without
locality. EisENTRAUT (1964) did not catch any specimens on his expeditions and even
doubted its presence. In his Myonycteris revision, BERGMANS (1976) studied only the skin
of KRUMBIEGEL’s specimen and raised the possibility that the bat belongs to M. brachy-
cephala, an endemic from nearby Säo Tome island (BocAGe 1889).

The Baney specimen confirms the presence of Myonycteris torquata on Bioko island. It
was netted in a secondary forest surrounded by agrıcultural crops and cocoa plantations.
Although still subadult, the skull clearly shows the typıcal M. torgquata shape, mainly ın its
wide post-dental palatum bone and weak “fascıa temporalis” ın the zygomatic arches.
Furthermore, the specimen does not share any of the characteristic dental features
described by AnDERSEN (1912a) for M. brachycephala. Myonycteris torquata has not been
found again, in spite of intense netting efforts. It can be considered the most rare fruit bat
on Bioko island. Moreover, the fact that the species is subadult and was captured in the
western point closest to the mainland would support the possibility of M. torgnata being
represented on Bioko only by vagrant or migrant individuals that sporadically reach the
island.

Emballonuridae
Taphozous mauritianus Geotfroy, 1818

Material [13]: All obtained in Malabo: 6 adult dd, 2 subadult 8, 3 adult 22 and 2
subadult ?2.

Selected measurements [9]: FA mean 61.2 (58.7-65.0); GSL mean 22.1 (21.8-22.5); CCL
mean 20.6 (20.2-21.2); ZW mean 12.9 (12.4-13.4).

Remarks: Taphozous manritianus was apparently first cıted from Bioko as early as 1876 by
PETERS (EISENTRAUT 1964). Another specimen is mentioned by BasıLıo (1962), but
EISENTRAUT did not find ıt. The presence of T. manritianus ıs confırmed by our series. The
bats were caught while roosting in coconut palms in Malabo city. Therefore, the species
probably occupies most of the coastal coconut tree plantations on the island. The
confirmation of this widespread Afrıcan species was expected since it has been recently
found on other ıslands in the Gulf of Guinea (JusTE and IBANEZ 1993).

Nycteridae
Nycteris arge Thomas, 1903

Material [42]: Malabo-Rıiaba road, km 3: 1 adult d. Malabo-Rıaba road, km 10: 1 adult &
and 2 adult 22. Malabo-Riaba road, km 61: 6 adult dd and 7 adult 22. Malabo-Rıaba
road, km 62: 2 adult dd and 3 adult 22. Bantabare, Riaba: 1 adult 2. Malabo-Luba road,
km 3: 1 adult d and 3 adult ??2. Malabo-Luba road, km 13:1 adult d and 1 adult 2.
Malabo-Luba road, km 15: 1 adult 2. Elgorriaga, Malabo-Luba road, km 19: 1 adult 4.

276 J. Juste B. and C. Ibanez

Malabo-Luba road, km 40: 2 adult dd. Malabo-Luba road, km 42: 2 adult d& and 2 adult
22. Batete, Luba: 1 adult d. Luba: 4 adult dd.

Selected measurements: FA [42] mean 42.1 (36.448); GSL [24] mean 19.5 (18. 2-21.1);
CCL [23] mean 16.9 (15.9-18.4); ZW [24] mean 11.4 (10.2-12.6); DCC [24] mean 4.7
(4.3-5.5).

Remarks: The only known specimen was obtained at the beginning of the century on the
southwestern coast (Bantabare) and recorded by Anpersen (1912b) and considered as N.
hispida (EISENTRAUT 1973). Our series confirms the presence of N. arge on Bioko island as
a common species in both forest and cocoa plantations all over the island. It has been found
inside tree holes and drains. The wide varıation in the measurements of the population is
outstandig; even so, our figures fıt well within the ranges given for the species (KooPMAN
1975; VAN CAKENBERGHE and DE VREE 1985).

Hipposideridae
Hipposideros commersoni (Geoffroy, 1813)

Material [22]: Ericorico river, Malabo-Rıaba road, km 3: 2 adult dd, 1 subadult &.
Grande rıver, Rıaba: 1 adult 2. Timbabe river, Malabo-Luba road, km 3: 1 adult d.
Basupu, Malabo-Luba road, km 14: 1 adult ?. Bosao river, Malabo-Luba road, km 20:
3 adult SS and 3 adult ??. Apü rıver, Malabo-Luba road, km 21: 1 subadult & and 1
subadult ?. Eolä river, Patio Vivancos, Baney: 3 adult dd, 1 subadult d, and 2 subadult
22. Borabuope river, Mallo plantation, Malabo: 2 adult SG.

Selected measurements: Males: FA [12] mean 104.0 (102.0-105.8); GSL [11] mean 38.7
(36.1-39.2); CBL [11] mean 33.5 (31.3-34.7); DCC [111] mean 11.2 (10.5-11.8); ZW 11]
mean 22.0 (21.4-23.2). Females FA [6] mean 99.0 (96.4-100.4); GSL [4] mean 37.2
(36.5-38.6); CBL [4] mean 32.4 (31.7-33.1); DCC [4] mean 10.3 (9.8-11.1); ZW [4] mean
20.6 (19.9-21.8).

Remarks: Hipposideros commersoni was fırst cited for Bioko by CABRERA (1912) by means
of a purchased male specimen. It was not considered by EISENTRAUT (1964, 1973) and
Hayman and Hırr (1971) but was mentioned again in a recent inventory of the bat
collection of the Museo de Ciencias Naturales in Madrıd (IBANEZ and FERNANDEZ 1989).
The specimen (MNCN N?°128) has been re-studied, and turned out to be an adult
Hipposideros male, but noticeably smaller (FA 84; GSL 30.1; CBL 26.3; DCC 7.0; ZW
16.2) than specimens of our H. commersoni series. The body fur (in alcohol) is yellowish
orange and lacks the A. commersonı’s typical dark spots on the shoulders. The MNCN
N°128’s skull, ıs clearly weaker than a typical A. commerson:’s, without any crest and
showing duller edges of the rostrum. We now conclude that the specimen MNCN N?°128
does not belong to H. commersoni; rather, it ıs a mislabelled Aipposideros, probably
belonging to the Asıan ‘diadema’ group of Hııı (1963).

Therefore, our series represents the first real A. commersoni specimens from Bioko.
Their measurements fit will into the range of H. commersoni gigas (ROSEVEAR 1965). They
were netted across rivers in both cocoa plantations and rain forest and no shelters were
found.

Hipposideros cyclops (Temminck, 1853)

Material [18]: Bioko (unknown locality): 1 adult 2. Malabo-Rıaba road, km 2: 6 adult 22.
Ericorico river, Malabo-Riaba road, km 3: 1 adult 2. Vda. Mera plantation, Malabo-Luba
road, km 5: 1 adult d and 4 adult ?2. Luba: 2 adult dd and 1 adult 2. Eola river, Patıo
Vivancos, Baney: 1 adult 2. Borabuope river, Mallo plantation, Malabo: 1 adult S

Selected measurements: Males: FA [5] mean 67.2 (66.0-68.3); GSL [2] mean 29.3; CCL [2]
mean 25.9 (25.7-26.0); DCC [2] mean 8.2 (8.1-8.3); MW [2] mean 12.2 (12.1-12.4); ZW

Contribution to the knowledge of the bat fauna of Bioko island U

[2] mean 16.2 (16.1-16.3). Females: FA [13] mean 70.1 (69.0-72.5); GSL [9] mean 29.4
(28.4-30.5); CCL [9] mean 26.0 (25.1-27.0); DCC [9] mean 8.1 (7.7-8.7); MW [9] mean
12.6 (12.3-12.9); ZW [9] mean 16.5 (16.0-17.0).

Remarks: EisENTRAUT (1964, 1973) included this species in the bat fauna of Bioko based on
a photograph by BasıLıo (1962), but without having seen any specimens. Our series
confirms the presence of Hıipposideros cyclops on Bioko as a quite common bat, and as a
typical dweller of trunk holes in both forest and cocoa plantations throughout the island.

Vespertilionidae
Glauconycteris poensis (Gray, 1842)

Material [20]: Ericorico rıver, Malabo-Riaba road, km 3: 1 d and 1 2. Vda. Mera
plantation, Malabo-Luba road, km 5: 1 ?. Bosao rıver, Malabo-Luba road, km 20: 1 2.
Apü rıver, Malabo-Luba road, km 21: 2 22. Oprocage farm, Moka, Luba: 2 dd and 12.
Matadero river, Malabo: 1 ?. Borabuope river, Mallo plantation, Malabo: 1 d and 7 92.
Ela Nguema, Malabo: 1 ?. BasileE peak road, km 1:1.

Selected measurements: Males: FA [5] mean 38.5 (37.5-39.6); GSL [1] 13.0; CCL [1] 12.9;
ZW [1] 9.7; MW [1] 8.5; DCC [1] 4.8. Females: FA [15] mean 39.3, (36.2—41.3); GSL [8]
mean 13.0 (12.7-13.5); CCL [8] mean 12.8 (12.5-13.2); ZW [6] mean 9.5 (9.3-9.9); MW
[8] mean 8.3 (7.8-8.5); DCC [8] mean 4.7 (4.5—4.9).

Remarks: The species was named G. poensis because Bioko island (formerly Fernando
Poo) was thought, by mistake, to be the type localıty instead of Abo, lower Niger (AELLEN
1952). EISENTRAUT (1964) mentioned only a young Glauconycteris tentatively ıdentified as
a G. poensis, but questioned its presence on Bioko. All the captured specimens show a
tawny yellowish fur with trıcoloured hairs, typical white flank-stripes and shoulder-spots.
The FA measurements fit well within the range of the G. poensis given by ROSEVEAR
(1965), the skulls from Bioko being a little bigger. Therefore, Glauconycteris poensıs ıs
confirmed on Bioko island, where ıt has been netted up to 1300 m a.s.l. (Moka) and where
it ıs apparently common, including within cocoa plantations.

Glauconycteris beatrix Thomas, 1901

Material [4]: Basupu, km 14 Malabo-Luba road: 1 2. Matadero river, Malabo: 1 adult 2.
Borabuope river, Mallo plantation , Malabo: 2 adult 22.

Selected measurement: FA mean 38.6 (36.4-40.2).

Remarks: This is the first record of Glanconycteris beatrix from Bioko island, known
already from Cameroon (AELLEN 1952). The specimens differ from the former species in
that they have differently shaped ears and tragus. The fur colour is also notably darker and
the white flank-stripes are absent, although one specimen shows small white shoulder-
spots (EBD 20503). Glauconycteris beatrix has been netted in the same habıtats as G.
poensis in both cocoa plantations and forests, but it is much rarer.

Pipistrellus (Pipistrellus) kuhlii (Natterer, 1817)

Material [4]: Lake Biao, Moka: 3 adult dd and 1 adult 2.

Selected measurements: FA [4] mean 35.5 (34.4-37); GSL [4] mean 13.4 (13.0-13.8); CCL
[3] mean 12.4 (12.1-12.8); MW [4] mean 7.3 (7.2-7.4); DCC [4] 4.3 (4.1—4.6).

Remarks: All four specimens show the distinctive conic shape of the upper inner incisors
and the upper premolar, clearly noticeable from the outside. No white along the wing
membrane border, or along the uropatagium is noted, and the measurements match those
given for Cameroon specimens (Hırr 1968).

278 J. Juste B. and C. Ibanez

This newly recorded species from Bioko island is apparently montane since it has been
collected only above 1300 m. Pıpistrellus kuhlii shows a continuous distribution along a
coastal fringe from northern Africa southward through the eastern coast and reaching as far
as Cape Province (Hırr 1968) P. kuhli is scattered throughout West-Africa. It has been
recorded from the Canary Islands (TrujıLLo 1991); La’youn (El Aaiün), Western Sahara
(IBANEZ and FERNÄNDEZ 1989); the Cape Verde Islands (AzzaroLı PuccETTi and ZAva
1988); and from some mountain areas of western tropical Africa, such as Mount Nimba
(Hırr 1982) and Mount Cameroon (Hırr 1968).

Pipistrellus (Neoromicia) tennipinnis (Peters, 1872)

Material [4]: Bosao river, Malabo-Luba road, km 18: 1 adult 2. Borabuope& rıver, Mallo
plantation, Malabo: 3 adult 22.

Selected measurements: FA [4] mean 31.6 (31.0-32.0); MW [1] 7.0; ZW [1] 8.0; DCC [1]
ia

Remarks: P. tennipinnis ıs wıdespread ın western and central Afrıca (RosEvEAR 1965) and
it is well known on nearby Mount Cameroon (AELLEN 1952). The specimens collected
show typically whitish and translucent wings. They represent the first record of P.
tenuipinnis trom Bioko, where they have been netted mainly in cocoa plantations.

Pipistrellus (Neoromicia) cf. capensis (Smith, 1829)

Material [18]: Ericorico rıver, Malabo-Riaba road, km 3: 1 adult d, 1 adult 2. Grande
rıver, Riaba: 1 adult 2. Vda. Mera plantation, Malabo-Luba road, km 3: 2 adult 22.
Sampaca, Malabo-Luba road, km 7: 2 adult ?2. Bosao rıver, Malabo-Luba road, km 20: 1
adult ?. Musola rıver, Luba: 2 adult ??. Eola river, Patio Vivancos, Baney: 1 adult 2.
Borabuope rıver, Mallo plantation, Malabo: 1 adult d and 5 adult 2. Basıle village,
Malabo: 1 adult d.

Selected measurements: Males [3]: FA mean 34.8 (34.5-35.0); GSL mean 13.9 (13.8-14.0);
CBL mean 12.9 (12.8-13.0); DCC mean 4.3 (4.2—4.4). Females: FA [15] mean 35.4
(34.0-37.0); GSL [13] mean 13.9 (13.6-14.4); CBL [13] mean 13.1 (12.7-13.6); DCC [13]
mean 4.4 (4.2-4.6); ZW [6] mean 9.0 (8.8-9.3).

Remarks: All Bioko specimens have dull brown fur. Their average measurements are
slightly larger than the values given by Koopman (1975) for P. capensis, but agree with
those given by RosEvEAR (1965). The systematics of this group remain very entangled,
especially regarding the West African forms. We therefore consider our specimens at
present as Pipistrellus (N.) cf. capensis, which is the first record of the species from Bioko
island.

Molossidae
Mops (Xiphonycteris) spurrelli (Dollman, 1911)

Material [32]: Ericorico river, Malabo-Riaba road, km 3: 3 adult dd and 17 adult 22.
Grande river, Riaba: 2 adult d&. Timbabe river, Malabo-Luba road, km 3: 1 adult ?. Apu
rıver, Malabo-Luba road, km 21: 1 adult d and 7 adult 22.

Selected measurements: Males: FA [6] mean 28.7 (28.0-29.7); GSL [2] mean 15.9
(15.9-16.0); CBL [2] mean 14.7 (14.7-14.8); CCL [2] 14.5; DCC [2] mean 4.6 (4.54.8);
ZW [2] mean 10.2 (10.0-10.5) MW [2] mean 9.5 (9.4-9.6). Females: FA [26] mean 28.2
(27.0-29.6); GSL [12] mean 15.4 (15.1-15.8); CBL [12] mean 13.9 (13.6-14.3); CCL [12]
13.7 (13.3-14.1); DCC [12] mean 3.9 (3.5—4.1); ZW [12] mean 9.6 (9.3-10.0); MW [12]
mean 9.3 (9.1-9.3).

Remarks: This species had been known from Bioko by one specimen from Banapä

Contribution to the knowledge of the bat fauna of Bioko island 27

(BasıLıo 1962) and another specimen without specific locality (Kock 1969). Its presence
on the island was, therefore, questioned (Hayman and Hırr 1971; EisEntrAUT 1973). A
new series confirms the presence of Mops spurrelli on the island; surprisingly, it seems to
be the most common molossid. It was captured along most of the rıvers suitable for the
species on the island.

Mops (Xiphonycteris) thersites (Ihomas, 1903)

Material [24]: Grande river, Riaba: 2 adult dd, 9 adult ?? and 1 subadult ?. Ericorico
river, Malabo-Rıiaba road, km 3: 4 adult ??. Apü river, Malabo-Luba road, km 21:
1 adult ?. Lopesi rıver, Malabo: 2 adult dd and 4 adult ??. Matadero river, Malabo:
I ackulle 2;
Selected measurements: Males [4]: FA mean 38.7 (38.0-39.5); GSL mean 20.3 (18.9-21.3);
CBL mean 18.1 (17.6-18.5); CCL mean 17.4 (16.9-17.7); DCC mean 5.6 (5.4-6.0); MW
mean 11.3 (10.9-11.8); ZW mean 12.5 (11.9-12.8). Females: FA [19] mean 38.6
(372.049.2); GSL [18] mean 19.1 (18.5-20.0); CBL [13] mean 17.4 (16.7-17.9); CCL [18]
mean 16.8 (16.3-17.2); DCC [18] mean 5.2 (4.9-5.4); MW [18] mean 11.0 (10.6-11.3);
ZW [18] mean 11.9 (11.7-12.2).
Remarks: There has been some confusion about the middle-sized molossids of Bioko
island. A single specimen from Bantabare was first ıdentified as Chaerephon pumila
(Dogson 1878) and subsequently as Mops leonis (Tmomas 1908), at present M. brachy-
pterus. Consequently, both species were long accepted as inhabitants of the island, (e.g.
Hayman and Hırr 1971). EISENTRAUT (1964) mentioned both species from the ısland but
later considered that the specimen may actually represent Mops thersites (EISENTRAUT
1973), although the record ıs still accepted as Chaerephon pumila elsewhere (KooPMAN
ID).

The new series confirms the presence of Mops thersites on the island as acommon bat, at
least in the lowland zones and cities.

Discussion

Among Bioko’s previously known bat species, Eptesicus platyops, Chaerephon pumila,
Mops brachypterus and Nycteris hıspida have not been captured in spite of our intense
netting efforts. The absence of C. pumila and M. brachypterus seems to confırm EIsEn-
TRAUT’sS (1973) statement of a systematic mistake. We agree that these last two species
should not be included in the checklist of Bioko. Finally, only three Nycteris hispida
specimens are known from Bioko, all of them from the 19th century. We have checked the
identity of one of them (MNCN N?° 76) and it apparently belongs to N. hispida, although
its tragus is not semilunate and only one incisor can be considered trifid because of wear.
The remaining two specimens have recently been studied by VAN CAKENBERGHE and DE
VREE (1993).

The newly found bat species, and the confirmed ones, strengthen the resemblance of
Bioko’s bat fauna to that of the Mount Cameroon zone. Almost all of these bats have
previously been recorded there. Finally, the absence of endemism in the bat fauna of Bioko
is apparently confirmed. This ıs in contrast to other mammal groups like such as primates,
which reach up to 70% ot endemism at a subspecific level (Burynskı and KosTEr 1986).
Bioko, a typical landbridge island, was connected with the mainland relatively recently,
about 6,000 years ago (IHYS VAN DEN ANDENAUERDE 1967). This fact, coupled with the
high vagility of bats, is likely to have hampered any speciation process among the group on
Bioko.

280 J: Juste B. and C. Ibanez

Acknowledgements

We wish to thank ANTONIoO AyonG NGUEMA for his help, support, and friendship throughout so
many years of field work, and Mr. LEANDRO MBomfo Nsu£, former Minister of Culture of
Equatorial Guinea, for his encouragement of and confidence in our work. AGIE REEvEs, Anpy,
CAROLINE GREEN and CLAUDIA KELLER for their kind help with the English and German transla-
tions. The work was supported by the Oficına de Cooperaciön con Guinea Ecuatorial [at present,
Instituto de Cooperaciön para el Desarrollo] of the Spanish Ministerio de Asuntos Exteriores with the
cooperation of the Asociacıon de Amigos de Donana.

Zusammenfassung

Beitrag zur Kenntnis der Chiropterenfauna der Insel Bioko, Äquatorial-Guinea

Die Fauna der Insel Bioko (ehemals Fernando Poo, Äquatorial-Guinea) ist schon Objekt ver-
schiedener Studien gewesen, aber dennoch nicht vollständig bekannt, besonders was die Chiropteren-
fauna angeht. Die Fledermausarten Arpposideros commersoni, Glauconycteris beatrix, Pipistrellus
kuhlii, P. tenuipinnis und P. cf. capensis werden zum erstenmal für die Insel erwähnt. Das Vorkom-
men der Arten Myonycteris torquata, Taphozous maurıtianus, Nycteris arge, Hipposideros cyclops,
Glauconycteris poensis, Mops spurrelli und M. thersites, in früheren Veröffentlichungen als zweifelhaft
erwähnt, wird nun auf Bioko bestätigt. Mit diesen Resultaten wird die Liste der Fledermausarten für
die Insel Bioko um 25 % erweitert und zählt hiermit 26 Arten.

References

AELLEN, V. (1952): Contribution & l’etude des chiropteres du Cameroun. Me&m. Soc. Neuchät. Sci.
Nat. 8, 1-121.

ANDERSEN, K. (1912a): Catalogue of the Chiroptera in the collection of the British Museum. 1.
Megachiroptera. London: British Museum (NH).

— (1912b): Brief diagnoses of eight new Petalia, with a list of known forms of the genus. Ann. Mag.
Nat. Hist. (8) 10, 546-550.

AZZAROLI PucceTTI, M.L.; Zava, B. (1988): Nouvelles donnees sur les chiropteres des iles du Cap-
Vert. Boll. Mus. reg. Scı. nat. Torino 6, 603-615.

Basırıo, A. (1962): La vida anımal en la Guinea Espanola. Sec. Ed., Madrid: I.E.A. (C.S.1.C.).

BERGMANS, W. (1976): A revision of the African genus Myonycteris Matschie, 1899 (Mammalıa,
Megachiroptera). Beaufortia 24, 189-216.

BocaGe, J. V. (1889): Chiropteres de l’ile St. Thome. J. Sci. Math. Phys. Nat. 2, 197-199.

Burynsk1, T.; KosTEr, S. (1986): The status and conservation of forest and primates on Bioko Island
(Fernando Poo), Equatorial Guinea. WWF Uganda: Impenetrable Forest Conservation Project
Report.

A. (1912): Catälogo de los mamiferos del Museo Nacional de Ciencias Naturales. Trab.
Mus. Cienc. Nat., Madrid 11, 1-147.

CoRBET, G. B.; Hırı, J. E. (1986): A world list of mammalıan species. Sec. Ed., London: Brit. Mus.
(Nat. Hist.).

Dosson, G. E. (1878): Catalogue of the chiroptera in the collection of the British Museum. London.

EISENTRAUT, M. (1964): La faune de chiropteres de Fernando-Po. Mammalıa 28, 529-552.

— (1973): Die Wirbeltierfauna von Fernando Poo und Westkamerun. Bonner Zool. Monogr. 3,
1428.

Hayman, R. W.; Hırı, J. E. (1971): Order Chiroptera. In: The mammals of Africa. An identification
manual. Ed. by J. MEEsTErR and H. W. SETZER. Washington: Smithsonian Inst. Press. Pp. 1-73.

Hııı, J. E. (1963): A revision of the genus Hipposideros. Bull. Br. Mus. Nat. Hist. (Zool.) 11, 1-129.

— (1968): Bats from the Cameroons, with the description of a new species of Pipistrellus. Bonn. zool.
Beitr. 19, 43-48.

— (1982): Records of bats from Mount Nimba, Liberia. Mammalıa 46, 116-120.

Hırı, J. E.; Harrıson, D. L. (1987): The baculum in the Vespertilioninae (Chiroptera: Vesper-
tilionidae) with a systematic review, a synopsis of Pipistrellus and Eptesicus, and the descriptions
of anew genus and subgenus. Bull. Br. Mus. Nat. Hist. (Zool.) 52, 225-305.

IBANEZ, C.; FERNANDEZ, R. (1989): Catälogo de murcielagos de las colecciones del Museo Nacional de
Ciencias Naturales. Monografias, Mus. Nac. Cienc. Nat., Madrid 2, 1-54.

IBANEZ, C.; VALVERDE, J. A. (1985): Taxonomic status of Eptesicus platyops (Thomas, 1901)
(Chiroptera, Vespertilionidae). Z. Säugetierkunde 50, 241-242.

Juste B., J.; IBANEZ, C. (1993): A new Tadarida of the subgenus Chaerephon (Chiroptera: Molos-
sıdae) from Säo Tome Island, Gulf of Guinea (West Africa). J. Mammalogy 74, 901-911.

Contribution to the knowledge of the bat fauna of Bioko island 281

Kock, D. (1969): Über ein weibliches Exemplar der afrikanischen Bulldoggfledermaus, Xiphonycteris
spurrelli, von Fernando Poo. Senckerbergiana biol. 50, 9-13.

Koorman, K. F. (1975): Bats of the Sudan. Bull. Amer. Mus. Nat. Hist. 154, 355-443.

— (1993): Order Chiroptera. In: Mammals species of the world. A taxonomic and geographic
reference. 2nd ed. Ed. by D. E. Wırson and D. M. REEDER. Washington: Smithsonian Institution
Press. Pp. 147-241.

KRUMBIEGEL, I. (1942): Zur Kenntnis der Säugetierfauna von Fernando Poo. Archiv f. Naturge-
schichte 11, 305-349.

ROSsEVEAR, D. R. (1965): The bats of West Africa. London: Brit. Mus. (Nat. Hist.).

THomas, ©. (1908): New bats and rodents in the British Museum Collection. Ann. Mag. Nat. Hist. 2
(8), 370-375.

THyYs VAN DEN ÄNDENAUERDE, D. F. E. (1967): The freshwater fishes of Fernando Poo. Ver-
haudelingen der Koninklijke Nederlandse Akademie van Wettenschappen. Afdeeling Natuur-
kunde 29 (100), 1-167.

TruJrLLo, D. (1991): Murcielagos de las Islas Canarıas. Madrid: Colecciön Tecnica, ICONA.

VAN CAKENBERGUE, V.; DE VREE, F. (1985): Systematics of African Nycteris (Mammalıa: Chiropt-
era). In: African Vertebrates. Ed. by. K. SCHUCHMANN. Bonn: Zoolog. Forschungsinst. u. Mus.
A. Koenig. Pp. 53-90.

— (1993): Systematics of African Nycteris (Mammalıa: Chiroptera) part II. The Nycteris hispida
group. Bonn. zool. Beitr. 44, 299-332.

Authors’ address: Dr. JavIER JustE B. and Dr. CarLos IBANEZ, Estacıön Biolögica de Donana,
C.S.1.C., Aptdo. 1056, E-41080 Sevilla, Spain

Z. Säugetierkunde 59 (1994) 282-288
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Distribution of the Cabrera water shrew (Neomys anomalus)
in Northeastern Spain

By I. TorrRE and J. L. TELLA

Departamento de Biologia Animal (Vertebrados), Facultad de Biologia, Universidad de Barcelona,
Barcelona, Spain

Receipt of Ms. 20.4. 1993
Acceptance of Ms. 5. 2. 1994

Abstract

This study deals with the range of the Cabrera water shrew (Neomys anomalus) in the northeast
Iberian peninsula and the ecological factors that may determine its distribution. Information is
provided on 26 new sıtes, including the first records of the species on the southern slope of the
Pyrenees. In these mountains, where the water shrew (Neomys fodiens) lives, the Cabrera water shrew
(N. anomalus) was found at high altitudes and also below the range occupied by the former species. In
the remainder of the study area its distribution seems to be conditioned by the hydrographic network
characteristics, showing mediterranean preferences.

Introduction

The geographical distribution of the Cabrera water shrew in the Iberian peninsula has not
been thoroughly studied. Although a number of references suggest a widespread range
(Faus 1991), in the NE Iberian peninsula this species has been found only occasıonally,
perhaps due to its scarcity or low detectability through sampling methods (LöPEz-FUSTER
et al. 1992). Its distribution and ecological requirements have been supposed on the basis of
a small number of locations (GosALBEz 1987; JIMENEZ et al. 1989; Faus 1991), or
generalization from extensive data of this species over the entire range ot ıts European
distribution (SPITZENBERGER 1990). The aim of this study was to describe these aspects ın
greater detail, by analysing both information on new and previously known locations.

Material and methods

The study was carried out in the autonomous regions of Aragön, Catalonia and in the province of
Castellön. The altitude of the regions studied ranged from sea level to 3404 m above sea level. There
are sıx bioclimatic zones belonging to the Eurosiberian and Mediterranean regions (Rıvas-MARTINEZ
1983). Therefore, almost all the climatic and ecological characteristics of the Iberian peninsula can be
found in the study area. Figure 1 shows the principal geographic characteristics of the study area.

A review of the literature revealed 46 reports of the Cabrera water shrew from 24 locations
(MiLLerR 1912; Napa and Paraus 1967; GARZöN et al. 1971; Sans-ComA 1973; VERICAD and
MEyLAn 1973; GonZALEZ 1975; PELAYo 1979; GOSALBEZ et al. 1985; Rusıo 1985; ARRIZABALAGA et
al. 1986; GosALBEz 1987; JIMENEZ et al. 1989; SPITZENBERGER 1990; Faus 1991; LÖPEZ-FUSTER et al.
1992; TORRE et al. 1992).

Our prospections were centred on wide areas where the presence of this species was unknown.
The analysıs of barn owl (Tyto alba) pellets collected from 1985 to 1992 provided 15,486 small-
mammal preys from 33 different sites in the Ebro depression. Specimens were also captured at three
different sites: Alt Aneu (Pyrenees), by placing 100 mouse traps for four days (December 1987) along
the Noguera Pallaresa river and in alpine meadows; Hijar (Ebro depression), wıth the same number of
mouse traps for three days (February 1988) along the Martin riverside and some irrigation channels;
Ateca (on the Iberian range spurs), using 75 “pitfall” mouse traps at the Manubles riverside for 6 days
(March 1990). We also obtained information about the species from two nature magazines and by
interviewing some specialists working on small mammals ın the study area.

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5905-0282 $ 02.50/0

Distribution of the Cabrera water shrew in Northeastern Spain 283

Fig. 1. Orography, hydrographic basıns, location of the study area and sites where Neomys anomalus
were recorded (Black circles = new localities; white circles = localities cited in literature; lightly dotted
area = >500 m.a.s.l.; densely dotted area = > 1000 m.a.s.l.)

Those Cabrera water shrews trapped or found dead were identified by morphological characteris-
tics (CABRERA 1914; SPITZENBERGER 1990). The skulls found in the pellets were identified using the
discriminant formula given by BÜHLER (1964). The skulls obtained by collaborators have also been
determined by us. Reports of observations of live water shrews were only considered if the informant
was an experienced person.

From each location we recorded the altıtude above sea level, river flow, rainfall and annual average
temperature (FORTEZA 1985; LEON et al. 1987; M.O.P.U. 1988; BosQquE and Vıra 1992). We have
only treated the number of localities because there is a narrow relation between number of localities
and number of Neomys anomalus for every varıable considered: altitude (r = 0.94, p<0.001),
temperature (r = 0.97, p<0.001), rainfall (r = 0.95, p< 0.0001), and water flow (r = 0.95, p< 0.002).
Since every sampling method itself provides little information about the species (LÖPEZ-FUSTER et al.
1992) we treated all localities together. The results obtained may correspond to a biased sample, but
could be considered as an approximation of the real ranges occupied by this species.

Results and discussion

The presence of Neomys anomalus was recorded at 26 new sites and two previous ones. We
found 45 anımals, added to those already described ın the literature, totals 91 Cabrera
water shrews, belonging to 50 different sites in the NE of the Iberian peninsula. Figure 1
shows the situation of records of Neomys anomalus in the study area, and the table
provides details on records sites, altitude, associated rivers, and number as well as origin of
Neomys anomalus.

With regard to the altitude (Fig. 2A), the species was irregularly distributed between
180 and 1,850 m.a.s.l., but most of the records were made below 1,000 m (96 %), ın
contrast with its preference for highlands in central Europe (Fons et al. 1980; van LAAR
1983; TABERLET 1984). The species was also found at variable river flows (Fig. 2B), from
rivers with less than 2 m’/s to the Ebro river (261 m/s), suggesting that its presence was

I. Torre and J. L. Tella

284

("wos 'sıad) OAVTaAd "I
(7661) "Te 19 vaısnJ-ZadoT
("wo9 'sıad) ZaanyNaad ‘A
("wo9 'sıad) ZAanyNaad 'q
("wos 'sıad) ZAanyNaad 'q
("wo9 'sıad) ZAanyNaad 'A

("wos 'sıad) Tvaıny '[
("wo9 'sIod) ZIHONYS 'W '[

(6861) "Te 19 zanamwı[

("wo9 'sı9d) TIHIIOAAAI "A
(jgndun) Tvaınd pue z1ansung
("wo9 'sı9d) ZIHONYS 'W '[
(7661) Te 7 auNO],

("wo9 'sıod) ZAHONYS 'W '[
("wos 'sıad) vaTog
(gndun) ‘fe 19 zagTyso9
(‘jgndun) ‘fe 19 zaaTys09)
(gndun) ‘fe 19 zag1yso9
("wo9 'sıod) OnIg 'I '[
(4pnas sıyp)

("wos 'sıad) olaaway '[
("wo9 'sıad) olaaway '[
(6861) "Je 39 zanamı[

("wo9 'sıad) olaayay '[
(£/61) NVIaaW pur avomay
(S/61) ZIIyZNON

(TZ6T) Te 19 NOZUVO)
(ZI6I) yaTıım

("wos 'sıad) olaaway '[
(6861) "fe 32 zanamı[
(6861) "Te 99 zanawı[
(6861) ‘Te 9 zanamı[
(6861) "Te 9 zanawı[
(6861) "Te 9 zanawı[
(1661) savq

aa
tell
Ad
TO
Ad
TO
1: (®)
do
Io
Ad
Ad
Ad
Ad
Ad
a4
Ad
Ad
Ad
UL
Ad
IO
1 (@)
Io
1: ®)
UL
Ad
tal
UL
Io
do
Ad
Io
as!
Io
Ad

X m um u NN mm NT ind io dry mm m [md NO or rm rm nm rm m m nm

sıoyIny

991NOS

skwoaN SN

eysond]
soflond)

o1g4

oigY

il

oigA

eıpotg

eıpaIg

voor

uogef

uoref

eJpsuog
Ppn>219157
oppdojepeng,
odofepeng
unıeW

unieW

unıeW

soIeur]

soseur]
Jeraejepend
Jerarjepend
Jeraefepend
(ss1ely) eIOoW
(ss1eh) eIoW

ersuefeg
ersuepeg
eruefeg
ersuefeg
eruefed
esounisyelfä
soyueZisg

ursegq Sıydes3oapıfy

0983
08%
081
081
081
081
008
008
088
0001
0001
VER
809
r59
056
6lZ/
r6r
00%
00r
0,74
O8L1
Ocel
0/8
Orll
OOLI
0EOI
0E0I
186
008
008
2
22
Och
087
[)

Spnamıy

IELIWXLOE
OrSOWX,LOE
CHOENNX.LOE
SOIENXLOE
SOIENX,LOE
SOIENX,LOE
I9TOIX.LOE
I9TOIX.LOE
eeer IXLOE
CHE IX.LOE
CHOEIX,LOE
LOZOTX.LOE
9LIOTIXLOE
909 NXLOE
6166 IX,LOE
KEIL
roce IALOE
8rrO’IA.LOE
TSZO’IA.LOE
e991 IALOE
VILONA.LOE
SILOMALOE
SI6SAX.LOE
VLIENXLOE
PLIEIXLOE
8STEAXLOE
8STEAXLOE
CEI6HX,LOE
OTIENXLOE
OTI6EMX,LOE
6I00MA.LOE
6L0OOMALOE
6I60OMA.LOE
SSEIMA.LOE
CI6E IA LOE

WLN

snjewoue s{woaN] Fo ul110 pur Iaqwınu “I9ALı Payemosse “apnyne ‘sayıs Sunpdwes

(Z) u9ay

(Z) euozeues,

(Z) oqusnf

(Z) eine) e7

(Z) eoueagpy e7

(Z) eoueigpy e7

(Z) eıparg oLISISeUOW
(Z) eapaıq OLISISeUoW
(Z) eypowegey

(Z) eauesogpeg

(Z) aur9oJfeg

(Z) ep

VARZEIN,

(Z) 'D Kay [pp sos
(a1) :O pop arzıed
(11) ng

(AL) sese se] ap sew
(1,1) 'v PP »rfegjv
(11) 'V pp arjegıy
(AL) elıy

(Z,L) &IOW >p soseur
(IL) eIOoW >p saseurT
(ZL) PnıoL

(ZL) umersegfv

(ZL) unerseggvy

(7L) sop'qny >p eIoW
(11) sopıqny ap eıoW
(SJ) sesesseg

(SD) sog

(SI) stlog

(SI) esa1a],

(SI) esa1a],

(SO) esuıaf

(SZ) org esounssyerfiä
(SO) se [pp eILIOZ

Arıpe9o7

285

Distribution of the Cabrera water shrew in Northeastern Spain

(E/61) VWOD-SNYS
(S86T) Te 19 zaaTysoH

("wo9 'sıad) 'ıl aaTvoavW "N

(EZ61) VNOD-SNVS
(S861) oraay

(9861) "Te 12 vovıvavzıyay
(9861) ‘Te 19 vovIvavzıaay

("uo9 'sıad) asay "Vf

("wos 'sıad) OTIIav[ 'y N

(Apnas sıyp)
(0661) AIONIENAZLIAS

(/961) sAvIvq pue IvavN

(ZI61) vaTııW
("wo9 'sıad) svaIy 'T '[
("wo9 'sıad) Tvaınd '[

-(jgndun) anovınoW pue vOvIvavzıaay

(7661) "Te 32 waısng-zaa97]
("wo9 'sıad) ZIHONYS 'W '[

(6/61) OAVIEq

(6/61) OAVIaq
("wo9 'sıad) OAvITad

4d
Ad
UL
Ad
Io
Ad

[4
I
I
l
I
I
€
I
I
I
Ä
[4
6
I
I
€
I
I
I
I
I

soIdng-IV Ip eIory

(1503u09) PUunIeW
(sosag) saua],
STUIOJJEA TU9LIO],
eIOPIO]

eIOPIO],

U)

pooue1g

smopesw Jurdıy
(31395) puuey> Jean
91399

smopeau aurdıy

eysanf]
eysand
(eysanyy) eIIOW
OUTAILIIPTEA
OUTAIAIIPTEA
eypand

'pesp punoJ = (4 ‘podden = yL "uoneArssgo :gO

S6c
006
90r
IICL
1443
09T
097
005
06€
0581
Ost
Irc
ecc
0991
059
NL
9€8
00€
088
0001
ss

6C6SDCLLIE
ICDDOamS
reLEeDALIE
SCICDITIIBE
9I9YDALIE
8IrSOqLLIE
rIcsdaaLle

II0EIDLIE

gS/SyIDLIe
IESEIOIREE
CRIDONELZ
OIFrcDDLTIE
OITODDITE
OSESNX.LOE
00I6NX.LOE
LC8OWXK.LOE
ITIOWNX.LOE
SCCONX.LOE
OC6OMX.LOE
6I60OWX.LOE
VEICNXLOE

‘sIO[fpd [Mo uIeq = Ad :92.1N0g "TUN = 9 !euopsieg = g ‘euodene], = ] ‘eplaJfJ = T ‘sand = H ‘rzodenez = Z ‘oO, ], = ,L “UOIJISe) = SD !S99UTAOIA

(9) saIangay

(q) 9 pp mawoqueg 'S
(g) epaıpen3Ay

(q) [Od 1J[paseId

(g) sPAourJZ

(q) surdurey

(g) eIapıoynepeg 'W 'S
(I) 9 ap emadıag vis
(L) 322Aejr\-egrg ©
(T) Buy ıy

(T) s2109UOM >p I
(T) essnaopfoW

(T) ep

(H) osuy

(H) enderereg "pie
(Z) oAe9uoM ap ees]V
(Z) uouy

(Z) uouy

(Z) sayueuege],

(Z) sagueurege ],

(zZ) elıoq

286 I. Torre and J. L. Tella

not conditioned by water flow. As has been stated before (Amores 1975; GosALBEZ 1987;
SPITZENBERGER 1990), the Cabrera water shrew lives distant from the streams, its
distribution perhaps being favoured by the presence of irrigation channels (CorTEs and
Gir 1984). The Cabrera water shrew evidently has a wide climatic range (Fig.2C, D),
showing clear mediterranean trends: annual average temperatures were between 10 and
16°C for 90% of the sites, and rainfall was lower than 800 mm in 92 % of the sites. The
raınfall index does not constrain its distribution; these results are in contrast to those
observed by Trıano (1985) in SW Spain. Our results agree with those found by GoONZALEZ
and RomAn (1988) ın the province of Burgos.

Concerning its geographical distribution, CABRERA (1914) pointed out its presence in
the Spanish Pyrennes, but the lack of evidence available for this statement shed some
doubts on its authenticity. Later researchers (e.g. VERICAD 1970; Sans-CoMA and MAaR-
GALEF 1981; GiL et al. 1986; MORENO and BAarRBosA 1992) did not find this species in these
mountains, and a recent review (SPITZENBERGER 1990) did not consider the presence of the
Cabrera water shrew in the Spanish Pyrenees. We provide the first confirmed records for
their presence here. These may suggest a low density distribution along the axıal Pyrenees,
where the species may occupy alpıne meadows far from streams over 1,600-1,800 m.a.s.l.,
and streams of the Prepyrenean foothills below 1,000 m.a.s.l. This distribution could be
conditioned by the presence of the water shrew (Neomys fodiens), which lives ın this area
over 900-1,000 m.a.s.l. (GOsSALBEZ 1987), as a result of the altırudınal and habitat

A. ALTITUDE C. TEMPERATURE

00-2 2-4 4-6 6-8 8-10 10-1212-1414-1616-18 >18 6-8 8-0 10-12 12-14 14-16 16-18
m (x100) °C
B. WATER FLOW D. RAINFALL

5

0-2 2-4 4-8 8-16 16-32 32-6464-188 >188 3-4 4-5 6-6 6-7 7-8 8-9 9-10 10-1111-1212-13 >13
m3/s mm (x100)

Fig. 2. Frequency of distribution of the localities where Neomys anomalus was found according

to altitude (A), river water flow of the nearest station to the site of sampling (B), annual average
temperature (C), and rainfall (D)

Distribution of the Cabrera water shrew in Northeastern Spain 287

segregation sometimes reported between both species (SPITZENBERGER 1990). Otherwise,
the information recorded is insufficient to confirm this hypothesis, more detailed studies
being required about the species in the Pyrenees. In the remainder of the study area, where
N. fodiens is absent, N. anomalus is widely distributed along the Iberian range, most
probably reaching the arid regions of the Ebro depression through the river systems. The
presence of the Cabrera water shrew in two separated centres of the Catalan range confirms
its distribution along these mountains, discussed by JIMENEZ et al. (1989) and Faus (1991),
but we were unable to determine whether a continuity exists from the Iberian range to the
Pyrenees through the Catalan range.

Acknowledgements

We are very grateful to A. ARRIZABALAGA, ]J. VERDEJO, E. PELAYO, F. HERNANDEZ, J. M. SANCHEZ,
L. BoLEA, ]. GUIRAL, R. MARGALEF Jr., J. MArTin, E. MONTAGUD, V. PEDRoccHI, A. P£RrEz, ]. 1.
Pıno, J. A. Rey, J. L. Rıvas and R. R. JarıLLo for allowing us to use their unpublished data.

Drs. J. GosALBEzZ, M. J. LÖPEZ-FUSTER, ]. VENTURA, E. CAsTIEn, J. NADAL, and an anonymous
reviewer provided comments on an earlier draft. We also thank T. BALLESTEROS, T. BORAU, $. CARBO,
A. FARNöS, F. V. Faus, M. A. Garcia, F. GONZÄALEZ, I. GRABULOSA, V. FERNÄNDEZ, J. JIMENEZ, E.
MOoRENoO, R. MırLarA, A. REQUEJO, J. Rurz-OrLmo and ]J. SoL£ for their help. C. GORTAZAR
translated the abstract into German, and M. VILLARROEL ıimproved the English text.

Zusammenfassung

Verbreitung der Sumpfspitzmaus (Neomys anomalus) im Nordosten Spaniens

Beschrieben werden die Verbreitung der Sumpfspitzmaus im Nordosten Spaniens sowie die ökologi-
schen Faktoren, die ihr Verbreitungsgebiet bestimmen. Wir informieren über 26 neue Fundorte.

Auf der Südseite der Pyrenäen, wo die Sumpfspitzmaus zum ersten Male gefunden wurde, lebt die
Art auf alpinem Grasland, wahrscheinlich aufgrund einer Kompetenzbeziehung zu N. fodiens. Im
restlichen Studiengebiet, wo Neomys fodiens nicht auftritt, findet man die Sumpfspitzmaus in
verschiedenen Klima- und Höhenlagen, obwohl mediterrane Umgebungen bevorzugt werden. Die
Sumpfspitzmaus findet man in den axialen Pyrenäen, im Ebrotalkessel und der Umgebung der
iberischen Gebirgskette. Noch ist nicht klar, ob N. anomalus vom Küstengebirge bis ın die östlichen
französischen Pyrenäen eine zusammenhängende Verbreitung aufweist. Die Anwesenheit der
Sumpfspitzmaus wird andererseits von dem hydrographischen Netz bestimmt. Die Spitzmaus breitet
sich über Ströme und Bewässerungskanäle aus und erreicht mitunter Trockenbiotope, die aber immer
feuchte Mikrohabitate aufweisen.

References

AMORES, F. (1975): Neomys anomalus: Nueva localıdad en el Suroeste de Espana. Donana, Acta Vert.
2, 285-286.

ARRIZABALAGA, A.; MONTAGUD, E.; GOSALBEZ, ]. (1986): Introducciö a la biologia ı zoogeografia dels
petits mamifers (Insectivors i Rosegadors) del Montseny (Catalunya). Barcelona: Generalitat de
Catalunya, CIRIT.

BosQue, J.; VıLa, J. (1992): Geografia de Espana. Vols. 1, 6, 9, 10. Barcelona: Ed. Planeta.

BÜHLER, P. (1964): Zur Gattungs- und Artbestimmung von Neomys-Schädeln. Gleichzeitig eine
Einführung in die Methodik der optimalen Trennung zweier systematischer Einheiten mit Hilfe
mehrerer Merkmale. Z. Säugetierkunde 29, 65-93.

CABRERA, A. (1914): Fauna Iberica: mamiferos. Madrid: Museo Nacional de Ciencias Naturales.

CoRTES, J. A.; GıL, J. M. (1984): Neomys anomalus, nuevas citas para Granada. Donana, Acta vert.
1558150155:

Faus, F. V. (1991): Nouvelle donnee sur la crossope de Cabrera, Neomys anomalus (Cabrera, 1907),
dans l’est de l’Espagne. Mammalia 55, 452-456.

Fons, R.; LıBo1s, R.; SAINT-GIRoNs, M. C. (1980): Les micromammiferes dans le department des
Pyrenees Orientales. Vie et Milieu 30, 285-299.

FORTEZA, M. (1985): Caracterizacıön agroclimätica de la provincia de Teruel. Zaragoza: M. A.P.A.

GARZÖN, J.; CASTROVIEJO, S.; CASTROVIEJO, J. (1971): Notas preliminares sobre la dıstribuciön de
algunos micromamitferos del Norte de Espana. Säugetierkde. Mitt. 19, 217-222.

GiL, J.; GONZALEZ, F.; Puıc, D. (1986): Alimentaciö de l’öliba (T'yto alba): Distribuciö dels mamifers
Insectivors ı Rosegadors al Ripolles. Butll. Centre d’Estudies del Ripolles 10, 22-33.

GONZALEZ, J. M. (1975): Descripciön de la fauna de vertebrados de la zona de Mora de Rubielos
(Teruel). Bol. Est. Cent. Ecol. 4, 63-78.

288 I. Torre and J. L. Tella

GONZALEZ, ].; ROMAN, J. (1988): Atlas de los micromamiteros de la provincia de Burgos. Burgos: Ed.
: Gonzalez and J. Roman.

GOSALBEZ, J.; LÖPEZ-FUSTER, M. J.; GÖTZENS, G.; Sans-CoMma, V. (1985): El poblament ak petits
mamifers (Insectivora ı Rodenuia) a Catalunya. Requeriments ambientals ı Distribuciö geogräfica.
Bull. Inst. Cat. Hist. Nat., sec. zool. 6, 209-230.

GOSALBEZ, J. (1987): Insectivors ı rosegadors de Catalunya. Barcelona: Ketres $. A.

JIMENEZ, ]J.; GUILLEM, P.; MARTINEZ, J. (1989): Nota sobre la distribuciön en el extremo meridional
del Sistema Iberico de Neomys anomalus y Microtus arvalıs. Medi. Natural 1, 121-123.

LEÖN, A. DE; ARRIBA, A.; PLazA, M. DE LA (1987): Caracteristicas agroclimäticas de la provincia de
Zaragoza. Zaragoza: M.A.P.A.

LÖPEZ-FUSTER, M. J.; VENTURA, J.; GISBERT, J. (1992): Caracteristicas craneometricas de Neomys
anomalus Cheese, 1907 (Insectivora, Soricinae), en la Peninsula Iberica. Donana Acta vert. 19,
115-121.

NapaL, J.; Paraus, X. (1967): Micromamiferos hallados en egagröpilas de Tyto alba. P. Ins. Biol.
Apl. 42, 5-15.

MILLER, G. S. (1912): Catalogue of the mammals of Western Europe in the collection of the British
Museum. London: British Museum (N.H.).

M.O.P.U. (1988): Aforos y datos hidrogräficos de aguas subterräneas. Vols. 8, 9, 10. Madrid.

MORENO, E.; BARBOSA, A. (1992): Distribution patterns of small mammal fauna along gradients of
latitude and altitude in Northern Spain. Z. Säugetierkunde 57, 169-175.

PErLayo, E. (1979): Aves en el barranco de Valdetrevino y Collado del Campo (Talamantes).
Cuadernos de Estudios Borjanos 4. Centro de Estudios Borjanos. Instituciön Fernando el
Catolico.

Rıvas-MARTINEZ, S. (1983): Pısos bioclimäticos de Espana. Lazaroa 5, 33-43.

Rug1o, P. (1985): Observaciön de Neomys anomalus. El Medi Natural del Valles. Sabadell: Annals del
CEEM 1, 234.

Sans-CoMA, V. (1973): Contribuciön al conocimiento de los micromamiferos del nordeste de la
Peninsula Iberica y su interes biolögico. Insectivoros y Roedores en la egagröpilas de Tyto alba.
Ph. D. Thesis, Univ. Barcelona.

SAnS-CoMA, V.; MARGALEF, R. (1981): Sobre los Insectivoros (Mammalıa) del Pirineo Catalan.
Pirineos 113, 93-111.

SPITZENBERGER, F. (1990): Neomys anomalus Cabrera, 1907, Sumpfspitzmaus. In: Handbuch der
Säugetiere Europas. Ed. by J. NIETHAMMER and F. Krapp. Wiesbaden: Aula-Verlag. Vol. 1,
317-333.

TABERLET, P. (1984): La musaraigne de Miller Neomys anomalus. In: Atlas des mammiferes sauvages
de France. Ed. by Societe frangaise pour l’Etude et la protection des Mammiferes, 4041.

TORRE, 1.; MaArRTin, J.; P£EREZ, A. (1992): Presencia del musgano de Cabrera (Neomys anomalus) en
Ateca: primer registro en la provincia de Zaragoza. Ateca 1, 38—41.

Trrano, E. (1985): Situaciön actual del musgano de Cabrera (Neomys anomalus) en las sierras
subbeticas cordobesas. Oxyura 2, 65-73.

Van LaaRr, V. (1983): A record of Neomys anomalus Cabrera, 1907 from the Vosgues. Mammalıa 47,
123-125.

VERICAD, J. R. (1970): Estudio faunistico y biolögico de los mamiteros montaraces del Pirineo. Pub.
Cent. Pir. Biol. Exp. 4, 1-231.

VERICAD, J. R.; MEyLan, A. (1973): Resultats de quelques piegeages de micromammiferes dans le sud-
est de ’Espagne. Mammalıa 37, 333-341.

Authors’ addresses: IGnacıo TORRE, Departamento de Biologia Anımal (Vertebrados), Facultad de
Biologia, Universidad de Barcelona, Avda. Diagonal 645, E-08028 Barcelona and
Jost L. Terra, Estaciön Biolögica de Donana (CSIC), Avda. Ma Luisa s/n,
Pabellön del Peru, E-41013 Sevilla, Spain

Z. Säugetierkunde 59 (1994) 289-298
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

A report on the distribution of small mammals from Namibia
By J. ©. Matson and B. R. BLoop

Department of Biological Sciences, San Jose State University, San Jose; Department of Physical
Therapy, Mt. St. Mary’s College, Los Angeles; and Research Assoc., Sect. Mammals, Natural History
Museum, Los Angeles, California, USA

Receipt of Ms. 4. 2. 1994
Acceptance of Ms. 28. 4. 1994

Abstract

A sample of small mammal species from 37 pooled localıties throughout Namibia was used to assess
ecological and zoogeographical distribution patterns. Multivariate techniques (Cluster Analysıs and
Principal Components Analysıs) revealed an essentially ecological distribution based upon climate/
vegetation types within Namibia. Five major ecological areas are defined which correspond to
previously described vegetation and zoogeographic regions.

Introduction

Searching for and recognizing patterns in nature ıs one of the fundamental endeavors of
scientists. Spatial patterns of species distributions belong to the realms of ecology and
biogeography. While the ultimate causes of species distributions are evolutionary and
require historical explanations, identifying the patterns are an essential first step in the
process.

Mammalian taxonomy and distribution in southern Africa are poorly understood
(DiPpENAAR et al. 1983; SKINNER and SMITHERS 1990). Analyses ın southern Africa
(RAUTENBACH 1978) and especially of Namibia (CoETZEE 1983) have greatly increased our
understanding of mammalıan distribution patterns ın this part of the African continent.
Namibia represents a major portion of the Southwest Arıd Biogeographic Zone. The fauna
of this zone is considered to be quite distinct and to have had a long evolutionary history
(BIGALKE 1972).

The Natural History Museum of Los Angeles County (LACM) has an extensive
collection (over 6,500 specimens) of mammals from Namibia. While these specimens have
never been reported upon in the literature, CoETZEE’s (1983) did use the field ıdentifica-
tions of some of these as a basıs for his analysis of distribution patterns of mammals ın
Namibia. In our curation of thıs collection we noted that there were numerous misidentifi-
cations and errors. We have corrected these and use the revised material as the basıs for the
following analyses.

The major question we ask ıs “What geographic and/or ecological patterns can be
discerned from the specimens from Namibia housed in the LACM?“ Three Biotic Zones
are recognized in Namibia by SKINNER and SMITHERS (1990), Namib Desert, Southwest
Arıd, and Southern Savanna Woodland. These are the same as defined by RAUTENBACH
(1978). CoETZEE (1983) also recognized and defined these zones as mammalıan Zoogeo-
graphical Provinces wıthin Namıbia. Since CoETZEE (1983) had presented an analysıs of
mammalıan distribution patterns in Namibia, we wanted to use the current material as a
comparable sample. If the faunal areas recognized by CoETZEE (1983) have biological
validity, then an analysis of a subsample of mammals from Namibia should also reflect
sımilar patterns. In addition we use the results of this analysis to suggest testable
hypotheses concerning mammalıan distribution patterns.

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5905-0289 $ 02.50/0

290 J. ©. Matson and B. R. Blood

Material and methods

Because the LACM collection concentrated upon small mammals, we restricted our analysis to the
following orders, Insectivora Macroscelidea, Lagomorpha, Rodentia, and Hyracoıdea. Mammalian
taxonomy used herein follows that of SKINNER and SMITHERS (1990) and MEESTER et al. (1986).
However, these authors recognize but a single species of hyrax from Namibia, Procavia capensis.
Because there are definite morphological differences in the specimens we have on hand we prefer to
recognize two species for our analyses, P. capensis and P. welwitschu.

Localities were used as operational taxonomic units (OTU’s sensu SoKAL and SNEATH 1963). Since
several localities were represented by only a few species, we pooled nearby localities. Figure 1 is a map
of the pooled localıties. A list of the localities and species is given in tables 1 and 2. The pooled
localities were used as the OTU’s for the analyses.

Fig. 1. Map of Namibia showing the pooled localities used in the analysis of mammalıan distributions

Cluster analysis

Similarity between OTU’s was calculated separately using the squared Euclidian distances and the
“Faunal Resemblance Factor” (FRF) utilized by CoETzEe (1983). The OTU’s were clustered using the
unweighted pair group method with arithmetic averages as suggested by SNEATH and SokaL (1973).
While both similarity matrices produced relatively equivalent groupings, the results presented here are
based upon the squared Euclidean distances.

Principle components analysis (PCA)

OTU’s were subjected to PCA, based upon a correlation matrix, to select an optimal subset of species
(variables) from our sample that would bring about an ordination of the pooled localities. In order to
simplify interpretation of the analysis, low PC scores are those OTU’s which have a cummulative
loading of less than -0.99 while high PC scores are those greater than +1.00.

All computations were done using an IBM/PC 386 using the SPSS statistical packages (NORUSIS
1988).

SET IITI SS

A report on the distribution of small mammals from Namibia

Table 1. Collecting localities in Namibia
Number in parentheses at end of locality indicates the pooled localıty of Fig. 1

Kaokaland Area I; Opuwa, 106 km N, 65 km W Epupa Valley (1).
Kaokaland Area I; Opuwa, 66 km N, 84 km W Etengua (1).
Kaokaland Arae I; 63 km N, 55 km W, near Otjıangasemo (1).
Kavango Area I; Nkurrenhura, South Africa Police Camp (3).
Kaokaland Area I; 39 km N 130 km W Marienfluss (2).
Kavango Area I; 25 km N, 113 km W Rundu, Kanana (3).
Kaokaland Area I; Otju (2).

Kavango Area I; 9 km S, 84 km E Rundu, Shitemo (4).
Kaokaland Area I; Orumpebe (2).

Kavango Area I; 4 km $S, 18 km E Andara, Bagani (5).
Kavango Area I; 81 km $, 73 km W Rundu, Tsotsana (6).

Kavango Area I; 66 km S, 113 km E Rundu, s fork nr N fork, Omuramba (4).

Tsuemb Dist; 45 km N, 33 km E Tsuemb, Wildernis 882 (7).
Grootfontein Dist; 80 km N, 93 km E Grootfontein, Tiervlei 1022 (6).
Damaraland; 6 km SE Sesfontein 207 (8).

Outjo Dist; 37 km N, 28 km W Kamanjab, Ermo 646 (9).

Tsumeb Dist; 40 km WNW Grootfontein, Ghaub Farm 47 (7).
Grootfontein Dist; 30 km ENE Otavi, Sumas Farm 746 (7).
Grootfontein Dist; 35 km E Omkrap 218 (11).

Outjo Dist; 41 km N, 45 km E Outjo, Pirre 345 (10).

Damaraland; Palmwag 702 (8).

Outjo Dist; Westfallen 245 (9).

Grootfontein Dist; 8 km $, 4 km W Otavi, Elephantenberg 792 (7).
Grootfontein Dist; 25 km $, 36 km E Grootfontein, Okamaruru 220 (11).
Damaraland; Krone 721 (8).

Hereroland East; 58 km N, 108 km E Otijinunu (12).

Otijıwarongo Dist; 52 km ESE Otijiwarongo, Okosongomingo Farm 148 (13).

Otijıwarongo Dist; 9 km $, 11 km E Kalkfeld, Elshorst 90 (14).
Otijıwarongo Dist; 78 km $S, 48 km E Otijiwarongo, Ousema 201 (13).
Omaruru Dist; 35 km N, 27 km W Omaruru, Eausiro W 100 (15).
Hereroland East; 7 km N, 21 km E Otijinunu (12).

Okahanhdja Dist; 75 km N, 117 km E Okahanhdja, Kalidona 277 (16).
Omaruru Dist; 8 km N, 9 km W Uitspan 59 (= Kompanenosiid 59) (15).
Omaruru Dist; 18 km $, 6 km E Omaruru, Kamombande 86 (15).
Gobabis Dist; 117 km N, 39 km E Gobabis, Gelukwater 681 (17).
Omaruru Dist; Erongo West 83 (15).

Gobabis Dist; 67 km N, 82 km W Gobabis, Kambingana 204 (18).
Gobabis Dist; 64 km N, 103 km E Gobabis, Oostenwald 447 (19).
Karıbib Dist; 17 km $, 7 km W Usakos, Naob 69 (21).

Karıbib Dist; 47 km S Wilhelmstal, Okandukaseibe Farm 27 (22).
Windhoek Dist; 24-30 km N, 68 km E Windhoek, Okatumba South 149 (23).
Gobabis Dist; 41 km N, 2 km W Gobabis, Eava 383 (20).

Karıbib Dist; Nordenberg 76 (21).

Karıbib Dist; 35 km $S, 3 km W Usakos, Dorstrivier 13 (21).
Windhoek Dist; 71 km ENE windhoek, Muambo 130 (23).

Windhoek Dist; 10 km N, 68 km E Windhoek, Springbock Valley 132 (23).
Karibib Dist; 73 km $S, 3 km E Bethal Farm (24).

Gobabis Dist; 40 km $S, 88 km E Gobabis, Uithou 366 (27).
Swakopmund Dist; 8 km E Swakopmund, Swakopmund River (28).
Windhoek Dist; 81 km SW Wasservallei 382 (25).

Windhoek Dist; 110 km E Windhoek, Arnhem Farm 9 (26).
Windhoek Dist; Autabiıb 100 (26).

Windhoek Dist; 10 km N, 31 km W Rehoboth, Naos 46 (29).
Rehoboth Dist; Wostel 256 (29).

Windhoek Dist; 9 km S, 59 km W Rehoboth, Isabis Farm 19 (29).
Gobabis Dist; 75 km S, 24 km W Gobabis, Mentz 65 (30).

Rehoboth Dist; Nauzerus West 229 (32).

Windhoek Dist; Solitare 412 (31).

Rehoboth Dist; Billisport 172 (32).

Mariental Dist; Mibela 200 (30).

291

292 J- ©. Matson and B. R. Blood

Table 1 (continued)

Maltahoe Dist; 53 km $, 110 km W Maltahoe, Gorrasis 99 (34).

Mariental Dist; Asanıb 294 (33).

Mariental Dist; vicinity of Twee River (33).

Luderitz Dist; 77-81 km WNW Helmeringhausen, edge of Kanaan Farm 104 (34).
Bethanie Dist; 23 km WNW Helmeringhausen, Barby Farm 26 (35).
Keetsmanshoop Dist; 89 km ENE Koes, Welverdiend Farm 328 (33).

Bethanie Dist; Odendorf 43 (36).

Keetsmanshoop Dist; Spitzkoppeost 159 (36).
Keetsmanshoop Dist; Gaibis 226 (33).
Keetsmanshoop Dist; Reinfels 125 (36).
Keetsmanshoop Dist; Naute 119 (36).
Keetsmanshoop Dist; Kochena 74 (37).
Keetsmanshoop Dist; Warmfontien 280 (37).

Results and discussion
Cluster analysis

Inspection of the dendrogram (Fig. 2) indicates that there are two major clusters and five
minor clusters. Geographically, these are more easıly shown in fıgures 3 and 4. Essentially,
the two major clusters (Fig. 3) separate the country into west/southwest and northeast
regions. The five minor clusters shown in figure 4 correspond fairly well with the Faunal
areas described by CoETZEE (1983) and to the vegetation types of Gizss (1971). These fıve
clusters are as follows (using terminology from Giızss 1971): 1) Namib Desert, Southern
Kalaharı Desert, and Dwarf Shrub areas; 2) Escarpment and Mopane Savanna areas; 3)
Mopane Savanna and Thornbush Savanna areas; 4) Highland Savanna; and, 5) Northern
and Central Kalaharı Desert and Karstveld areas. As can be seen in figure 4, there ıs a
definite west to east gradient which reflects the climatic, vegetational, and geological
differences in these areas (COETZEE 1983; GıeEss 1971).

Principle components analysis

The PCA, using 38 species (10 species could not be used because they were represented ın
only one locality each) as varıables, was able to extract 33 principle axes. The first axis
accounts for 15.3 % of the total variation. The second and third axes account for 9.4 % and
7.7% of the variation, respectively. The first ten axes are necessary to represent 70 % of
the variation. This indicates that the total variation ıs spred over almost all of the species
distributions more or less equally. However, certain patterns can be recognized using the
first axis alone.

As can be seen in figure 5, Principal Component (PC) I appears to be an east/west
component. Low PC scores for the pooled localities occur in the west and south while high
PC scores occur in the northeast. The distribution of PC scores corresponds with the
distribution of clusters as discussed previously. Rainfall increases from west to east and
south to north in Namibia (CoETZEE 1983; Gıess 1971). This would indicate that the First
Principal Component is related to the ecological distribution of various species.

The distribution of representative high and low “loading” species on PC I are shown in
figure 6. Species (variables) in this analysıs had PC loading coefficients (Tab. 3) ranging
from -0.61 (Petromus typicus) to +0.79 (Mastomys natalensis). Species with negative
loadıng coefficients tend to be in the west and south while those with positive loadıng
coefficients are in the northeast.

As we interpret these results, species with very low (negative) PC loading are more arıd
adapted while those with very high (positive) PC loading are more mesic. Species with low

A report on the distribution of small mammals from Namibia 233

Table 2. Checklist of species
Species names followed by a list of locality numbers from table 1

INSECTIVORA
1. Crocidura cyanea - 36, 40, 50, 55, 56, 57, 65.
2. Crocidura Me eorrna — 14, 29, 32, 38.
3. Crocidura hirta - 13, 14, 35, 37, 38, 46.
4. Crocidura mariquensis — 20.

MACROSCELIDEA
5. Macroscelides proboscideus 5 43, 61.
Gm Blephantulussintufu = 1, 9,712516,118,5227247265927,28529, 30, 32, 34,35, 36, 37,38, 40, 41,
46, 47, 48, 51, 54, 55, 59, 60, 65, 70.
7. Elephantulus rupestris - 9, 16, 39, 43, 44, 50, 55, 37, 58, 59, 67, 68, 72.
LAGOMORPHA
8. Lepus capensis - 10, 37, 39, 41, 44, 46, 47, 48, 56, 62, 66, 73.
9. Lepus saxatılıs — 2, 12, 35, 40, 41, 46, 47, 48, 65, 68, 72.
10. Pronolagus randensis — 47.
RODENTIA
Cryptomys hottentotus - 6, 8, 9, 12, 14, 41, 42, 45, 48, 52.
Hystrix afrıcaeaustralis — 40, 41, 46.
IRedetesscapensisı 2, 19, 12513, 142 16,26, 29732, 34, 35, 37, 38, 39, 40, #1, 42, 47,548, 51,
550553.0309,.00,108,.09,.73.
Graphiurus murinus — 12, 14, 26, 34, 53.
Graphiurus platyops — 44.
Xerus inaurıs - 16, 24, 32, 35, 41, 42, 46, 51, 55, 56, 58, 65, 66, 69, 72, 73.
Funisciurus congicus — 2, 15, 17, 20.
Paraxerus cepapı — 13, 17, 20.
Petromus De — 5540547505 55, 574 65,1674873:
Parotomys brantsı — 66, 69.

Parotomys lıttledaleı — 49, 64, 65.

Lemniscomys rosalia — 6, 10, 14, 26, 27, 28, 31.

Rhabdomys pumilio — 25, 28, 36, 37, 40, 41, 43, 46, 47, 49, 51, 52, 53, 54, 55, 56, 57, 58, 59,
609629647652.00%.072..682.09,.70,,723 73:

Ep woosnami — 26, 38.

MuSsmRautuUs AROSB LE 16 1952. 07233926, 27,28, 29,30, 32,33, 34,355 36,37, 38, 49,
41, 42, 46, 47, 48, 49, 50, 51, 54, 55, 56, 65, 66.
Mus minutoides - 8, 12, 37.
Mus setzerı - 12.
Mastomys natalensis — 4, 6, 8, 10, 11, 12, 13, 14, 16, 17, 19, 20, 22, 23, 24, 26, 27, 28, 30, 31,
32, 34, 35, 36, 37, 38, 40, 41, 42, 46, 49, 51, 52, 56.
Mastomys shortridgei — 20.
Thallomys paedulcus — 4, 6, 10, 13, 14, 15, 30, 34, 36, 40, 42, 47, 48, 55, 57, 59, 65, 66.
Aethomys chrysophilus - 1, 6, 8, 10, 11, 13, 14, 16, 17, 19, 24, 27, 29, 31, 32, 35, 37, 38, 41,
462505555.
AlethomySmamaquensıs 1, 2, 3,5, 7. 3, 16,17,20%21,22,23, 24,27, 28, 29, 30, 31, 34, 36,37,
38, 39, 40, 41, 42, 43, 44, 46, 47, 48, 50, 51, 52, 54, 55, 56, 57, 58, 59, 60, 65, 67, 68, 72, 73.
Desmodillus auricularıis - 5, 32, 35, 51, 62, 64, 65, 66, 73.
Gerbillurus paeba - 5, 12, 13, 24, 25, 29, 30, 31, 32, 35, 38, 39, 40, 41, 42, 43, 44, 47, 49, 51,
56, 58, 60, 64, 66, 68, 69, 71, 73.
Gerbillurus setzeri - 40.
Gerbillurus tytonis — 64.
Gerbillurus vallinus - 79.
Iareranleuco 2aster, 412954 648-210211,125153,514.215,416,,19720922923724727,,28,,29,538,
SIE 3AN53D3B37340741,42743516477 50, 5115 534555 503 58, 00,69, 72:
Tatera brantsii - 31, 56, 66.
SACCoStomUSKeammpestns] —010,127153,,14,16,02203,26252 33053234, 355 37535 39549,
41, 47, 51, 52, 56.
Malacothrix typica — 16, 37, 42, 51, 55, 56, 65.
Dendromus melanotis - 12, 26, 31, 41, 46.
Steatomys parvus — 12.
Steatomys pratensis - 10, 14, 16, 26, 27, 28, 29, 41, 50, 55, 64.
Petromyscus collinus - 5, 7, 21, 22, 28, 39, 40, 44, 50, 57, 58, 67.
Petromyscus monticularis — 65.

HYRACOIDEA

47. Procavia capensis — 40, 51, 56, 65, 72, 73.

48. Procavia weltwitschü - 3,5.

294 J. ©. Matson and B. R. Blood

0 b) 10 15 20 25
a TE Te Er N TE

Fig. 2. Dendrogram of pooled localities showing recognized clusters. Numbers of minor clusters on
the left. Major Cluster I includes 1, 2, and 3. Major Cluster II includes 4 and 5. See text for
explanation

PC loadings include Petromus typicus (-0.61). Elephantulus rupestris (-0.56), Petromyscus
collinus (-0.56), and Rhabdomys pumilio (-0.54). RAUTENBACH (1978) indicated that the
first three of these species were restricted to the Southwest Arıd and/or Namib Desert
Biotic Zones. Species with high PC loadings include Mastomys natalensis (+0.79),
Aethomys chrysophilus (+0.73), Saccostomus campestris (+0.72), and Crocıdura hirta
(+0.54). These species, while occurring in the Southwest Arıd Biotic Zone, are also found
in more mesıc regions (RAUTENBACH 1978).

When the results of both the Cluster and PC analyses are considered together, the
distribution patterns of small mammals are more clear. Cluster 1 (Fig. 3) represents the arıd
west and south of Namibia. Low PC scores for localities in these areas support this idea
(Fig. 5), as well as, the low PC loadings for species (Fig. 6) that are considered to be
restricted to arıd regions. These species have a major portion of their distribution ın the
localities of Cluster I. Cluster II (Fig. 3) represents localities in the north and east of
Namibia and have relatively high PC scores (Fig. 5). The species associated with this
cluster are those with positive loadings and a more mesic ecological distribution.

A report on the distribution of small mammals from Namibia 295

ns CLUSTER

Fig. 4. Map of Namibia showing distribution of five minor clusters corresponding to vegetation and
biotic areas. See text for explanation

296 J. ©. Matson and B. R. Blood

PCI

I) G - 0.99

Scores

Fig. 5. Map of Namibia showing distribution of high and low PC scores for each pooled locality

Localities within Namibia that have intermediate PC scores contain a mixture of
species. In addition, those species with intermediate loadıngs (between +/-0.50) are either
widely distributed or have few locality records. Widely distributed species include
Gerbillurus paeba (PC loading of 0.00 and occurring in 23 of the 37 pooled localities).
Species with few locality records include Parotomys hittledalei (PC loading of -0.31 and

Table 3. First Principal Component loading coefficients for 38 species of small mammals in

Namibia

Coefficient/Species Coefficient/Species

Petromus typicus +0.11 Paraxerus cepapı
Elephantulus rupestris +0.13 Mastomys shortridgei
Petromyscus collinus +0.21 Tatera lencogaster
Rhabdomys pumilio +0.23 Tatera brantsü
Crocidura cyanea +0.25 Elephantulus intufi
Parotomys lttledaleı +0.28 Mus minutoides
Aethomys namaquensis +0.32 Steatomys pratensıs
Procavıa capensıs +0.37 Zelotomys woosnami
Desmodillus auricularis +0.40 Graphiurus murinus
Macroscelides proboscideus +0.45 Cryptomys hottentotus
Procavıa welwitschii +0.45 Dendromus melanotis
Lepus saxatılis +0.47 Crocidura fuscomurina
Xerus inauris +0.48 Pedetes capensıs
Thallomys paedulcus +0.50 Lemniscomys rosalıa
Funisciurus congicus +0.50 Mus indutus
Malacothrıx typica +0.54 Crocıdura hirta
Lepus capensıs +0.72 Saccostomus campestris
.00 Gerbillurus paeba +0.73 Aethomys chrysophilus
+0.09 Aystrix afrıcaeaustralis +0.79 Mastomys natalensıs

A report on the distribution of small mammals from Namibia 297

Small Mammal PC I
Loading

Low PC
Coeffients

Mixture

High PC
Coeffients

Fig. 6. Map of Namibia showing distribution of small mammal species that have high or low loadıng
coefficients on the First Principal Component. See text and table 2 for species and explanation

found in only three localities of Cluster I) and Zelotomys woosnami (PC loading of +0.37
but found in only two localities of Cluster II).

While these results are based only upon a sample of the mammals from Namibia, they
do reveal trends similar to those described by CoETZEE (1983) and RAUTENBACH (1978).
BIGALKE (1972) and MoraIn (1984) considered southwestern Afrıca (including Namibia,
parts of Botswana, and most of South Africa) to be an important biogeographic region
because of the high amount of endemism. While our analyses would appear to represent
more of an ecological distribution, ıt does confirm the zoogeographic analyses of COETZEE
(1983) and RAUTENBACH (1978).

Acknowledgements

We thank JOHN HEYNING, SARAH GEORGE, and LINDA BARKLEY of the LACM for allowing us to
work with the Namibia Mammal Collection. A major portion of this collection was aquired through
the efforts of Lant LESTER. The original expedition, October and November 1972, and a 1974
expedition were sponsored by Mrs. REEsE TAayror. Additional specimens were obtained between
1974 and 1977. Travel funds to visit the LACM were made available through a San Jose State
University Foundation, Faculty Development Grant and the Department of Biological Sciences, San
Jose State University.

Zusammenfassung

Verbreitungsmuster von Kleinsäugern in Namibia

Eine Sammlung von über 6500 Kleinsäugern aus Namibia wurde genutzt, um ökologische und
zoogeographische Verbreitungsmuster zu ermitteln. Über das ganze Land verteilte Einzellokalitäten
wurden zu 37 Fundgebieten zusammengefaßt. Multivarıate Auswertungstechniken (Cluster Analysıs,
Principal Components Analysıs) ergaben klare ökologische Verbreitungsmuster, die den Klima- und
Vegetationstypen von Namıbia folgen. Fünf ökologische Hauptregionen lassen sich definieren, die
mit bereits von früheren Autoren charakterisierten Vegetationszonen und zoogeographischen Regio-
nen korrespondieren.

298 J: ©. Matson and B. R. Blood

Literature

BIGALKE, R. C. (1972): The contemporary mammal fauna of Africa. In: Evolution, Mammals, and
Southern Continents. Ed. by A. Keast, F. C. Erk, and B. Grass. New York: State University of
New York Press. Pp. 141-194.

COETZEE, C. G. (1983): An analysis of the distribution patterns of the Namibian terrestrial mammals
(bats excluded). Ann. Mus. Roy. Afr., Centr., Sc. Zool. 237, 63-73.

DiPPENAAR, N. J.; MEESTER, J.; RAUTENBACH, I. L.; WOLHUTER, D. A. (1983): The status of southern
Afrıcan mammal taxonomy. Ann. Mus. Roy. Afr. Centr., Sc. Zool. 237, 103-107.

Gizss, W. (1971): A preliminary vegetation map of South West Africa. Dinteria 4, 5-14.

MEESTER, J. A. J.; RAUTENBACH, I. L.; DiPpEnAAR, N. ]J.; BAKER, C. M. (1986): Classification of
southern African mammals. Transvaal Mus. Monogr. no. 5, 359 pp.

Moran, $. A. (1984): Systematic and Regional Biogeography. New York: Van Nostrand Reinhold
Coane

Norusıs, M. ]J. (1988): SPSS/PC+ Advanced Statistics V2.0. Chicago: SPSS, Inc.

RAUTENBACH, 1. L. (1978): A numerical re-appraisal of the southern African biotic zones. Bull.
Carnegie Mus. Nat. Hist. 6, 175-187.

SKINNER, J. D.; SMITHERS, R. H.N. (1990): The mammals of the southern African subregion. Univ.
Pretoria, RSA.

SNEATH, P. H. H.; SokaL, R. R. (1973): Numerical Taxonomy. San Francisco: W. H. Freeman.

SokAL, R. R.; SNEATH, P. H. H. (1963): Principles of numerical Taxonomy. San Francisco: W. H.
Freeman.

Authors’ addresses: JOHN ©. MaTson, Department of Biological Sciences, San Jose State University,
San Jose, CA 95192 and Brap R. BLooD, Department of Physical Therapy, Mt.
St. Mary’s College, Los Angeles, CA 90049, USA

Z. Säugetierkunde 59 (1994) 299-308
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Cytogenetic diversity and evolution of Andean species
of Eligmodontia (Rodentia, Muridae)

By A. E. SPOTORNO, J. SUFAN-CATALAN, and LAURA I. WALKER

Laboratorio de Citogenetica Evolutiva, Departamento de Biologia Celular y Genetica, Facultad
de Medicina, Universidad de Chile, Santiago, Chile

Receipt of Ms. 4.2.1994
Acceptance of Ms. 31.5.1994

Abstract

The standard and G-banded chromosomes of northern species of the phyllotine genus Eligmodontia
were investigated. E. puerulus showed 2n = 50, NFa = 48 ın bone marrow cells of three males and four
females from northern Chile, and E. moreni 2n = 34, NFa = 48 in three males from northern
Argentina. Comparisons showed extensive conservation of G-band patterns, including those of 15
telocentric chromosomes of the former with the arms of eight metacentrics of the latter; these
characteristics suggest seven centric fusions and one pericentric ınversion in E. moreni. C- bands were
small in E. puerulus, as well as in the related Andinomys edax 2n = 54, NFa = 54 (two males and one
female from northern Chile). All these northern species have chromosome arm sızes smaller than 9 %
of the total karyotype, in contrast to some longer arms reported in the southern E. typus and
E. morgani; the latter were probably derived by tandem fusions. Thus, southern species comprise a
derived phyletic line, probably evolved from a primitive northern ancestor having 2n = 50 and NFa =
48. The role of geographic and cytogenetic factors in this speciation pattern, similar to that of the
related Auliscomys species living in the same area, is discussed.

Introduction

Among the nine genera included in the tribe Phyllotini ot South American murid rodents,
Eligmodontia is clearly distinct and specialized, given its adaptations to life in arıd zones.
Although ıt was considered monotypic by some authors, two recent cytogenetic studies
(ORTELLS et al. 1989; KELT et al. 1991) support the idea that at least three species should be
distinguished, on the basıs of striking chromosomal differences. These are: Elhgmodontia
puerulus, 2 n = 50, NFa = 48 (PEARson and PaTTon 1976; ORTELLS et al. 1989), reported
for populations living in the Altiplano (the highlands of southern Peru, Bolivia and
northern Argentina); E. typus, 2n = 44, NFa = 44 (ORTELLS et al. 1989) for populations
from central Argentina; and E. morganı, 2n = 32, NFa = 32 (ORTELLS et al. 1989; KELT et
al. 1991) for those from southern Argentina. A cytologically unknown fourth species from
northern Argentina, E. moreni, has also been included in the most recent world species list
(Musser and CARLETON 1993), but these authors stated (p. 701) that “The differentiation
of moreni from morgani in the S Andes and from typus ın the Pampas warrants further
study”.

We present here cytogenetic data on specimens of this fourth species. We also compare
all standard chromosomes described for these four species, as well as the G-banded
karyotypes of Elıgmodontia puerulus from northern Chile and E. moreni from northern
Argentina. In addition, we present C banded karyotypes of the first species and those of
the karyotypically related phyllotine Andinomys, both from northern Chile, to date not
yet described in these terms. The latter is a monotypic genus of the Altiplano, considered
to be primitive on the basis of protein electrophoresis (SPOTORNO 1986).

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5905-0299 $ 02.50/0

300 A. E. Spotorno, J. Sufan-Catalan, and Laura I. Walker

Material and methods

Specimens

All anımals were collected in the field. Skulls and skins were prepared as voucher specimens and
deposited in the collection of the Laboratorio de Citogenetica, Facultad de Medicina, Universidad de
Chile (LCM).

Taxa (taxonomic names according to Musser and CARLETON 1993), original localities, number and
sex of specimens (LCM numbers in parenthesis) were as follows. E. puerulus: Parinacota, 110 km NE
Arıca, I Regiön de Tarapacä, CHILE, 1 male (LCM 650); Choquelimpie, 114 km NE Arica, I Regiön
de Tarapaca, CHıte, 2 males (LCM 1183-1193) and 4 females (LCM 1184-1283-1438-1439).
E. morent: Cauchari, Provincia de Salta, ARGENTINA, 3 males (LCM 1702-1703-1704). Andinomys
edax: Murmuntanı, 110 km E Arıca, I Regiön de Tarapaca, Chile, 1 male (LCM 243); Pampa Yuscuni,
100 km NE Arıca, I Regiön de Tarapaca, Chile, 1 male and 1 female (LCM 678-677).

Chromosome analysis

Chromosomes were obtained from bone marrow cells using the conventional in vivo colchicine
hypotonic technique, preceded by a yeast injection to improve the mitotic index (LEE and ELDER
1980). Total chromosome counts per cell were made in at least five good quality metaphases per
specimen. NFa is the number of visible autosomal arms per cell.

Chromosome measurements were based on photographic enlargements, using the best single
chromatid per pair (best meaning easiest to measure). Values were transformed to percentages of the
total haploıd female set. Relative values, together with those from ıdiograms already published, were
displayed in a scatter diagram, which we have called the karyo-idiogram (SPOTORNO et al. 1987); each
chromosome is represented by a single point according to its arm lengths. This is a useful device which
allows the simultaneous description, comparison and eventual distinction of all chromosomes from
many species. Two derived morphological variables can be evaluated: total chromosome size (short
arm plus long arm lengths) and centromeric index (100 times short arm length divided by total
chromosome length). Such a procedure assumes that the total genome size is conserved among the
species compared. Although this assumption is generally true for mammals, it may be validated by C-
banding techniques, which detect heterochromatin-containing satellite DNA, or by the use of marker
chromosomes when available (SPOTORnNo et al. 1987).

Chromosomes were classified according to morphology (centromere position) using the nomen-
clature of LEvan et al. (1964), and also according to size. We distinguished large, medium or small
chromosomes when their relative lengths were >9 %, 9-5.5 % or <5.5 % of the female haploid set,
respectively (see Massarıntı et al. 1991).

Chromosome bands were obtained by treating metaphase cells with G-banding (CHIARELLT et al.
1972) and C-banding techniques (CRossEn 1972; SUMNER 1972). Comparisons of G-banded karyoty-
pes were made in at least three selected metaphases from each taxon. Using shared G-band patterns,
chromosomes from two or more species were classified as totally corresponding, partially correspond-
ing or unique (WALKER et al. 1979).

Results

The karyotypes of the two main geographic populations were clearly different. All the
specimens of E. puerulus from Chile exhibited cells with 2n = 50 and NFa = 48. All
chromosomes were telocentric, with no visible short arms (Fig. 1a). This karyotype was
essentially identical to those described for populations from Peru, Bolivia and northern
Argentina (ORTELLS et al. 1989; Kerr et al. 1991). Chromosome sizes graded from
medium to small. The largest telocentric chromosome (pair 1) was clearly identified by
size, amounting to 8.3 % of the total karyotype length. Pairs 3 and 8 showed a secondary
constriction in the middle of their long arms (arrows in Fig. 1a), although the latter was not
clearly visible in short metaphases. Sex chromosomes were difficult to identify in standard
Giemsa karyotypes; the X was the third largest and the Y was among the many small
chromosomes.

All the Eligmodontia moreni specimens from northern Argentina consistently showed
karyotypes with 2n = 34 and NFa = 48 (Fig. 1b). The eight largest chromosomes displayed
metacentric or submetacentric shapes, with pairs 3 and 6 exhibiting a clear secondary

Cytogenetic diversity and evolution of Andean species of Eligmodontia 301

Fıg.1. Standard karyotypes of: a) Eligmodontia puerulus female (LCM 1193), X chromosome
localized according to its size; and b) E.moreni male (LCM 1704). Arrows point to secondary
constrictions

constriction in the middle of their long arms (arrows in Fig. 1b). The largest chromosome
arm comprised 8.5% of the karyotype length. The X chromosome was the largest
telocentric, and the Y a small telocentric with an extremely short arm.

G-bands allowed the identification and comparison of every chromosome from both
karyotypes. The X of E. puerulus was the third in size (Fig. 2a), showing two dark bands in
the middle of its arm, a pattern already described for most mammals (PATHAK and STOCK
1974). It was almost identical to the X of E. moreni (Figs. 2b and 3). The Y chromosomes
from both species were also very similar in bands and size, except for the presence of a
short arm unique to E. moreni (Fig. 3).

When G-banded karyotypes were compared side by side (Fig. 3), all chromosomes or
chromosome arms were found to correspond in bands and sızes. The single exception was
chromosome 13 from E. pnerulus, which appears to be unique. In particular, the arms of
the largest seven metacentric chromosomes of E. moreni had corresponding telocentric
chromosomes in E. puerulus, suggesting the occurrence of Robertsonian fusion/fission
processes during the evolution of these species. The remaining metacentric pair 8 corre-
sponded in bands and size to the single telocentric pair 5 from E. puerulus, suggesting the
occurrence of a pericentric inversion.

C-bands were very small and confined to centromeric positions in most, ıf not all, the
chromosomes of both E. puerulus (Fig. 4a) and Andinomys edax (2n = 56, NFa = 56, Fig.
4b). The latter karyotype was very similar to that reported for a single female from
northern Argentina (PEARsoN and ParTon 1976), with the exception that only pairs 1, 27
and Y had short arms in these Chilean specimens. A faint intercalar C-band was observed
in the middle of pair 8 in Andinomys, and at a similar site of pair 3, or perhaps 4, ın
E. puerulus; this happens to be the usual localization of secondary constrictions in its
Giemsa standard chromosomes (Fig. 1a).

302 A. E. Spotorno, J. Sufan-Catalan, and Laura I. Walker

Fig. 2. G-banded karyotypes of: a) Eligmodontia puerulus female (LCM 1283); and b) E. moreni male
(LCM 1704). Arrows point to secondary constrictions

Fig.3. Correspondence of G-band patterns between the chromosomes of Eligmodontia moreni
(aligned large numbers at bottom) and E.puerulus (small numbers above). Chromosome 13 of the
former not included, since it was difficult to match

Cytogenetic diversity and evolution of Andean species of Eligmodontia 303

_

a ‚ih, 2“. en

Ba

N
S

10 um

Fıg.4. C-banded metaphases of: a) Eligmodontia puernlus female (LCM 1283), and b) Andinomys
edax male (LCM 678), with 2n=56, NFa=56. Sex chromosomes and the smallest metacentric
autosome are indicated

Discussion

Our results provide data allowing for a reasonably complete view of the extreme cytogene-
tie diversity and the still obscure phylogenetic relationships of Eligmodontia species.

But firstly, we wıll now evaluate empirically the assumption that, despite the large
changes in 2n and NFa, the total genome sizes of Eligmodontia species have been
conserved since their last common ancestor. On the one hand, there are no large amounts
of heterochromatın in E. puerulus NFa = 48, or ın E. typus NFa = 44 (ORTELLS et al. 1989),
or in the related phyllotine Andinomys edax NFa = 56. On the other hand, if relative
lengths of all chromosomes from the fıve species are compared ın a single karyo-ıdiogram,
as shown in figure 5, at least two marker chromosomes, the X and Y, clearly retain their
relative lengths at roughly 6 to 7% and 3.4 to 3.9 %, respectively, despite the most
probable pericentric inversion which gave rise to the derived typus X and Y, as well as the
morganı Y chromosome. If large changes in total genome size had occurred, such length
ranges should be expected to exhibit larger varıations than those observed here. G-bands,
where available, also document the size constancy of the identified sex chromosomes. In
summary, gross constancy among genome sızes may be accepted as a reasonable assump-
tion for further chromosome comparisons based on relative lengths.

Inferences about species distinctions, chromosomal changes and phylogenetic relation-
ships can be made from our results when compared with those already published. It is
immediately obvious that at least seven Robertsonian fissions or fusions and one pericen-

304 A. E. Spotorno, J. Sufan-Catalan, and Laura I. Walker

us
®
4 e
2
%/o 3
2 ®
4a
e 1
5 v
9 ®
EN
IN 2
BEEEN a
6 SUR
g 9 IE
| SS
ER Ines
. &®) -. BB -
2 N .
ze x - (m)
X) \ 5
x -
D
®
6 =
jan.
5 29
2
Dv
IN
Aue;
ya
1107
oO
s Sn.
wWY .:
oM:
=
2 25 er
* Andinomys edax
DO Eligmodontia puerulus
1 moreni
typus
morgani
10)

pP %o

Fig. 5. Karyo-idiogram (bivarıate plot) showing relative sizes of short arm (p) versus long arm (g)
chromosomes from four Ehigmodontia species and Andinomys (not all chromosomes actually shown).
Total chromosome size may be read on the diagonal; chromosome classifications by morphology and
size follow LEvan et al. (1964) and ORrTELLS et al. (1989), respectively. Eligmodontia typus and
E.morgani data are from ORTELLS et al. (1989)

Cytogenetic diversity and evolution of Andean species of Eligmodontia 305

tric inversion strongly support the separate species status of E. puerulus and E. moreni. The
eventual hybrid produced from a cross among individuals with such karyotypes most
probably would exhibit abnormal meiosis and reduced fertility. Therefore, the chromo-
some differences detected substantiate the morphological and ecological distinctions
previously noted between populations of both nominal species (Mares et al. 1989).

At the same time, the NFa = 48 shared only by these two species suggests a primitive
condition close to the NFa = 54 shown by Andinomys. These three species wıth only
medium or small chromosomes (see Fig. 5) have in fact close and exclusive geographic
ranges in the Altiplano, and within the northeastern side of the xeric diagonal which
divides South America in two parts (SPOTORNO and VELoso (1989). This geographic sub-
region has been considered the center of origin of the phyllotine group (REıG 1986).
Independent evidence based on electrophoretic analysıs of proteins also placed the
monotypic Andinomys at the base of the phyllotine radıation (SPOTORNO 1986).

The southernly distributed FE. typus and E. morgani seem to belong to a different and
derived phyletic line. Their divergent karyotypes NFa = 44 and 32 are characterized by
chromosome arms longer than 9 %, which are absent ın the NFa = 48 karyotypes: the long
arm of E. typus metacentric pair 1 and the long arms of E. morgani telocentric pairs 1
through 5 (Fig. 5). These have been interpreted as being produced by tandem transloca-
tions (ORTELLS et al. 1989). As such, they must represent derived conditions that arose
from the primitive ones presently seen in the northern NFa = 48 karyotypes.

For instance, chromosome comparisons based on the karyo-ıdıogram (Fig. 5) allow the
proposal of the following most simple transformation series. The largest metacentric pair 1
of E. typus, having arms of 9.3 and 8.2%, ıs surely a unique and extremely derived
condition. It was probably formed from the centric fusion of two derived telocentric
elements of correspondant large sizes, similar to pairs 5 and 6 of E. morgani (pair 5 being
the intermediate condition shared by E. typus). Finally, the derived large telocentric
morganı chromosome 5 probably arose through tandem fusion of two small telocentric
elements such as those found in E. puerulus (the most ancestral condition, also shared by
Andinomys), or ın the small q arm of metacentric 1 of E. moreni (an additionally derived
condition). Such a transformation series for thıs character may be linearly written through
the following species tree: (/typus, morganı] pnerulus [moreni]). This topology ıs also a
reasonable summary of cytogenetic data, as well as of geographic data, since it ıs consistent
with the southern (typus-morganı) — northern (puerulus-moreni) axıs of species distribu-
tion; therefore, ıt is a good candidate for a reasonable estimate of the real species
phylogeny.

A strikingly similar northern-southern geographic pattern has been postulated for the
chromosomal evolution of four related phyllotine species of the genus Aulıscomys
(WALKER and SPOTORNO 1993). Among them, “an ancestral telocentric karyotype would
have undergone three consecutive tandem fusions” in southern species; later, three centric
fusions probably occurred in northern species. The paleogeographic model proposed
there, based on the assumption that actual biotic patterns ın South America were
determined by Quaternary geological and climatic changes (VUILLEMIER 1971), seems to
be also valıd for Eligmodontia species.

Moreover, the extreme cytogenetic diversity of Eligmodontia, as well as the occurrence
of tandem fusions rarely documented in mammals, suggest an active role of chromosomal
changes in the speciation process. Although this has been a subject of renewed interest and
controversy (CAPANNA 1982; PATTON and SHERWOOD 1983; BAKER and BICKHAM 1986;
Sıres and Morrtz 1987), recently Reıcg (1989) has contributed importantly to clarıty such
causal relationship; he suggested that explosive speciation processes were triggered by
chromosomal rearrangements. However, fertility studies on the heterozygotes for diffe-
rent chromosomal rearrangements indicate that meiotic and evolutionary consequences are
drastically different, depending upon the type of rearrangement. Thus, while single centric

306 A. E. Spotorno, J. Sufan-Catalan, and Laura I. Walker

fusions would have little or no reproductive isolation effects, because the fertility of
heterozygotes ıs modified slightly or not at all (BickHam and Baker 1979; Jonn 1981;
PaATTon and SHERwooD 1983), single tandem fusions would have drastic consequences,
severely reducing the fertility of heterozygotes (WHITE 1973; JoHN 1981; for a recently
reported case where tandem fusions are involved in hybrid infertility, see RyDer et al.
1989). The fact that both types of chromosomal changes have occurred independently
within two different but related phyletic lines evolving within the same subregions, invites
consideration of geographic factors in contrast to cytogenetic factors.

The following hypothetical sequence of events would be consistent with our cytogene-
tıic data, the present species ranges and geography of the region. An ancestral, perhaps late
Miocene species with NFa = 48, having a wide northern-southern range, was separated by
the Plio-Pleistocene xeric diagonal. The northern ancestor would have evolved into the
present E. puerulus, maintaining its karyotype morphology, and secondarily to E. moreni,
mainly by centric fusions. Here, the ice barriers generated by cyclic glacial warming and
freezing within the Andean valleys (VUILLEMIER 1971) would be an associated requirement
for speciation processes that maintained the NFa = 48; i.e. geographic factors would be
sufficient for biological isolation, and cytogenetic isolation was not required. By contrast,
the populations on the flat landscapes of southern Argentina were less prone to be affected
by such ice barriers. Here, tandem fusions, with drastic meiotic isolating consequences,
could probably be sufficient to change NFa. In other words, geographic factors would be
insufficient for isolation, and cytogenetic isolation was required in the southern subregion.
A similar pattern of chromosome divergence seems to have occurred in the species of
Auliscomys (WALKER and SPOTORNO 1993). Therefore, the isolation required for specia-
tion might be the product of subsidiary or complementary actions between extrinsic
(geographic) factors and intrinsic (chromosome) mechanisms. These would explain the
high degree of chromosomal divergence observed among these Biyllseme rodents as well
as within many mammalıan groups.

Acknowledgements

This work was supported by Grant 92-1186 from the Fondo Nacional de Ciencia y Tecnologia, Chile.
We thank Dr. Luis ConTRERAS and Dr. PaBLo MARQUETTE (PSP/WWF 7578 and Lincoln Zoo
Neotropical Fund) for providing some field specimens, Mr. GERMAN MANRfQUEZ for help with
German translation and Mr. JUAN OYaRrce for assistance in the collection and care of the animals.

Zusammenfassung

Cytogenetische Vielfalt und Evolution von Eligmodontia-Arten in den Anden (Rodentia, Muridae)

Bei zwei Arten der Gattung Eligmodontia wurden Standard- und G-gebänderte Chromosomen aus
Knochenmarkszellen untersucht. Drei Männchen und vier Weibchen von E. puerulus aus dem
Norden Chiles zeigten 2n = 50, NFa = 48. Drei Männchen von E. moreni aus dem Norden
Argentiniens zeigten 2n = 34, NFa = 48. Vergleichende Untersuchungen ergaben eine starke
Konservierung der G-Bandenmuster. Bei insgesamt 15 telozentrischen Chromosomen von E. pueru-
lus stimmten die Bandenmuster mit jenen von 8 metazentrischen Chromosomen von E. moreni
überein. Dies kann als das Ergebnis von sieben zentrischen Fusionen und einer perizentrischen
Inversion bei E. moreni interpretiert werden. Ähnlich wie bei der verwandten Art Andinomys edax
(2n = 54, NFa = 54, zwei Männchen und ein Weibchen aus dem Norden Chiles untersucht), waren die
C-Bänder von E. puerulus schmal. Im Gegensatz zu den südlichen Arten E. typus und E. morgani
betrugen bei den nördlichen Arten die längsten Chromosomenarme weniger als 9% des diploıden
Karyotyps. Das Auftreten erheblich längerer Chromosomenarme bei den südlichen Arten deutet auf
das Vorliegen von Tandemfusionen hin. Die südlichen Arten stellen demnach eine eigene phylogeneti-
sche Linie dar, die sich aus einem primitiven nördlichen Vorfahren mit 2n = 50 und NFa = 48 ableitet.
Die Bedeutung geographischer und cytogenetischer Faktoren im Artbildungsprozeß bei Zligmodontia
wird unter Bezugnahme auf Auliscomys-Arten aus demselben Verbreitungsgebiet diskutiert.

Cytogenetic diversity and evolution of Andean species of Elıgmodontia 307

References

BAKER, R. J.; BICKHAM, J. W. (1986): Speciation by monobrachial centric fusions. Proc. Natl. Acad.
Scı. USA 83, 8245-8248.

BIcKHAMm, J. W.; BAkER, R. ]J. (1979): Canalization model of chromosomal evolution. Bull. Carnegie
Mus. Nat. Hist. 13, 70-84.

CAPANNA, E. (1982): Robertsonian numerical varıation in anımal speciation: Mus musculus, an
emblematic model. In: Mechanisms of speciation. Ed by C. Barigozzi. New York: A. R. Liss. Pp.
155-177.

CRroSssEn, P. E. (1972): The Crossen procedure. Mamm. Chrom. Newsl. 13, 40.

CHIARELLI, B. A.; SARTI-CHIARELLI, M.; SHAFER, D. A. (1972): Chromosome banding with trypsin.
Mamm. Chrom. Newsl. 13, 44-45.

KELT, D. A.; Parma, R. E.; GALLARDO, M. H.; Cook, J. A. (1991): Chromosomal multiformity ın
Eligmodontia (Muridae, Sigmodontinae), and verification of the status of E. morganı. Z.
Säugetierkunde 56, 352-358.

Joun, B. (1981): Chromosome change and evolutionary change. In: Evolution and speciation. Ed. by
W.R. ATCHLEy and D. WooDRUFF. Cambridge: Cambridge Univ. Press. Pp. 23-51.

Lee, M. R.; ELper, F. F. B. (1980): Yeast stimulation of bone marrow mitoses for cytogenetic
investigations. Cytogenet. Cell Genet. 26, 36-40.

Levan, A.; FREDGA, K.; SANDBERG, A. (1964): Nomenclature for centromeric position on
chromosomes. Hereditas 52, 201-220.

Mares, M.; OJEDA, R. A.; BarQauzz, R. M. (1989): Guide to the mammals of Salta Province.
Norman, USA: Univ. Oklahoma Press.

MAsSsARINTI, A. I.; BARROS, M. A.; ORTELLS, M. O.; Reıc, ©. A. (1991): Chromosomal polymorph-
ism and small karyotypic differentiation in a group of Ctenomys species from Central Argentina
(Rodentia: Octodontidae). Genetica 83, 131-144.

MusseEr, G. G.; CARLETON, M. D. (1993): Family Muridae. In: Mammal species of the world. Ed. by.
D. E. Wırson and D. M. REEDER. Washington, London: Smithsonian Institution Press. Pp.
501-756.

ORTELLS, M. O.; Reıc, ©. A.; WEINBERG, R. L.; HURTADO DE CATALFO, G. E.; GENTILE DE
Fronza, T. M. L. (1989): Cytogenetics and karyosystematics of phyllotine rodents (Cricetidae,
Sigmodontinae). II. Chromosome multiformity and autosomal polymorphism in Eligmodontia.
Z. Säugetierkunde 54, 129-140.

PATHAR, $.; STOck, A. D. (1974): The X chromosome in mammals: karyological homology as
revealed by banding techniques. Genetics 78, 703-714.

PATTon, J. L.; SHERWOOD, $. W. (1983): Chromosome evolution and speciation in rodents. Ann.
Rev. Ecol. Syst. 14, 139-158.

PEARsoNn, O. P.; PATToNn, J. L. (1976): Relationships among south american phyllotine rodents based
on chromosome analysis, J. Mammalogy 57, 339-350.

Reıc, ©. A. (1986): Diversity patterns and differentiation of high Andean rodents. In: High Altitude
Tropical Biogeography. Ed. by. M. MoNASTERIO and F. VUILLEUMIER. New York: Oxford Univ.
PresssP pr 404 439.

— (1989): Karyotypic repatterning as one triggering factor in cases of explosive speciation. In:
Evolutionary biology of transıent unstable populations. Ed. by A. FontpeviLa. Heidelberg,
Berlin: Springer-Verlag. Pp. 246-289.

Ryper, O. A.; KumamoTo, A. T.; DURRANT, B. $.; BENIRSCHKE, K. (1989): Chromosomal diver-
gence and reproductive isolation in dik-diks. In: Speciation and its consequences. Ed. by D. OTTE
and J. A. EnDLER. Sunderland, Massachusetts: Sinauer. Pp. 208-228.

Sıres, J. W.; Morıtz, B. (1987): Chromosomal evolution and speciation revisited. Syst. Zool. 36,
153-174.

SPOTORNO, A. E. (1986): Systematics and evolutionary relationships of Andean phyllotine and
akodontine rodents. Unpubl. Diss., Univ. California, Berkeley, USA.

SPOTORNO, A.; BRum, N.; Dı Tomaso, M. (1987): Comparative Cytogenetics of South American
deer. Fieldiana (Zool.) 84, 473-483.

SPOTORNO, A. E.; VELoso, A. (1990): Flora and Fauna of the altiplano. In: Strategies in human
adaptation to a rigorous environment: the Aymara. Ed. by. W. J. SchuLL and F. ROTHHAMMER.
Dordrecht: Kluwer Acad. Publ. Pp. 19-32.

SPOTORNO, A. E.; WALKER, L. I. (1979): Analysis of chromosomal similarıty according G bandıng
patterns in four chilean species of Phyllotis (Rodentia, Cricetidae). Arch. Biol. Med. Exper. 12,
8399.

SUMNER, A. T. (1972): A simple technique for demonstrating centromeric heterochromatin. Exp. Cell
Res. 75, 304-306.

VUILLEMIER, B. S. (1971): Pleistocene changes in the fauna and flora of South America. Science 173,
771-780.

WALKER, L. 1.; SPOTORNO, A.E. (1993): Tandem and centric fusions in the chromosomal evolution of

308 A. E. Spotorno, J. Sufan-Catalan, and Laura I. Walker

the South American phyllotines of the genus Auliscomys (Rodentia, Cricetidae). Cytogen. Cell
Genet. 61, 135-140.
WALKER, L. I.; SPOTORNO, A. E.; FERNANDEZ-DOoNOoso, R. (1979): Conservation of whole arms
during chromosomal divergence of phyllotine rodents. Cytogen. Cell Genet. 24, 209-216.
WHITE, M. J. D. (1973): Anımal cytology and evolution. Cambridge: Cambridge Univ. Press.

Authors’ address: ANGEL E. SPOTORNO, JUAN SUFAN-CATALAN, and LAURA I. WALKER, Laboratorio de
Citogenetica Evolutiva, Departamento de Biologia Celular y Genetica, Facultad de
Medicina, Universidad de Chile, Casılla 70061 - Santiago 7, Chile

Z. Säugetierkunde 59 (1994) 309-316
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Influence of photoperiod and temperature on moulting processes
in Microtus brandti (Radde, 1861)

By ANNEGRET STUBBE and SABINE WIEGAND

Institute of Zoology, Martin-Luther-University, Halle/Saale, FRG

Receipt of Ms. 29. 1. 1993
Aceptance of Ms. 25. 1. 1994

Abstract

Voles were kept under different environmental conditions (light and temperature regime). The
duration of seasonal moults is closely connected with temperature changes. Nevertheless seasonal hair
changes seem to be fixed endogenously because they also occur under constant light and temperature
conditions. However, the endogenous rhythm can be forced by varyıng environmental conditions.

An intermediate moult was observed in Microtus brandti between the two winter furs. This could
be induced by shortening day length. An increase in haır density is a result of short day length and low
temperatures.

Introduction

The description of the course of moulting within mammal populations has often led to
conflicting information. There is a need to shed some light on these complex processes in
populations with regard to age structure and in the consequent differentiation between
dependent, so-called mature moults (STUBBE and WIEGAND 1994), and seasonal hair
changes in adult anımals induced by environmental conditions (e.g. day length and
temperature alterations, especially). The aim of the present study in the vole Microtus
brandti was to find a model for the moulting processes not only for the genus Microtus but
also for small rodents in general.

Material and methods

For species description, anımal husbandry and feeding, as well as staining treatment, see STUBBE and
WIEGAND (1994).

Hair density was determined by a modification of the method of BAake (1966). Six anımals were
investigated from each maintenance condition (MC). We used samples from the ventral side of winter
furs, taking three samples per anımal in each case. Pile and guard hairs were counted separately from
woolly hairs in an area of I mm’. During the present study the voles were kept under different
maintenance conditions (MC):

IMG 1er ale 2 lee ed, Ts ars)

MC 2: 20 # 2°C, L:D adapted to the natural photoperiod (L:D —, T = const.)
MC 3: TandL:D adapted to the naturally changing conditions (L:D, T —)
M@A2OEFP2ZEITL (permanentiliche).

In addition some data on pelts taken from animals caught between 1988 and 1990 in their natural
environment near Ulan-Bator (Mongolia) were analysed.
Results
Seasonal influence on moulting cycles

Under all MC except MC 4 we observed that moults occur throughout the entire year. To
assist our understanding, we determined so-called “moulting types” for characteristic hair
coat changes of spring- and autumn-born animals (see Tab. 1).

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5905-0309 $ 02.50/0

310 Annegret Stubbe and Sabine Wiegand
Table 1. Moulting types of spring and autumn generation

Moulting Transition from ... fur into... fur
type Spring generation Autmn generation

first into second juvenile plumage first insto second juvenile plumage

second juvenile into first mature second juvenile into first mature
coat (= summer fur) coat (= winter fur)

summer into first winter fur winter into summer fur
(MC 1: only summer into winter fur)

first into second winter fur summer into winter fur

(missed ın MC 1)

second winter into summer fur
(MC 1: winter into summer fur)

Figure 1 explains the moulting courses of the spring and autumn generations under
different MC throughout the year, including moult types. The bars show moult duration,
beginning with the earliest, and ending with the latest of all twenty anımals investigated at
each MC. The different starting points of moulting types I and Il are the result of several
different birth times. :

In spring-born anımals we found five moults, whereas voles born in late summer and
autumn only showed four hair changes during the first year of life. Moult courses in the
spring generation are more complex than those of the autumn generation. Regular overlaps
are typical of the spring generation but the moults of autumn-born anımals are always
temporally separated. Spring-born animals kept under MC 1 only showed four moults,

ZUIIIIIEESIHHII.
VUIHTRSSOHIKITIIIIDDDG

J FM TIAM J I AS © .N DE JE TEZMT ANSETZEN SEE SZENE

PZZZAIZZZA WZUZZ

EEE

MC 1 (L:D,T = constant )

MC 2 (L:D= varying, T = constant )
MC3 (L:D,T = varying )

MC4 (L:L,T = constant )

IUEN

ASMEZETAT SEO HN EDEN A

Fig. 1. Moulting courses of spring and autumn generation under different maintenance conditions
throughout the year including moulting types

ON DJ TE MANS

Influence of photoperiod and temperature on moulting processes 311

whereas voles under MC 2 and MC 3 passed through five hair changes and developed a
second winter pelt.

The duration of the moult into winter fur is abridged, and can be prolonged during the
transition from winter into summer coat under the influence of temperature (MC 3) in the
spring generation, whereas it is very short ın the autumn generation. Under constant
conditions (MC 1) we also found a winter and a summer coat. Thus, we can conclude that
moulting is fixed endogenously, but influenced by photoperiod and temperature. This
opinion is supported by the results of anımals kept under permanent lıght and at constant
temperature (MC 4). The moults of these adult voles were only observed for twenty
weeks, but they started at the same time as ın the anımals of MC 1-3 (see Fig. 1).

Duration of moulting

Figure 2 shows the duration of the moult under different MC. The time of change from the
first into second juvenile pelage ıs very short (average 10 days). Most of the time required is
in moulting from the second juvenile to the first mature coat (average 40 days). The
following seasonal moults (type III-IV) proceed at a faster rate again. Generally the
duration of the moult (all moult types) is longer for spring-born anımals than for the
autumn generation.

Altogether the hair change from the winter to summer fur occurs more rapidly than the
stronger temperature-dependent changes from the summer into winter coat (MC 3).
Under constant temperature conditions (with or without the influence of photoperiod;
MC 1 and MC 2) the moult from summer into winter fur takes longer than the change
from the winter into summer coat.

Comparative studies on pelts

Table 2 shows the number of pelts investigated in addition to the observations on living
anımals and the courses of seasonal moulting under different environmental conditions. All
pelts were derived from anımals older then 60 days, therefore we found exclusively

Moult duration (d)

Maintenance conditions

Generations

spring

I IE IV V I I I IV V I IE I IV V IVEEV

Moulting type

Fıg. 2. Average moult duration under different maintenance conditions

312 Annegret Stubbe and Sabine Wiegand

Table 2. Distribution of pelts and moulting patterns throughout the year under different
maintenance conditions (MC)

Month

January
February

March
April
May
June

July
August
September
October
November
December

Total

N = number of pelts investigated; n (%) = percentage of pelts with hair changes; W = pelts of voles
caught in the wild (Mongolıa). -

seasonal moult patterns. Moults appear throughout the year. Under MC 1 the moults were
spread comparatively unıformly over several months, showing minimum in February,
July, and November. Thus we can assume a slight increase in moulting anımals in spring
and autumn. However, its induction must be endogenous, because of the absence of any
“zeitgeber”. Under constant temperature and the influence of photoperiod (MC 2) we
found a high percentage of shedding trom November to February; the intermediate winter
moult occurring at this time. We had no winter pelts from field investigations, so the
interpretation of these data ıs difficult. Normally, the winter coat should be developed at
the end of September, and the summer fur at the beginning of June under the extreme
conditions of the natural environment; therefore the results in table 2 (W) correspond well
with anticipated conclusions. No data are available with regard to a second winter fur ot
voles living in the steppes of Mongolıa.

Hair density dependence on abiotic factors

The results (Tab. 3; 18 samples from sıx voles under each MC) show an increasing number
of hairs from MC 1 to MC 3. There ıs a significant difference (t-test, & = 0.05; WEBER
1972) between the number of guard and pile hairs under MC 1 and MC 3, MC 2 and MC3.
The voles kept under the influence of photoperiod and changing seasonal temperatures
developed the highest hair densities.
Animals from MC 1 and MC 2 were kept permanently at 20 #2°C. Here, the
differences in hair density may be dependent on light conditions. The regulation of fur
density could be a primary reaction to
Table 3. Average number of hairs/mm? and temperature alterations, whilst light regime
mean variation (winter fur) has an accelerating effect. Therefore, hair
density may be regarded as an indicator of

ME guard and woolly soraliharr changes in abiotic factors.
pile hairs hairs

Discussion

Publications commenting on the start and
cessation of seasonal moults are rare, de-

Influence of photoperiod and temperature on moulting processes 918

scriptions of the course of moulting in whole populations throughout the year being more
frequent. Here, the different age classes of anımals within populations are rarely consid-
ered.

RowseEmITT et al. (1975) described a peak of moulting activity in “autumn” (October
until January) and a “spring” moult between June and August in Microtus breweri ın
Massachusetts. KEMPER (1976) observed seasonal moults between February and April in
Pseudomys ın Australia, which agrees with the autumn moult of the temperate zone of the
northern hemisphere using the same point of time. BüHLow (1970) indicated an increasing
proportion of moulting Arvicola terrestris in September and again in April. For Cle-
thrionomys glareolus, LEHMANN (1958) noticed that the highest percentage of moulting
animals occurred in March and October, and BAkE (1981) found a peak between March
and May in Microtus arvalıs.

We were able to show that, in Microtus brandti especially, the autumn hair change ıs
strongly linked to definite times, whereas the time of the spring moults can partly continue
until June. Clear differences in moult duration, dependent on sımulated photoperiod and
temperature changes, were established for Microtus brandti. Thus, the coat change to
summer fur occurs always more slowly than that to winter fur. This ıs evident because of
the necessity to end the period of higher energy turnover quickly before the onset of
winter. Also, the anımals need a dense winter coat within a short time ıf temperatures
decrease suddenly; BAkE (1981) noted the faster occurrence of the moult in autumn,
compared with spring.

Another problem ist the interpretation of all moulting processes wıthin a population
throughout the year. Most authors have observed that moulting anımals are distributed
comparatively evenly throughout the year, but the explanations of this phenomenon are
very different. BECKER (1952) doubted that there was any seasonal influence. He stated
that the start of a single moult cycle was dependent on individual maturation states. As
wild rat populations are composed of anımals of different ages, he saw no parallel between
hair changes and season. The same opinion was expressed by LANGENSTEIN-IssEL (1950)
and OLIVIERA et al. (1992). However, ıf we see each individual of a population as passing
through the birth-time and age-dependent first and second juvenile moults (see STUBBE and
WIEGAND 1994) — and this is proven for Rattus norvegicus for nearly the whole year - and
only then occur seasonal moults, we obtain the appearance of regularly distributed
moulting over all months of the year within a population. Thus the reference of BECKER
(1952) to the birth date distribution over the entire year is the key for the year-long
occurrence of moulting.

STEIN (1960) considered a type of temporary, indeterminate, but endogenously induced
moulting sequence as the original, prımary type, but he presumed a secondary parallel
between season and moulting cycles. Vıro and KoskeELA (1978) found relationships
between the 0 °C-isotherm, the occurrence of compact snow cover, and the development
of winter fur in Micromys minutus. They reported the completion of the winter pelt for all
the anımals in a population, one month before attaining the 0 °C-isotherm. LEHMANN
(1958) observed dependencies ot the temporal course of moulting on habitat: he noticed
that the hair change of epigean-living anımals was more rapid than that of voles and moles,
in which it seemed slow and “sluggish”.

RowsEMITT et al. (1975) maintained that Microtus breweri shows a dependence of moult
cycles on reproductive activity. In Apodemus sylvaticus, RÖBEN (1969) described a direct
seasonal link, but in this species he missed any summer coat, so that the summer-winter
rhythm was broken. In our study we have only integrated anımals born in March and the
first halt of April (spring generation) or in the second half of September until first halt of
November (autumn generation). Only individual mean variations in moult times lead to
overlaps by successive moults. In natural populations, anımal birth dates may occur from
February until October, thus, both different age classes, and numerous moulting states

314 Annegret Stubbe and Sabine Wiegand

will be found. Consequently, we can merely state that first and second juvenile moults
(often called mature moults) are strongly age dependent, whereas the hair changes of adult
anımals that follow are more or less subject to season (seasonal moults). We envisage a
primarily endogenous rhythm, which can be influenced by environmental factors. STEIN
(1954), BüHLow (1970) and BAkE (1981) gave sımilar explanations.

Numerous publications have refered to light regimes as releasers of moulting. Thus, by
shortening daylıght duration, the development of real winter fur was always observed
(BıssonnETTE 1935; BissonNETTE and BALeY 1944; HarvEY and MACFARLANE 1958;
BELJAEVv et al. 1964; AL-KHATEEB and JOHNnsoNn 1971; Dusy and Travıs 1972; HOFFMANN
1978; LyncH and GENDLER 1980; RoUGEOT et al. 1984; Rmopes 1989). Early studies
doubted the influence of temperature; BissonNETTE and Wırson (1939) reported a
minimal influence of temperature changes on moulting processes. HEATH and LyncH
(1983) kept Peromyscus lencopus under short-day conditions in both warm and cold
environments. Both anımal groups grew a winter coat, but the proportion of moulting
mice was higher at lower temperaturs. A regulation of moult speed and variation by
temperature was found by Rust (1962) in short-tailed weasels. Furthermore, JACKEs and
Watson (1975) described a statistical correlation between moult, light duration, tempera-
ture, and snow cover by parameter-free correlation coefficients in Lepus timıdus. In
Mustela vison, BELJAEV (1976) reported light to be of the greatest importance in the
moulting processes. Our investigations on Microtus brandti lead us to the conclusion that
there ıs an endogenously fixed rhythm of moult cycles driven by light duration and
temperature. This assumption is validated by the fact that anımals kept under long-day
conditions and at constant temperature (MC 1) develop both a summer and a winter coat.
Otherwise, shortening light duration seems to be the releaser of the intermediate winter
hair change. This moult is missed under constant light and temperature conditions (MC 1).
We believe that temperature and temperature alterations, in particular, have regulatory
effects on moult duration. Voles kept under a seasonal temperature regime (MC 3) display
a moult duration limited to the minimum time. Moulting in these anımals is better adapted
to season than is that of anımals living without temperature alterations. Anımals kept
under constant temperature conditions show a higher temporal variability at the start and
end of moulting. Also, the results from voles kept in permanent lıght and at constant
temperature (MC 4) support the hypothesis of an endogenous control of the moulting
processes. These anımals showed the same moult course as anımals living under varying
environmental conditions.

In our study moulting processes within a single litter seemed to be synchronized; in the
natural environment a “priming” effect of some plant compounds, especially in spring,
may be possible (SANDERS et al. 1981).

In the case of seasonally dependent aberrations in fur density, two theories exist:

1. The complete replacement of old fur (SrErIN 1960, Microtinae)

2. The breakoff of hair points of the winter coat and partial change of woolly into pile

hairs (IvANTER et al. 1985, Soricidae).

More is known about the induction of fur density aberrations, therefore we observed the
minimum hair density in voles kept under constant light and temperature. Shortening day
length leads to increases in fur density, which reaches its highest level if a temperature
decrease is added. Significant differences in total hair numbers are noted between constant
environmental conditions (MC 1) and a simulated seasonal course of photoperiod and
temperature (MC 3).

Furthermore, some authors have found several other factors to influence fur density.
HATTLINGER (1968) described a correlation between hair density and body length ın
Apodemus, and SEALANDER (1972) noticed a 78% increase of hair weight in winter ın
Clethrionomys rutilus. A higher winter fur density was observed by ArL-KHATEEB and
Jornson (1971) in Microtus agrestis. Here, the voles were kept under natural and long day

Inflnence of photoperiod and temperature on moulting processes 315

conditions. “Long-day-anımals” did not develop a winter coat. The regulation of fur
density by day length has also been described in Sorex araneus (BoRrowskı 1958), ferrets
(Harvey and MACFARLANE 1958), Micromys minutus (Vıro and KoskELA 1978), Microtus
arvalıs (BAKE 1981), Phodopus sungorus (MAsupA and OısHı 1988), Microtus pennsyl-
vanıcus (RHODES 1989) and varıous other mammals (JoHunson 1984). While these studies
attrıbute a decisive influence to light, our investigations show that temperature also has a
regulating function upon fur density.

Acknowledgements

This publication is no. 226 of the “Results of Mongolian-German Biological Expeditions since 1962”.
We thank Dr. H. GrIFFITHS from the Departement of Genetics of the University of Leeds for
linguistie corrections.

Zusammenfassung

Der Einfluß von Photoperiode und Temperatur auf Fellwechselprozesse bei
Microtus brandti (Radde, 1861)

Die Steppenwühlmäuse wurden bei unterschiedlichen Umweltbedingungen (Licht- und Temperatur-
regime) gehalten und der Haarwechsel an lebenden Tieren über ein Jahr verfolgt. Die Dauer der
saısonalen Härungen ist temperaturabhängig, sie verkürzt sich bei schnellem Absinken der Tempera-
turen deutlich. Saisonale Fellwechsel scheinen endogen fixiert zu sein, da sie auch bei konstanter
Temperatur und gleichbleibender Tageslänge auftreten. Allerdings wird die endogene Rhythmik
durch varıierende Umweltbedingungen verstärkt.

Bei Verkürzung der Tageslänge kann eine intermediäre Härung zwischen zwei Winterfellen
ausgelöst werden. Die Haardichte erhöht sıch signifikant bei Kurztag und niedrigen Temperaturen.

References

AL-KHATEEB, A.; JOHNSON, E. (1971): Seasonal changes of pelage in the vole (Microtus agrestis). II:
The effect of daylength. Gen. Comp. Endocrinol. 16, 229-235.

BAKE, U. (1966): Untersuchungen zum Haarkleidaufbau und zum Haarwechsel der Feldmaus,
Microtus arvalıs (Pallas, 1779) und der Rötelmaus, Clethrionomys glareolus (Schreber, 1780).
Diss., Univ. Halle.

BAkKE, U. (1981): Beitrag zum Aufbau des Haarkleides und zum Haarwechsel der Feldmaus Microtus
arvalıs (Pallas, 1779). Säugetierkdl. Inf. 5, 3-49.

BECKER, K. (1952): Haarwechselstudien an Wanderratten (Rattus norvegicus Erzl.) Biol. Zbl. 71,
626-640.

BELjJaev, D. K. (1976): Teoreticeskie osnovy ı prakticeskie aspekty ispolzovanıja fotoperiodisma v
razvedenii pusnych zverej. Svetovoj faktor v povyzenii pusnych zverej. Moskva: Nauka.

BELJAEv, D. K.; Urkın, L. G.; Kurıckov, B. A. (1964): Vlijanıe svetogo reZima na razvitie
mechogopokrova u norok (Mustela vıson SCHR). Izv. sıb. otd. Akad. Nauk. SSSR 4, 91-100.

BIssoNNETTE, T. H. (1935): Relations of haır cycles in ferrets to changes in the anterior hypophysis
and to light cycles. Anat. Rec. 63, 159-168.

BissoNNETTE, T. H.; Barey, E. E. (1944): Experimental modification and control of molts and
changes of coat-color in weasels by controlled lighting. Ann. New York Acad. Sci. 45, 221-260.

BIsSONNETTE, T. H.; Wırson, E. (1939): Shortening daylight periods between May 15 and September
12 and the pelt cycle of the mink. Science 89, 418-419.

Borowskt, $. (1958): Varıiations in density of coat during the life cycle of Sorex araneus aranens L.
Acta theriol. 2, 286-289.

BÜHLow, E. (1970): Untersuchungen über den Haarwechsel bei Schermäusen, Arvicola terrestris (L.,
1758). Zool. Anz. 184, 18-32.

Dusgy, R. T.; Travıs, H. F. (1972): Photoperiodic control of fur growth and reproduction in the Mink
(Mustela vison). J. Exp. Zool. 182, 217-226.

HAITLINGER R. (1968): Seasonal varıation of pelage in representatives of the genus Apodemus Kaup,
1829, found in Poland. Zool. Polon. 18, 329-345.

Harvey, N. E.; MACFARLANE, W. V. (1958): The effects of day length on the coat-shedding cycles,
body weight, and reproduction of the ferret. Aust. J. Biol. 11, 187-199.

HEATH, H. W.; LyncH, G. R. (1983): Intraspecific differences in the use of photoperiod and
temperature as environmental cues in White-footed mice Peromyscus leucopus. Phys. Zool. 56,
506-512.

316 Annegret Stubbe and Sabine Wiegand

HorFrMann, K. (1978): Effects of short photoperiods on puberty, growth and moult in the
Djungarıan hamster (Phodopus sungorus). J. Reprod. Fert. 54, 29-35.

IVANTER, E. V.; IVANTER, T. V.; Tumanov, I. L. (1985): Adaptivnye osobenosti melkich mlekopita-
juScich. Leningrad: Nauka.

Tees A. D.; Watson, A. (1975): Winter whitening of Scottish mountain hares (Lepus timidus
scoticus) ın relation to daylength, temperature and snow-lie. J. Zool. (London) 176, 403409.

Jounson, E. (1984): Seasonal adaptive coat changes in mammals. Acta Zool. Fennici 171, 7-12.

KEMPER, C. M. (1976): Maturational and seasonal moult in the New Holland mouse, Pseudomys
novaehollandiae. Aust. Zool. 19,017

LANGENSTEIN-IssEL, B. (1950): Biologische und ökologische Untersuchungen über die Kurzohrmaus
(Pitymys subterraneus DeSelys-Longchamps). Z. Pfanzenbau und Pflanzenschutz 1, 145-183.

LEHMANN, E. von (1958): Zum Haarwechsel deutscher Kleinsäuger. Bonn. zool. Beitr. 9, 10-23.

LyncH, G. R.; GENDLER, $. L. (1980): Multiple responses to different photoperiods occur in the
mouse, Peromyscus leucopus. Oecologıa (Berlin) 45, 318-321.

Masupa, A.; OısHı, T. (1988): Effects of photoperiod and temperature on body weight, food intake,
food storage, and pelage color in the Djungarıan hamster, Phodopus sungorus. J. Exp. Zool. 248,
133-139.

OLIVIERA, J. A. De; Lorını, M. L.; Persson, V. G. (1992): Pelage varıation in Marmosa incana
(Didelphidae, Marsupialia) wıth notes on taxonomy. Z. Säugetierkunde 57, 129-136.

RHonpss, D. H. (1989): "The influence of multiple photoperiods and pinealectomy on gonads, pelage
and body weight in male Meadow voles, Microtus pennsylvanıcus. Comp. Biochem. Physiol. 93 A,
445—449.

RöBeEn, P. (1969): Ein für europäische Kleinsäuger neues Haarwechselschema. Zur Gattung
Apodemus ım Rhein-Neckar-Gebiet. Säugetierkdl. Mitt. 17, 3142.

ROUGEOT, J.; ALLAIN, D.; MARTINET, L. (1984): Photoperiodic and hormonal control of seasonal coat
changes in mammals with special reference to sheep and mink. Acta Zool. Fennica 171, 13-18.
RowsEMITT, C.; Kunz, T. H.; Tamarın, R. H. (1975): The timing and patterns of molt in Microtus

breweri. Occ. Papers Mus. Nat. Hist. Univ. Kansas 34, 1-11.

Rust, C. C. (1962): Temperature as a moditying factor in the spring pelage change of short-tailed
weasels. J. Mammalogy 43, 323-328.

SANDERS, E. H.; GARDNER, P. D.; BERGER, P. J.; Nesus, N. C. (1981): 6-Methoxybenzoxazolinone:
A plant derivative that stimulates reproduction in Microtus montanus. Science 214, 67-69.

SEALANDER, J. A. (1972): Circum-annual changes in age, pelage characteristics and adıpose tissue in
the Northern red-backed vole, Clethrionomys rutilus dawsoni, ın Interior Alaska. Acta theriol. 17,
1-24.

Stein, G. H. W. (1954): Materialien zum Haarwechsel deutscher Insectivoren. Mitt. Zool. Mus.
Berlin 30, 12-34.

— (1960): Zum Haarwechsel der Feldmaus (Microtus arvalıs Pallas, 1779) und weiterer Muroidea.
Acta theriol. 3, 27-44.

STUBBE, A.; WIEGAND, $. (1994): Ontogenesis of pelage and course of moulting in Microtus brandti
(Radde, 1861). Z. Säugetierkunde 59, 199-208.

Vıro, P.; KoskELa, P. (1978): Moult topography, moulting and the structure of the fur in the Harvest
mouse. Acta theriol. 23, 503-517.

WEBER, E. (1972): Grundriß der biologischen Statistik. Jena: G. Fischer Verlag.

Authors’ address: Dr. ANNEGRET STUBBE and Dipl.-Biol. SABINE WIEGAND, Institut für Zoologie,
Martin-Luther-Universität Halle-Wittenberg, Domplatz 4, D-06099 Halle/Saale,
FRG

Z. Säugetierkunde 59 (1994) 317-320
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

\WISSEINS@ ELAUSTEIEIL EN TRUNVZMIIFTESTORUNNKE

Chromosomes of two rare species of neotropical mammals:
Southern pudu (Pudu puda) and Bush dog (Speothos venaticus)

By A. SCHREIBER and R. DmocH

Zoologisches Institut I der Universität Heidelberg and Zoologischer Garten, Frankfurt, FRG

Receipt of Ms. 10. 2. 1994
Acceptance of Ms. 13. 4. 1994

For several years European zoos (Europäisches Erhaltungszuchtprogramm) have coordi-
nated captive breeding projects for two neotropical mammals, the southern pudu (Pudu
puda) and bush dog (Speothos venaticus), including specimens imported from various,
mostly unknown geographic orıgins. Mammalıan species are frequently comprised of
chromosomally polymorphic or polytypic populations, thus cytogenetic investigations are
a useful tool when establishing breeding populations in zoos (e.g. RYDEr et al. 1989;
DATHE et al. 1989; SCHREIBER et al. 1993). Since both pudus and bush dogs are faırly
delicate mammals to handle for blood sampling, we profited from rare occasıons of blood
sampling to check whether the initial phase of the captive conservation breeding project
had led to a hybridization of unrecognized cytotypes.

The southern pudus kept in European zoos descended from wild captured founders
imported in 1966, 1970, and 1972 from unidentitied areas of the comparatively small native
range of this species. The exact geneology of the four investigated pudus kept at the
Cologne Zoo is unknown. Of the sıx karyotyped bush dogs kept at the Frankfurt Zoo, one
male is a first-generation offspring of wild captured parents imported from French Guiana.
The other five investigated specimens descended from seven imported anımals (with
varying percentages of founder representation): They possessed individuals from Paraguay
(two wild captured founders) and French Guiana (three wild captured founders) in their
ancestry, plus two ancestors acquired from the animal trade without proper information as
to their origins. Three-day cultures of peripheral Ilymphocytes from heparinized blood
samples were grown in Chromosomen-Medium B (Seromed, Berlin). After 3-12 hours
incubation with colcemide, the preparation of chromosomes was performed as previously
described (SCHREIBER et al. 1993). Per specimen, 10-20 metaphases were photographed
and analysed.

The one female and three male karyotyped southern pudus contained a complement of
2n = 70 (NF = 74). The X chromosomes were large metacentrics, whereas all autosomes,
with the exception of one pair, were acrocentrics. The largest acrocentric pair of autosomes
possessed conspicuous satellite appendages (Fig. 1).

The three female and three male investigated bush dogs showed an identical karyotype
of 2n = 74 (NF = 76). Other than the relatively large female sex chromosomes, all
autosomes were acrocentrics (Fig.2). Ihe Y chromosome was a small metacentric.

The karyotype monomorphism found in both sample series does not indicate hybridi-
zation of cytotypes during the breeding project. Previous investigations reported karyoty-
pes from one male (KoULIsScHER et al. 1972), and from one male and one female southern
pudus to be comprised of 70 chromosomes each (NF = 74) (SPOTORNO and FERNANDEZ-

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5905-0317 $ 02.50/0

318 A. Schreiber and R. Dmoch

IMIWEHIBIEIEY
ABIEEIEZIEEIBIEE
u rn wu
Me U mu EEE Ey u wu KW
aan nn nA HN no

N.

Fig. 1. G-stained karyotype of a male southern pudu (P. puda)

An AA AN nA HR ON AA
an Ar na na Ark Am ar

AA LAN AA Yo de AN
GR.) aA a an ww www. ww‘

ee ww.‘ AB AR LCDR BCE an

Fig. 2. G-stained chromosomes of a female bush dog (Speothos venaticus)

Chromosomes of Southern pudn and Bush dog 99

Donoso 1975), findings that are extended to a sample size of seven pudus by the present
report. However, cervids, including neotropical species, are notorious for their cytogene-
tic variability (NEITZEL 1982, 1987). A single northern pudu (Pudu mephistophiles)
contained 69 chromosomes, including two acrocentrics corresponding to a metacentric,
and presumably represented the heterozygote of a translocation polymorphism (NEITZEL
1979). Some authorities include pudus in the genus Mazama, the brocket deer (HERSH-
KoVITZ 1982; CzErNY 1987), whose species display elevated cytogenetic varıation, the
taxonomic background of which remains to be clarıfied. In single individual deer classified
as Mazama americana, the following karyotypes were found: 2n = 68, NF = 74 (TayLor et
al. 1969; unknown geographic origin of specimen), 2n = 49/50, NF = 72 (JORGE and
BENIRSCHKE 1977, M. americana temana), 2n = 54 plus two B-chromosomes (NEITZEL
1982, hybrid zoo specimen with parents imported from two regions in Paraguay), and 2n =
52 (NF = 56) plus 4-5 B-chromosomes (NEITZEL 1987, female originating from Paraguay).
Interbreeding of the latter cytotype with a male Mazama gonazoubira (2n = 70) resulted in
presumably infertile offspring with 2n = 61 (plus two B-chromosomes) (NEITZEL 1987).
While M. americana ıs so polymorphic, M. gomazoubira retains a very conservative
karyotype of 2n = 70 (NF = 70), considered to be ancestral for Cervidae (NEITZEL 1982).
The bewildering chromosomal varıation in the closely related “species” Mazama
americana, which appears to have evolved within only 2 million years since the presumed
colonization of South America (NEITZEL 1987), and the odd-numbered karyotype of the
only Pudu mephistophiles investigated to date (NEITZEL 1979), suggest that the present
sample sıze of seven karyotyped southern pudus is insufficient to exclude the existence of
local chromosomal populations in this species. Pudus are solitary, sedentary deer (HERSH-
KOVITz 1982; CzZERNY 1987) with probably quite limited interbreeding between regional
stocks. Cervids with similar lifestyles, e.g. muntjacs (Muntiacus), have fixed so many
chromosomal mutations between populations that they represent the ungulates with the
most extensive cytological varıation (NEITZEL 1982). In Bovidae, dik-diks (Rhynchotra-
gus) provide another example of extensive regional chromosomal diversity ın a philopatric,
territorial ruminant with obviously limited gene flow between stocks (DATHE et al. 1989;
RyDEr et al. 1989).

Bush dogs are a rare species which avoıd human settlements and are not frequently
observed (THORNBACK and JENKINS 1982; GINSBERG and MAacDonarD 1990). Details of
their taxonomy are not well known. CABRERA (1957) recognized three subspecies, $.v.
venaticus, S.v. wingei and S.v. panamensis, which range widely, though sparsely through
evergreen rain forests of tropical South America. However, there is no revision of these
subspecies, and their exact distributional ranges are unknown. $.v. venaticus ıs believed to
inhabit both Guiana and Paraguay, the two countries from where the documented founder
anımals of the zoo population originated. WURSTER-HiLL and CENTERWALL (1982)
karyotyped one bush dog of unknown geographic origin and found 2n = 74. Canidae, like
deer, include species wıth microchromosomes or B-chromosomes (e.g. Nycterentes,
Vulpes), and ın Nyctereutes procyonides, the racoon dog, chromosome numbers range
from 2n = 42 to 2n = 56/57, the species including chromosomal mosaic individuals, and B-
chromosomal polymorphism (WUuRSTER-Hirr et al. 1986). WURSTER-HiLL et al. (1988)
encountered as many as eight Robertsonian translocation differences between N.p. procy-
onides from China (2n = 54 plus B-chromosomes), and N.p. viverrinus from Japan (2n =
38 plus B-chromosomes). Yosıpa and WAapAa (1985) reported numerous variable
chromosomal fissions from the racoon dog population living in Central Honshu, Japan.
We have no evidence for B-chromosomes in bush dogs.

320 A. Schreiber and R. Dmoch

Acknowledgements

Colleagues from the Cologne and Frankfurt Zoos supplied us with blood samples. This study was
financially supported by a grant by the Cologne Zoo (Prof. Dr. G. NOGGE) to promote investigations
on species bred within the Europäisches Erhaltungszuchtprogramm.

References

CABRERA, A. (1957): Catalogo de los maniferos de America del Sur. Revista del Museo Argentino do
Cıencıas Naturales Bernadivo Rivadıvıa 4, 1-307.

CZERNY, $. (1987): Die Spießhirsche und Pudus. Wittenberg Lutherstadt: A. Ziemsen.

DATHE, H.; BELITZ, B.; KÖRNER, H.; Pırra, C. (1989): Ein Beitrag zur Kennzeichnung zytogenetisch
unterschiedener Formen von Kirk-Dikdiks, Madoqua (Rhynchotragus) kirki (Günther, 1880).
Zool. Garten N.F. 59, 21-26.

DmochH, R. (1990): Internationales Zuchtbuch für den Waldhund, Speothos venaticus (Lund, 1842).
Frankfurt: Zoologischer Garten.

GINSBERG, J. R.; MAcDonALD, D. W. (1990): Foxes, wolves, jackals, and dogs. An action plan for the
conservation of canıds. Gland: IUCN.

HERSHKOVITZ, P. (1982): Neotropical deer (Cervidae). Part I. Pudus, genus Pudu Gray. Fieldiana,
New Series 11, 1-86.

JORGE, W.; BENIRSCHKE, K. (1977): Centromeric heterochromatin and G-banding of the red brocket
deer, Mazama americana temana (Cervoidea, Artiodactyla) with a probable non-Robertsonian
translocation. Cytologia 42, 711-721.

KOULISCHER, L.; TySKENS, J.; MORTELMAnS, J. (1972): Mammalian cytogenetics. VII. The
chromosomes ot Cervus canadensis, Elaphurus davidianus, Cervus nippon, and Pudu pudu. Acta
Zool. Pathol. Antv. 56, 25-30.

NEITZEL, H. (1979): Chromosomenevolution in der Familie der Hirsche. Bongo 3, 27-38.

— (1982): Karyotypenevolution und deren Bedeutung für den Speziationsprozeß der Cerviden. PhD
Thesis, Freie Universität Berlin.

— (1987): Chromosome evolution of Cervidae: Karyotypic and molecular aspects. In: Cytogenetics.
Ed. by G. Oge and A. Basrer. Berlin, Heidelberg: Springer-Verlag, Pp. 90-112.

RyDER, ©. A.; KUMAMOTO, A. T.; DURRANT, B. $.; BENIRSCHKE, K. (1989): Chromosomal diver-
gence and reproductive isolation in dıkdiks. In: Speciation and its consequences. Ed. by D. OTTE
and J. A. EnDLER. Sunderland (Mass.): Sinauer Ass. Pp. 208-225.

SCHREIBER, A.; HELD, M.; NöTzoLp, G. (1993): Molecular and chromosomal evolution in anoas
(Bovidae: Bubalus spec.). Z. Zool. Syst. Evolut.-forsch. 31, 64-79.

SPOTORNO, O.; FERNANDEZ-DoNoso, F. (1975): The chromosomes of the Chilean dwarf-deer
“pudu”, Pudu pudn (Molina). Mamm. Chromosome Newsletter 16 (1), 17-18.

TAyLoR, K. M.; HUNGERFORD, D. A.; SNYDER, R. L. (1969): Artiodactyl mammals: Their chromo-
some cytology in relation to patterns of evolution. In: Comparative mammalıan cytogenetics. Ed.
by K. BENIRSCHKE. New York: Springer-Verlag. Pp. 346-356.

THORNBACK, J.; JENKINS, M. (1982): The IUCN mammal red data book. Part I: Threatened
mammalıan taxa of the Americas and the Australasian zoogeographic region (excluding Cetacea).
Gland: IUCN.

WURSTER-Hırı, D. H.; CENTERWALL, W.R. (1982): The interrelationships of chromosome banding
patterns in canıds, mustelids, hyena, and felids. Cytogenet. Cell Genet. 34, 178-192.

WURSTER-HiLL, D. H.; WARD, ©. G.; Davıs, B. H.; PARK, J. P.; Moyzıs, R. K.; MEYNE, ]J. (1988):
Fragile sites, telomeric DNA sequences, B chromosomes, and DNA content in racoon dogs.
Nyctereutes procyonides, with comparative notes on foxes, coyote, wolf, and racoon. Cytogenet.
Cell. Genet. 49, 278-281.

WURSTER-HiLı, D. H.; Warn, ©. G.; Kapa, H.; WHITTEMORE, $. (1986): Banded chromosome
studies and B-chromosomes in wild-caught racoon dogs, Nyctereutes procynides viverrinus.
Cytogenet. Cell Genet. 42, 85-93.

Yosıpa, T. H.; Wapa, M. Y. (1985): Cytogenetic studies on the Japanese racoon dog. Proc. Jap.
Acad. B 61, 451-454.

Anschriften der Verfasser: Dr. ARND SCHREIBER, Zoologisches Institut I der Universität, Im
Neuenheimer Feld 230, D-69120 Heidelberg, FRG; Dr. RÜDIGER
DmochH, Zoologischer Garten Frankfurt, Alfred-Brehm-Platz 16, D-
60316 Frankfurt, FRG

Erscheinungsweise und Bezugspreis 1994: 6 Hefte bilden einen Band. Jahresabonnement Inland:
DM 378,- zuzüglich DM 13,80 Versandkosten; Jahresabonnement Österreich: 65 2949,- zuzüg-
lich 6S 164,- Versandkosten; Jahresabonnement Schweiz: sfr 364,— zuzüglich sfr 21,— Versand-
kosten; Jahresabonnement EG-Binnenmarkt-Länder mit USt-ID-Nr.: DM 353,27 zuzüglich
DM 19,63 Versandkosten; Jahresabonnement EG-Binnenmarkt-Länder ohne USt-ID-Nr. und
Drittländer: DM 378,- zuzüglich DM 21,- Versandkosten. Das Abonnement wird zum Jahres-
anfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens
am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhand-
lung oder bei der Verlagsbuchhandlung Paul Parey GmbH & Co. KG, Spitalerstraße 12,
D-20095 Hamburg, Bundesrepublik Deutschland, bestellt werden. Die Mitglieder der „Deut-
schen Gesellschaft für Säugetierkunde“ erhalten die Zeitschrift unberechnet im Rahmen des
Mitgliedsbeitrages.

Z. Säugetierkunde 59 (1994) 5, 257-320

Einem Teil dieser Auflage liegt der Bericht 1993 des Schatzmeisters der DGS bei.

ÖL 6),321-384, Dezember 1994 ISSN 0044-3468 C 21274 F
460

ZEITSCHRIFT FÜR
SAUGETIERKUNDE

INTERNATIONAL JOURNAL
DF MAMMALIAN BIOLOGY

rgan der Deutschen Gesellschaft für Säugetierkunde

Juri, H.; Schönherr, R.: Neue Nachweise der Schabrackenspitzmaus (Sorex coronatus) in Baden-Württemberg
durch Polyacrylamidgel-Elektrophorese. — New records of the Jersey shrew (Sorex coronatus) in Baden-
Württemberg by polyacrylamide gel electrophoresis 321

\os, A. C.: Reproductive performance of the Red fox, Vulpes vulpes, in Garmisch-Partenkirchen, Germany,
1987-1992. — Die Reproduktion des Rotfuchses (Vulpes vulpes) im Landkreis Garmisch-Partenkirchen,
Deutschland, 1987-1992 326

urves, M. G.; Kruuk, H.; Nel, J. A. J.: Crabs Potamonautes perlatus in the diet of Otter Aonyx capensis and Water
mongoose Atilax paludinosus in a freshwater habitat in South Africa. — Krabben, Potamonautes perlatus, in der
Nahrung von Kapfingerotter, Aonyx capensis, und Wassermanguste, Atilax paludinosus, in einem Süßwasserha-

bitat in Südafrika 332
Jaucy, F.: Fates of fossorial Water voles, Arvicola terrestris, as revealed by radiotelemetry. — Erfassung von
. Einzelschicksalen bei Ostschermäusen (Arvicola terrestris) mit Hilfe der Radiotelemetrie 342
fyStufek, B.; Filippucci, Maria Grazia; Macholän, M.; Zima, J.; Vujosevic, M.; Simson, S.: Does Microtus majori
oceur in Europe? — Kommt Microtus majori in Europa vor? 349

zallardo, M. H.; Köhler, Nelida: Demographic changes and genetic losses in populations of a subterranean rodent
(Ctenomys maulinus brunneus) affected by a natural catastrophe. -— Demographische Veränderungen und
Verluste von genetischer Variabilität bei Populationen des grabenden Nagers (Ctenomys maulinus brunneus) als
Folge einer Naturkatastrophe 358
guilera, M.; Corti, M.: Craniometric differentiation and chromosomal speciation of the genus Proechimys (Ro-
| dentia: Echimyidae). —- Kraniometrische Differenzierung und chromosomale Artbildung in der Gattung Proechi-
_ mys (Rodentia: Echimyidae) 366

lissenschaftliche Kurzmitteilungen

leijaard, E.; Bree, P. J. H., van: Cases of dental malocclusion in populations of Red foxes (Vulpes vulpes) in the
| state of Victoria, Australia. — Fälle von Gebißanomalie in Populationen des Rotfuchses (Vulpes vulpes) aus dem
Staat Victoria, Australien 378

eumer, J. W. F.: Eliomys (Hypnomys) onicensis nomen novum, to replace the homonym Hypnomys intermedius
Reumer, 1981 (Rodentia: Gliridae) from Majorca. - Eliomys (Hypnomys) onicensis nomen novum als Ersatz für
| das Homonym Hypnomys intermedius Reumer, 1981 (Rodentia: Gliridae) von Mallorca 380

itteilungen der Gesellschaft ; a

verlag Paul Parey Hamburg n

FIERINÜUSEEBERYEDIERORS

P. J. H. van BrEE, Amsterdam - W. FIEDLER, Wien - H. Frick, München - G. B.

HARTL, Wien — W. HERRE, Kiel - R. HUTTERER, Bonn - H.-G. Krös, Berlin - H.-].

Kuhn, Göttingen — E. Kurzer, Tübingen - W. MAser, Tübingen — J. NIETHAMMER,

Bonn - O. Anne E. Rasa, Bonn — H. ReıcHstein, Kiel - M. Röhrs, Hannover -

H. SCHLIEMANnN, Hamburg — D. STARcK, Frankfurt a. M. - E. THEnıus, Wien - P. Vo-
GEL, Lausanne -— H. Wınkıng, Lübeck

SICEIRTEIEIEHTUNIGYE DEKOR ONE

D. Kruska, Kiel - P. LANGER, Gießen

This journal is covered by Biosciences Information Service of Biological Abstracts, and by Current Con-
tents (Series Agriculture, Biology, and Environmental Sciences) of Institute for Scientific Information

Die Zeitschrift für Säugetierkunde veröffentlicht Originalarbeiten und wissenschaftliche Kurzmittei-
lungen aus dem Gesamtgebiet der Säugetierkunde, Besprechungen der wichtigsten internationalen
Literatur sowie die Bekanntmachungen der Deutschen Gesellschaft für Säugetierkunde. Verantwort-
licher Schriftleiter im Sinne des Hamburgischen Pressegesetzes ıst Prof. Dr. Dieter Kruska.

Zusätzlich erscheint einmal im Jahr ein Heft mit den Abstracts der Vorträge, die auf der jeweiligen
Hauptversammlung der Deutschen Gesellschaft für Säugetierkunde gehalten werden. Sie werden als
Supplement dem betreffenden Jahrgang der Zeitschrift zugeordnet. Verantwortlich für ihren Inhalt
sind ausschließlich die Autoren der Abstracts.

Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z. Hd. Prof. Dr. Dieter
Kruska, Institut für Haustierkunde, Biologiezentrum der Christian-Albrechts-Universität, Am
Botanischen Garten 9, D-24118 Kiel, Bundesrepublik Deutschland. Für die Publikation vorgese-
hene Manuskripte sollen gemäß den „Redaktionellen Richtlinien‘ abgefaßt werden. In ihnen
finden sich weitere Hinweise zur Annahme von Manuskripten, Bedingungen für die Veröffent-
lichung und die Drucklegung, ferner Richtlinien für die Abfassung eines Abstracts und eine
Korrekturzeichentabelle. Die Richtlinien sind auf Anfrage bei der Schriftleitung und dem Verlag
erhältlich.

Sonderdrucke: Anstelle einer Unkostenvergütung erhalten die Verfasser von Originalbeiträgen und
Wissenschaftlichen Kurzmitteilungen 50 unberechnete Sonderdrucke. Mehrbedarf steht gegen
Berechnung zur Verfügung, jedoch muß die Bestellung spätestens mit der Rücksendung der
Korrekturfahnen erfolgen.

Vorbehalt aller Rechte: Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich
geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks,
des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der
Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei-
ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch im Einzelfall grund-
sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines
Beitrages ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber-
rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom
24. Junı 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen
den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind
mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen.

Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of
the fırst page of an article in this journal indicates the copyright owner’s consent that copies of the article may
be made for personal or internal use, or for the personal or internaluse of specitic clients. This consent is given
on the condition, however, that the copier pay the stated percopy fee through the Copyright Clearance
Center, Inc., 21 Congress Street, Salem, MA 01970, USA, for copying beyond that permitted by Sections 107
or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying
for general distribution, for advertising or promotional purposes, for creating new collective, or for resale.
For copying from back volumes of this journal see “Permissions to Photo-Copy: Publisher’s Fee List” of the
TC,

Fortsetzung 3. Umschlagseite

© 1994 Paul Parey. Verlag: Paul Parey GmbH & Co. KG, Hamburg. Anschrift: Spitalerstr. 12, D-20095 Hamburg,
Bundesrepublik Deutschland.
Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., Heide/Holst.

Z. Säugetierkunde 59 (1994) 321-325
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Neue Nachweise der Schabrackenspitzmaus (Sorex coronatus) in
Baden-Württemberg durch Polyacrylamidgel-Elektrophorese

Von H. Turnı und R. SCHÖNHERR
Zoologisches Institut, Abt. Physiologische Ökologie, Universität Tübingen, und Max-Planck-Gruppe
für molekulare und zelluläre Biophysik, Jena, Deutschland

Receipt of Ms. 12. 4. 1994
Acceptance of Ms. 23. 6. 1994

Abstract

New records of the Jersey shrew (Sorex coronatus) in Baden-Württemberg by
polyacrylamıde gel electrophoresis
Ascertained are new records and the eastern distribution of Sorex coronatus in Baden- Württemberg.
The determination of the species Sorex araneus/Sorex coronatus was realized by polyacrylamide gel
electrophoresis (PAGE) of total blood proteins. 69 anımals from different regions of Baden-
Württemberg were identified: 43 Sorex aranens and 26 Sorex coronatus. First records of Sorex
coronatus exist now for the region Tauberbischofsheim, Mögglingen, Geislingen, Isny and Wangen/
Allgäu. Certainly Sorex coronatus ıs spread to the eastern border of Baden-Württemberg.

Einleitung

Die Existenz zweier morphologisch unterschiedlicher Sorex araneus-Typen wies bereits
VON LEHMANN (1955) u. a. für das Rheinland nach, doch erst Meyran (1964) gelang mit
seiner karyologischen Untersuchung ein wichtiger Schritt zur Klärung der Artproblema-
tik. Anschließende Untersuchungen ergaben, daß Sorex araneus eın instabiler Chromoso-
menrassen-Komplex ist (bedingt durch Robertsonsche Translokationen), von dem sich -
neben der Schabrackenspitzmaus (Sorex coronatus) - noch ein weiterer, stabiler Karyotyp,
die ın Spanien und Portugal vorkommende Art Sorex granarius, abgrenzen ließ. Die
verwandtschaftlichen Verhältnisse innerhalb der Sorex araneus-Gruppe konnten schließ-
lich durch VOLOBOUEvV und CATZEFLIS (1989) sowie von WOojJcık und SEARLE (1988)
geklärt werden.

Die große Variabilität von Sorex araneus und die enge Verwandtschaft zu Sorex
coronatus lassen eine zuverlässige Trennung beider Arten nach äußeren Merkmalen kaum
zu (u. a. NEET 1992). Mehrere Trennmethoden wurden entwickelt, die eine Artdiagnose
mit Hilfe von Schädelmerkmalen ermöglichen sollen (u. a. Hausser und JAMMOT 1974;
Mys et al. 1985; HANDWERK 1987). Obwohl diese Trennmethoden regional 90-96 %
richtig zuordnen, muß berücksichtigt werden, daf einige dieser Schädelmaße geographi-
schen Schwankungen unterliegen. Um jede Restunsicherheit auszuschließen, erfolgt die
Artdiagnose durch Karyotypbestimmung oder durch Analyse von Serumproteinen mit
Polyacrylamidgel-Elektrophorese (PAGE). In letzterem Falle sind beide Arten an ihren
Albuminbanden klar zu unterscheiden (HAussER und ZUBER 1983; BRÜNNER 1988).

Die Schabrackenspitzmaus ist ein atlantisches Faunenelement. Sie kommt in Nordspa-
nıen, Frankreich, in der Schweiz, den Niederlanden und in Belgien vor. Ihre genaue
östliche, vermutlich durch Deutschland verlaufende Verbreitungsgrenze ist noch nicht
bekannt. Den Nachweisen ım Rheinland folgten — meist auf morphologischen Merkmalen
basierend — weitere in Niedersachsen, Westfalen, Thüringen, Osthessen, Saarland, Rhein-
land-Pfalz, Unterfranken und Baden-Württemberg (Hausser 1990; HERRMANN 1991;
MeınıG 1991; SCHELPER 1988; SCHLEGEL und BECKER 1990).

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5906-0321 $ 02.50/0

922 H. Turni und R. Schönherr

Für Baden-Württemberg liegen außer aus Nordbaden (Braun und Kischnick 1987)
auch biochemisch (PAGE) ermittelte Nachweise vor. Diese betreffen den Raum Südbaden
(BRÜNNER 1988), Nordbaden (BRÜNNER 1990) und Schönbuch/Tübingen (KuLzer et al.
1993), insgesamt also die westlichen und zentralen Landesteile.

Ziel der vorliegenden Untersuchung war, mit Hilfe einer sicheren biochemischen
Methode (PAGE) die gegenwärtige Verbreitung der Schabrackenspitzmaus in Baden-
Württemberg zu erfassen.

Material und Methoden

Für die vorliegende Untersuchung standen 69 Tiere zur Verfügung. Das für die PAGE benötigte Blut
(2 ul) wurde mit heparinisierten Mikropipetten aufgenommen und ın ein Eppendortfgefäß ausgeblasen;
es ist bei 20°C ca. 10 Stunden, im Kühlschrank (4-8°C) ca. eine Woche und im Tiefgefrierschrank
(-80°C) mindestens ein Jahr haltbar. Dieser Umstand macht es möglich, bereits verstorbene Tiere
(Katzenbeute, Totfunde), die erst Stunden nach ihrem Tod eingefroren wurden, für die PAGE
heranzuziehen.

Mit Hilfe der von Speckkäfern gesäuberten und anschließend vermessenen Schädel wurde eine auf
Schädelmerkmalen beruhende, vorläufige Artzuordnung unternommen, um die Blutproben sinnvoll
auf dem Elektrophoresegel anordnen zu können.

Die PAGE wurde nach der Methode von HAusser und ZUBER (1983) durchgeführt (ausführliche
Anleitung bei BRÜNNER 1988). Bei Sorex araneus und Sorex coronatus ergibt sich am Serumprotein
Albumin eine deutliche Trennung infolge der unterschiedlichen Ladungen (wie eine SDS-PAGE
ergab, ist das Molekulargewicht beider Albumintypen identisch). Während die Acrylamid-Konzen-
trationen von Sammelgel, Trenngel sowie die Zusammensetzung des Elektrophoresepuffers im
wesentlichen beibehalten wurden, erwies sich eine Verdünnung der Blutproben als sehr vorteilhaft.
Wurden die Proben in der von Hausser und ZUBER (1983) und BRÜNNER (1988) angegebenen
Konzentration verwendet, so erschienen die Banden zu dick. Deshalb wurde für eine optimale
Bandendicke folgende Blutprobenkonzentration ermittelt:

Lösung I: 3 ul Blut in 20 ul Salıne

Lösung II: 3 ul von Lösung 1 ın: 8 ul Saccharoselösung (40 %) + 50 ul 0,075M Tris/HCl pH 8,9 + Spur
Bromphenolblau

Probenauftrag: 15 ul Lösung II

Zwischen den Albuminbanden von Spitzmäusen, die erst wenige Minuten bis Stunden tot waren
und von Individuen, die bereits 5 Jahre (!) bei -30 °C eingefroren waren, gab es keine Qualitätsunter-
schiede.

Ergebnisse und Diskussion

Von den 69 Individuen, deren Fundorte sich auf 33 Meßtischblätter von Baden-Württem-
berg verteilen, konnten durch PAGE 43 Waldspitzmäuse (Sorex araneus) und 26 Schabrak-
kenspitzmäuse (Sorex coronatus) sicher identifiziert werden.

Erstmals konnte die Schabrackenspitzmaus ım Allgäu (Wangen, Isny) auf der Ostalb
(Geislingen, Mögglingen) und bei Tauberbischofsheim (Heckfeld, Boxberg) nachgewiesen
werden. Sie erreicht somit überall die Ostgrenze von Baden-Württemberg. In den Abbil-
dungen 1 und 2 wurden alle bisher durch PAGE ermittelten Nachweise von Sorex araneus
und Sorex coronatus kartographisch dargestellt.

Mit diesen Nachweisen ist auch der Anschluß an die auf morphologischen Merkmalen
basierenden Funde in Osthessen (PiEPER 1978; MEınıG 1991) und Unterfranken (PiEPER
1978) gelungen.

Obwohl bisher nur wenige vergleichende Daten zur Ökologie beider Spitzmausarten
vorliegen, wird angenommen, daß die Verbreitung von Sorex coronatus eng an Landschaf-
ten mit atlantischem Charakter (ausgeglichene Temperatur- und Feuchtigkeitsverhältnisse)
gebunden ist. Hingegen bevorzugt Sorex aranenus Gebiete mit kontinentalen Verhältnissen
(hohe Bodenfeuchte, starke Temperaturschwankungen). Überall dort, wo „kleinräumige
Mischklimate mit atlantischen und kontinentalen Charakteristika“ (MEınıG 1991) vorherr-
schen, dürften beide Arten gemeinsam vorkommen. Solche Gebiete erwähnte bereits
BRÜNNER (1990) auch für Baden-Württemberg. Ebenso konnten im Schönbuch beı

Neue Nachweise von Sorex coronatus in Baden-Württemberg 029

Sorex coronatus Millet, 1828

20 |21 22 123 24 |25 |26 |27 |28

5

Ei
=
a
Bi
w
>
[on
ST
1%
©

|

82 IN = Si
83 I
84 S
85
1o|ıı [12 |ı3 |14 18 | 19 |20 |2ı

Abb. 1. Erstnachweise von Sorex coronatus in Baden-Württemberg durch PAGE. © BrRÜNNER (1988,
1990); © KULZER et al. (1993); @ Neue Befunde dieser Studie

Tübingen beide Arten sogar in den gleichen Biotopen gefangen werden (KuULZER et al.
1993).

Wie aus den beiden Verbreitungskarten (Abb. 1 und 2) hervorgeht, sind Arealüber-
schneidungen (oft gleiche Biotope) keine Seltenheit. Eine Auswertung der ökologischen,
geographischen und klimatischen Daten aller baden-württembergischen Funde könnte
einen interessanten Beitrag zur Ökologie beider Arten liefern. Dies wird im Zusammen-
hang mit der geplanten flächendeckenden Kartierung in Baden-Württemberg möglich.

Danksagung

Für die finanzielle Unterstützung der Untersuchung im Rahmen des Projektes „Wildlebende Säuge-
tiere in Baden-Württemberg“ danken wir dem Ministerium für Umwelt Baden-Württemberg, Stutt-
gart, und der Landesanstalt für Umweltschutz Baden-Württemberg, Karlsruhe. Für die Hilfe bei der
Beschaffung der Spitzmäuse danken wir R. DESCHLE, Dr. F. DIETERLEN, E. PALIOCHA, U. SCHARDT
und W. ScHhLunn. Für die Durchsicht des Manuskriptes danken wir Prof. Dr. E. Kurzer. Ganz
besonderer Dank gilt Dr. R. FLössEr, der diese Untersuchung anregte und förderte. Die Herstellung
der Verbreitungskarten erfolgte mit Hilfe eines Programms von N. HiRNEISEN (Oko-Software
Tübingen).

324 H. Turni und R. Schönherr

Sorex araneus Linnaeus, 1758

17 118 |19 |20 |21 |22 53 [24 25 [26 |27 |28

Faf tunen:
Dam Da

& Ara

c—

4
\ DD
O
DO \ BEE
OD S
68 S I 68
69 % | 69
5 L |
709 | ) O y ® 70
(®) ©
7 o lo 71
72 (eI6) 72
73 | RS S | 73
74 74
4
Ko BEER I A 75
76 | ei 11700
77 a,
78 2 78
(@) &
79 | 9 \ 79
80 (eI6) (@) 80
[/(©)
8 81
s \
82 | IR N i
Dur \
83 \ ep I 83
84 5 N j |
| zT If 4— 0 IN &
85 | | 85
oO HE B22] ler] za AZ AI 2 ZI 2228 | 25 227 28

Abb. 2. Identifizierung von Sorex araneus in Baden-Württemberg durch PAGE. © BRÜNNER (1988,
1990); © KULZER et al. (1993); @ Neue Befunde dieser Studie

Zusammenfassung

Das Vorkommen der Schabrackenspitzmaus (Sorex coronatus Millet, 1828) in Baden-Württemberg,
insbesondere in den östlichen Landesteilen wurde untersucht. Die Trennung der Arten Sorex araneus
und Sorex coronatus erfolgte durch Polyacrylamidgel-Elektrophorese (PAGE) am Gesamtbluteiweiß.
Von 69 Spitzmäusen wurden 43 als Sorex araneus und 26 als Sorex coronatus identifiziert. Erstnach-
weise von Sorex coronatus liegen nun für die Gebiete Tauberbischofsheim, Mögglingen, Geislingen,
Isny und Wangen/Allgäu vor. Somit ist die Schabrackenspitzmaus mindestens bis an die Ostgrenze
von Baden-Württemberg vorgedrungen.

Literatur

Braun, M.; Kıschnick, P. (1987): Spitzmäuse und ihre Nahrung in einem Buchenwald. Carolınea 45,
159-161.

BRÜNNER, H. (1988): Untersuchung zur Verbreitung, Ökologie und Karyologie der Waldspitzmaus
(Sorex araneus Linne, 1758) und der Schabrackenspitzmaus (Sorex coronatus MILLET, 1828) ım
Freiburger Raum mit Bemerkungen zu einigen anderen Spitzmausarten. Dipl.-Arbeit Univ.
Freiburg.

BRÜNNER, H. (1990): Zur Verbreitung von Waldspitzmaus (Sorex araneus LinNE, 1758) und Scha-
brackenspitzmaus (Sorex coronatus MILLET, 1828) in Oberrheinebene und Schwarzwald. Unver-

Neue Nachweise von Sorex coronatus in Baden-Württemberg 925

öff. Bericht im Auftrag des Forschungsvorhabens „Wildlebende Säugetiere in Baden-Württem-
berg“, Karlsruhe.

HANDWERK, J. (1987): Neue Daten zur Morphologie, Verbreitung und Ökologie der Spitzmäuse
Sorex araneus und $. coronatus ım Rheinland. Bonn zool. Beitr. 38, 273-297.

Hausser, J. (1990): Sorex coronatus MiLLET, 1882 — Schabrackenspitzmaus. In: Handbuch der
Säugetiere Europas. Hrsg. von: J. NIETHAMMER und F. Krapp, Wiesbaden: Aula-Verlag. Pp.
279-286.

HaAusSER, J.; JAMMOT, D. (1974): Etude biometrique des machoires chez les Sorex du groupe araneus
en Europe continentale (Mammalia, Insectivora). Mammalia 38, 324-343.

HaussER, J.; ZUBER, N. (1983): Determination specifique d’individus vivants des deux especes
jumelles Sorex araneus et Sorex coronatus, par deux techniques biochimiques (Insectivora,
Soricidae). Revue suisse Zool 90, 857-862.

HERRMANN, M. (1991): Säugetiere im Saarland. Schriftenreihe des Naturschutzbundes Saarland e. V.
(DBV).

KULZER, E.; LINDEINER, A. v.; WOLTERS, 1.-M. (1993): Säugetiere im Naturpark Schönbuch. Beih.
Veröff. Naturschutz Landschaftspflege Bad.-Württ. 71, 1-212.

LEHMANN, E. von (1955): Über die Untergrundmaus und Waldspitzmaus in NW-Europa. Bonn.
zool. Beitr. 6, 8-27. r

MEınıg, H. (1991): Zur Verbreitung und Okologie von Sorex araneus L., 1758 und Sorex coronatus
Millet, 1828 (Mammalıa, Insectivora) im Kreis Mettmann und in der Stadt Wuppertal. Jahresber.
naturwiss. Ver. Wuppertal 44, 5-14.

Meyran, A. (1964): Le polymorphisme chromosomique de Sorex araneus L. (Mammalıa, Insecti-
vora). Revue suisse Zool. 71, 903-983.

Mys, B.; VAN DER STRAETEN, FE; VERHEYEN, W. (1985): The biometrical and morphological ıdentifica-
tion and the distribution of Sorex araneus L., 1758 and $. coronatus Millet, 1828 in Belgium
(Insectivora, Soricidae). Lutra 28 55-70.

NEET, ©. R. (1992): The use of fur colour characters to distinguish the sibling species Sorex araneus
and Sorex coronatus (Insectivora, Soricidae): a field test in a zone of parapatrıc contact. Z. Säuge-
tierkunde 57, 176-178.

PIEPER, H. (1978): Zur Kenntnis der Spitzmäuse (Mammalıa, Soricidae) in der Hohen Rhön. Beitr.
Naturkde. Osthessen 13/14, 101-106.

SCHELPER, W. (1988): Nachweis der Schabrackenspitzmaus Sorex coronatus Millet, 1828 ın Südnieder-
sachsen. Beitr. zur Naturkunde Niedersachsens 41, 89-91.

SCHLEGEL, B.; BECKER, K. (1990): Ein aktueller Nachweis der Schabrackenspitzmaus (Sorex coronatus
Millet, 1828) in Niedersachsen. Beitr. zur Naturkunde Niedersachsens 43, 7.

VOLOBOUEV, V.; CATZEFLIS, F. (1989): Mechanısm of chromosomal evolution in three European
species of the Sorex araneus-arcticus group (Insectivora: Soricidae). Z. zool. Syst. Evolut.-forsch.
27, 252-262.

WOJCIK, J. M.; SEARLE, ]J. B. (1988): The chromosome complement of Sorex granarius — the ancestral
karyotype of the common shrew (Sorex araneus)? Heredity 61, 225-229.

Anschriften der Verfasser: HENDRIK TURNI, Universität Tübingen, Zoologisches Institut, Abt.
Physiol. Okologie, Auf der Morgenstelle 28, D-72076 Tübingen, und Dr.
ROLAND SCHÖNHERR, Max-Planck-Gruppe für molekulare und zelluläre
Biophysik, Jena, Drackendorfer Str. 1, D-07747 Jena

Z. Säugetierkunde 59 (1994) 326-331
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Reproductive performance ofthe Red fox, Vulpes vulpes,
in Garmisch-Partenkirchen, Germany, 1987-1992

Dysaa@s\os

WHO Collaborating Centre for Rabies Surveillance and Research at the Federal Research Centre for
Virus Diseases of Animals, Tübingen, Germany

Receipt of Ms. 2.11.1993
Acceptance of Ms. 28.7.1994

Abstract

The reproductive performance of the red fox Vulpes vulpes was studied after the disappearance of
rabies between 1987 and 1992 in the county of Garmisch-Partenkirchen. The uteri of 452 adult vixens
from the study area, and another 603 vixens from rabies-endemic and rabies-free areas in Bavaria were
examined. No significant difference in embryonic litter size was found between 74 rabid and 34 non-
rabid vixens in 1987. Furthermore, no difference in mean litter size was found between a rabies-
endemic area and the rabies-free study area, Garmisch-Partenkirchen. Litter size and the proportion
of barren vixens dıd not show any significant yearly varıation during 1988-1991 in the study area. No
difference in litter size (number of placental scars) between the varıous age classes was found.
However, the differences in productivity between the different age-classes were significant, due to a
higher proportion of barren yearlings. Although the fox-density increased continually during this
field study, apparently the density is still below its carrying capacıty. Therefore, no density-
dependent effect on the productivity of the red fox population in Garmisch-Partenkirchen could be
shown.

Introduction

When a population ıs reduced below its carrying capacity, an increased productivity may
compensate for these losses. MAcDonAaLD (1987) described this possibility for the red fox
(Vulpes vulpes), larger litters being typical for areas with heavy mortality. Could,
therefore, a decrease in the productivity be expected in areas where the fox density has
increased as a result of the disappearance of rabies? This would be ot interest ın areas where
this important mortality factor of the fox population has disappeared after application of
oral vaccination. Data from areas ın Switzerland indicate that rabies can kill over 50 % ot a
local fox population during the height of the epidemic (WANDELER et al. 1974a).

The productivity of a fox population depends on two factors: lıtter size and the
proportion of adult vixens reproducing (LLoyD et al. 1976; MacDonarp and VoıGT
1985). This study describes the reproductive performance of the fox in the alpine county
(“Landkreis”) Garmisch-Partenkirchen, Germany. The present European rabies epizootic
reached Garmisch-Partenkirchen in the autumn of 1965. For the next twenty years, wıth
the exception of 1970-1971, the county was infected. The rabies-incidence ın thıs area was
classified as type four; high oscillations with repeated peaks of rabies-occurrence, often 10
or more cases per 100 km? and per year (JACKSON and SCHNEIDER 1984). Since the
application of oral vaccination against rabies in the spring of 1985, no rabid fox has been
found here up to the end of this field study in March 1992. The fox density in Garmisch-
Partenkirchen increased continually during this study, although it seems that it had not yet
reached its carrying-capacity during this period, 1987-1992 (Vos 1993).

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5906-0326 $ 02.50/0

Reproductive in Vulpes vulpes 927

Study area

The study area, the county of Garmisch-Partenkirchen (1012 km?) in Bavaria is bounded on the south
by the Austrian Alps. Approximately 46 % of the county is covered with forest, and 21.5 % is used
for agrıcultural purposes (mainly pastures). The northern part (altitude 600-850 m) of Garmisch-
Partenkirchen consists of hilly landscape, covered with a mixture of woods and grassland. The central
and southern parts are characterized by mountain-ranges, up to 2964 m, interspersed with valleys and
mountainslopes covered with forest. Above tree-line, steep and ragged landscape dominates, with a
few scattered alpine meadows.

Material and methods

The data presented here came from a study on fox population dynamics after the disappearance of
rabies in Garmisch-Partenkirchen. Uteri were obtained from carcasses of adult vixens (n = 452) killed
or found dead in the study area between November 1987 and March 1992. The vixens were collected
throughout the year. Litter size was determined from counts of implanted embryos and, after
parturition, by placental scar counts, as described by Vos (1993). Since the placental scars persist, they
were also used to determine whether or not a vixen had been reproductively active during the last
breeding season. The term “barren” includes all adult vixens that failed to reproduce; i. e., those that
did not ovulate or failed to implant or lost the entire set of implanted embryos before parturition. To
examine the relationship between rabies and the reproductive performance of foxes, vixens (n = 603)
from rabies-endemic and rabies-free areas in Bavaria were examıned. These vixens consisted of animals
delivered for rabies diagnosis collected annually from February to April 1987 through 1991. The age
of the vixens was estimated by radiography of canine teeth and/or by counting cementum annuli in the
first premolars or canınes (Vos 1993).

Results

In spring 1987, ın Bavarıa, the average number of implanted embryos of rabid vixens (X =
5.57, n = 74) did not differ significantly from non-rabid vixens (x = 6.26, n = 34)
(Kolmogoroff and Smirnoff Test, D = 0.28, n. s.). Also, no significant difference could be
observed between the average number of ımplanted embryos in vixens from a rabies-
infected area, the district (“Regierungsbezirk”) of Schwaben (x = 5.68, n = 29), and those
from the rabies-free county of Garmisch-Partenkirchen since 1985 (x = 5.75, n = 16)
(Student’s t-Test, t = 0.16, df = 15). Apparently rabies has no direct influence on litter size.
The frequency distribution of the number of placental scars in vixens from Garmisch-
Partenkirchen between 1988 and 1991 ıs presented in table 1. No significant differences in
the average number of placental scars (Student’s t-Test) and the frequency distribution
(Kolmogoroft and Smirnoft Test) could be observed among years.

The average litter sıze based on the number of embryos was higher than the average
litter sıze based on the number of placental scars in vixens from Garmisch-Partenkirchen.
This difference was a result of the period in which the samples were taken. Vixens with
implanted embryos were delivered only during early-pregnancy stages; only prenatal
losses during ımplantation and these early pregnancy stages could be considered. How

Table 1. Number of placental scars in vixens in Garmisch-Partenkirchen, 1987-1991

(x - mean number of placental scars, SD - standard deviation)

Number of placental scars

328 AR@EVoS

ever, by the evaluation of the placental scars, all visible losses between implantation and
birth were taken into account. In early summer, on average 4.6 cubs were observed at the
den sites (n = 14). The difference between the number of embryos of early pregnancy
stages and the number of cubs observed at the den sites, indicated a loss of around 20 %.
Litter size of foxes from different areas ıs shown in table 2.

The productivity of the fox is not only determined by litter size, also the proportion of
reproductively active vixens plays an important role. No differences in the proportion of
barren vixens could be observed for Garmisch-Partenkirchen between the years 1988-1991
(n’-Test, x” = 4.27, df = 3, n.s.). The yearly productivity of the vixens (i.e. including
barren anımals) in Garmisch-Partenkirchen was estimated on the basis of the mean number
of placental scars and the proportion of barren vixens (Tab. 3). For productivity, no
differences among years could be shown (analysis of varıance, F = 0.72, df[3,135], n.s.).
15.3% of the examined adult vixens (n = 170) were barren. Of these 26 barren vixens,
69.2 % were yearlings, 19.2 % were 2-year olds and two vixens were over four years of
age. The proportion of barren yearlings was significantly higher compared to the other age
elassesa x 2 lese 65xdt _ ur)!

In table 4 the mean lıtter size (placental scar counts) and the proportion of barren vixens
are presented for the different age classes. No age-dependent effect on litter size could be
shown (analysıs of varıance, F = 1.79, df[3.98], n. s.). Also no age-dependent litter size
(number of embryos) could be found for vixens in the Bavarıan sample (analysis of
varıance, F = 0.65, df[2,83], n. s.). However, the productivity of yearlings in Garmisch-
Partenkirchen was sıgnificantly lower than of vixens 3-years of age and older (Duncan-
Test, P< 0.05). The 2-year-old vixens showed the highest mean litter size. As a result of

Table 2. Litter size of foxes in different study areas

Study-area

rabies-free
rabies infec.
Oberlausitz
Midwest
London
Bristol

N-Dakota
Bavarıa

Wales
Wriezen

Garmisch-P.

558

99202
4.6

Author

WANDELER et al. (1974)
WANDELER et al. (1974)
ANSORGE (1990)

STORM et al. (1976)
Harrıs and SMITH (1987)
Harrıs and SMITH (1987)
STUBBE (1980)

ALLEN (1984)

LrovD et al. (1976)
LrovD et al. (1976)
LrovD et al. (1976)
LrovD et al. (1976)
GORETZKI et al. (1981)
Vos (1993)

PS = placental scars; E = Embryos; YF = cubs at den. In some samples data of PS and E are

combined.

Table 3. Yearly productivity of vixens in Garmisch-Partenkirchen, based on the number of
placental scars and the proportion of barren vixens

Litter sıze
Barren vıxens (%)
Productivity

Reproductive in Vulpes vulpes 9)

Table 4. Productivity of vixens

Mean number of placental scars (litter size) and the proportion of barren vixens for different age
classes in Garmisch-Partenkirchen

Age class eitter size Barren vixens Productivity
(year) (%)

the relativly high proportion of barren vixens, the productivity of this age class was below
that of older vixens.

Discussion

Like WANDELER et al. (1974b), no difference ın the mean number of ımplanted embryos
could be observed between a highly infected area and the rabies-free study area, Garmisch-
Partenkirchen since 1985. After the disappearance of the important density-dependent
mortality factor, rabies, the red fox density increased continually in Garmisch-Partenkir-
chen (Vos 1993). However, even the growth rate of the population of this generalist, the
red fox, is not unlimited. The population threshold (carrying-capacity) of a certain area ıs a
result of density-independent factors, e.g. habitat structure. However, the population
numbers are regulated by density-dependent events (WEHNER and GEHRING 1990).

Negative density-dependent processes limit the growth rate of the fox population. The
occurrence of these feedback mechanisms could indicate that the fox population
approaches the carrying capacity. These mechanısms do not necessarıly mean an increased
mortality rate; a decrease in reproduction performance could also influence the growth-
rate. In Garmisch-Partenkirchen, no yearly difference was observed in the two important
reproduction parameters; litter-size and the proportion of barren vixens. The foxes ın this
county lıve in a very stable environment with a high food supply. Apparently the increased
fox density after the disappearance of rabies has not reached its carrying capacity.
Therefore, no decrease or yearly fluctuations ın the reproductive performance were
observed during this field study. This is in contrast to several other studies, where the
fluctuating numbers of barren vixens can be seen as a density-dependent regulation
mechanism in areas wıth limited or strongly fluctuating food supplies (EnGLunD 1980;
MacDonauD 1980; SCHANTZ 1981). EnGLunD (1980) found ın the northern corniferous
belts of Sweden not only large annual variatıon in the number of barren vixens, but also in
the number of cubs per litter. In these areas foxes depend for food almost exclusively on
the fluctuating rodent populations. The observed spatial and temporal differences in the
reproductive performance of foxes in the different areas are to a large extent a result of
varlatıons in the socıal structure of the fox populations. In some habıtats foxes live ın social
groups comprised of one adult male and several adult vixens (MAacDonarD 1979; SCHANTZ
1981), whilst elsewhere foxes lıve in territorial pairs (STORM et al. 1976). Observations of
the above-mentioned famıly groups have been made in areas of high population density
and stable food availability (MacDonALD 1980), but also in areas where fox populations
experience strong food-resource fluctuations (ENGLUND 1980; SCHANTZ 1981). The subor-
dinate vixens of these groups reproduce only in years when there are abundant food
resources. At low or intermediate levels of food abundance only the dominate alpha vixen
reproduces (MAcDonALD 1983).

Like AnsorGe (1990), litter sıze did not show an age-dependent effect in this study.

330 AN@r Vos

Contrary to these results, ALLEN (1984) found an increase in ovulation rate and embryonic
litter size as a function of increasing female age of foxes in North Dakota. However,
HARRIS and SMITH (1987) could only observe a decrease in litter size in extremely old
vixens ın the London area. In accordance with other studies (Harrıs 1979; ENGLUND
1980) most barren vixens ın Garmisch-Partenkirchen were yearlings, hence the productiv-
ity of thıs age class was lower than that of older vixens.

Acknowledgements

I am most grateful to all the hunters who assısted with field work and to Dr. GrÖTscH and his staff
(Landesuntersuchungsamt für das Gesundheitswesen in Südbayern) in Oberschleißheim for their
support in the laboratory. Financial support was provided by Klocke Verpackungs-Service GmbH,
Weingarten-Baden.

Zusammenfassung

Die Reproduktion des Rotfuchses (Vulpes vulpes) im Landkreis Garmisch-Partenkirchen,
Deutschland, 1987-1992

Untersucht wurde die Reproduktionsleistung des Rotfuchses Vxlpes vulpes nach dem Verschwinden
der Tollwut zwischen 1987 und 1992 im alpinen Landkreis Garmisch-Partenkirchen. Uteri von 452
adulten Fähen aus dem Untersuchungsgebiet und die von 603 Fähen aus ganz Bayern wurden auf
Anwesenheit von plazentalen Narben oder Embryonen untersucht. Keine signifikanten Unterschiede
in der durchschnittlichen Zahl der Embryonen zwischen 74 tollwutpositiven und 34 tollwutnegativen
Fähen konnten festgestellt werden. Kein signifikanter Unterschied ın der durchschnittlichen embryo-
nalen Wurfgröße konnte während dieser Untersuchung zwischen den Fähen aus dem tollwutinfizier-
ten Regierungsbezirk Schwaben und dem seit 1985 tollwutfreien Landkreis Garmisch-Partenkirchen
nachgewiesen werden. Keine signifikanten jährlichen Differenzen im Anteil nicht-reproduzierender
Fähen und der Wurfgröße wurden während 1988 und 1991 gefunden. Auch unterschied sich die
Wurfgröße nicht zwischen den verschiedenen Altersklassen. Die Unterschiede ın der Produktivität
der Altersklassen als Folge des höheren Anteils nicht-reproduzierender ljähriger Fähen waren jedoch
signifikant. Die Ergebnisse dieser Untersuchung deuten darauf hin, daß die Fuchsdichte im Landkreis
Garmisch-Partenkirchen sieben Jahre nach dem Verschwinden der Tollwut anscheinend ihren krıiti-
schen Grenzwert noch immer nicht erreicht hat.

References

Auren, $. H. (1984): Some aspects of reproductive performance in female red fox in North Dakota. ].
Mammalogy 65, 246-256.

ANSORGE, H. (1990): Daten zur Fortpflanzungsbiologie und Reproduktionsstrategie des Rotfuchses,
Vulpes vulpes, in der Oberlausitz. Säugetierkd Inf. 3, 185-199.

EnGLun, J. (1980): Population dynamics of the red fox (Vulpes vulpes L., 1758) ın Sweden. In:
Biogeographica, The Red Fox. Ed. by E. Zımen. Den Haag: Junk Publ. Vol. 18, 195-207.

GORETZKI, J.; PausTtian, K.-H. (1981): Untersuchungen zum Lebendfang von Jungfüchsen (Vxlpes
vulpes [L]) als Grundlage für Reduzierungs- und Impfmaßnahmen. Zool. Garten. N F Jena 51,
250-263.

Harrıs, $. (1979): Age-related fertility and productivity in red foxes, Vulpes vulpes, in suburban
London J. Zool. (London) 187, 195-199.

HaRRıs, $.; SMITH, G. C. (1987): Demography of two urban fox (Vulpes vulpes) populations. J. Appl.
Ecol. 24, 75-86.

Jackson, H. C.; SCHNEIDER, L. G. (1984): Rabies in the Federal Republik of Germany, 1950-81: the
influence of landscape. Bull. WHO 62, 99-106.

Lroyp, H. G.; JENnsEn, B.; HAAFTEN, J. L. van; NIEWOLD, F. ]J. J.; WANDELER, A.; BöceEL, K.;
Arata, A. A. (1976): Annual turnover of fox populations in Europe. Zbl. Vet. Med. B. 23,
580-589.

MacDonaıp, D. W. (1979): Helpers ın fox society. Nature 282, 69-71.

MacDonALD, D. W. (1980): Rabıes and Wildlife. Oxford: University Press.

MacDonaup, D. W. (1983): The ecology of carnıvore socıal behaviour. Nature 301, 379-384.

MacDonALD, D. W. (1987): Running with the fox. London, Sydney: Unwin Hyman.

MacDonarp, D. W.; VoıGT, R. (1985): The biological basıs of rabies models. In: Population
dynamics of rabies in wildlife. Ed. by P. J. Bacon. London: Academic Press. Pp. 71-103.

Reproductive in Vulpes vulpes RN!

SCHANTZ, T. von (1981): Evolution ot group living, and the importance of food and social
organization in population regulation; a study on the red fox (Vulpes vulpes). Thesis, Univ. Lund.

STORM, G. L.; ANDREWS, R. D.; PHıLLıps, R. L.; BısHoPr, R. A.; SINEFF, D. B.; TESTER, J. R. (1976):
Morphology, reproduction, dispersal and mortality of mid-western red fox populations. Wildl.
Monogr. 49, 1-82.

STUBBE, M. (1980): Population ecology of the red fox (Vulpes vulpes L., 1758) in the G. D. R. In:
Biogeographica, The Red Fox. Ed. by E. Zımen. Den Haag: Junk Publ. Vol. 18, 71-96.

Vos, A. C. (1993): Aspekte der Dynamik einer Fuchspopulation nach dem Verschwinden der
Tollwut. Thesis, Univ. München.

WANDELER, A.; WACHENDÖRFER, G.; FÖRSTER, U.; KREKEL, H.; SCHALE, W.; MÜLLER, J.; STECK, F.
(1974a): Rabies in wild carnıvores in Central Europe. I. Epidemiological studies. Zbl. Vet. Med.
B. 21, 735-756.

WANDELER, A.; MÜLLER, J.; WACHENDÖRFER, G.; SCHALE, W.; FÖRSTER, U.; STECK, F. (1974b):
Rabies in wild carnivores in Central Europe. III. Ecology and biology of the fox in relation to
control operations. Zbl. Vet. Med. B. 21, 765-773.

WEHNER, R.; GEHRING, W. (1990): Zoologie. 22. Aufl. Stuttgart: Georg Thieme Verlag.

Author’s address: Dr. A. C. Vos, Impfstoffwerk Dessau-Tornau GmbH, Postfach 214, D-06855
Roßlau, Germany

Z. Säugetierkunde 59 (1994) 332-341
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Crabs Potamonautes perlatus in the diet of Otter Aonyx capensis
and Water mongoose Atılax paludinosus in a freshwater habitat
in South Africa

By M. G. Purvss, H. Kruuk, and ]J. A. J. NEL

Department of Zoology, University of Stellenbosch, Stellenbosch, South Africa

Receipt of Ms. 23. 3. 1994
Acceptance of Ms. 22. 6. 1994

Abstract

Studied the feeding ecology of sympatric Cape clawless otter and water mongoose at the Olifants
River, Cape Province. Crab was an important prey in the diet of both species and faecal analysis was
used to determine the extent of dietary overlap. Mean crab size taken by water mongoose was larger
than that taken by otters. Fish was the second most important prey in the diet of otters, whereas
terrestrial prey comprised a major part of the diet of the water mongoose. Different foraging patterns
of the predators and habitat selection by crabs of different sizes could explain the observed variation.

Introduction

Over most of their range in southern Africa the two species of otter - Cape clawless Aonyx
capensis and spotted-necked Lutra maculicollis —- coexist with the water mongoose Atılax
paludınosus. They share the same habıtat, albeıt that the otters are more aquatic and the
water mongoose ranges more often away from water; they show temporal overlap in
activity; and some prey items are common to two or all three species (SKINNER and
SMITHERS 1990). Atılax paludinosus ıs the most widespread, followed by A. capensıs; L.
maculıcollis has a more easterly distribution and unlike the others does not utilize marıne
habıtats.

Both otter species have been little studied. RowE-Rowe (1977a, b) documented their
feeding behaviour in captivity and under natural conditions in Natal, while van DER ZEE
(1981, 1982), ARDEN-CLARKE (1986) and VERWOERD (1987) reported on the food, status,
population density and spatial organızation of A. capensis in marine habıtats. BARKER (1989)
compared the feeding habıts of A. paludinosus in the wıld and in captıvity and MADDOcCK
and PERRIN (1993) reported on their role in an assemblage of viverrids.

Freshwater crabs Potamonautes spp. (Family Potamonidae) contributed the major
portion of the diet of A. capensıs in Natal (RowE-Rowe 1977a), in Lake Victoria (KRUUK
and GOUDSWAARD 1990) and in various rivers ın the SW Cape (unpubl. data). Du Torr
(1980), Louw and Ne (1986), RowE-Rowe (1977a) and BAkER (1989) found crab to be
important prey of A. paludinosus in freshwater habitats, while WHITFIELD and BLABER
(1980), Louw and Ner (1986), and MAacDonaıD and Ner (1986) found crustaceans of
importance in the diet of water mongooses in estuarine and in marine habıtats.

Crabs constitute the highest macroinvertebrate biomass in some South African rivers
(Hırn and O’Kerrre 1992), as do crayfish in some rıvers ın Italy (GHERARDI et al. 1989)
and high crab densities have also been recorded in trout streams in Zimbabwe (TURNBULL-
Kemp 1960). Crustaceans can, therefore, either seasonally or year-round provide an
important food base for both otters and water mongoose.

This study discusses the utilization of this resource by sympatric A. capensis and
A. paludinosus. It reports on their diet, the use of crabs and seasonal changes in diet ın a

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5906-0332 $ 02.50/0

Crabs in the diet of Aonyx capensis and Atılax paludinosus 333

freshwater habitat, and it discusses differences ın foraging patterns underlining the
observed varıatıon.

Study area

The study was conducted along a 5 km stretch of the Olifants River ca. 8 km downstream from the
Bulshoek Dam (31°58’ S; 18°45’ E), some 100 m above sea level and ca. 32 km north and downstream
of Clanwilliam. Here the river is slow-flowing with marshlands, reeds or grassy verges, oxbows, large
basins and with rapids or rocky pools up to 6 m deep. Vegetation on the banks were mainly grasses
and sedges e.g. Paspalum sp., Juncus sp. and Prionium sp. Sesbania punicea trees formed thickets right
up to the river edge in some places.

Average rainfall at the dam site is 265 mm p.a. (1981-1991), while the total rainfall for 1991 was
275 mm (Weather Bureau, in lit.). There is a rise of approximately 2-3 m in the water level at the
study site during June and July, when rainfall is at its peak and the sluices of the Bulshoek Dam are
opened. The minimum and maximum air temperatures vary from 19-34 °C in January (summer) to
9-19 °C in July (winter).

The fish species in the study area are mostly exotic, e.g. smallmouth bass Micropterus dolomien, or
translocated species, e.g. banded tilapıa Tilapia sparrmanii, while indigenous species (e.g. Barbus
capensis, B. serra and Labeo seeberi), of which most are endemic, are rare. Juvenile fish (2-8 cm in
length) were more abundant during the summer, with a rapıd decline in numbers from May/June
onwards (SWIEGERS, pers. comm.).

Material and methods

Collection and analysis of scats

Aonyx scats were usually found < 10 m from the water edge, often in exposed areas. Flat rocks were
commonly used for collective sprainting, which probably had a communicatory function (KRUUK
1992), as in the European otter. Such latrines were often used repeatedly over several visits to the
study area. Individual scats were also scattered around in areas ranging from flat rocks to grassy
banks. The mean diameter of the scats collected (when measurable) was 22.2 mm (SD = 3.63, range =
15.3-30.0, n = 50). No otter scats were found in September.

Atılax scats were also deposited at latrines as well as singly on different substrates, but usually in
areas with vegetational cover, often >10 m from the water edge. The mean diameter was 15.8 mm
(SD = 2.2, range = 13.0-20.0, n = 23). No water mongoose scats were collected in April.

Both Cape clawless otter Aonyx capensis and water mongoose Atılax paludinosus scats were
collected during February and November (summer), April (autumn), June (winter) and September
(spring) 1991, from latrines as well as individual defecation sites. Otter scats were identified and
distinguished from water mongoose scats by size, form, smell and the absence of banded haır as found
in water mongoose scats. Only relatively fresh (unbleached) and intact scats were collected; those not
positively ıdentified were rejected. Scats were placed separately in paper bags and labelled with the
species name, date and site of collection.

Air-dried scats were teased apart and all crab eyestalks, intact or broken, extracted. Other
diagnostic prey remains, e.g. fish scales and bones were also removed. Fish were identified using scale
characteristics and frogs were ıdentified from skeletal remains.

Crabs were ıidentified as Potamonautes perlatus (Milne Edwards). A total of 66 crabs of varying
sızes (mean carapace width = 31.6 mm, SD = 13.8, range = 8.5-60.1 mm) were trapped or collected by
hand in the study area. Both eyestalks of each crab were measured with an occular ruler in a stereo
microscope. The length (L) of an eyestalk was taken as the longest axis from anterior to posterior
when viewed from the lateral side. There was no consistent difference in the length of the two
eyestalks of individual crabs and ıt was assumed that they were the same length.

The maximum width of the carapace was measured with calıpers and correlated with the eyestalk
length. There was a clear numerical linear relationship (Fig. 1), tor which

C = 8.33 E-11.48 (1)

In this C = carapace width (mm) and E = eyestalk length (mm). The correlation was highly significant
(r? = 0.99, p< 0.001).
The crabs were also weighed, and regressions of wet weight (dependant variable) and eyestalk
length (independent variable) were calculated. A log-linear relationship was found
log W = -0.72 + 0.319E (2)

where W = wet weight (g) and E = eyestalk length (mm); (r? = 0.96, p<0.001).

334 M. G. Purves, H. Kruuk, and J. A. J. Nel

Carapace width in mm

Eyestalk length in mm

Fig. 1. The relationship between carapace width and eyestalk length in the freshwater crab
Potamonautes perlatus (r” = 0.988, P < 0.001, n = 66, slope = 8.33)

Quantification of data

Prey remains found in scats were expressed as relative percentage of occurrence, calculated by totalling
all the occurrences of all the prey items and expressing the actual occurrence of each prey item as a
percentage of the total. For each scat dominance of prey type was also found by volumetric analysis of
prey items. For that analysis, in a scat consisting mainly of crab remains, crab would be the dominant
prey type and the other prey items in this scat would be ignored. The different dominant prey items
were than totalled and expressed as a percentage of the total number of scats. |

The maximum number of crabs potentially represented by the eyestalks in each scat was taken as
the number of eyestalks found, i.e. each eyestalk represented a crab. Half of the number of eyestalks
found in each scat was taken as indicating the minimum number of crabs in that scat, implying that a
pair of eyestalks represented a crab.

Results
Diet composition

The proportions of prey remains from scat collections at different times of year ıs shown in
figure 2. In both predators, crabs were the most abundant prey at all times, and especially
in February. From then until November this proportion in the diet decreases. At all times
the diet of Atılax was more varıed than that of Aonyx, with slightly fewer crabs, and also
fewer fish, but more terrestrial prey such as rodents and insects. A sımilar picture emerges
from the analyses of scats by estimated bulk of prey remains (Fig. 3).

In total 70% of the otter scats analysed were dominated by crab and 13.1 % by fısh
(Tab. 1). Tılapıa sparrmanıl was by far the dominant fish species (85 % occurrence) re-
corded, with smallmouth bass Micropterus dolomieu being the only other fish preyed

Crabs in the diet of Aonyx capensis and Atılax paludinosus 335

AONYX

FEBRUARY N=27

70

60

50

40

3%

20

10

0

70

60

50

40

30

20

10

0 ZZ,
CR FI FR IN MA BI PL
JUNE N =27 JUNE N=2

58833285

Fıg. 2. Relative percentage of occurrence showing seasonal variation of different prey items in the diets
of Aonyx capensis and Atilax paludinosus. (CR = crab, FI = fish, FR = frog, IN = insect, MA = small
mammal, BI = bird, PL = plant, MO = mollusc, SC = scorpion, RT = reptile). N = number of scats

336

age domin

M.G. Purves, H. Kruuk, and J. A. J. Nel

sonal varıation of different prey items in the

ance of occurrence showing sea
x capensis and Atılax paludinosus. N = number of scats

Crabs in the diet of Aonyx capensis and Atılax paludıinosus 93m

upon. No remains of indigenous fish
species were recorded. The frogs were
Strongylopus grayıi and Rana fuscigula.
A variety of insect taxa were rep-
resented, but only those with a vo-
lumetrie contribution of >5 % (suggest-
ing deliberate ingestion) were taken into
consideration. These included: Odona-
ta-Aeschnidae nymphs (dragonflies),
which were the volumetric dominant
insect prey, Coleoptera-Dytiscidae and
Scarabaeıdae (beetles), Hymenoptera-
Formicidae (ants) and Orthoptera-
Acrididae (grasshoppers. A small
number of spiders (Subclass Arachnida)
also occurred. Otomys saundersiae was
the only small mammal that could be
identified from A. capensıs scats.

In 65.7 % of scats of A. paludinosus
crab was the dominant food (Tab. 2).
Atılax preyed on a greater varıety of
inseets than Aonyx. Coleoptera (beetles)
were the most abundant insects found in
the diet of Atılax. Orthopteran (grass-
hoppersa „loeusts; and Zerickets),

Table 1. Summary of prey items recorded
in 132 Aonyx capensis scats

Item Occurrence Relative% % Dominance

Crab 123
Fish

Frog
Insect
Plant
Mammal

Bird

Table 2. Summary of prey items recorded
in 70 Atilax paludinosus scats

Item Occurrence Relative% % Dominance

Crab
Fish
Frog

Insect

Plant
Mammal

Bird

Reptile
Miscellaneous

Lepidopteran (moths and butterflies)

and Isopteran (Termites) remains were also identified. Odonata nymphs were less impor-
tant than ın the diet of otters. Mammal prey that could be identified were O. saundersiae
and Aethomys namaquensis.

Size of crabs eaten

Some 1860 eyestalks found ın Aonyx scats were undamaged enough to be measured. Using
equation (1), carapace widths could be calculated (Fig. 4a). The sıze of crabs varied from a
minimum carapace width of 4.3 mm to a maximum of 61.0 mm (mean carapace width =
28.5 mm, SD = 8.88). Of the eyestalks extracted 75 % represented crabs with carapace
widths of 15-35 mm (Fig. 4a).

Some scats contained only small crabs; e.g. 27 eyestalks found in one scat all rep-
resented crabs with carapace wıdths <30 mm. Others consisted of only larger sızed crabs,
e.g. 6 eyestalks found in one scat represented crabs all wıth carapace width > 30 mm. Very
few (1.7 %) crabs found in the scats had a carapace width >45 mm.

In the Atilax scats analysed 85 eyestalks were found, representing crab sizes ranging
from a minimum carapace width of 9.3 mm to a maximum of 47.7 mm (mean carapace
width = 29.4 mm, SD = 9.16). Most (62 %) of the crabs had a carapace width of 15-35 mm
(Fig. 4b). A significant difference (p < 0.001) was found between the sizes of crab preyed
upon by otters and water mongoose (Mann-Whitney Test, T = 6494). Atılax ate more
larger sızed crabs, although none of the remains in their diet represented crabs as large as
those eaten by Aonyx. The distribution of crab-sizes eaten by Atzlax was bi-modal. The
average wet weight of crabs taken was 8.33 g (range: 0.27-74.73 g) tor Aonyx capensıs and
8.98 g (range: 0.75-35.3 g) for Atılax paludinosus.

The foraging behaviour of two groups of otters, active at dawn, were observed during
February. One group comprised four individuals and the other two. Only crabs were seen
to be eaten. The 30 crabs taken were all caught by the otters diving and all were eaten in the

338 M. G. Purves, H. Kruuk, and ]J. A. J. Nel

% Frequency

SI
AISSSSSY
0-55

0
0-10 10-15 15-20 20-25 25-30 30-35 35-40 40-45 45-50 5

@!

arapace width in mm

(BD) 25

5-30 30-35 35-40 40-45 45-50 50-55

N
2

0 IN
0-10 10-15 15-20 20-25

Carapace width in mm

Fig. 4. Frequency of different sizes of crabs, as deducted from the length of eyestalks extracted from
the scats of (a) Aonyx capensis and (b) Atılax paludinosus

water. Larger crabs were eaten with the otter lying on ıts back and the pincers being fed
into the mouth first. Crab catching was concentrated in grassy areas where the depth of the
water was <1 m. No observations were made on the foraging behaviour of water

mongoose.
Discussion

Although the diets of both Aonyx and Atılax varıed during different seasons, crab
remained the single most important prey type for both, and both often ate rather small
sızed crabs. On average, Atılax took larger crabs than Aonyx.

Atilax occupies a wider range of habitats than Aonyx (RowE-RoweE 1978; WHITFIELD
and BLABER 1980; STUART 1981; SKINNER and SMITHERS 1990), it is more mobile on land
and wanders greater distances away from water (RAUTENBACH and NEL 1978). Although
there is considerable dietary overlap, terrestrial prey seem to be of more importance in the
diet of Atılax and the hunting of aquatic prey could be restricted to the shallows
(RAUTENBACH and NEr 1978).

Crabs in the diet of Aonyx capensis and Atilax paludinosus 339

In a marine habitat Louw and NEL (1986) found virtually no dietary overlap between
the two species, with Atzlax concentrating on shore crabs and other terrestrial species,
while Aonyx took mainly benthic prey. Thus there was a degree of spatial as well as food
segregation between the two sympatric species, with A. paludinosus foraging along the
water edge intertidally and A. capensis foragıng in the sea itself. ROwE-Rowe (1977a) found
a dietary overlap of 65% between the two species ın a freshwater habitat. Both species
seem to have much the same activity regimen, being mostly nocturnal and crepuscular
(RowE-RowE 1978; MAapDDock and PERRIN 1993; own observations).

Predatory and feeding behaviour have been described for the clawless otter (RowE-
Rowe 1977b) and water mongoose (BAKER 1989). Aonyx swims underwater, turning the
head from sıde to side and often feeling under stones for prey with the fore-feet (RowE-
Rowe 1977b). RowE-Rowe (1977b) also found that captured crabs were held with both
fore-feet and eaten in the water and that the whole crab was eaten. Our observations ın the
field confirmed thıs.

BAkER (1989) found that Atzlax sıghted crabs when swimming or when walking past
water. Their feet were then used in feeling for the crabs, but their heads were not
immersed. Once the crabs had been located, however, they ducked their heads under water
and caught the crabs in the mouth (BAKER 1989). Crabs were taken from the water and
eaten on land, and in large specimens parts of the carapace were often discarded (BAKER
1989). Empty carapaces, with intact eyestalks, were often seen near the Olıfants River and
if these were discarded Atılax prey, there might be a degree of bias ın the sizes of crabs
represented in the scats. However some birds e.g. Giant and Pied kingfishers, are also
known to prey on crabs and could have been responsible for the empty carapaces.

A difference in size of crab eaten by the two species could be expected, because of the
difference ın feeding behaviour of the two predators, and differences in movement patterns
between small and larger crabs (HıLL and O’KEeErFE 1992). Smaller crabs were more often
found in the grassy verges and muddy banks of the river and ın secluded pools away from
the main flow. Larger crabs appeared to prefer deeper water. Based on the differences ın
foraging areas of the otter and mongoose, and therefore availability of different sızed crabs,
one would expect Aonyx to consume larger crabs than Atzlax. Sımilarly, the disparity ın
body sıze (ca. 3 kg for Atılax, 10-12 kg for Aonyx) favours Aonyx takıng larger crabs.
However our results do not bear out this prediction.

In Israel GHERARDI and MicHELI (1989) found larger crabs straying up to 40 m away
from the water at night, while the smaller ones hid under rocks and in crevices at this time
and did not venture out of the water as often. In the Jonkershoek valley near Stellenbosch
large crabs are also occasionally seen venturing far (up to >500 m) away from the Eerste
River. If the same applies to the crabs in the Olıfants River, Atılax would more readily
encounter such crabs as they forage not only along the water edge, but also further afıeld.
This could explaın why the average size of crabs found in their diet was larger than that
found in the diet of otters.

Atılax utilized a wider range of prey, of which about 45 % was terrestrial. This and the
fact that tracks of Atılax were often concentrated along the side of the river and at shallows
and pools away from the main flow, indicates that they do not forage in water as much as
Aonyx.

In Thailand the SE Asıan small clawed otter Aonyx cinerea and the crab eating
mongoose Herpestes urva occur sympatrically. Both species feed on freshwater crabs
Potamon smithianus. Here it was found that the mongoose took significantly smaller crabs
than the otter (Kruuk et al. 1994 and unpubl. obs.). In contrast to the situation in South
Africa the size disparity between these two species is much smaller with A. cıinerea
weighing 5 kg, H. urva 3.4 kg. Both are nocturnal and solitary (MAacDonALD 1984) or
occur ın small groups (A. cinerea — pers. obs.). EwER (1973) mentions that A. urva feeds
largely on similar prey as Atılax paludinosus, i.e. frogs and Crustacea.

340 M. G. Purves, H. Kruuk, and J. A. J. Nel

Thus overlap ın resource utilization between crab-eating otters and mongooses does not
appear to be confined to Africa. However, despite the high proportion of crab in the diet
of these various species, there is no evidence to suggest that actual competition occurs. To
investigate that important question, further information is required about possible limiting
roles of the different prey species in the lives of these carnıvores.

Acknowledgements

We thank the Foundation for Research Development and the University of Stellenbosch for providing
funds for fieldwork, and travel funds for H. K.; P. R. SwiEGERS for field assistance; B. Sachse for her
help in analyzing scats and assisting in various other ways; H. HuysamEn for permission to work on
his farm; E. van Dijk for identifying anuran remains; H. GEERTSEMA for identifying insect remains
and I. Cordes for translating the German summary.

Zusammenfassung

Krabben, Potamonautes perlatus, in der Nahrung von Kapfıngerotter, Aonyx capensis, und
Wassermanguste, Atılax paludinosus, in einem Süßwasserhabitat in Südafrika

Die Nahrungsökologie zweier syntoper Arten, Kapfıngerotter und Wassermanguste, wurde am
Olifants-Fluß in der Kapprovinz, Südafrika, untersucht. Süßwasserkrabben waren ein wichtiger
Bestandteil in der Nahrung beider Arten. Anhand von Kotanalysen wurde ermittelt, in welchem
Maße sich die Nahrungsspektren überschneiden. Wassermangusten ernähren sich von durchschnitt-
lich größeren Krabben als Otter. Nach Krabben war Fisch der zweitwichtigste Bestandteil der
Nahrung der Otter, während terrestrische Tiere einen größeren Teil der Nahrung von Wassermangu-
sten ausmachen. Unterschiede ın der Ernährungsweise der beiden Raubtierarten und ın der Wahl von
Krabben verschiedener Größe können die gefundenen Unterschiede erklären.

References

ARDEN-CLARKE, C. H. G. (1986): Population density, home range size and spatial organization of the
Cape clawless otter, Aonyx capensis, in a marine habitat. J. Zool. (London), 209, 201-211.

BAkER, C. M. (1989): Feeding habits of the water mongoose (Atilax paludinosus). Z. Säugetierkunde
54, 31-39.

Du Toıt, C. F. (1980): The yellow mongoose Cynictis penicillata and other small carnıvores in the
Mountain Zebra National Park. Koedoe 23, 179-184.

EWER, R. F. (1973): The Carnivores. New York: Cornell Univ. Press.

GHERARDI, F.; MICHEL, F. (1989): Relative growth and population structure of the freshwater crab,
Potamon potamios palestinensis, in the Dead Sea area (Israel). Isr. J. Zool. 36, 133-145.

GHERARDI, F.; Tarpuccı, F.; MiıcHeLı, F. (1989): Energy maximization and foraging strategies in
Potamon fluviatile (Decapoda, Brachyura). Freshwater Biology 22, 233-245.

Hıuı, M. P.; O’Keerre, J. H. (1992): Some aspects of the ecology of the freshwater crab
(Potamonantes perlatus Milne Edwards) in the upper reaches of the Buffalo River, eastern Cape
Province, South Africa. S. Afr. J. Aquat. Scı. 18, 42-50.

Kruuk, H.; GoUDSwAARD, P. C. (1990): Effects of changes in fish populations ın Lake Victoria on
the food of otters (Lutra maculicollis Schinz and Aonyx capensis Lichtenstein). Afr. J. Ecol. 28,
322-329.

Kruux, H. (1992): Scent marking by otters (Zutra lutra): Signalling the use of resources. Behav. Ecol.
3, 133-140.

Kruuk, H.; KANCHANASAKA, B.; O’SULLIVAN, S.; WANGHONGSA, $. (1994): Niche separation in three
sympatric otters Lutra perspicillata, L. lutra and Aonyx cinerea in Huai Kha Khaeng, Thailand.
Biol. Conservation 69, 115-120.

Louw, C. J.; Ner, J. A. J. (1986): Diets of coastal and inland-dwelling water mongoose. S. Afr. ].
Wildl. Res. 16, 153-156.

MacDonaıp, D. W. (ed.) (1984): The encyclopaedia of mammals. Vol. 1. London: George Allen and
Urman.

MacDonatp, ]J. T.; NEL, J. A. J. (1986): Comparative diets of sympatric small carnıvores. $. Afr. ].
Wildl. Res. 16, 115-121.

Mappock, A. H.; PErRIN, M.R. (1993): Spatial and temporal ecology of an assemblage of viverrids ın
Natal, South Africa. J. Zool. (London) 229, 277-287.

RAUTENBACH, 1. L.; NEL, J. A. J. (1978): Coexistence in Transvaal Carnıvora. Bull. Carnegie Mus.
nat. Hist. 6, 138-145.

Crabs in the diet of Aonyx capensis and Atılax paludinosus 341

Rowe-RowE, D. T. (1977a): Food ecology of otters ın Natal, South Africa. Oikos 28, 210-219.

— (1977b): Prey capture and feeding behaviour of South African otters. Lammergeyer 23, 13-21.

— (1978): The small carnıvores of Natal. Lammergeyer 25, 1-48.

SKINNER, J. D.; SMITHERS, R. H. N. (1990). The mammals of the Southern African subregion.
Pretoria: University of Pretoria.

STUART, C. T. (1981): Notes on the mammalıan carnıvores of the Cape Province, South Africa.
Bontebok 1, 1-58.

TURNBULL-KEMP, P. St. J. (1960): Quantitative estimations of populations of the rıver crab, Potamon
(Potamonautes) perlatus (M. Edw.) in Rhodesian trout streams. Nature 185, 481.

VAN DER ZEE, D. (1981): Prey of the Cape clawless otter (Aonyx capensis) ın the Tsitsikama Coastal
National Park, South Afrıca. J. Zool. (London) 194, 467483.

— (1982): Density of the Cape clawless otter Aonyx capensıs (Schinz, 1821) ın the Tsitsikama
National Park. S. Afr. J. Wildl. Res. 12, 8-13.

VERWOERD, D. ]J. (1987): Observations on the food and status of the Cape clawless otter Aonyx
capensis at Betty’s Bay, South Africa. S. Afr. J. Zool. 22, 33-39.

WEHITFIELD, A. K.; BLABER, $. J. M. (1980): The diet of Atılax paludinosus (water mongoose) at St
Lucia, South Africa. Mammalıa 44, 315-318.

Authors’ addresses: M. G. Purvzs and Prof. J. A. J. Ner, Department of Zoology, University of
Stellenbosch, PB X5018, Stellenbosch 7599, R. S. A.; Dr. H. Kruuk, Institute of
Terrestrial Ecology, Banchory, Kincardineshire AB31 4BY, Scotland

Z. Säugetierkunde 59 (1994) 342-348
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Fates of fossorial Water voles, Arvicola terrestris, as revealed by
radiotelemetry

By F. Saucy

Institute of Zoology, University of Fribourg, Fribourg, Switzerland

Receipt of Ms. 28. 2. 1994
Acceptance of Ms. 18.7. 1994

Abstract

The fates of 94 fossorial water voles were studied in Switzerland from 1984 to 1986 using
radıotelemetry. Fifty-six individuals were recaptured on the study plots at the end of the radiotracking
sessions. Taking into account that 34 individuals either died, dispersed or were killed by predators, the
fates of 95.7 % of the voles could be assessed, as compared to only 59.6 % if losses were tallied by
classical capture-recapture analysıs. In most cases, death in situ (10 cases) could be distinguished from
the 13 recorded cases of predation (including 4 cases of predation on dispersers). Only 4 tags (4.3 %)
were lost. Moreover, the position of the tags and the marks left on the radio-collars allowed the
identification of mammalıan and avian predators in most cases of predation. Finally, the fates of the 15
dispersers could also be assessed.

Introduction

Dispersal and predation strongly affect population dynamics of small rodents (LiDICKER
1985; Kress 1992; PEARson 1985; Korpımäkı 1993). Unfortunately, classical trappıng
data do not allow the distinction between losses due to mortality and to dispersal.
Moreover, mainly because of technical limitations, actual rates of dispersal or of predation
are difficult to estimate. Therefore, there is still little data reported from natural popula-
tions. However, the recent wıdespread use of radıotelemetry ın small mammal studies
allows one to investigate these questions in greater detail (Mapıson et al. 1985; MCSHEA
1990; McSHEA and Mapıson 1992).

The water vole, Arvicola terrestris, is a widespread Palaearctic species that presents
varıous ecological forms (review in REICHSTEIN 1982). Contrasting with the lowland
populations of A. terrestris that are semi-aquatic and live in wet habitats (WIJNGAARDEN
1954; PELıkan and HouısovAa 1969; STODDART 1970; WIELAND 1973), the fossorial
populations (A. t. scherman) occur ın dry habitats ın mountainous parts of Central Europe
(MeEyLan 1981). Fossorial water voles live permanently in underground burrows ın
grasslands and meadows and they differ from the aquatic populations by many features,
including body sıze, population dynamics, social and mating behaviour (reviews ın
REICHSTEIN 1982; Saucy 1994a), as well as genetic varıabılıty (Saucy et al. 1994).
Furthermore, the fossorial populations undergo multiannual fluctuations ın density with
an unusually long cycle that lasts for 6 years, on average (Saucy 1988a, 1988b, 1994b).

The aim of this study was to investigate the fates of radıo-collared fossorial water voles
under natural conditions during the course of a multiannual cycle. In this study, data are
reported that were collected during experiments conducted in fossorial populations of the
water vole in the Jura mountains, as well as the Swiss Alps. In the Jura mountains, the fates
of radio-monitored voles were assessed during a decline, a period of low numbers, and the
beginning of the increase phase, whereas in the Swiss Alps voles were studied during
phases of high density.

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5906-0342 $ 02.50/0

Fates of fossorial Water voles, Arvicola terrestris 343

Material and methods

Population estimates

Before the beginning of a radiotracking session, population numbers were estimated during two-day
live-trapping sessions using the capture-mark-release sampling technique adapted for A. t. scherman
by Aıroıpı (1976, 1978). In all cases, densities (individuals/ha) were computed as the number of
captured voles divided by the surface of the plots actually sampled (Tab. 1). Because of the diversity in
density conditions (<1 ind/ha-> 300 ind/ha), varying areas have been sampled during the different
experiments. At Les Cluds (near Bullet, Jura Mountains, Switzerland, 1200 m, Tab. 1), where
radıotelemetry was applied as a complementary technique during the course of a long-term population
study (1982-1986; Saucy 1988a), population changes were monitored monthly (April-November) on
a 0.16ha quadrat (40. mx40 m) and three times a year on a larger rectangular 0.5 ha grid
(20 mx 200 m) (May, August and October; Saucv 1988a). Following the decline of the population
that occurred in 1984 (the populations became extinct in the above grids), trapping was extended to
the remainder of the grassland (15 ha). In the four localities from the Swiss Alps, population densities
were estimated on rectangular grids ranging from 0.13 to 0.20 ha.

Table 1. Study sites, dates, duration of the experiments, density, and phases of the population cycle

Exp} Location Duration ofthe Density Phase of
No. (coordinates) experiments (ind/ha) the cycle

(days)

Les Cluds 06-2510: 35 Decline

(6° 34’ E 46° 51’ N)

Les Cluds ED. BE 56 Low numbers
Les Cluds WS 49. 19 Increase
Flendruz “SD, 28. 7 Peak

(7° 10’ E 46° 30° N)

La Comballaz 5 10), 26 Peak

(7° 04’ E 46° 23’ N)

Le Sepey EDLER 7 Peak

(7° 03’ E 46° 23' N)

Rougemont 7 0,8. Peak
(7° 14’ E 46° 29' N)

Radiotelemetry

Following the technique described by several authors (BROooKs and Banks 1971; Banks et al. 1975;
Mapıson 1977; Mıneau and Mapıson 1977) 94 voles were equipped with radio-collar transmitters.
Ten SM-1 (AVM Instrument Co., Ltd, Livermore, USA) and 10 SS-1 (Biotrack, Wareham, UK)
transmitters were used during this study. The tags, weighing 3-4 g and transmitting in the 148-149
MHz frequency band, were powered by 85 mAh batterıes that lasted for 35 days, on average. They
were shaped with a dental acrylic resin to fit the throat of the voles. Spent batteries were replaced by
dissolving the resin in acetone. The radio-collars were fitted around the necks of the anımals under
ketamine anaesthesia (5 mg/100 g of body weight). The voles were then held in cages overnight to
permit recovery from the anaesthesia. Two 3-element Yagi antennae and a LA12-DS (AVM) radio-
receiver were used to detect transmitting voles at a range of 50 m to several hundred meters depending
on the local conditions. Implant-type transmitters, used by other authors (MıHor et al. 1988;
Mapıson et al. 1985), were tested and soon discarded because of an approximately 10-fold reduction
in detection range. In order to minimise disturbance, the two Yagi antennae were mounted on tripods
and were placed within a range of 20 to 30 m oftthe voles in such a position that the bearings formed an
approximately 90° angle. Localisations of voles were mapped using a reference grid of stakes placed at
2.5 m intervals. Variation in the signal intensity and in the rhythm of transmission enabled determina-
tion of whether an anımal was actıve or resting. Voles found to be inactive for more than 5 successive
localisations were considered to be dead animals. The radio-collars were then searched out and the
causes of deaths were determined. Individuals that left their home-ranges to settle at a distance in new
burrows and that were never relocated in their former home-range were classified as dispersers.

344 F. Sancy

Radiotracking sessions

Voles were radıiotracked during long-term (up to 8 weeks), as well as short-term (1—4 weeks) sessions
(Tab. 1). Long-term sessions involved the few and isolated individuals that could be trapped during
the population decline (1984) and the period of low numbers (1985) at Les Cluds (Exp. 1-2; Tab. 1).
The anımals were monitored once a day or every second day and the sessions were terminated by the
death or the disappearance of the anımals. During short-term sessions, a selection of 10 to 20 voles,
including all those individuals living in a restricted area, were studied simultaneously. These short-
term sessions were carried out in the four Alpine localities in 1985 and 1986 and in the Jura mountains
in 1986 (Exp. 3-7; Tab. 1). During these experiments, the voles were monitored 3—4 times each day
during the 1-week sessions or every second day during the 3- or 4-week experiments.

Finite survival and mortality rates have been estimated for each session. According to KrEBs
(1989), all rates have been converted into standard time intervals (7 days). The following rates have
been considered: 1) Gross survival rates = number alive at the end of a radiotracking session/number
of radıocollared individuals released at the beginning of the experiment, 2) Disappearance rate = I -
gross survival rate. The latter has subsequently been partitioned into death, dispersal and predation
rates according to the primary causes of disappearance from the study plots.

Results

At Les Cluds, the population peaked in 1983 (150 ind./ha), declined in 1984 and was nearly
extinct ın 1985 (Saucy 1988a). No voles could be caught on the 0.5 ha grid in 1985 and
extensive additional trapping carried out in the 15 ha neighbouring the grid did not yield
more than 7 individuals (0.5 ind/ha; Saucy 1988a). In the Alps, the high densities
(> 150 ind/ha) recorded in the four localities suggest that these populations were in a peak
phase (Tab. 1).

The fates of the 94 voles that were monitored ın all five localities during seven distinct
experiments are shown in table 2. Globally, 56 voles (59.6 %) were recaptured alive within
the limits of their trappıng grid at the end of the experiments, while 10 anımals (10.6 %)
died in their burrows, 9 individuals (9.6 %) were killed on the study site by a predator and
15 anımals (16.0 %) dispersed. The fates of only 4 voles (4.3 %) remained unexplained.
These anımals disappeared from their burrows and the transmitters could not be located
over several square kilometres of surrounding terrain.

Including 4 anımals that were killed after having dispersed, 13 cases of predation were
recorded during the study (Tab. 2). The level of predation was possibly highest at Les
Cluds in 1985 (Exp. 2). Two of the 3 monitored voles were killed by stoats and the fate of
the third anımal, although unknown, might also be explained by predation (a foraging
stoat was observed in the close vicinity of the study plot a few hours before the vole

Table 2. Fates of the 94 radiotracked voles

The primary causes of disappearance from the study plots are indicated in columns 4-6, while the fates
of the dispersers are given in column 8

Experiment Number of Alive Dead Killed Dispersed Unknown Killed or
No. voles radio- on site in situ bya fate disappeared
tracked predator after dıspersal

m m ON VID W
7 ,o-o000-
ı oOOOoO00O—-NMN

1
1
1
0
4
1
j
9

N
(0 7

Fates of fossorial Water voles, Arvicola terrestris 345

Table 3. Finite weekly rates of gross survival and of apparent mortality
Disappearance rates have been partitioned into 3 components (death, dispersal and predation)
according to the primarily ıdentified causes of disappearance (i1.e. columns 4-6 of Tab. 2). All rates

have been converted into standard 7-day time intervals

Gross survival Disappearance Dispersal Predation
rate

disappeared). Several cases of predation were also observed at Les Cluds ın 1984 and at La
Comballaz (sessions 1 and 5) where 3 and 4 of the monitored voles were killed by
predators. In the former case, two individuals were caught by predators after having
dispersed. Overall, 4 of the 15 dispersers (26.7 %) were killed or dısappeared soon after
dispersal.

Voles disappeared from the study plots at an average rate of 0.19 per week (Tab. 3;
extremes ranging between 0.13 and 0.36). On average, deaths, as well as predation on site
contributed to approximately 20% of weekly losses, while dispersal accounted for
approximately twıce thıs amount.

In most cases the predators could be identified owing to the localisation of transmitters
and to marks left on the tags. In seven cases, predation was attrıbuted to domestic cats Felıs
silvestris f. catus, which, when predating on voles, left many clearly visible biting or
chewing marks (little holes) on the collar. The tags were often relocated in barns or houses.
Predation by the stoat, Mustela erminea, was suspected on several occasions, but could be
established beyond doubt in only two cases (at Les Cluds in 1985). In both occurrences,
biting marks, similar to those left by cats, were found on the collars. The prey was either
devoured in the nest or conveyed over a distance and hidden under a mound of stones.
Finally, four cases of predation by avıan predators were recorded on the basis of the
distinctive marks left by the birds’ beaks on the collars. In one case, the transmitter was
relocated on a tree. The actual avian predator could be visually ıdentified in three cases.
Predation was then attrıbuted to the common buzzard, Buteo buteo, in two cases, to the
black kite, Milvus migrans, ın one case, and once to an unknown nocturnal avıan predator.

Discussion

It has been frequently reported from vole studies, that a high proportion of anımals
disappears between successive trapping sessions (reviews in KREBS and Myers 1974; TAITT
and Kress 1985). Few studies, however, have attempted to quantify the relative impact of
predation and of dispersal on the vole population dynamics under natural conditions.
Among others, HıLBorn and Kress (1976), using radioactive tags, were mostly
unsuccessful in explaining the fates of disappearing meadow voles (M. pennsylvanicus)
during a population decline, while MıHok et al. (1988) were unable to determine the death
causes of radio-tracked individuals. In contrast, McSHEA (1990), in a pooled analysıs of
three different studies on M. pennsylvanicus, reported estimates of losses to predation

346 F. Saucy

varying between 14 and 41% of the radiotracked voles. Correspondingly, unexplained
losses, however, ranged between 22 % and 36 % (McSHeaA 1990).

Apart from this study, radıotelemetry has been used in previous studies on Arvicola
terrestris by LEUZE (1976) and by JErpsson (1986, 1990). Both authors, however, studied
aquatic populations (in Scotland and Sweden, respectively) and only the former used this
method in a demographic perspective. Therefore, the results reported here provide the first
estimates of predation and of dispersal in A. t. scherman using radiotelemetry.

In this study, I was able to explain the fates of 90 out of the 94 (95.7 %) individuals that
were monitored. Moreover, the predator could be identified in 12 of the 13 (92.3 %)
instances of ascertained predation. Besides the 19 individuals that died or were killed by a
predator in situ, 15 voles dispersed and had left the study plot by the end of the
experiment. If these populations had been monitored by trapping, only 56 individuals
(59.6 %) would have been recaptured within the study area. Therefore, the fates of 40.4 %
of the voles would have remained unexplained instead of only 4.3 % ın thıs study.

Moreover, and in spite of the few numbers involved, these results indicate that dispersal
is a rısky event in Arvicola terrestris, since 4 of the 15 dispersers (26.7 %) were eventually
killed by predators. This finding confirms observations made by LeuzE (1976) who
recorded even higher losses due to predation (mainly from herons) during the dispersal
phase of young water voles (up to 50 % of dispersing females).

Several interpretations can be ınvoked to explain the fates of the voles that diısappeared.
Transmitter faılures can be dismissed, as the voles would have been recaptured in their
burrows. It ıs possible, although unlikely, that the tags suddenly stopped functioning
following the death or the dispersal of an anımal. Long distance dispersal movements are
sımilarly unlikely. Fossorial water voles usually disperse within 30 to 100 m (Saucy 1988a)
and transmitters were searched over large areas, encompassing several hundreds of metres
around the study plots. A likely explanation ıs that the voles were killed by predators that
either damaged the transmitters or conveyed them over large distances. Foxes, Vulpes
vulpes, which were common on the study sites could be, among others, responsible for
damaging tags when preying on A. t. scherman. The transports ot tags over long distances
(200-800 m) by varıous predators (including domestic cats, stoats and avian predators)
were recorded on several occasions.

There is a strong trophic relationship between the stoat and the fossorial form of the
water vole (DEBROT 1981). Stoat populations undergo cyclic fluctuations lagging 1 year
behind those of A. terrestris. The impact of the stoat on populations of A. terrestris ıs likely
to be especially high during declines and periods of low numbers. Although the data are
few, this study supports the hypothesis that stoats, which reach their highest densities at
the end of the population peaks of A. t. scherman (DEBRoT 1981), might drive already
declining populations to extinction. During the course of this study, the few voles that
were found to survive the decline at Les Cluds were suspected to have been killed by stoats
(Saucy 1988a).

In conclusion, the present study provides preliminary estimates of survival, dispersal
and predation rates in A. t. scherman using radiotelemetry. It also confirms the potentially
strong impact of predators on the population dynamics of this vole. This suggests a role for
predation in the unusually long population cycle reported ın this species.

Acknowledgements

This study was carried out during the course of a PhD thesis supported by the Station Federale de
Recherches Agronomiques de Changins, Nyon, Switzerland. I am very grateful to Dr. A. MEyLAN
who supervised the study and to G. Mayor and J.-L. BERTHOLET for their technical and field
assistance. I also would like to thank Drs. ]J.-P. Arrorpı, R. K. Rose and A. BucHara who read and
greatly improved earlier drafts of the manuscript. Anonymous referees made helpful comments. The
summary was translated into German by M. Krıec.

Fates of fossorial Water voles, Arvicola terrestris 347

Zusammenfassung

Erfassung von Einzelschicksalen bei Ostschermäusen (Arvicola terrestris) mit Hilfe der Radiotelemetrie

Mittels Radiotelemetrie wurde in der Schweiz von 1984 bis 1986 das Leben von 94 Ostschermäusen
(Arvicola terrestris scherman) verfolgt. Am Ende der Radiotelemetrie-Versuche konnten in den
Untersuchungsparzellen 56 Individuen wieder gefangen werden. In Anbetracht der Tatsache, daß 34
Individuen entweder starben, auswanderten oder durch Prädation getötet wurden, konnte das
Schicksal von 95.7 % der Schermäuse beurteilt werden, im Gegensatz zu 59.6 %, wenn die Verluste
mittels der klassischen Fang-Wiederfang-Methode erfaßt worden wären. In den meisten Fällen konnte
der Tod in sıtu (10 Fälle) vom Tod durch Prädation (13 Fälle) unterschieden werden. Nur vier
Radiosender (4.3 %) gingen verloren. Darüber hinaus konnten in zwölf von 13 Fällen die Art der
Säuger- und Vogelprädatoren bestimmt werden. Zur Identifizierung der Prädatoren wurden die
Position der Radiosender und die auf den Halsbändern hinterlassenen Spuren verwendet. Schließlich
wurde auch das Schicksal der 15 Auswanderer beurteilt.

References

Aırorpı, J.-P. (1976): Experiences de capture et recapture chez le Campagnol terrestre, Arvicola
terrestris scherman Shaw (Mammalıa, Rodentia). Terre Vie 30, 31-51.

— (1978): Etude par capture et recapture d’une population de Campagnols terrestres, Arvicola
terrestris scherman Shaw. Terre Vie 32, 3-45.

Banks, E. M.; BRooKs, R. ].; SCHNELL, J. (1975): A radıotracking study of home range and activity of
the brown lemming (Lemmus trimucronatus). J. Mammalogy 56, 888-901.

Brooks, R. J.; Banks, E. M. (1971): Radio-tracking study of lemming home range. Commun. Behav.
Biol. 6, 1-5.

DEBRoOT, $. (1981): Trophic relations between the stoat (Mustela erminea) and its prey, mainly the
water vole (Arvicola terrestris scherman). In: Worldwide Furbearer Conf. Proc. Ed. by ]. A.
CHAPMAN and D. Purstey. Frostburg, Maryland: Worldwide Furbearer Conf. Inc. Vol. 2,
1259-1289.

HıLBorn, R.; Kress, C. J. (1976): Fates of disappearing individuals in fluctuating populations of
Microtus townsendü. Can. J. Zool. 54, 1507-1518.

Jerpsson, B. (1986): Mating by pregnant water voles (Arvicola terrestris): a strategy to counter
infanticide by males? Behav. Ecol. Sociobiol. 19, 293-296.

— (1990): Effects of density and resources on the socıal system of water voles. In: Socıal systems and
population cycles ın voles. Ed. by R. H. TamArın, R. OSTFELD, $. Puch, and G. BuJALska.
Basel: Birkäuser Verlag. Pp. 213-226.

Korrımäkı, E. (1993): Regulation of multiannual vole cycles by density-dependent avian and
mammalıan predation? Oikos 66, 359-363.

Kress, C. J. (1989): Ecological methodology. New York: Harper and Row.

— (1992): The role of dispersal in cyclic rodent populations. In: Animal dispersal. Small mammals as
a model. Ed. by N. C. STENSETH und W. Z. Jr. LiDicker. London: Chapman and Hall. Pp.
160-175.

Kress, C. J.; Myers, ]J. H. (1974): Population cycles in small mammals. Adv. Ecol. Res. 8, 267-399.

LEuZE, C. C. K. (1976): Social behaviour and dispersion in the water vole, Arvicola terrestris,
Lacepede. Ph. D. thesis, Univ. of Aberdeen.

LIDICKER, W. Z. Jr. (1985): Dispersal. In: Biology of new world Microtus. Ed. by R. H. Tamarın.
Spec. Publ. Amer. Soc. Mammal. Vol. 8, 420-454.

Mapıson, D. M. (1977): Movements and habitat use among interacting Peromyscus lencopus as
revealed by radiotelemetry. Can. Field. Nat. 91, 273-281.

Mapıson, D. M.; FizGEraLd, R. W.; McSHeA, W. J. (1985): A user’s guide to the successful
radiotracking of small mammals in the field. In: Proc. Fifth Int. Conf. on Wildlife Biotelemetry.
Ed. by R. W. Weeks and F. N. Long. Laramie: Univ. of Wyoming Press. Pp. 28-39.

McSHEA, W. J. (1990): Predation and its potential impact on the behavior of microtine rodents. In:
Social systems and population cycles in voles. Ed. by R. H. TamArın, R. OsTFELD, $. PucH, and
G. Bujaıska. Basel: Birkäuser Verlag. Pp. 101-110.

McSHEA, W. J.; Mapıson, D. M. (1992): Appendix 2. Alternative approaches to the study of small
mammal dispersal: insights from radiotelemetry. In: Animal dispersal. Small mammals as a model.
Ed. by N. C. STEnsETH and W. Z. Jr. Livicker. London: Chapman and Hall. Pp. 319-332.

Meyran, A. (1981): Bilan de quelques annees de recherches fondamentales et appliquees sur le
Campagnol terrestre, Arvicola terrestris scherman (Shaw). Defense des Vegetaux 208, 143-154.

MIHoR, S.; LawTon, T.; SCHWARTZ, B. (1988): Fates and movements of meadow voles (Microtus
pennsylvanicus) following a population decline. Can. J. Zool. 66, 323-328.

MineaAu, P.; Mapıson, D. (1977): Radio-tracking of Peromyscus leucopus. Can. J. Zool. 55, 465-468.

348 F. Saucy

PEARson, O. P. (1985): Predation. In: Biology of new world Microtus. Ed. by R.H. TAMARIN. Spec.
Publ. Amer. Soc. Mammal. Vol. 8, 535-566.

PELIKAN, J.; HoLısovA, V. (1969): Movements and home ranges of Arvicola terrestris on a brook.
Zool. Listy 18, 207-224.

REICHSTEIN, H. (1982): Arvicola terrestris (Linnaeus, 1758) — Schermaus. In: Handbuch der
Säugetiere Europas. Ed. by. J. NIETHAMMER and F. Krapp. Vol. 2/I: Nagetiere II. Wiesbaden:
Akademische Verlagsges. Pp. 217-252.

Saucy, F. (1988a): Dynamique de population, dispersion et organisation sociale de la forme fouisseuse
du Campagnol terrestre (Arvicola terrestris scherman [Shaw], Mammalia, Rodentia). PhD thesis,
Neuchätel, Switzerland.

— (1988b): Description des cycles pluriannuels d’Arvicola terrestris scherman (Shaw) en Suisse
occıdentale par la methode de Panalyse des series temporelles. EPPO Bull. 18, 401413.

— (1994a): Population dynamics of fossorial populations of Arvicola terrestris in Western Europe
with particular reference to life history varıation and population ecology between fossorial and
aquatic forms. In: Arvicola. Ed. by P. PAnTELEYEv. Moscow: Nauka (in press).

— (1994b): Density dependence in time series of the fossorial form of the water vole, Arvicola
terrestris. Oikos (in press).

Saucy, F.; Wust-Saucy, A.-G.; PeLz, H.-]J. (1994): Biochemical polymorphism and genetic variabil-
ity in aquatic and fossorial populations of the water vole, Arvicola terrestris, in western Europe.
Pol. ecol. Stud. (in press).

STODDART, D. M. (1970): Individual range, dispersion and dispersal in a population of Water voles
(Arvicola terrestris [L.]). J. Anım. Ecol. 39, 403-425.

Taıtt, M. ]J.; Kress, C. J. (1985): Population dynamics and cycles. In: Biology of new world
Microtus. Ed. by R. H. Tamarın. Spec. Publ. Amer. Soc. Mammal. Vol. 8, 567-620.

WIELAND, H. (1973): Beiträge zur Biologie und zum Massenwechsel der großen Wühlmaus (Arvicola
terrestris L.). Zool. Jb. Syst. 100, 351-428.

WIJNGAARDEN, A. Van (1954): Biologie en bestrijding van de Woelrat, Arvicola terrestris terrestris L.
in Nederland, PhD thesis, Eindhoven, The Netherlands.

Author’saddress: Dr. Francıs Saucy, Institute of Zoology, University of Fribourg, Perolles,
CH-1700 Fribourg, Switzerland

Z. Säugetierkunde 59 (1994) 349-357
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Does Microtus majori occur in Europe?

By B. KrySturEk, MARIA GRAZIA FILIPPUCCI, M. MACHOLAN, J. ZIMA, M. VUJOSEVIC,
and S. SIMSON

Slovene Museum of Natural History, Ljubljana, Slovenia; Dipartimento di Biologia, II Universita di
Roma “Tor Vergata”, Rome, Italy; Institute of Animal Physiology and Genetics, A.S.C.R., Brno,
Czech Republic; Institute for Biological Research, Belgrade, Yugoslavia

Receipt of Ms. 11.4. 1994
Acceptance of Ms. 9. 6. 1994

Abstract

Voles from Mt. Pelister, Macedonia, which are known from the literature as Microtus majori, were
subjected to morphometric, karyotypic and electrophoretic analyses, and compared with Microtns
subterraneus from Slovenia and Montenegro, and Microtus major from Asıa Minor. The diploid
chromosome number of the Pelister voles (2n = 52) is the same as ın M. subterraneus from the Balkans.
Genetic distances, as revealed by electrophoretic analysıs of 27 gene locı between voles from Mt.
Pelister and M. subterraneus from Slovenia and Montenegro correspond to those generally observed
among subspecies of Arvicolidae. Discriminant analysıs of 12 raw skull measurements separated
successfully M. majorı from M. subterraneus. According to this, the Pelister population should be
allocated to the latter, which means that there ıs no reason to include M. mayori ın the list of the
European fauna.

Introduction

Microtus majori Thomas, 1906 ıs considered to inhabit the Caucasus, northern Turkey and
north-western Iran (probably incorrectly given as north-eastern Iran by Gromov and
BARANOVvA 1981), but it was recently also reported from the Balkans (MussEr and
CARLETON 1993). Namely, MArLEc and STORCH (1963) and later FELTEN and STORCH
(1965) point out the existence of two sıze classes amongst voles belonging to the Microtns
subterraneus group from Macedonia. The larger morphotype, from Mt. Pelister, was first
ascribed to Pitymys multiplex (Fatio, 1905) but later ıdentified as Pitymys majori (FELTEN
et al. 1971) or Microtus majori (STORCH 1982). Subsequent authors reported additional
localıties tor Microtus majorı ın the Balkans: Kıvanc (1986) ın European Turkey and
NIETHAMMER (1986) in northern Greece. However, other recent studies of the rodent
fauna of the Balkans have not confirmed the presence of this species in Greek Macedonia
(VOHRALIK and SoFIANIDOU 1987) or failed to mention it within the territory of the former
Yugoslavia (PETRoV 1992).

The aım of the present study is to reevaluate the identity of the voles from Mt. Pelister
which form the basis of the inclusion of M. major: in the list of European mammals. This
population was compared with M. majori from Asıa Minor, and with two populations of
M. subterraneus (de Selys Longchamps, 1936) from south-eastern Europe by biomerric,
karyotypic and electrophoretic analyses.

Material and methods

We examined 107 voles. Material included voles from museum collections, as well as freshly collected
anımals. Standard museum specimens are housed in the following collections (acronyms in brackets):
British Museum (Natural History), London (BMNH); Forschungsinstitut und Natur-Museum
Senckenberg, Frankfurt am Main (SMF); Naturhistorisches Museum Wien, Vienna (NMW); and
Slovene Museum of Natural History, Ljubljana (PMS). Material was pooled into four samples: sample

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5906-0349 $ 02.50/0

350 B. Krystufek et al.

1: Microtus majori, Asıa Minor (vicinity of Trabzon, including the type of majorı; BMNH); sample 2:
M. subterraneus, Slovenia (PMS); sample 3: Mt. Pelister, Macedonia (NMW, PMS, SMF); sample 4:
M. subterraneus, Mt. Lovcen, Montenegro (PMS).

All our specimens from sample 2 were collected at Kopanki, Mt. Pelister, or from its vicinity, i.e.
in the same area from which voles ascribed to M. majorı originated.

Chromosomes: Nine individuals were examined karyologically (Pelister 7; Lovcen 2). Standard
flame-dried preparations were made directly from the bone marrow of colchicined animal (Forp and
HAMERTON 1956). In most anımals, metaphase spreads were differentially stained by following the
slightly modified G-banding and C-banding methods of SEABRIGHT (1971) and SuMNER (1972).
Nucleolar organiser regions (NORSs) were revealed by the silver staining technique of HoweELL and
BLack (1980).

Allozyme analysıs: Tissue samples from 14 fresh specimens (Pelister 6; Lovcen 2; Slovenia 6) were
collected and transported in liquid nitrogen to the laboratory, where they were preserved at -80 °C
until processed. Homogenates for electrophoresis were obtained from portions of muscle tissue
crushed in distilled water. Electrophoretic analysıs was carried out on 27 locı, encoding 21 enzymes:
a-Glycerophosphatase dehydrogenase (E.C. 1.1.1.8; «Gpdh), Sorbitol dehydrogenase (E.C. 1.1.1.4;
Sdh), Lactate dehydrogenase (E.C. 1.1.1.27 Ldh-1 and Ldh-2), Malate dehydrogenase (E.C. 1.1.1.37;
Mdh-1 and Mdh-2), Malic enzyme (E.C. 1.1.1.40; Me-1 and Me-2), Isocitrate dehydrogenase (E.C.
1.1.1.42; Idh-1), 6-Phosphogluconate dehydrogenase (E.C. 1.1.1.44; 6Pgdh), Glucose-6-phosphate
dehydrogenase (E.C. 1.1.1.49; G6pdh), Indophenol oxidase (E.C. 1.15.1.1.; Ipo), Nucleoside
phosphorylase (E.C. 2.4.2.1; Np), Glutamateoxaloacetate transaminase (E.C. 2.6.2.1; Got-1 and
Got-2), Hexokinase (E.C. 2.7.1.1; Hk), Creatine kinase (E.C. 2.7.3.2; Ck), Adenylate kinase (E.C.
2.7.4.3; Adk), Phosphoglucomutase (E.C. 2.7.5.1; Pgm-1 and Pgm-2), Esterases (E.C. 3.1.1.1; Est-1
and Est-3), Acıd phosphatase (E.C. 3.1.3.2; Acph). Adenosine deaminase (E.C. 3.5.4.4; Ada),
Aldolase (E.C. 4.1.2.13; Aldo), Mannose phosphate isomerase (E.C. 5.3.1.8; Mpı), Glucose phos-
phate isomerase (E.C. 5.3.1.9, Gpi).

Electrophoretic procedures follow those described in Fırıppuccı et al. (1988). Isozymes were
numbered in order of decreasing mobility from the most anodal. Allozymes were designated
numerically according to their mobility relative to the most common allele in the reference population
from Mt. Pelister. Allozymic data were analysed as genotype frequencies with the BIOSYS-1 program
of SWOFFORD and SELANDER (1981). The amount of genetic divergence between populations was
estimated using the indices of standard genetic identity (I) and distance (D) proposed by NEi (1978). A
dendrogram of the genetic relationships between populations was obtained using unweighted pair-
group analysıs, UPGMA (SokAL and SNEATH 1963).

Morphometric analysis: Specimens were preserved as skulls wıth skins or in alcohol (skulls
extracted). Only adult, undamaged skulls (total 53: sample 1: 11; sample 2: 16; sample 3: 24; sample
4:2) were used for multivariate analysıs. Twelve skull measurements were taken from each skull
(Fig. 1) using a vernier calliper, accurate to the nearest 0.1 mm. The abbreviations used were: CbL -
condylobasal length; RoL - rostrum length; NcL - neurocranial length; DiL - diastema length; MxT -
maxillary toothrow length; ZgB — zygomatic breadth; BeB - braincase breadth; IoC - interorbital
constriction; BcH - braincase height per bullae; Bc - braincase height without bullae; RoH1 - height
of rostrum at the anterior alveoli of the first upper molar; RoH - height of rostrum across the second
upper molar.

Variations in metrical characters among samples were analysed by discriminant analyses of raw
data using the Statgraphics statistical program (version 5).

Results and discussion
Karyotype

The diploid chromosome number of voles from Pelister (sample 2) is 2n = 52. The
karyotype comprises one pair of large subtelocentric and one pair of large submetacentric
autosomes. The remaining autosomes are acrocentrics of decreasing size, except for the
smallest pair which is metacentric. Distinct short arms can be seen in most of the
acrocentric chromosomes. The X chromosome is a large metacentric and the Y ıs a large
acrocentric (Fig. 2).

There are at least four pairs of acrocentric chromosmes possessing NORs. These are
located in the telomeric region of one pair of medium-sized chromosomes, and ın the
pericentromeric area of one small pair. In two pairs of autosomes they are displayed on the
apparent short arms (Fig. 3a). The G-banded sex chromosomes are shown in figure 3b.
Whereas the X chromosome seems to be of a standard type as far as both sıze and bandıing

Does Microtus majorı occur in Europe? 3a

Fıg. 1. Cranıal measurements of voles used in this study. See text for abbreviations

Te Tem TWETmET

Au BR vn a vn du Ba

un v8 Aa na an 7 88 Mm

6% an en 2x 3

Fig. 2. Conventionally stained karyotype of Microtus subterraneus from Mt. Pelister

352 B. Krystufek et al.

b_ c

U 3

N \
Ss

Fig. 3. Chromosomes of Microtus subterraneus from Mt. Pelister. a: Four chromosome pairs carrying
NORSs; b: G-banded; c: C-banded sex chromosomes (the Y on the right)

pattern are concerned, the Y is conspicuously large (nearly as large as the X) and without
any obvious G-banding pattern. Only weak C-bands were revealed in some acrocentric
autosomes ın C-banded metaphases. The X chromosome displays no C-bands. In contrast,
the Y chromosome is entirely heterochromatic (Fig. 3c).

The karyotype of the Lovcen specimens appears to be sımilar to that of the Pelister
voles: 2n = 52, with one pair of large subtelocentrics, one pair of large submetacentrics,
and one pair of small metacentrice chromosomes. Since C-banding was not successful in the
only male available from Lovcen, the question of Y chromosome sıze remains open.

The same diploid number of chromosomes (i.e. 2n = 52) has already been reported for
Microtus subterraneus from the former Yugoslavia, including populations from Slovenia
(Zıveovi6 et al. 1975), and from Mt. Pelister (PETRov and Zıvkovi@ 1979). Microtus
majori from the Caucasus displays a different diploıd number, 2n = 54 (Zıma and Krau
1984), while the karyotype of M. majorı from its type localıty (in the vicinity of Trabzon,
northern coast of Asıa Minor) has not been studied yet. A karyotype of 2n = 52
chromosomes was also found in M. subterraneus from varıous parts of Europe (see
NIETHAMMER 1982; SaBLINA et al. 1989; Zıma and KrarL 1984, for review). A large Y
chromosome was reported from the Austrian Alps (GAMPERL et al. 1982) whereas, ın
certain other regions of Europe, only the standard, smaller Y has been found (ZımA 1984;
SABLINA et al. 1989). It should be noted, however, that the Y chromosome of the Alpine
population is only slightly different from the standard Y chromosome, while the element
found in the Pelister population is about twice as large as the standard Y. Thus, this
phenomenon cannot be interpreted as being a consequence of a varyıng degree of
chromosome spiralisation ın individual preparations, but should be considered to be a
specific feature of the study population.

Electrophoretic analysis

Seventeen of the twenty-seven loci analysed were monomorphic and fixed for the same
allele in all the populations studied: «Gpdh, Sdh, Ldh-1, Ldh-2, Mdh-1, Mdh-2, Me-2,
Idh-1, Ipo-1, Np, Got-2, Hk, Ck, Adk, Pgm-2, Aldo, Pgi. The allele frequencies of the
polymorphic locı in the populations analysed are given ın table 1.

Two locı (Got-1 and Pgm-1) partially discriminated the Mt. Pelister population from
M. subterraneus from Mt. Lovcen and Slovenia.

From the allele frequencies at the 27 locı tested, Nei’s values of genetic identity and
distance were calculated amongst populations using all pairwise comparisons (Tab.2). An
UPGMA dendrogram summarızing the genetic relationships between the samples ıs given
in figure 4.

The lowest genetic distance value found was between populations from Slovenia and
Mt. Lovcen (D = 0.011). The population from Mt. Pelister displayed higher values of

Does Microtus majori occur in Europe? 558

Table 1. Allelic frequencies observed at the polymorphic loci analysed in Balkan populations of
M. subterraneus.

See text for explanation
Pelister

Slovenia Lovcen

0.83 1.00 1.00

0.17 — =

0.80 1.00
0.20

0.92
0.08

= 0.08
1.00 9.92

0.08 1.00
0.92 _

0.08 0.75
0.92 0.25

0.33
0.67

0.17
0.83

1.00

0.08
0.92

0.08
0.92

genetic distance with those from Slovenia (D = 0.067) and Montenegro (D = 0.062). These
values correspond to those generally observed among subspecies of Arvicolidae (D =
0.064; Grar 1982) and more generally in other rodents (Fırıppuccı et al. 1991).

Phenetics

Microtus majori possesses three pairs of teats, two inguinal and one pectoral, while the
pectoral teats are absent in M. subterraneus (NIETHAMMER 1972). All of our seven lactatıng
females from Mt. Pelister had only the two inguinal pairs. In contrast, one standard
museum skin from Mt. Pelister (SMF 23,585) clearly shows an additional pectoral pair of
teats; this is the only female in the SMF collection which was obviously lactating. The
possibility that the number of teats may be polymorphic in the marginal population of M.
subterraneus is further supported by a per-

sonal communication from B. PrErrov. Table 2. Values of genetic identity (Nei’s I,

Amongst four females that he collected on
Mt. Orjen, Montenegro, from which only
the M. subterraneus karyotype has been
reported (Zıvkovi6 et al. 1975), the two
lactating females had an additional pair of
pectoral teats. A polymorphism in this
character has also been reported in Microtus
savıı (de Selys Longchamps, 1838), which
has either 2 or 3 pairs of teats (Krapp 1982).

above the diagonal) and distance (Nei’s D,
below the diagonal), between Balkan samples
of M. subterraneus, based on 27 loci

Pelister

Slovenia Lovcen

0.989 0.935
0.011 _ 0.940
0.067 0.062 _

Slovenia —

Lovcen
Pelister

354 B. Krystufek et al.

I 7
0.05 6) D

Fig. 4. UPGMA dendrogram summarizing genetic relationships between three populations of Mic-
rotus subterraneus. D = Ner’s (1978) unbiased genetic distance, based on 27 enzyme locı. The
cophenetic correlation coefficient is 0.997

The tail appears to be relatively longer in M. majori from Asıa Minor than in M.
subterraneus from Slovenia. In the voles from Mt. Pelister the taıl is as short as in Slovenian
M. subterraneus (Fıg.5). However, since external measurements were taken by different
collectors, they are likely to have been affected by differences in measuring techniques.

The first two functions resulting from the discriminant analysis of 12 raw skull
measurements of four samples (which were responsible for 91.8 % of the varıance) clearly
distinguished European M. subterraneus from Asıan M. majori (Fig. 6). According to their
skull morphology, the Pelister population should thus be allocated to M. subterraneus. All
specimens of M. majorı were classified correctly, whilest there was some overlap between
samples 2 to 4 (Tab. 3). A total of 9 specimens (= 17%) was misclassified.

In the next step, samples 2 to 4 (M. subterraneus) were pooled and discriminant analysis
was repeated. Specimens of both taxa were allocated into their actual groups. Removing
five cranial varıables (NcL, Dil, NcL, BcH, Mxt) from the discriminant analysis did not
affect the classification results. The discriminant function, based on 7 raw skull measure-
ments, could be useful in distinguishing the two species in museum material (Fig. 7):

DF = -0.60888x CbL - 1.19402 x ZgB + 3.08761 x BcB - 3.00499 x [IoC
+3.18741x Bc +2.79813 x RoH1 - 3.81314x RoH - 25.8104

The discriminant function has values lower than 1.1 in M. subterraneus and higher than 1.5

in M. magori.
Sample 1
) Hulp
& | Sample 3
0)

Sample 2

oz

‚vod, ragen Weir 7 me Tilgrpea Bere
25 30 35 40 45%

Fig. 5. Frequency histogram of the relative tail length (100x tail / head and body length) in Microtus
subterraneus (sample 2), M. majori (sample 1) and voles from Mt. Pelister (sample 3)

Does Microtus majori occur in Europe? 935

DF1

Fig. 6. Projection of four samples of 53 voles on the first two discriminant variates. Polygons enclose
scores for all individuals within a locality group, and crosses are placed on group centroids. See text
for identifying numbers

Table 3. Classification table for analysis based on four groups of voles

Rows are actual and columns are predicted groups (in %)

Predicted group
Actual group p Sample size

1 M. majorı 0.0
2 Slovenia $ 75.0
3 Pelister Ä 16.6
4 Lovcen i 0.0

Taxonomic conclusions

The present evidence does not suggest the inclusion of the Pelister voles into Asıa Minor’s
M. majori. For the time being, it seems much more appropriate to include them in M.
subterraneus. This also means that there is no reason to include M. majori ın the European
fauna. Anyhow electrophoretic, as well as karyological data, indicate that the Pelister

N IT DRG

10

-3 .-2 -1 0 1 2 3 4 5 DF

Fig. 7. Distribution of the specimens of Microtus majori (black) and Microtus subterraneus (white) in
the discriminant function axis. Stars represent placement of group centroids

356 B. Krystufek et al.

population may be distinct from M. subterraneus from the rest of Europe. Its taxonomic
relations to other large-sızed Balkan voles, usually ascribed to M. subterraneus (her-
cegovinensis, brauneri), remain open. According to the electrophoretic analysis, very large
voles from Mt. Lovcen, which are even bigger than M. majori, are genetically closer to
small M. subterraneus from Slovenia than to the Pelister voles. The fact that M.s.
hercegovinensis (Martino, 1940) and other populations of similarly large M. subterraneus
from Bosnia and Herzegovina have recently been placed into Microtus multiplex (PETROV
1992) suggests that the taxonomy of large “M. subterraneus” in the Balkans continues to
remain a source of debate.

Acknowledgements

We thank the following persons for allowing us to examine specimens under their care: Drs. K.
BAUER, P. JENKINs, F. SPITZENBERGER, and G. STORCH. Dr. G. STORCH also provided much
stimulating discussion and H. GRIFFITHS improved the English text. This research was partly
supported by the Ministry of Science and Technology of Slovenia (grant P1-5079-0614 to BK).

Zusammenfassung

Kommt Microtus majorı ın Enropa vor?

Wühlmäuse aus dem Pelister-Gebirge in Makedonien, die in der Literatur als Microtus majorı geführt
werden, wurden morphologischen, karyologischen und elektrophoretischen Untersuchungen unter-
worfen und mit M. subterraneus aus Slowenien und Montenegro sowie M. majori aus Kleinasien
verglichen. Die diploide Anzahl der Chromosomen (2n = 52) der Wühlmäuse vom Pelister stimmt mit
M. subterraneus vom Balkan überein. Die genetischen Distanzen, ermittelt durch elektrophoretische
Analyse von 27 Genloci, zwischen Wühlmäusen vom Pelister und M. subterraneus aus Slowenien und
Montenegro entsprechen denen, die gewöhnlich zwischen Subspezies von Arvicolidae festgestellt
werden. Mittels einer Diskriminanzanalyse von 12 Schädelmaßen ließ sich M. majori und M.
subterraneus erfolgreich trennen. Danach muß die fragliche Population aus dem Pelister-Gebirge
letztgenannter Art zugeordnet werden, was bedeutet, daß M. majori aus der Liste der europäischen
Fauna zu streichen ist.

References

FELTENn, H.; STORCH, G. (1965): Insektenfresser und Nagetiere aus N-Griechenland und Jugoslawien.
Senck. biol. 46, 341-367.

FELTEN, H.; SPITZENBERGER, F.; STORCH, G. (1971): Zur Kleinsäugerfauna West-Anatoliens. Teil I.
Senck. biol. 52, 393424.

Fırıppuccı, M. G.; Ropıno’, E.; NEvo, E.; CAPAnNA, E. (1988): Evolutionary genetics and
systematics of the garden dormouse, Eliomys Wagner, 1840. 2. Allozyme diversity and differentia-
tıon of chromosomal races. Boll. Zoll. 55, 47-54.

Fııppuccı, M. G.; FADDA, V.; KRYSTUFER, B.; Sımson, $.; Amorı, G. (1991): Allozyme varıatıon
and differentiation in Chzonomys nivalıs (Martins, 1842). Acta Theriol. 36, 47-62.

Fornp, C. E.; Hamerton, J. L. (1956): A colchicine, hypotonic citrate, squash sequence for
mammalian chromosomes. Stain Technol. 31, 247-251.

GAMPERL, R.; EHMAnnN, C.; BACHManN, K. (1982): Genome size and heterochromatin varıation in
rodents. Genetica 58, 199-212.

GROMOov, I. M.; BarANova, G. I. (1981): Catalog of the mammals of the U.S.S.R., Pliocene - recent.
Leningrad: Nauka (In Russian).

Grar, J. D. (1982): Genetique biochimique, zoogeographie et taxonomie des Arvicolidae (Mammalıa,
Rode), Revue suisse Zool. 89, 749-787.

Howeıı, W. M.; BLack, D. A. (1980): Controlled silver-staining of nucleolus organizer regions with
protective collodial developer: a 1-step method. Experientia 36, 1014-1015.

Kıvang, E. (1986): Microtus (Pitymys) majori Thomas, 1906 in der europäischen Türkei. Bonn. zool.
Beitr. 37, 39-42.

Kraı, B.; MıTEv, D. B. (1976): Karyotypes of Pitymys subterraneus (Microtidae, Rodentia) from the
mountains of southern Bulgaria. Zool. Listy 26, 27-31.

Krapp, F. (1982): Microtus savii (de Selys-Longchamps, 1838) — Italienische Kleinwühlmaus. In:
Handbuch der Säugetiere Europas, Rodentia II. Ed. by J. NIETHAMMER and F. Kraprp. Wiesba-
den: Akademische Mn Pp. 429-437.

Marec, F.; STorcH, G. (1963): Kleinsäuger (Mammalia) aus Makedonien, Jugoslawien. Senck. biol.
44, 15541783

Does Microtus majori occur in Europe? 33

Musser, G. G.; CARLETON, M. D. (1993): Family Muridae. In: Mammal species of the world: A
taxonomic and geographic reference. 2. ed. Ed. by D. E. Wırson and D. M. REEDER. Washington,
London: Smithsonıan Inst. Press. Pp. 501-755.

Neı, M. (1978): Estimation of average heterozygosity and genetic distance from small number of
individuals. Genetics 89, 583-590.

NIETHAMMER, J. (1972): Die Zahl der Mammae beı Pitymys und bei den Microtinen. Bonn. zool.
Beitr. 23, 49-60.

— (1982): Microtus subterraneus (de Selys-Longchamps, 1836) — Kurzohrmaus. In: Handbuch der
Säugetiere Europas. Rodentia II. Ed. by J. NIETHAMMER and F. Krapp. Wiesbaden: Akademische
Verlagsgesellschaft. Pp. 397418.

— (1986): Über griechische Nager ım Museum A. Koenig ın Bonn. Ann. Naturhist. Mus. Wien 88/
89 (B), 245-256.

PETRov, B. (1992): Mammals of Yugoslavia. Insectivores and Rodents. Natural History Museum in
Belgrade, Suppl. 37, 1-186.

PETRov, B.; Zıvkovic, $. (1979): Present knowledge on the systematics and distribution of Pitymys
(Rodentia, Mammalia) in Yugoslavia. Biosistematika 5, 113-125.

SABLINA, ©. V.; ZıMa, J.; RADJABLI, $. I.; KRYSTUFEK, B.; GOLENISCEV, F. N. (1989): New data on
karyotype variation in the pine vole, Pitymys subterraneus (Rodentia, Arvicolidae). Vest. £s.
Spole£. zool. 53, 295-299.

SEABRIGHT, M. (1971): A rapıd bandıng technique for demonstrating centromeric heterochromatın.
Lancet 1, 971-972.

SoKAL, R. R.; SNEATH, P. H. A. (1963): Principles of numerical taxonomy. San Francisco: W. H.
Freeman.

STORCH, G. (1982): Microtus majorı Thomas, 1906. In: Handbuch der Säugetiere Europas, Rodentia
II. Ed. by J. NIETHAMMER and F. Krapp. Wiesbaden: Akademische Verlagsgesellschaft. Pp.
452-462.

SUMNER, A. T. (1972): A simple technique for demonstrating centromeric heterochromatin. Exptl.
Cell Res., 75, 304-306.

SWOFFORD, D. L.; SELANDER, R. B. (1981): BIOSYS-1: aFORTRAN program for the comprehensive
analysıs of electrophoretic data ın population genetics and systematics. J. Hered. 72, 281-283.
VOHRALIK, V.; SOFIANIDOU, T. (1987): Small mammals (Insectivora, Rodentia) of Macedonia,

Greece. Acta Univ. Carolinae - Biologica 1985, 319-354.

Zıma, J. (1984): A chromosomal banding study of Pitymys subterraneus (Arvicolidae, Rodentia). Folıa
Zool. 33, 223-228.

Zıma, J.; Krar, B. (1984): Karyotypes of European mammals II. Acta Sc. Nat. Brno 18, 1-62.

Zıvkovid, S.; PETROv, B.; Rımsa, D. (1975): New data on the taxonomy of Balkan Pitymys
representatives (Mammalıa, Rodentia). Biosistematika 1, 31-42 (in Serbian with a summary ın

English).

Authors’ addresses: BoRIS KRYSTUFER, Slovene Museum of Natural History, PO Box 290, SLO-
61001 Ljubljana, Slovenia; MARIA GRAZIA FILIPPUCccI and SHIMON SIMSON,
Dipartimento dı Biologia, II Universita di Roma “Tor Vergata”, via O.
Raımondo, I-00173 Roma, Italy; MrLoS MACHOLAN and Jan ZıMmA, Institute of
Anımal Physiology and Genetics, A.S.C.R., Veveri 97, CZ-60200 Brno, Czech
Republic; MrLADEN VUJoSEVIC, Department of Genetics, Institute of Biological
Research, 29 novembra 142, YU-11060 Beograd, Yugoslavıa

Z. Säugetierkunde 59 (1994) 358-365
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Demographic changes and genetic losses in populations of a
subterranean rodent (Ctenomys maulinus brunneus) affected by
a natural catastrophe

By. M. H. GALLARDO and NELIDA KÖHLER

Instituto de Ecologia y Evoluciön, Universidad Austral de Chile, Valdivia, Chile

Receipt of Ms. 21. 1. 1994
Acceptance of Ms. 21. 6. 1994

Abstract

Studied the demographic and genetic effects of a volcanic eruption on two local populations (Rio
Colorado, Las Raices) of the fossorial rodent Ctenomys maulinus brunneus, in the Andes. Our data
represent a unique contribution because the pre-existing demographic data and levels of genetic
variation were contrasted with the changes monitored afterwards. Comparative census data of the
breeding population sıze before and after the volcanic eruption, revealed a population decline of about
90% ın Rio Colorado. An electrophoretic survey of 23 presumptive enzyme loci detected a
considerable reduction of genetic variability in both populations. In Rio Colorado, the proportion of
polymorphic locı (P) decreased from 47.8% to 17.4 %, and the expected heterozygosity (H) from
8.9% to 2.8%. In Las Raices, P decreased from 50 % to 0% and H from 13.2 % to 0%. Although
low genetic varıability in fossorial mammals is generally assumed to reflect an adaptation to the stable
subterranean niche, in some cases ıt may be merely the result of genetic drift.

Introduction

Demographie bottlenecks have received significant attention since they have important
evolutionary implications for assessing the genetic consequences of reduced population
sıze (Ner and TayımA 1981; NEI et al. 1975; CHAKRABORTY and NEI 1977). These estimates
also have implications for conservation projects of mammalian species in fragmented
habitats (LANDE and BARROWCLOUGH 1987; MArUYAaMA and KımURrA 1980; VARVIO et al.
1986) and for speciation events promoted by founder effects (BARTON 1989). Genetic
diversity ıs crucial in an evolutionary sense, and bottlenecks are the quickest means
available for loosing genetic varıation ın natural populations through random fluctuations
of population size (LEBERG 1992). There is considerable theoretical support for the
hypothesis that a natural population passing through a bottleneck should loose genetic
varlatıon in direct proportion to the severity of such an event (CHAKRABORTY and NEI
1977; MAaRrUYAMA and FUERST 1985a, b; Ner et al. 1975). Despite the theoretical attention
and well developed models predicting bottleneck effects (McCommAas and BRYANT 1990),
no actual hints of both demographic and genetic data before and after a bottleneck ın the
wild have yet been reported. Exemplary studies in mammals (BONNELL and SELANDER
1974; HARwooD and Harr 1990; O’BrıEn and EVERMAN 1988; O’BrıEn et al. 1987;
PAckER et al. 1991) have used reduced genetic varıation to infer historical bottlenecks due
to the absence of previous suitable parameters. To assess the degree of demographic
modifications after a volcanic eruption affecting the fossorial rodent Ctenomys maulinus
brunneus in southern Chile, we present comparative census data. By comparing elec-
trophoretic data from the same 23 presumptive gene locı before and after the bottleneck,
we aim to test the hypothesis that bottlenecks lead to predictable decreases ın allozyme
varlation.

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5906-0358 $ 02.50/0

Demographic changes and genetic losses in populations of a Ctenomys 959

Material and methods

Field studies were conducted in the austral summer 1986-1987, and 1991-1992, at the type localıty of
C. maulinus brunneus (Rio Colorado, Malleco province, Chile, 38° 25’ S, 71° 32’ W; 1,450 m altitude)
and ın Cordillera Las Raices, seven km southeast of the topotype locality (Fig. 1). Rio Colorado is a
flat 4 km? high Andean steppe located 3.5 km SW of the Lonquimay volcano. This area sustains a large
population of Ctenomys, faırly isolated from other local demes by physiographic and vegetational
gaps. The Holocenic Lonquimay volcano was dormant until December 1988 when its 50-year period
of inactivity was interrupted by a 6-months eruptive phase (MORENO and GARDEWEG 1989; BARRIEN-
ros and ACEVEDO-ARANGUIZ 1992).

Pre-eruption estimates of the number of breeding individuals (N) in Rio Colorado were obtained
from censuses conducted within four grids differing in size, vegetation type, and plant coverage. All
animals were removed from the grids, and their reproductive condition was assessed according to
GALLARDO and AnRIQUE (1991).

Site A: a 50x70 m grid on consolidated friable soils which had an 80-90 % cover of bunchgrass
interspersed with annual plants. Site B: two grids (40x60 m, and 50x50 m) on loose volcanic
sediments with a 40-50 % cover of annual plants. Site C: a single 60x 140 m grid on loose volcanic
soils which had a 20-25 % cover of annual plants. Field observations confirmed a lack of vegetation
from sites Band C owing to the deposition of volcanic ash and scoria following the eruption. There
was also a lack of Ctenomys activity except in sıte A where a one hectare grid was sampled in the
austral summer 1991.

Pre-eruption estimates of allozymic variation obtained from proteins encoded by 23 locı of 51 pre-
eruption specimens from Rio Colorado and nine from Las Raices have been previously reported
(GALLARDO and KÖHLER 1992; GALLARDO and ParmA 1992). The locı assayed were: isocitrate
dehydrogenase (ICD-1, ICD-2, Enzyme Commision No. 1.1.1.42), malate dehydrogenase (MDH-1,
MDH-2, 1.1.1.37), glutamate-oxaloacetate transaminase (GOT-1, GOT-2, 2.6.1.1), glycerol-3-

68°

Lolco Vn.

La Holandesa Vn.

Tolhuaca Vn.
# Laguna Verde\n.

Caracol Vn.
Lonquimay Vn.,
RIO Colorado

aglas Raices

1 Study sites

Fıg. 1. Map of the study areas depicting the volcanoes (Vn.) that form the Lonquimay Volcanic Front
in the Andes of southern Chile

360 M. H. Gallardo and Nelida Köhler

Table 1. Changes in gene frequencies of polymorphic loci in two populations of Ctenomys m.
brunneus affected by the eruption of volcano Lonquimay, Chile

Pre-eruption data for Rio Colorado were obtained from GALLARDO and Parma (1992), and for Las
Raices, from GALLARDO and KÖHLER (1992)

Rio Colorado Las Raices
Post Pre

Sample sıze (n) 58
ICD-1

Mean Polymorphism
Mean N’

Allele/locus

Mean Heterozygosity
(direct count)
(Hardy-Weinberg)

* Genotypes for this locus and population deviate significantly from Hardy-Weinberg expecta-
tions.

Demographic changes and genetic losses ın populations of a Ctenomys 361

phosphate dehydrogenase (GPD, 1.1.1.8), glucose 6-phosphate dehydrogenase (Gd, 1.1.1.49),
glucose dehydrogenase (GDH, 1.1.1.47), lactate dehydrogenase (LDH-1, LDH-2, 1.1.1.27), xan-
thine dehydrogenase (XDH), phosphoglucomutase (PGM-2, PGM-3, 2.7.5.1.), glucose isomerase
(GPI, 5.3.1.9), phosphogluconate dehydrogenase (PGD, 1.1.1.44), albumin (ALB), transferrin
(IRFER), hexokinase (HK, 2.7.1.1), acıd phosphatase (ACP-1, ACP-2, 3.1.3.2), and malic enzyme
(ME, 1.1.1.40). Fifty eight anımals from Rio Colorado, and 23 from Las Raices were live-trapped and
screened for the same locı after the eruption. Tissue homogenates, buffer systems, migration
conditions and mixtures were prepared according to the methods of SELANDER et al. (1971). Estimates
of expected heterozygosity values were compared to the observed numbers and tested for significance
by the nonparametric Wilcoxon two-sample test and by a t-test of arcsin transformed genetic data
(ArcHıe 1985). All computations of genetic varıation were performed with Biosys-1 (SwOFFORD and
SELANDER 1989).

Results

Twenty three anımals captured ın site A gave an estimate of 66 anımals/ha. An additional
estimate of 35 anımals/ha was generated by pooling density data from the two grids in site
B, whereas an estimate of 8.3 anımals/ha was obtained from site C. Direct estimates of the
breeding population size obtained from census data ıindıcated 1930 anımals before the
eruption. Based on a density estimate of 20 anımals/ha after the eruption in sıte A, the
breeding population sıze consisted of 168 anımals, which corresponds oo 2.
reduction ın population sıze after the catastrophe. This estimate provided a rough
approximation of the effective population size; at best, it was an upper limit, below which
Ne decreases depending on departures from ıidealized structure (HusBanD and BARRETT
1992).

There was a severe loss of electrophoretic varıatıion ın both populations following this
bottleneck episode (Tab. 1). Mean polymorphism was significantly reduced; it dropped
from 47.8% to 17.4% ın Rio Colorado and reached ıts minımum value in Las Raices.
Eighty percent of previously polymorphic locı ın Rio Colorado, and all varıable locı ın Las
Raices became monomorphic after the bottleneck. As theory predicts, the most frequent
allele was more lıkely to become fixed, but the reverse held for locı GDH and PGM-2 in
Las Raices. Sımilarly, locus XDH-1, al-
though still polymorphic ın Rio Colorado,
changed significantly in frequency after the

Table 2. Observed pre- and post-bottleneck
mean heterozygosity values of polymorphic
loci in Ctenomys maulinus brunneus

eruption (Tab. 1). (Rio Colorado)

The difference between the pre- and The non parametric comparison of mean values
post-eruption frequency of heterozygotes is based on the Wilcoxon two-sample test
was highly significant (P < 0.0056; Tab. 2).

Further indications for rejecting the null Dreier NED se lereele
hypothesis of equality between initial and Heterozygosity _ Heterozygosity
final mean heterozygosities were obtained
by congruently significant results of angu- ee Be
lar-transformed data (P < 0.004; not GOT-1 | 0.000
shown). No analogous comparisons were GPD ; 0.000
conducted in the Las Raices sample where LDH-2 0.000
the final absence of heterozygotes resulted a an
from a generalized monomorphism. PGM-3 0.000
PGI-1 3 0.000
PGD 3 0.340
Discussion Rx Eh

ACP-2 5 0.051

Total 2 0.054

Our estimations of density fall within the
range reported for other subterranean Neon re Ze a P = 0.0056
mammals (NEevo 1979), implying a direct

362 M. H. Gallardo and Nelida Köhler

correlation between number of anımals and food supply. Estimates ranging from 347
anımals/ha ın C. peruanus (PEARson 1959) to 218 anımals/ha in C. talarum (PEarson et al.
1968) indicate the large varıance of demographıc attributes in the genus (Reıc et al. 1990).

Theoretically, the major genetic consequence of a bottleneck involves a reduction in the
number of alleles, because the varıants most at risk are those in low frequency (CHak-
RABORTY and NEr 1977). This prediction, supported by our data, reached its maximum
expression ın the Las Raices sample. Mean heterozygosity was also affected, especially ı in
Las Raices where all estimates of genetic diversity reached the most extreme minimum
values. Although theory predicts a significant reduction of the number of alleles with less
than 10 founders, and even fewer pioneers to affect the average heterozygosity (CHAK-
RABORTY and NEr 1977; McCommAas and BRyANT 1990), significant losses in genetic
diversity occurred with a larger number of founders, provided a slow population recovery
(LANDE 1987).

In connection with the genetic effects of demographic bottlenecks, deficiencies in
sampling may distort the assessment of heterozygosity levels. Although these estimates
appear to be more affected by the number of locı screened than by the number of
individuals analyzed (GoRrMANn and Renzı 1979). Considering that the analysis of 8-12
individuals yıelds, on the average, a heterozygosity estimate wıthın 1% of the value
calculated using larger numbers of individuals, sampling biases can be confidently disre-
garded in our estimations (GORMAN and Renzı 1979). Another possible bias stems from
the assumption that losses result from the bottleneck itself (BoILEAU et al. 1992), with no
additional genetic cost because a fast demographic recovery is expected. These predictions
result in underestimates of the absolute losses, as Ctenomys ıs a k-strategist exhibiting a
low instrinsice growth rate (Reıc et al. 1990).

Considering the small period of time that has elapsed since the bottleneck, drastic
declines in heterozygosity ın relation to expected values probably do not result from
consanguineous matings, but from a generalized form of inbreeding associated with limited
population sıze (CHESSER and Ryman 1986). In this respect, panmictic conditions do not
hold since considerable levels of intrademic genetic differentiation strongly suggest popula-
tion subdivision (GALLARDO unpubl. results). Thus, agonistic behaviour and limited
dispersal in Ctenomys (GALLARDO and AnRIQUE 1991) are not consistent with the
assumption of a panmixtic unit of 168 individuals, but support an explanation based on the
sustained effects of drift acting on disrupted breeding assemblages.

From a conservation viewpoint, bottleneck-flush cycles are major factors contributing
to species vulnerability (O’BRıEn and EvERMAN 1988). Limited population sızes threaten
demic persistence when a threshold density value defining mating success is reached
(LAnDE 1987, 1988). Besides, the genetic unıformity attained through repeated bottleneck
events also affects demic survival by inbreeding depression (PACkER et al. 1991; WILDT et
al. 1987), or by a homogeneous inmunological response to environmental disturbances
(O’BRIEN et al. 1985).

Although current interpretations of genic and chromosomal varıation patterns ın
Ctenomys emphasıze a causal relationship with spatial components (Reıc et al. 1990),
sustained vulcanısm and the tectonically-derived seismic activity ın the Andes heighten a
component of temporal stochasticity. In this line, previous eruptions of the Lonquimay
volcano took place in 1940, 1887 and 1853 (MORENO and GARDEWEG 1989). Assuming
similar environmental effects, population survivorship, and a one-year generation time
(GALLARDO and ANRIQUE 1991) the genetic pool of these, and probably other local
populations has been altered in generations 135, 101, and 48 before the present.

While contrasting with the multiple karyotypic forms observed in the low land
Ctenomys species (GALLARDO 1991), the karyotypic uniformity reported for species from
the Andes (GALLARDO 1979) conflicts with the optimal conditions for chromosomal
differentiation that stem from limited population sıze (LAnDE 1979, 1985; WRIGHT 1941).

Demographic changes and genetic losses in popnlations of a Ctenomys 363

Apparently, chromosomal conservatism may be better explained by recurrent vulcanısm
for when extinctions and recolonizations are frequent, subpopulation divergence is pre-
vented (MAaruyamA and Kımura 1980) and new colonıes are likely to descend from a single
ancestral deme (WRIGHT 1941). Considering that the activity of the nearby volcanoes
Antuco, Llaima and Villarrica total 36 eruptive events since 1640 (VEBLEN 1985), the long-
term karyotypic unıformity ın Andean Ctenomys populations appears to be coupled to the
structure of the environment. Furthermore, low levels of genetic varıability attained by
recurrent catastrophes may imitate an adaptive response to the stable subterranean niche
(Nevo 1979, 1990) although no selective pressure can counterbalance the effects of drift
when populations recover from precarious levels of genetic varıability (MARUYAMA and
FuersTt 1985a, b).

Acknowledgements

This research was supported by grant FN 92/0178 and DID UACH grant S-94-29 to MHG.

Zusammenfassung

Demographische Veränderungen und Verluste von genetischer Variabilität bei Populationen des
grabenden Nagers (Ctenomys manlinus brunneus) als Folge einer Naturkatastrophe

Es wurden die demographischen und genetischen Auswirkungen eines Vulkanausbruchs auf zwei
lokale Populationen (Rio Colorado, Las Raices) des grabenden Nagers Ctenomys maulınus brunneus
in den Anden untersucht. Ein Vergleich der Populationsgrößen vor und nach dem Vulkanausbruch
auf der Grundlage von Zählungsergebnissen in Rio Colorado ergab einen Rückgang der Individuen-
zahlen um etwa 90%. Die elektrophoretische Untersuchung von 23 Enzymlocı zeigte in beiden
Beständen eine starke Reduktion der genetischen Variabilität. In Rio Colorado sanken die Polymor-
phierate (P) von 47,8 % auf 17,4% und der durchschnittliche erwartete Heterozygotiegrad von 8,9 %
auf 2,8%. In Las Raices sanken P von 50 % auf 0% und H von 13,2 % auf 0%. Unsere Daten liefern
einen Beitrag zu den wenigen Fällen, in welchen bei natürlichen Populationen Verluste von geneti-
scher Variabilität direkt einem Engpaß in der Populationsgröße zugeordnet werden können. Obwohl
die geringe genetische Variation bei grabenden Nagern in der Regel als Anpassung an die stabilen
Lebensbedingungen der unterirdischen Nische zugeschrieben wird, kann sie ın einigen Fällen lediglich
auf genetische Drift zurückzuführen sein.

Literature

ARCHIE, J. W. (1985): Statistical analysıs of heterozygosity data: independent sample comparisons.
Evolution 39, 623-637.

BARRIENTOS, $. E.; ACEVEDO-ARANGUIZ, P. S. (1992): Seismological aspects of the 1988-1989
Lonquimay (Chile) volcanic eruption. J. Volcan. Geothermal Res. 53, 73-87.

Barton, N. H. (1989): Founder effect speciation. In: Speciation and its consequences. Ed. by D.
OTTE and J. A. ENDLER. Massachusetts: Sinauer Ass. Pp. 229-256.

BoILEAU, M. G.; HEBERT, P. D. N.; ScHwarTZ, $. S. (1992): Non-equilibrium gene frequency
divergence: persistence founder effects in natural populations. J. Evol. Biol. 5, 25-39.

BonnELL, M. L.; SELANDER, R. K. (1974): Elephant seals: genetic variation and near extinction.
Science 184, 908-909.

CHAKRABORTY, R.; NEı, M. (1977): Bottleneck effects on average heterozygosity and genetic distance
with the stepwise mutation model. Evolution 31, 347-356.

CHESSER, R. K.; Ryman, N. (1986): Inbreeding strategy in subdivided populations. Evolution 40,
616-624.

GALLARDO, M. H. (1979): Las especies chilenas de Ctenomys. I. Estabilidad cariotipica. Arch. Biol.
Med. Exper. 12, 71-82.

— (1991): Karyotypic evolution in Ctenomys (Rodentia, Ctenomyidae). J. Mammalogy 72, 11-21.

GALLARDO, M. H.; AnRIQUE, J. (1991): Populational parameters and burrow systems in Ctenomys
maulınus brunneus (Rodentia, Ctenomyidae). Med. Ambiente 11, 48-53.

GALLARDO, M. H.; KöHLer, N. (1992): Genetic divergence in Ctenomys (Rodentia, Ctenomyidae)
from the Andes of Chile. J. Mammalogy 72, 99-105.

GALLARDO, M. H.; Parma, R. E. (1992): Intra- and interspecific genetic varıabilıty in Ctenomys
(Rodentia, Ctenomyidae). Biochem. Syst. Ecol. 20, 523-534.

GORMAN, G. C.; Renzı, J. (1979): Genetic distance and heterozygosity estimates ın electrophoretic
studies: effects of sample size. Copeia 2, 242-249.

364 M. H. Gallardo and Nelida Köhler

HARWOoOD, J.; Harz, A. (1990): Mass mortality in marine mammals: its implications for population
dynamics and genetics. Trends Ecol. Evol. 5, 254-257.

HuskanD, B. C.; BARRETT, $. C. H. (1992): Effective population size and genetic drift in Eichhornia
paniculata. Evolution 46, 1875-1890.

LAnDE, R. (1979): Effective deme sizes during long-term evolution estimated from rates of
chromosomal rearrangements. Evolution 33, 234-251.

— (1985): The fixation of chromosomal rearrangements in a subdivided population with local
extinction and colonization. Heredity 54, 323-332.

— (1987): Extinction thresholds in demographic models of territorial populations. Amer. Naturalist
130, 624-635.

— (1988): Genetics and demography in biological conservation. Science 241, 1455-1460.

LANDE, R.; BARROWCLOUGH, G. F. (1987): Effective population size, genetic variation, and their use
in population management. In: Viable populations for conservation. Ed. by. M. E. SouL£. New
York: Cambridge Univ. Press. Pp. 87-123.

L£Berg, P. (1992): Effects of population bottlenecks on genetic diversity as measured by allozyme
electrophoresis. Evolution 46, 477—494.

MARUYAMA, T.; FuErst, P. A. (1985a): Population bottlenecks and nonequilibrium models in
population genetics. III. Genic homozygosity in populations which experience periodic bot-
tlenecks. Genetics 111, 691-703.

— (1985b): Population bottlenecks and nonequilibrium models in population genetics. II. Number of
alleles in a small population that was formed by a recent bottleneck. Genetics 111, 675-689.
MARUYAMA, T.; Kımura, M. (1980): Genetic varıability and effective population size when local
extinction and recolonization of subpopulations are frequent. Proc. Natl. Acad. Sci. USA 77,

6710-6714.

McCommas, $. A.; BRYANT, E. H. (1990): Loss of electrophoretic varıation in serially bottlenecked
populations. Heredity 64, 315-321.

MORENO, H.; GARDEWEG, M. C. (1989): La erupciön reciente en el complejo volcänico Lonquimay
(Diciembre 1988), Andes del Sur. Rev. Geol. Chile 16, 93-117.

NEI, M.; MARUYAMA, T.; CHAKRABORTY, R. (1975): The bottleneck effect and genetic varıability in
populations. Evolution 29, 1-10.

Neı, M.; Tajıma, F. (1981): Genetic drift and estimation of effective population size. Genetics 98,
625-640.

NEvo, E. (1979): Adaptive convergence and divergence of subterranean mammals. Ann. Rev. Ecol.
Syst. 10, 269-368.

— (1990): Genetic diversity and its ecological correlates in nature: comparisons between subterra-
nean, fossorial, and aboveground small mammals. In: Evolution of subterranean mammals at the
organısmal and molecular levels. Ed. by. E. Nevo and ©. A. Reıc. New York: A.R. Liss. Pp.
347-366.

O’BRIEN, S$. J.; EvERMAn, ]. F. (1988): Interactive influence of infectious disease and genetic diversity
in natural populations. Trends Ecol. Evol. 3, 254-259.

O’BRIEN, $.J. ; ROELKE, M. E.; MARKER, L.; NEWMAN, A.; WINKLER, C. A.; MELTZER, D.; CoLLy,
L.; EVERMANN, J. F.; BusH, M.; WıLDT, D. E. (1985): Genetic basıs for species vulnerability ın the
cheetah. Science 227, 1428-1435.

O’BRIEN, S. J.; WILDT, D. E.; BusH, M.; Caro, T. M.; FITZGIBBON, C.; AGGUNDEY, 1.; LEAKEY, R. E.
(1987): East Afrıcan cheetahs: evidence for two population bottlenecks. Proc. Natl. Acad. Scı.
USA 84, 508-511.

PAcKER, C.; Pusey, A. E.; RowLEy, H.; GILBERT, D. A.; MARTENSON, J.; O’BRIEn, S. J. (1991): Case
study of a population bottleneck: lions of the Ngorongoro crater. Conserv. Biol. 5, 229-237.
PEARson, O. P. (1959): Biology of subterranean rodents, Ctenomys, ın Peru. Mem. Mus. Hist. Nat.

“Javier Prado” 9, 1-56.

Pearson, ©. P.; Bıinstein, N.; Boıry, L.; Bush, M. C.; Dı Pace, M.; GALLOPIN, G.; PENCHAS-
ZADEH, P.; PIANTANIDA, M. (1968): Estructura social, distribuciön espacial y composiciön por
edades de una poblaciön de tuco-tucos (Ctenomys talarum). Invest. Zool. Chile. 13, 47-80.

Reıc, ©. A.; Bush, C.; ORTELLS, M. O.; ConTEras, J. R. (1990): An overview of evolution,
systematics, population biology, cytogenetics, molecular biology, and speciation in Ctenomys. In:
Evolution of subterranean mammals at the organısmal and molecular levels. Ed. by E. Nevo and
O. A. Reıc. New York: A. R. Liss. Pp. 71-96.

SELANDER, R. K.; SmItH, M. H.; Yang, $. Y.; Jounson, W. E.; GENTRY, ]J. B. (1971): Biochemical
polymorphism and systematics in the genus Peromyscus. I. Variation in the old-field mouse
Peromyscus polionotus. Univ. Texas Publ. 7103, 49-90.

SWOFFORD, D. L.; SELANDER, R. (1989): Biosys-1. A computer program for the analysis of allelic
variation in population genetics and biochemical systematics. Illinois Natural History Survey.
VARVIO, $. L.; CHAKRABORTY, R.; NEı, M. (1986): Genetic variation in subdivided populations and

conservation genetics. Heredity 67, 189-198.
VEBLEN, T. T. (1985): Stand dynamics in Chilean Nothofagus forests. In: The ecology of natural

Demographic changes and genetic losses in populations of a Ctenomys 365

disturbance and patch dynamics. Ed. by $S. T. A. Pıck£TT and P. S. WHITE. London: Acad. Press.
Pp. 35-51.

mo DIR BUSE, NEREGOODEROWELK-T. PACKERLGE. Posex, A. Es; BROwN, |. L.; Jostın, P.;
O’BrIEn, $. J. (1987): Reproductive and genetic consequences of founding isolated lion popula-
tıons. Science 329, 328-330.

WRIGHT, S. (1931): Evolution in mendelian populations. Genetics 16, 97-159.

— (1941): On the probability of fixation of a reciprocal translocation. Amer. Naturalist 75, 513-522.

Authors’ address: MıL Ton H. GALLARDO and N£LıDA KÖHLER, Instituto de Ecologia y Evoluciön,
Universidad Austral de Chile, Casilla 567, Valdivia, Chile

Z. Säugetierkunde 59 (1994) 366-377
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Craniometric differentiation and chromosomal speciation of the
genus Proechimys (Rodentia: Echimyidae)

By M. AcuiLerA and M. Corri

Departamento Estudios Ambientales, Universidad Simon Bolivar, Caracas, Venezuela, and
Dipartimento di Biologia Anımale e dell’Uomo, Universita di Roma ‘La Sapienza’, Roma, Italy

Receipt of Ms. 11. 3. 1994
Acceptance of Ms. 16. 6. 1994

Abstract

Multivariate morphometrics have been used to investigate the systematics and geographic variation of
the spiny rats of the genus Proechimys oceurring in Venezuela. These populations exhibit extensive
differences ın their karyotype from 2n = 24 to 2n = 62, and patterns of differentiation in morphologi-
cal traits of the skull and of the mandible are consistent with a phylogenetic hypothesis suggesting
successive events of speciation coupled with increase in diploid numbers. Although no single character
by itself can be used to discriminate between taxa, all species and subspecies are clearly distinguishable
from each other in a multivariate context.

Introduction

Spiny rats, Proechimys (Rodentia: Echimyidae), is a wıdely distributed genus occurring in
the lowland and premontane Neotropical forests of South America, with many extant
named species (PATTon 1987; Reıc et al. 1980). The entire genus is characterized by a high
chromosomal heterogeneity, with diploid numbers ranging from 2n = 14 to 2n = 64 (REIG
et al. 1980), and it has been considered as some of the best evidences for genetic change and
chromosomal speciation (Kıng 1993).

An exhaustive review of the genus has been provided by Patron (1987), who
recognized nine groups of species, primarıly based upon morphological traits. According
to the classıfication of PATTON (1987) three of these species groups occur in Venezuela, ı.e.
trinitatus, canicollis and guyannensis. The range of both canicollis and trinitatus lies north
of the Orinoco river, wıth only one species of trinitatus ın the southern part of the basın
(P. hoplomyoides). The upper Orinoco river groups comprise different species, P. canıcollis
(2n = 24), the only species of the canicollis group, and, for the trinitatus group, the species
P. trinitatis (2n = 62), and the nominate superspecies P. guairae, with three closely related
allospecies: P. poliopus (2n = 42), P. guairae (2n = 44, 46, 48, 50 and 52) and a third,
referred by AGUILERA et al. (1994) as Proechimys sp. (2n = 62) (Fig. 1).

The superspecies P. guairae represents a remarkable example of a typical rassenkreis,
occurring in parapatric contiguous ranges skirting the Maracaibo lake and the mountains of
Cordillera de la Costa and Cordillera Andina (Fig. 1). The rassenkreis ıs characterized by
karyomorphs with increasing diploid numbers, from P. poliopus (2n = 42) to Proechimys
sp. (2n = 62), providing evidence that led Reıc (1980) to hypothesize a model of
chromosomal speciation via centric fissions from lower to higher chromosomal numbers.

The species P. guairae is polytypic, and comprises different karyotypic races with
diploid numbers corresponding to 2n = 44, 46, 50 and 52 (REıc 1989; AGUILERA et al.
1994). Not all karyomorphs have been assigned a subspecific name, and some are known
from the locality from which chromosome preparations were available. Ranges of
chromosomal races are apparently contiguous and parapatric, from the lake of Maracaıbo
to the Unare basin, along the northern coast of Venezuela (Fig. 1). The chromosomal race

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5906-0366 $ 02.50/0

Craniometric differentiation and chromosomal speciation of Proechimys 367

iS

—ı:

Caribbean sea

oo

yl Il
Unare ano N2

TIDT7 Ss
n o
AU di

Ir aa

Proechimys guairae complex

B22] P. poliopus (2n=42, FN=76)

Frl PP. guairae (2n=44, 46, 48, 50, 52; FN=72-74)

Fammd P. barinas n. sp. (2n=62, FN=74)
Proechimys trinitatis complex

III Proechimys trinitatis (2n=62; FN=80)

E7Z72 Proechimys canicollis (2n=24; FN=44)

EEE Arcas above 1000 m

Fig. 1. Map of North Latin America, with the ranges of species and the approximate location of
chromosomal races. For the exact location of the localities sampled see Tab. 1

with the highest chromosomal number (2n = 52) occurs in a limited range out of the
rassenkreis in the east of Venezuela near P. trinitatis. The range ıs diısjunct and separated by
the Unare basın (Fig. 1).

There have been some studies on skull morphometrics of spiny rats that focused on
some non-geographic aspects of varıation, mainly related to growth (PATTon and RoGERS
1983; Pessoa and Dos Reıs 1991a), or on some aspects of intraspecific geographic
varıation (Dos Reıs et al. 1990; Pessoa and Dos Reıs 1990, 1991b; PessoA et al. 1990).
However, they refer to other species groups ot ParTron (1987) such as P. brevicanda,
P. iheringi, P. albispinus, P. dimidiatus, P. guyannensis, and no analyses are available on the
morphometric differentiation of populations and species, namely casiraguas, from Vene-
zuela.

The present report describes patterns of morphomerric relationships among species and
modes of geographic varıation in morphomerric traits within casıraguas, in an attempt to
relate these patterns to phylogeny or ecology. Furthermore, we attempt to answer two
basıc questions: are the karyomorphs morphometrically recognisable with sufficient
confidence? Is there any pattern of morphological differentiation that may be interpreted
in the light of the established model of chromosomal speciation?

Material and methods

Two-hundred and fifty-five specimens were analyzed representing five species and 12 populations of
the genus Proechimys ın Venezuela (Tab. 1). The species distribution is shown in figure 1 and the exact
location of the populations analysed is presented in table 1.

Since the complex displays wide karyotypic variation, most of the anımals used in this study were
previously karyotyped and correctly assıgned to a chromosomal group (Reıc et al. 1980; AGUILERA et
al. 1994) (Tab. 1). Therefore all populations are characterized by their karyotype. Those species and
races for which formal description has not yet been carried out will be indicated by us through their
diploid number and locality of origin and/or region. These are: Proechimys sp. (2n = 62), P. guairae
“Falcon” (2n = 46), P. guairae “Llanos” (2n = 50), and P. guairae “Oriente” (2n = 52). P. pohiopus,

368 M. Aguilera and M. Corti

P. guairae “Orıente” and
Proechimys sp. (2n = 62) Tierra
Buena — Las Matas are each rep-
resened by two populations
(Tab.1) and have been pooled to
increase sample size.

Nineteen distance characters
were recorded on the skull and four
distance characters on the mandible
with the aid of a digimatic caliper
(Mitutoyo, 0.Olmm precision).
The distance characters are as fol-
lows (Fig.2): Total length (TL):
from anterior point of nasal to the
sagıttal bulge of the occipital; Nasal
length (NL): from the anterior
point of the nasal to suture with
frontal; Basılar length (BL): from
the posterior point of the incisor at
its alveolus to the anterior border of
foramen magnum; Palatal length
(PL): from the posterior point of
the incisive foramen to the pos-
terior border of the palate; Palatine
length (PLL): from the posterior
point of incisor at its alveolus to the
posterior border of the palate; Up-
per diastema length (UDL): from
the posterior point of the incisor at
its alveolus to the anterior point of
alveolus M!; Incisive foramen
length (IFL); Alveolar length
(UAL): of the upper molar series;
Incisor-zygomatic length (IZL):
form the posterior point of the al-
veolus of the incisor to the pos-
terior border of the zygomatıc arch;
Bulla tympanica length (BTL):
longest length of the bulla taken
along the axis oblique to the skull
length; Fronto-maxillar suture
width (FMW); Bizygomatic width
(BZW): taken in the widest section;
Minimum _ interorbital width
(MLOW): taken over the frontal;
Palatal width (PAW): taken in the
middle part of the two M?; Incisive
foramen wıdth (IFW): maxımum
width; Bulla tympanıca width
(BTW): longest width of the bulla
taken along the axis oblique to the
skull length; Cranial width (CW):
the maximum width taken immedi-
ately on top of the external auditory
meatus; Maximum height of ros-
trum (MXHR); Cranial height
(CH).

Characters TE, NE BES DE
PLL, UDE EL, UARAIZEIBINIE
FMW, BZW, MLOW, CW,
MXHR, CH correspond to those
similarly recorded by PATTon and
RotGers (1983).

The following four distance
characters were recorded on the

Geographic location
(latitude and longitude)
10° 19’ N-67° 38’ W
10° 27° N-67° 50° W
09° 44’ N-68° 34’ W
09° 16° N-69° 04’ W
10° 10° N-64° 35’ W
09° 52° N-63° 53’ W
07° 27’ N-71° 207 W
09° 15° N-69° 39’ W
09° 117 N-69° 35’ W
07° 19’ N-71° 55’ W
10° 10° N-63° 33’ W

Some of the samples were collected from contiguous localities and pooled

Locality
Los Angeles del Tucuco

La Trilla
Cueva del Guacharo

Rio Cachıri
Kasmera

El Limon
Turiamo
Palmero

Turen

Cueva de Agua
San Juan de Areo
Guaquitas
Tierra Buena
Las Matas

La Nulita

»

Table 1. Species and subspecies, locality, diploid number, geographic location (latitude and longitude, acronym and male and female samples)

P. guairae “Falcon
P. guairae “Llanos”
P. guairae “Llanos”
P. guairae

P. canicolls

P. poliopus

P. g. gnairae
P. g. guairae
Proechimys sp.
(2n = 62)
Proechimys sp.
(2n = 62)
Proechimys sp.
(2n = 62)

P. trinıtatıs

“Oriente”

369

Craniometric differentiation and chromosomal speciation of Proechimys

(TVI) sıejow 19M0J aya Jo yıuay ze[osafe 10L13JUJ — EZ !( TAI) Y3uaf ewagserp 1oL1>FJu] - 7z °(HW) ySroy sepngqıpue — 17 :(IWN) yaSuaj sepngipue - 07 :(HD) Aysıay
jerue1d - 61 :(NHXIN) wnnso1 jo ıydıay umunxeWw — 8I (MD) QIpım JerueId — ZI (MALE) Yapım esruedurk eipng - 97 (Mal) YPPpIm usure1oF 9AIstau] — GI !(MVd)
ypım jerefeg - FI MOIN) Ypıa fergro-sun wnunum — EL (MZI) Yıpım smewosAzıg - ZI !(MINA) YIpım aanıns zeppixew-oruoNg — [I (IL) yaSuay esruedurä
eımg - 01 :(TZI) y3ua] onewosAz-ıosmuf - 6 !(TVN) yıSuay sejosape saddn - 8 ‘(TAI) YA3ua] uowe1oF JAısmu] - / (TAN) Su] eursıserp ssddn - 9 :(TId) yaSua]
suneeg — S (Id) YaSuay jerepeg — + :(T) YrSuaf zeftseg — € °(TN) Y33u>J Jesen - 7 ‘(TL) ya3ua] Jero,], — I 'spqıpueu aya pue [jnys aya uo painseaur sIagserey ‘7'317

370 M. Aguilera and M. Corti

mandible: Mandibular length (ML): from the posterior point of the incisor at its alveolus to the
posterior border of the mandible; Mandibular height (MH); Inferior diastema length (IDL): from the
posterior point of the incisor at ıts alveolus to the posterior border of the alveolus of M,; Inferior
alveolar length of the lower molars (IAL).

The two character sets were analysed separately. Although in rodents the skull and the mandible
form two structures that are highly integrated in their function and during growth, they were analysed
independently as they represent sets of genes reflecting different levels of co-adaptation.

Data were transformed into logarıthms to render character relationships linear.

The sample includes individuals from both sexes (Tab. 1) and we tested for possible significant
effects of sexual dimorphism over the characters by means of two way analysis of varıance (unbalanced
design), testing for population and sex differences, and for their interaction. A significant interaction
of sex with population would suggest that a particular character is sexually dimorphic; therefore, it
should be necessary to perform analyses independently for each sex, while no significance would
allow the data to be pooled irrespective of sex.

The sample comprises adult animals only, all satisfying 3 conditions: a) M? erupted (roughly
corresponding to age classes 7 up of Parron and Rocers 1983); b) over 200 g; and c) body length
over 200 mm. In doubtful cases, the choice was based upon a combination of aand b or aandc.

Since we recorded a certain varıation in size within each population representing static allometry
(KLINKENBERG and ZIMMERMANN 1992) it seemed essential to correct the data for a proper description
of the between group varıatıon. The generally known ‘BurnaBY’ procedure (BurnABY 1966) adjusts
original data according to the size vector from the pooled within-group covarıiance matrix (algorıthm
suggested by ROHLF and BooKsTEIN 1987). The pooled within-group covariance matrix was then
computed and eigenvectors extracted and examined. The first eigenvector was assumed to represent
within-group size and the data were adjusted following BuRNABY (1966).

Multivarıate analysis of varıance and canonical varıate analysıs (CVA) were used to test the
“BurnaBy’ adjusted data for differences between centroids and to depict a pattern of population
variation.

Mahalanobis distances were used to compute UPGMA and to test for congruencies in patterns
between the different character sets.

Differences between groups were also investigated for each character through analysıs of variance
and GT-2 test among means (SoKAL and ROoHLF 1981). For such practical purposes as the rapid and
precise identification of specimens in collections we used the ratios of an character to the total length
of the skull in order to identify appropriate measurements.

Univariate and multivariate analyses were performed with the SAS system for the PC (ver. 6.08)
using the procedures ANOVA, GLM, CANDISC, DISCRIM as variously modified in Marcus and
Corri (1989). A SAS IML procedure from AFEwoRK BEKELE et al. (1993) was used to adjust original
data following BurnaBy (1966).

Results

Two-way analysıs of varıance showed a significant effect of sexual dimorphism in only one
out of the 19 skull characters, ı.e. IFL (two further characters had p = 0.055) and ın one out
of 4 mandible characters (IDL, Tab. 2). These differences were accepted as negligible and
all further analyses were performed by pooling all individuals irrespective of sex. On the
contrary, all characters except PL revealed significant differences among populations
(Iab22)):

The eigenvectors were extracted from the pooled within-group covariance matrix of the
19 characters of the skull and examined. The coefficients associated with the first
normalised eigenvectors all have the same sign (Tab. 3), and so this vector was taken as
representing allometry. Raw data were adjusted following Burnagy (1966) and the new
data matrix excluding the allometric effect was subjected to canonical varıate analysıs
(CVA).

The first three canonical variates computed on the skull adjusted data express 59.85 %
of total variance (24.35 %, 19.1 % and 16.4 %, respectively). To obtain 90 % of varıance it
is necessary to reach the seventh canonical variate. However, the variance expressed by the
4th to 7th canonical variates decreases from 11.47% to 3.96%, and we accepted the
scenario depicted by the first three as representative of population variatıon.

The population ordination onto these first three variates is shown ın the stereogram in
figure 3. P. trinitatis has the highest score on CV1 and the lowest on CV2, P. canıcollis has

Craniometric differentiation and chromosomal speciation of Proechimys O7

Table 2. Skull and mandible characters

Analysis of variance testing for population and sex differences and for the interaction of sex with
populations

Character £ Character

NE ; B UDE

First row: population; second row: sex; third row: sex-population interaction, with number of
observations, F- value and probability. For character abbreviations see text

low score on CV2. CV3contributes in separate P. poliopus with the lowest value and the
three Proechimys sp. (2n = 62) which have the highest. The P. guairae populations have
intermediate scores on the three canonical variates. The latter are very similar except for
P. guairae “Oriente”, which has a low value onto CV2.

All between group comparisons are highly significant (p<0.001), as is shown by
Hotelling’s T? on Mahalanobis distances (Tab. 4).

The a posteriori probability of correct classification based upon Mahalanobis distances
from group centroids lies between 96% and 72.73% (average 87.11 %); most of the
incorrectly classified individuals fall within other populations of their own species, thus

372

Table 3. 19 eigenvalues extracted from the

pooled within-group covariance matrix of the

skull characters, and the character coefficients
associated with the first eigenvector

Character Eigenvalue 1

coefficients

Eigenvalues
19

106
NL
BZW

36.545
1.1404
0.6272
0.5703 MLOW
0.4911 CW
0.3790 BL
03525 IZL
0.3214 DIE
0.2980 Bele
0.2499 UDL
0.2204 DET
0.1913 UAL
0.1897 FMW
0.1561 PAW
0.1161 IFW
0.1018 BEE
0.0919 BTW
0.0807 MXHR
0.0402 CH

For abbreviation see text.

M. Aguilera and M. Corti

suggesting that these distances are a good
index of between-species differences. These
high values of correct classification decrease
when group membership is computed using
a Jack-knife restriction (crossvalidate op-
tion ın SAS) to a range Iying between
52.38% and 88% (average 69.97 %).
However, most of the incorrect classifica-
tions are still shared within the P. guairae
complex.

Although Mahalanobıs distances are not
characterized by a wide range of varıability
(2.87-6.04) (Tab.4), they nevertheless re-
flect species and population distinctions:
P. trinitatis and P. canicollis have, on aver-
age, the highest distances, and the lowest
are those between populations of the same
species (Tab. 4).

The UPGMA phenogram in figure 4
depicts population relationships based on
these distances: P. trinitatis and P. canicollis
are very different, the three Proechimys sp.
(2n = 62) populations are clustered together
and connected with the P. guairae complex,
which forms a homogeneous group. As
also shown by the plot in figure 3,

P. guairae “Oriente” ıs remarkably distinct form the other P. guairae populations, and ın

the UPGMA it ıs linked with P. canicollıs.

CVA was also performed on the log transtormed raw data of the mandible. The first
two canonical varıates account for 94.55 % of total varıation (82.47% and 12.07 %

2 Sg
D
1
P.g.0
CV 3
(6)
P. can
-1
-2
-3
-2 u
CV2

0

P.poli.

eva

2 3 -4

Fig. 3. Stereo scatter plot of population means for the skull onto the first three canonıcal varıates.
Units along axes are pooled within-group standard deviations. See Tab. 1 for population acronyms

318

Craniometric differentiation and chromosomal speciation of Proechimys

„IL Sup220H “2. = 10000 > d ‘..= [000 >d‘.=co0>d

10: rel TE gg°0 ;
«16 F 69V u > uur69 Gurt GG Ku vsrl6 Pc : e sıppula]
98€ 5% 16°0 EN) 28 WE le 5
ET Er: RE A a IE me rlC (704 '4) sndood 'd
xur0/r sEIE zueldErH 8% zunrlG Y 90 9
user ER (01 0 Ze 1 eo (O 34) .AUSUO,, grand]
ee 24:8 G'I IUOE E89 OO Ce
BO TE ee ler ee url ee (1 3°3°7) rund 3°]
re ee oz 0)
re ee BZ 5 (7 33:7) Yun 3°]
»Zl 1 Ken Eva | »OC 1 re } vrslZ 1
xurGl € 609 +69 C Kr xrx86 € xx«PrC (Z al 2d) .soUueTT,, ovaund ‘d
9] OO ale 90°]
ee EEE ne VE ne (1 T 34) „soue]],, avuwnd 'q
+00 C ehe sr lS C
BZ ar Re (€ 'q’d) (79 = ur) ds sawıg9904]

1: 9G "€ (134) „Uoseg,, oravnd 'q
279€
eG (uv3 'q) sıyjorurd ']
BERHEEeL OLcna |
zr28"7 (z '4’d) (79 = ur) 'ds sCung9aoug

= (1 'q’q) (79 = uz) 'ds saung9o04g

(M01 I9M0]J) sgipueur 9y3 pue (Mo1 ıaddn) jjnys 3y3 107 spro.1zua9 dno13 us9MJ3q sa9ueJsıp sigouejeyeg Jo xıayeur ayyJ, 'p oJgvL

374 M. Aguilera and M. Corti

Proechimys sp. Guaquitas (2n=62)

Proechimys sp. La Nulita (2n=62)
Proechimys sp. Tierra Buena (2n-62)
F. guairae ''Falcon" (2n=46)

P. guairae ''Llanos" Palmero (2n=50)
P. @. guairae El Limon (2n=48)

P. poliopus (2n=42)

P. guairae ''Llanos' Turen (2n-50)
P. g. guairae Turiamo (2n=48)

P. trinitatis (2n=62)

P. guairae ''Oriente'' (2n=52)

P. canicollis (2n=24)

Fig. 4. UPGMA phenogram computed from the Mahalanobis distances between population means for
the skull

P.b. 1
3 [) P.b. 3
[)
j P.g.O
P.trin. e,P.g.L1 o
ON ®
SR P.g.g. 1
©) P.g. F „P-poli.
o ()
o P.ge.L2
Pb.22 a Re =
I = ° P.can.
T l T T T
-2 -1 (6) 1 2 3 4
CV1

Fıg. 5. Scatter plot of population centroids onto first two canonical variates computed on mandible
characters. See Tab. 1 for population acronyms

respectively). The scatter plot of population centroids onto first two canonical variates is
given ın figure 5. CV1 produces a separation of P. canicollis and of P. guairae “Oriente”,
which have positive values, from all the others, which have negative values. Lengths of
diastema and of alveolar tooth row and mandible height contribute mainly to CV1 and
CV2, with pooled within-class standardised character coefficients that are at least 3 times
greater than the coefficients of mandible length.

P. guairae “Oriente” and P. canicollis have the highest Mahalanobıs distances and all are
highly significant (Hotelling’s T?, Tab. 4). Distances between the other populations vary
without regard to systematic relationships (Tab.4); for example, Mahalanobis distances
between populations of the same subspecies are usually high while distances between
populations of different subspecies or species may not be (Tab. 4).

However, a Mantel test between Mahalanobis distances derived from the skull and

Craniometric differentiation and chromosomal speciation of Proechimys 9715

those derived from the mandible shows that the two are significantly correlated (r = 0.388,
p = 0.0254), i.e. the analyses on the two skeletal components depict a congruent pattern of
population differentiation.

We also performed a Mantel test between the skull Mahalanobis distances and the linear
geographic distances measured between the collecting sites. The correlation proved to be
significant (r = 0.358, p = 0.0164). However, the correlation coefficient increases when the
test is performed within the guairae complex only (r = 0.40096, p = 0.0064), ı.e. when the
two allospecies P. trinitatis and P. canicollıs are excluded.

Analysis of variance on character ratios showed that they all differ significantly
(p<0.01) except BL, PL, IZL and CW. GT-2 comparisons revealed that significant
differences among means are differently scattered, some of the characters exhibiting a
homogeneous pattern of differences. There are few significant differences among means
within the P. guairae, and most of the significant differences are related to the other
species. Among these, NL and PAW distinguish P. poliopus, UAL, BTL and FMW are
unique for P. trinitatis, and BTW, IFW and IFL for P. canicollis. UDL and CH distinguish
P. canicollis and P. guairae “Oriente” and MXHR P. guairae “Llanos”, P. guairae
“Orıente” and P. trinitatıs.

Discussion

There is no completely positive answer to the first question on the morphomerric
identification of karyomorphs. Morphometric distinction ıs clear for those species that
have already been recognized and reported in the literature (e.g. PATTON 1987; GARDNER
and Emmons 1984; Reıc et al. 1980). Differences in morphomertric traıts are less evident
among the speciating taxa of the guairae complex. This ıs ın agreement with the allozyme
analysıs from BENADO et al. (1979) who showed that genetic distances within the
rassenkreis are low compared to those with P. trinitatis (their P. urichi; AGUILERA et al
1994).

There is a clear morphometric distinction between P. canicollis (the only species
forming the canicollis group) and the trinitatus group. Mahalanobis distances and the
UPGMA phenogram clearly highlight this differnce. This is in agreement with PATTon’s
(1987) hypothesis that P. canıcollis forms a well differentiated group.

Of greatest interest is the trinitatus group, ın which multivarıate morphometrics show a
high degree of differentiation among populations. P. trinitatis has the highest distinction ın
skull shape within this group. This species is restricted to a small area ın eastern Venezuela
and Trinidad island, and morphomertric differentiation may be a consequence of a longer
time of divergence resulting from the different routes of range expansıon of the genus.

Within the superspecies P. guairae, P. poliopus (already accepted as a different species
by Reıc et al 1980) shows a high morphometric differentiation, as well as P. guairae
“Orıente”, which occurs ın east Venezuela in a limited area. Moreover, it is interesting to
note that the UPGMA based upon skull Mahalanobis distances allows a distinction to be
made between the three populations of Proechimys sp. (2n = 62) and the other subspecies
of the rassenkreis. It has been proposed by AGUILERA et al (1994) that the former should be
considered as a new species, and their morphometric distinction suggests that the trend ın
the change of morphology of the skull is congruent with their chromosomal differentia-
tion. All populations of the P. guairae (i.e. P.g. guairae, P. guairae “Falcon”, P. guairae
“Llanos”) share the same sort of morphological modifications and the relationships among
them are partially congruent with modifications in karyotype.

Some of the character ratios (the ratio between each character and total length of the
skull) are of help in identifying the allospecies, ı.e. P. trinitatis, P. canicollis, and P. guairae
“Orıente”. It ıs not possible to perform a-posteriori identification of specimens from the
guairae complex using any individual character as ratio or as raw measurement, and

376 M. Aguilera and M. Corti

morphometric distinction ıs clear only in a multivariate context, as, for example, for the
species Proechimys sp. (2n = 62) which ıs clearly distinct. |

There is one question relative to morphometric differentiation in this speciose group:
Are the morphological changes reported here a by-product of chromosomal speciation, or
do they represent independent adaptation to local ecological conditions?

The fact that there is a significant correlation (although not particularly high) between
morphometrics and geography across all populations, and that this correlation is even
stronger within the guairae rassenkreis, indicates that morphomertric differentiation origi-
nated in the course of the successive events of speciation coupled with chromosomal
change. Therefore, our results are in favour of a phylogenetic cause for the morphometric
divergence among population and species.

Reıc etaal. (1980) hypothesized that speciation in the rassenkreis occurred following the
stasıpatric model of WHITE (1968). In this context, morphometric differences are believed
to have arısen in a clinal model where primary integration zones progressively evolved to
form tension zones across which gene flow ıs highly limited ıf not absent. However, the
following alternative model of chromosomal speciation adopted by Reıc (1980) and
subsequently accepted by AGUILERA et al. (1994) favours the hypothesis of speciation via
centric fission as essentially peripatric (Mayr 1982), following an increase in diploid
number. “This process was repeated several times under the influence of cycles of forest
retraction and expansıon determined by the Pleistocene climatic fluctuations” (Reıc et al.
1980, p. 308). If this is true, morphological differentiation in the superspecies Proechimys
guairae is a direct product of speciation of small peripheral isolates occurring over the last
50,000 years (BENADO et al. 1979).

Acknowledgements

This study has been supported by CONICIT, grant S1-1274 (©. A. Reıc and M. ‚AGUILERA), by
Decanado Investigacines-USB (M. Acustera), by CNR, Progetto bilaterale “Composizione
genomica e speciazione cromosomica nei roditori” no. 93.00233.CT04 (M. Corrr), and by MURST,
Fondi di Ateneo Universitä dı Roma ‘La Sapienza’, quota 60% “Meccanismi dı speciazione in
ambiente tropicale” (M. Corri).

Zusammenfassung

Kraniometrische Differenzierung und chromosomale Artbildung in der Gattung Proechimys
(Rodentia: Echimyidae)

Mittels multivarıater morphometrischer Methoden wurden die Systematik und die geographische
Variation der Gattung Proechimys in Venezuela untersucht. Die erfaßten Populationen zeigen
ausgeprägte Unterschiede hinsichtlich des jeweils vorkommenden Karyotyps (2n = 24 - 2n = 62). Die
auf Schädel- und Mandibelmerkmalen beruhende morphometrische Differenzierung in der Gattung
Proechimys unterstützt die Hypothese, daß aufeinanderfolgende Artbildungsereignisse mit einem
Anstieg der diploiden Chromosomenzahl einhergingen. Während sich die jeweiligen Einzelmerkmale
diesbezüglich als unzureichend erwiesen, waren alle Arten und Unterarten mit multivariaten Verfah-
ren klar unterscheidbar.

References

AFEWORK BEKELE; CAPANNA, E.; CoRTI, M.; Marcus, L. F.; SCHLITTER, D. A. (1993): Systematics
and geographic variation of Ethiopian Arvicanthis (Rodentia, Muridae). J. Zool. London 230,
117-134.

AGUILERA, M.; REıG, ©. A.; BARRoS, M. A.; BasANEz, M. G. (1979): Sistematica, cıtogenetics y datos
reproductivos de una poblacion de Proechimys canicollis del Noroeste de Venezuela. Acta Cient.
Venezolana 30, 408-417.

BENADO, M.; AGUILERA, M.; Reıc, O. A.; AyaLa, F. J. (1979): Biochemical genetics of chromosome
forms of Venezuelan spiny rats of the Proechimys guairae and Proechimys trinitatis superspecies.
Genetica 50, 89-97.

Burnasgy, T. P. (1966): Growth-invariant discriminant functions and generalized distances. Biomet-
rics 22, 96-110.

Craniometric differentiation and chromosomal speciation of Proechimys O7

Dos Reıs, $. F.; PEssoA, L. M.; Strauss, R. E. (1990): Application of size-free canonical discriminant
analysis to studies of geographic differentiation. Rev. Brasil. Genet. 13, 509-520.

GARDNER, A. L.; Emmons, L. H. (1984): Species groups in Proechimys (Rodentia, Echimyidae) as
indicated by karyology and bullar morphology. J. Mammalogy 65, 10-25.

Kınc, M. (1993): Species evolution. Cambridge: University Press.

KLINKENBERG, C. P.; ZIMMERMANN, M. (1982): Static, ontogenetic and evolutionary allometry: A
multivariate comparison in nine species of water spiders. Amer. Natur. 140, 601-620.

Marcus, L. F.; CorTtı, M. (1989): Data analysis in systematics. Int. Ther. Con. 5, pp 1-61.

Mayr, E. (1982): Processes of speciation in anımals. In: Mechanisms of speciation. Ed. by
C. Barıcozzı. New York: AlanRR. Liss. Pp. 1-19.

Parton, J. L. (1987): Species groups of spiny rats, genus Proechimys (Rodentia: Echimyidae). Field.
Zool. 39, 305-345.

PATTon, J. L.; ROGERs, M. A. (1983): Systematic implications of non-geographic varıation in the
Spiny rat genus Proechimys (Echimyidae). Z. Säugetierkunde 48, 363-370.

PEssoA, L. M.; DE OLivEIra, J. A.; Dos Reıs, $. F. (1990): Quantitative cranial character variation ın
selected populations of the guyannensis-group of Proechimys (Rodentia: Echimyidae) from Brazil.
Zool. Anz. 225, 396-400.

PessoA, L. M.; Dos Reıs, $S. F. (1990): Geographic varıation in Proechimys dimidatus (Guenther)
(Rodentia: Echimyidae). Zool. Anz. 225, 383-390.

— — (1991a): The contribution of cranıal indeterminate growth to non-geographic varıatıon in adult
Proechimys albispinus (Is. Geoffroy) (Rodentia: Echimyidae). Z. Säugetierkunde 56, 219-224.

— — (1991b): Cranial infraspecific differentiation in Proechimys ıheringi Thomas (Rodentia:
Echimyidae). Z. Säugetierkunde 56, 34-40.

Reıc, ©. A. (1980): Modelos de especiacion cromosomica en las casıraguas (genro Proechimys) de
Venezuela. In: Ecologia y genetica de la especiacion anımal. Ed. by ©. A. Reıc. Equinoccio:
editorial de la Univ. S. Bolivar. Pp. 149-190.

— (1989): Karyotypic repattering as one triggering factor in cases of explosive speciation. In:
Evolutionary biology of transıent unstable populations. Ed. by A. FONTEVIDELA. New York:
Springer-Verlag. Pp. 246-289.

Reıc, ©. A.; AGUILERA, M.; BARROS, M. A.; UsEcHE, M. (1980): Chromosomal speciation in a
rassenkreis of Venezuelan spiny rats (genus Proechimys, Rodentia, Echimyidae). Genetica 52/53,
291-312.

ROHLEF, F. J.; BOOKSTEIN, F. L. (1987): A comment on shearing as a method for ‘size correction’; Syst.
Zool. 36, 356-367.

SoKAL, R. R.; ROHLEF, F. ]J. (1981): Biometry. 2nd ed. New York: W. H. Freeman and Co.

WHITE, M. J. D. (1968): Models of speciation. Science 159, 1065-1070.

Authors’ addresses: MARISOL AGUILERA, Departamento Estudios Ambientales, Universidad Simon
Bolivar, Apdo. 89.000, Caracas, YV-1081-A, Venezuela, and Marco CoRrrI,
Dipartimento dı Biologia Anımale e dell’Uomo, Universita di Roma ‘La
Sapienza’, Vıa Borelli 50, I-00161 Roma, Italy

Z. Säugetierkunde 59 (1994) 378-379
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

WISSENSG HR EFEICHEZKURZI III EIINISEIN

Cases of dental malocclusion in populations of Red foxes
(Vulpes vulpes) in the state of Victoria, Australia

By E. MEIJAARD and P. J. H. van BREE

Department of Mammalogy, Zoological Museum, University of Amsterdam, The Netherlands

Receipt of Ms. 28. 1. 1994
Acceptance of Ms. 27. 7. 1994

BOUWMEESTER et al. (1989) described the high incidence ofa pronounced protrusion of the
maxillary incisors over the mandibular incisors which they found in skulls of red foxes
(Vulpes vulpes) in the North Holland Dune Reserve (NDH) in the Netherlands. This
aberration was present in 6.7 % of the skulls and proved, through skull measurements, to
be the result of a shortening of the front part of the mandibles. The aberration was thought
to be under monogenetic control. Its high incidence could be explained by the history of
the fox population in the NDH. Before 1968 the NDH was not inhabited by red foxes. In
that year four cubs from the same litter were set free, thus creating a small and isolated gene
pool. Compared to normal red foxes, affected anımals are likely to be at an ecological
disadvantage. BOUWMEESTER et al. (1989) expect the incidence of the aberration to decline
in the future. This view is supported by the fact that among several thousand red foxes
from England, where no such genetic bottleneck occurred, such anomaly was found only a
couple of times. Not one case was reported among several thousand red foxes from Sweden
(BOUWMEESTER et al. 1989).

In a study on skulls of red foxes from Australia three out of 39 skulls (= 7.7 %) showed

Fig. Skull of female fox, Vulpes vulpes, from around Geelong, Victoria, Australia, [-1989 (ZMA
24.065) with shortened mandibles

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5906-0378 $ 02.50/0

Dental malocclusion in Red foxes from Victoria, Australia 379

a similar shortening of the mandible. These skulls were collected around Geelong, Victoria
(38° 27’ S, 144°51’ E). In a collection of 22 skulls from the Western Australian Museum in
Perth one skull (= 4.5 %) showed the aberration. Palatal length was not measured on this
specimen. All the 56 fox skulls ın the collection of B. J. Coman (at Strathfieldsaye,
Australia) which were mostly from different areas ın Victoria, were normal, just like 42
skulls from the Museum of Victoria in Melbourne and 5 skulls from the Australian
Museum in Sydney.

Considering the history of Australian red foxes, which were introduced ın Australia
over 130 years ago (RoLıs 1969) and have been isolated since then, the incidence of the
aberration was thought to be remarkably high, i.e. 2.5 % of all the skulls mentioned above.

The Figure shows a specimen with shortened mandible out of the Geelong area.

To investigate the character of this aberration a number of skull measurements were
taken, namely condylo-basal length, palatal length and mandible length. By calculating the
palate/condylo-basal ratio and the mandible/condylo-basal ratio and by comparing these
calculations with the values for normal red foxes from the sample, it became clear that thıs
was a case of shortened mandibles. This is important to know because the aberration could
also have been caused by an elongated rostrum (BOUWMEESTER et al. 1989).

It goes beyond the scope of this contribution to speculate on causes of the established
high incidence of the aberration. Further research ıs needed on genetic and ecological
backgrounds to elucidate the reason for the described phenomenon. It seems, however,
that the small founder population in Australia has played an important role.

References

BOUWMEESTER, J.; MULDER, J. L.; BREE van, P. J. H. (1989): High incidence of malocclusion ın an
isolated population of the red fox (Vulpes vulpes) ın the Netherlands. J. Zool. (London) 219,
123-136.

Roıts, E. G. (1969): They all ran wild. Sydney: Angus and Robertson.

Authors’ address: E. MEIJAARD and Dr. P. J. H. van BREE, Institute for Systematics and Population
Biology (Zoological Museum), Mauritskade 61, NL-1092 AD Amsterdam, The
Netherlands

Z. Säugetierkunde 59 (1994) 380-381
© 1994 Paul Parey, Hamburg
ISSN 0044-3468

Eliomys (Hypnomys) onicensis nomen novum, to replace the
homonym Hypnomys intermedius Reumer, 1981 (Rodentia:
Gliridae) from Majorca

By J. W. F. REUMER

Natuurmuseum Rotterdam, Rotterdam, Niederlande

Receipt of Ms. 15. 3. 1994
Acceptance of Ms. 20. 7. 1994

In 1981, I described the endemic dormouse Hypnomys intermedius from the Pleistocene of
Majorca (REUMER 1981). The name was chosen for the intermediary position the species
takes in the evolutionary lineage leading from A. waldreni Reumer, 1979 to H. morpheus
Bate, 1944. The position of the Mediterranean island glirids within the trıbal framework of
the family was at that time unknown (Daams 1981). Some thirty years earlier, FRIANT
(1953) had described a subspecies of dormouse from the Ruscinian of Stte (southern
France), which ıs now known as the species Eliomys intermedius Friant, 1953.

ZAMMIT MAEMPEL and DE Brunn (1982) were the first to include the endemic
Mediterranean glirid genera as subgenera within Ehomys Wagner, 1840. This opinion is
since tollowed, e.g. by MoyaA-SoLa et al. (1984) and by ALcovEr and Acustt (1985).

Hypnomys intermedins Reumer, 1981 was considered by ZamMIT MAEMPEL and DE
Bru1pn (1982) to be a junior synonym of H. eliomyoıdes Agusti, 1980 from the Balearic
island of Menorca. However, AGcustı and MoyaA-SoLA (1990) considered A. intermedius
and H. eliomyoides to be two valıd species, based on morphological differences.

These opinions leave us wıth Ehomys (Hypnomys) intermedius Reumer, 1981 as a valid
species. It is then a homonym of Ehomys intermedius Friant, 1953. A nomen novum is
therefore necessary, for which I propose

Ehomys (Hypnomys) onicensis nomen novum

The name is derived from the type localıty (Sa Pedrera de S’Onix, Porto Cristo, Majorca).
It is noteworthy in this context that Acusti (1986) writes: “Hypnomys is considered to
evolve from Eliomys intermedius Friant.”

References

Aust, J. (1986): Dental evolution in the endemic glirids of the Western Mediterranean islands. In:
Teeth revisited. Ed. by D. E. Russe, J.-P. SANTORO, and D. SIGOGNEAU-RusseELL. Mem. Mus.
natn. Hist. nat., Parıs (serie C) 53, 227-232.

Acustı, J.; MovaA-SoLa, $. (1990): Neogene-Quaternary mammalıan faunas of the Balearics. Accad.
Naz. dei Lincei, Attı dei convegni Lincei 85, 459-468.

ALCOVER, J. A.; Acustı, J. (1985): Ehomys (Eivissia) canarreiensis n. sgen., n. sp., nou glırıd del
Pleistoce de la Cova de Ca Na Reıa (Pitiüses). Endins 10-11, 51-56.

Daams, R. (1981): The dental pattern of the dormice Dryomys, Myomimus, Microdyromys and
Peridyromys. Utrecht Micropal. Bull., spec. publ. 3, Utrecht.

FRIANT, M. (1953): Une faune du Quaternaire ancien en France mediterraneenne. Ann. Soc. geol.
Nord 73, 161-170.

Moyva-Sora, $.; Acustı, J.; Pons, J. (1984): The Mio-Pliocene insular faunas from the West
Mediterranean. Origin and distribution factors. Paleobiol. continent., Montpellier 14 (2),
347-357.

REUMER, J. W. F. (1981): The Pleistocene small mammals from Sa Pedrera de S’Onıx, Majorca
(Gliridae, Soricidae). Proc. Koninkl. Ned. Akad. Wetensch., B 84 (1), 3-11.

U.S. Copyright Clearance Center Code Statement: 0044-3468/94/5906-0380 $ 02.50/0

Eliomys (Hypnomys) onicensis nomen novum to replace the homonym Hypnomys intermedins 381

ZAMMIT-MAEMPEL, G.; DE Brurjn, H. (1982): The Plio/Pleistocene Gliridae from the Mediterranean
islands reconsidered. Proc. Koninkl. Ned. Akad. Wetensch., B 85 (1), 113-128.

Authors’ address: Dr. JELLE W. F. REUMER, Natuurmuseum Rotterdam, P. ©. Box 23452, NL-3001
KL Rotterdam, The Netherlands

MEEREIIEUNGENDDERFEESERESCHANET

Protokoll über die außerordentliche Mitgliederversammlung der Deutschen Gesell-
schaft für Säugetierkunde e.V. am 16. Juli 1994 im Großen Hörsaal des Zoologischen
Instituts der Universität Bonn

Der 1. Vorsitzende, Herr SCHMIDT, eröffnet die Versammlung um 13.00 Uhr.

1. Die Tagesordnung wird angenommen.

2. Zeitschrift für Säugetierkunde:
Herr SCHMIDT schildert die Chronologie der Aktivitäten der Gesellschaft seit dem
unrechtmäßigen Verkauf der Verlagsrechte durch den Parey-Verlag an den Verlag
Blackwell-Wissenschaft, Berlin. Er gibt bekannt, daß sechs Verlage daran interessiert
sind, die Herausgabe der tradıtionsreichen „Zeitschrift für Säugetierkunde“ ab Band 60
(1995) zu übernehmen. Es sind dies Aula/Wiesbaden, Birkhäuser/Basel, Blackwell/
Berlin, Ferdinand Enke/Stuttgart, Gustav Fischer/Jena, Walter de Gruyter & Co./
Berlin. Herr SCHMIDT schildert kurz die Vorzüge und Nachteile des jeweiligen Ange-
botes und gibt bekannt, daß der Vorstand unter Abwägung aller Gesichtspunkte (z.B.
Renommee des Unternehmens, Verlagsprogramm, Zuverlässigkeit in der Herstellung,
Kooperation mit der Schriftleitung, Preisgestaltung, Format) sich dazu entschlossen
hat, der außerordentlichen Mitgliederversammlung vorzuschlagen, der Reihenfolge
nach mit folgenden Verlagen Kontakt aufzunehmen: Fischer, Birkhäuser, de Gruyter,
Enke.
In der anschließenden Diskussion wird die Frage, ob es sinnvoll sei, die Zeitschrift in
Eigenregie herauszugeben, erörtert und abschlägig beschieden. Eine weitere Frage gilt
der wirtschaftlichen Situation des Verlags Fischer, Jena; sie wird vom Vorstand als
gesund bezeichnet. Geheim und schriftlich stimmen danach die Anwesenden über den
Vorschlag des Vorstandes ab, bei Parey zu kündigen und Verhandlungen mit anderen
Verlagen in der vorgeschlagenen Reihenfolge aufzunehmen. Das Ergebnis: 23 Ja-
Stimmen, keine Gegenstimme und keine Enthaltung.

3. Verschiedenes:
Hierzu wird die Frage erörtert, ob es sinnvoll sei, eine automatische Abbuchung der
Mitgliederbeiträge vorzunehmen. Die Mehrheit spricht sich aus Kostengründen da-

gegen aus.

Um 14.00 Uhr schließt der 1. Vorsitzende die Versammlung.

Prof. Dr. U. ScHMIDT Prof. Dr. H. ERKERT Dr. H. FRÄDRICH
1. Vorsitzender Geschäftsführer Schriftführer

Protokoll über die Mitgliederversammlung der Deutschen Gesellschaft für
Säugetierkunde e. V. am 26. September 1994 im Hörsaal des Biozentrums der
Universität Wien

Der 1. Vorsitzende, Herr SCHMIDT, eröffnet die Versammlung um 16.30 Uhr und gibt
unter dem Beifall des Auditoriums bekannt, daß Herr TEMBROcK, Berlin, zum Ehrenmit-
glied ernannt wurde.

1. Die Tagesordnung wird angenommen.

2. Der Geschäftsführer, Herr ERKERT, verliest den Bericht über das Jahr 1993. Die 67.
Hauptversammlung der Gesellschaft fand auf Einladung von Herrn MAIER vom 26.
September bis 1. Oktober 1993 in Tübingen statt; sie tagte dort gemeinsam mit der
Gesellschaft für Primatologie. Ein zusätzlicher Tag galt Fragen des Fledermausschut-
zes. Schwerpunktthemen waren „Offene Fragen der Phylogenie und Systematik der

10.

Jl,

383

Großgruppen“, „Geruchssinn und olfaktorische Kommunikation“ sowie „Biologie
der Primaten“. Mit 75 Vorträgen und 45 Postern war die Veranstaltung ein Erfolg. Der
FRITZ-FRANK-Förderpreis der DGS wurde an Herrn Dr. THomas MarTIn, Berlin, für
seine Arbeıt „Schmelzmikrostruktur ın den Inzisiven alt- und neuweltlicher hystrico-
gnather Nagetiere“ vergeben. Drei Poster wurden mit Buchpreisen bedacht, die der
Parey Verlag dankenswerterweise gestiftet hat. Herr ERKERT dankt den Veranstaltern,
Herrn MAIER und Herrn FISCHER, für die Ausrichtung, Herrn NIEMITZ und Herrn
MÜLLER für die organisatorische Unterstützung der erfolgreichen Tagung.
Im Berichtsjahr erschien der 58. Band der „Zeitschrift für Säugetierkunde“ in sechs
Heften mit insgesamt 384 Seiten; den beiden Schriftleitern und den aktiven Heraus-
gebern wird gedankt. Die Mitgliederzahl hatte sich bis Ende 1993 auf 622 geringfügig
erhöht. Durch den Tod verlor die Gesellschaft folgende Mitglieder:

Prof. Dr. Fritz Strauss, Wabern/Schweiz

Prof. Dr. MARTIN EISENTRAUT, Bonn

Herr HUBERT MERZ, Langenbach.

. Die Satzungsänderungen sind durch Eintragung in das Vereinsregister des Amtsge-

richtes Berlin-Charlottenburg nunmehr wirksam geworden. Die neue Satzung und das
Mitgliederverzeichnis gehen den Mitgliedern mit der nächsten Aussendung zu.

. Herr ERKERT erläutert den von Frau KÜHnrıchH abgefaßten detaillierten Kassenbericht

und dankt Frau KÜHnRICH für ihre sorgfältige und effektive Arbeit.

. Die Herren BOHLKEN und SCHLIEMANN haben die Konto-Unterlagen der Gesellschaft

in Hamburg geprüft und für korrekt befunden.

. Die Anträge auf Entlastung der Schatzmeisterin und des Vorstandes werden bei

Enthaltung des Vorstandes angenommen.

. Die Herren BOHLKEN und SCHLIEMANN werden bei einer Enthaltung als Kassenprüfer

für das Geschäftsjahr 1994 gewählt. Beide sind mit der Wahl einverstanden.

. Der Vorstand schlägt vor, die Mitgliedsbeiträge für 1995 unverändert zu lassen. Dies

wird bei einer Enthaltung angenommen.

. Die Mitgliederversammlung nimmt die Einladung von Herrn FIscHER an, die 69.

Jahrestagung vom 24. bis 28. September 1995 in Göttingen abzuhalten. Als Schwer-
punktthemen sind vorgesehen „Säugetiere in der Kulturlandschaft“, „Fortpflanzungs-
biologie“, „Chronobiologie/ Aktivitätsrhythmen“.

Per Akklamatıon wird dıe Einladung von Herrn KruskA angenommen, der für das

Jahr 1996 nach Kiel eingeladen hat.

Herr SCHMIDT berichtet, daß wegen der kritischen Situation der „Zeitschrift für

Säugetierkunde“ am 16. Juli 1994 eine außerordentliche Mitgliederversammlung in

Bonn stattfinden mußte, die leider nur schlecht besucht war. Herr ERKERT trägt den

gegenwärtigen Stand der Verhandlungen mit dem Parey Verlag vor und berichtet, daß

Kontakte mit dem Fischer Verlag, Jena, als möglichem künftigen Herausgeber aufge-

nommen wurden. Obgleich es derzeit noch ungeklärte juristische Fragen gibt, wird

der Vorstand alles versuchen, das Erscheinen des Bandes 60 zum frühestmöglichen

Zeitpunkt zu bewirken.

Die Kommissionen und Arbeitsgruppen der DSG berichten über ihre Tätigkeit.

a) Erneut wird kritisiert, daß von der Tierschutz-Kommission noch immer kein
ausführlicher Bericht vorliegt. Bei der anschließenden Diskussion wird deutlich,
daß die Arbeit gerade dieser Kommission zugegebenermaßen schwierig ist. Den-
noch drängen die Diskussionsteilnehmer darauf, daß die Tierschutzkommissions-
Mitglieder ihre Arbeit beschleunigen.

b) Herr ScHRÖPFER berichtet über aktuelle Probleme ım Hinblick auf Tier- und
Artenschutzämter und wird von der Versammlung beauftragt, eine Artenschutz-
Kommission zu gründen, welche die Zielsetzungen und Forderungen der DGS
definieren soll.

384

c) Herr HEIDECKE trägt die Aktivitäten der Biber- und Bisamgruppe vor.

d) Herr FrÄDrıcH berichtet über das erste Treffen der Arbeitsgemeinschaft Tiergar-
tenbiologie, das Ende 1993 ın Erlangen erfolgreich abgehalten wurde. Er gibt
bekannt, daß die zweite Tagung dieser Art auf Einladung von Herrn GANSLOSSER
vom 11. bis 13. November 1994 ebentalls ın Erlangen stattfinden soll.

e) Herr ScHmipr berichtet über die Aktivitäten der „Koordinationsgruppe Fleder-
mausschutz“ sowie über die erfolgreiche Tagung, die vom 22. bis 25. Juli 1994 ın
Bonn stattgefunden hat.

f) Die Versammlung begrüßt den Antrag von Herrn UHR, eine Arbeitsgemeinschaft
über das Thema Domestikation ins Leben zu rufen, und beauftragt ıhn, dazu
geeignete Schritte zu unternehmen.

12. Herr ScHmipr bittet die Anwesenden um Unterzeichnung eines Glückwunschbrietes
an Herrn HERRE, der im Maı 1994 85 Jahre alt wurde.

Der von Herrn HUTTERER formulierte Glückwunschbrief zum 75jährigen Jubiläum

der amerikanischen Säugetiergesellschaft ist dort mit Genugtuung aufgenommen

worden. Die amerikanischen Kollegen schlugen vor, zu gegebener Zeit beide Gesell-
schaften gemeinsam tagen zu lassen.
Die Sitzung endet um 18.35 Uhr.

Prof. Dr. U. SchmIDT Prof. Dr. H. ERKERT Dr. H. FRÄDRICH
1. Vorsitzender Geschäftsführer Schriftführer

10th International Bat Research Conference
25th North American Symposium on Bat Research

The 10th International Bat Research Conference and 25th North American Bat Research
Conference will be held at Boston University, Boston, Massachusetts, USA (7-12 August
1995). This joint conference will include: Plenary addresses, symposia, contributed papers,
poster papers, films/videos, workshops, and excursions. Questions concerning this con-
ference should be addressed to: 10th International Bat Research Conference, Department
ot Biology, Boston University, Boston, Massachusetts 02215, USA.

Erscheinungsweise und Bezugspreis 1994: 6 Hefte bilden einen Band. Jahresabonnement Inland:
DM 378,- zuzüglich DM 13,80 Versandkosten; Jahresabonnement Österreich: öS 2949,- zuzüg-
lich 6S 164,- Versandkosten; Jahresabonnement Schweiz: sfr 364,— zuzüglich sfr 21,— Versand-
kosten; Jahresabonnement EG-Binnenmarkt-Länder mit USt-ID-Nr.: DM 353,27 zuzüglich
DM 19,63 Versandkosten; Jahresabonnement EG-Binnenmarkt-Länder ohne USt-ID-Nr. und
Drittländer: DM 378,- zuzüglich DM 21,- Versandkosten. Das Abonnement wird zum Jahres-
anfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens
am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhand-
lung oder bei der Verlagsbuchhandlung Paul Parey GmbH & Co. KG, Spitalerstraße 12,
D-20095 Hamburg, Bundesrepublik Deutschland, bestellt werden. Die Mitglieder der „Deut-
schen Gesellschaft für Säugetierkunde“ erhalten die Zeitschrift unberechnet im Rahmen des
Mitgliedsbeitrages.

Z. Säugetierkunde 59 (1994) 6, 321-384

Mit dem Bandinhaltsverzeichnis

207 SKnauSOg|
Brvan/33 138 ST z

EI en

—

ADAZBRIATE HH Ha a

RA UWUAINT

—

zZ RUuaNT
IIRRARIFS SMITHSNNIAN

—

INSTITUTION

=

N

NOILMLILSN
LIBRARIE

> S31I4WUgI7T LIBRARIES SMITHSONIAN

zZ 3
= wm
> =
ae e)
r =
) m
z 2)
N
Z 02)
< <
== —|
(®) SIE
Io (77)
IE (©)
= 2
>
7 N
Ss S314V48911
1 m
zZ
= ad
& —
< —_
| c
cc 4
= fe)
= zZ
N
(m z
= oO
= >
=>)
„> er
= 5
m
7) zZ

NOILNLILSNI

LIBRARIES

S SJISVYSII_LIBRA

q

NVINOSHLINS

N INSTITUTION NOIlN

NOILNLILSNI

LIBRARIES SMITHSONIAN

= we
o ee)
>
= =
en e
hr &
w m
= (02)
N INSTITUTION
< N Z
= I
(@) Eu).
a in:
2 2
= z
5 3
Sr I
MM 7
= —
(a < ——
< ©
x 4
en [e)
“2 z
IN INSTITUTION
Re a RE

5 S314yYV4gdI9 LIBRARIES SMITHSONIAN

NOILNLILSNI

NOILNLILSNI

9 u

end [EB En
= — =
= = er
= a 5
° a °
er —_) =
INSTITUTION
ö TE:
E = E
DI. =)
Er 2 E
N = 5
= .# o z
—UNINeSHNIS EIN
Z Ss
zZ je
(©) a
(09) wn
I oO
= 2
P> > :
(dp) a
SMITHSONIAN INSTITUTION
2) zT N
u 7 u
= NIIT =
&. m &x
<<, = <
x = ©
2 ° =
—) 2 —s
s3lyvyg11
EI Zi le
= = =
> >
a > 2
m
u z &
INSTITUTION
a NN. 1)
= < =
z 5 T
ö 2 8
2 E z
2 >’
> 3 -
N fe a
6 wu 5
a he =
ja. IN. X 4
S RAS E S
(©) “ N Ex =
= er 3
INSTITUTION
5 _ 5
E ya =
250.4 = =
me HR ei >=

SMITHSON
NVINOSHLINS

SMITHSONIAN INSTITUTION

LIBRARIES

[3 EIBRARIES
)N NOILNLILSNI

NVINOSHLINS _S31UVY89 17

9 en

[1

NOILNLILSNI

SMITHSONIAN

NOILNLILSNI

SMITHSONIAN

NOILNLILSNI

LIBRARIES

LIBRARIES SMITHSONIAN

NOILNLILSNI

NVINOSHLINS S3IUVYAITI_LIBRARIES

we,

a
NOILNLILSNI

LIBRARIES

EEE

8IT_LIBRARIE
x

SMITHSONIAN

NOILNLILSN

NVINOSHLINS S:

INSTITUTION

SMITHSONIAN

N

NVINOSHLINS S314VAU

NVINOSHLINS S

A
SMITHSONIAN _ IN

ET,

ER
IN
ee

S314UV4g171 LIBRARIES SMITHSONIAN

“
N
N

INSTITUTION NOILNLILSNI

NVINOSHLINS _S’

NVINOSHLIWS

N

NOLLNLILSNI

NVINOSHLINS S

7
Bi

N
’
N
Y

R
%

INSTITUTION

SMITHSONIAN 1

NVINOSHLINS S31I4VYSIT_LIBRARIES

P4

=

=

oO

72}

ma

=

2.
N
_NVINOSHLINS _S
> N RN w '
= N >
4‘ N <h
Ss 8 €&
as, RR Dr
[@) —
= IN 1 A
SMITHSONIAN _I
za a

ION NOILNLILSNI NVINOSHLINS, S3IUVYAII_LIBRARIES INSTITUTION N

ag Km Ar pr a u.
u 7 uw MM) = w
= = Re = un „a £
= c 5% c | Be ie <
(66) = [6] a an? = 5
= S En 3 S er (®) =
25 2 ur
811 LIBRARI Da ul EL NOILNLILSNI NVINOSHLINS SI1IyYVyYgI? L
n m = r z u rm
a2 = 2 > =D . 5 | Do
7)
on zZ o z on z >
7)
=
=r
TE
77)
(e}
2
>
==

(ep) 2 aD >
2 < = EN < 2
2. 5 a 5: 5 3
I = 2 8: 2 2 :
i = E= 4 Lau 'r =
> - = a > = 5 = >
a z Z 7) zz (77) > - » >
IT _LIBRARIES SMITHSONIAN _ INSTITUTION NOILNLILSNI SHLINS S3ı9vagıT L
m “A = = SR = =
E E = EIS 5: E
= 5 5 3 Ss 5
= =] _ [e) Mia °
2 = r en 2 0 z
s3IuVygI1 LIBRARIES SMITHSONIAN N
z r 2 m 2 Bi =
9 5 ° Nemo. .5 °
= En = En = Def, 2 =
= Z =
> > = > 5 > =
> = > o E Ga: D =
7) m 22) = A = 72)
Me (7) = 7 = 7 =
3I1_LIBRARIES NOILNLILSNI NVINOSHLINS SI3IUVYSAIT_L
& a (47) 2 2) A zZ
N: = z EP - IP <
W. 2 | Baur) ee er DE Pr
2? ö E = 5 DA
j. = 3 “oO "x ®) fi %
1 = 4 > 2 ; =
= >» er my
7) z 3 = ro = 5
ION I LIBRARIES SMITHSONIAN N
u 77) 77, w 2 W
nn = # = z =
<<] <= ce X Ds. <<
2/ x —4 e &. c [a4
7 = ai em
_ = z a _ zZ ur
111 LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS SIIYVYEII L
\ 0 = u] = (00) = (00)
A 2 =! a =) Eau 5 a
E 2 = = = > ' > ®
7 = m = 7 = E z
PM z 7) = H . 2 n
ION NOILNLILSNI S31UVWYEAII_LIBRARIES SMITHSONIAN INSTITUTION N
2) zZ 2] 2 RZ Qn z Ne. (4)
2 < z „< Se < N z
Er z I zZ S zZ .SN 3
I. O L . oz SE ö IN =
e) = oO: = ö ERNNWEZ
> = = E = E =
a SMITHSONIAN _ INSTITUTION NOILNLILSNI NVINOSHLINS S3IUVY491I1I L
2 + z 1 za wu =
1 „2 a (e2) a
a 3 x a x = ne a
A = a
“| I x Ge! < — < —
E © 2. c e e e
7 2 en = E an) ur -
) = Z ö = | ° = °
2. ni ze = z un 2,
ION_ NOILNLILSNI _NVINOSHLINS_SI31IUVYY911 LIBRARIES SMITHSONIAN INSTITUTION Ni
z Mm: > P4 cm Fe Rz Ze

BEZ IT ” ME

Di TE ger

Zen
ge
mn

De u TE
N

N RR en

N a ae

Tanne

er 2.7

DE TEA
yarn gear a"

Dee re RR rt Lee mann er
4 VER ETEN

VER EERE

CR,

wa

a
DE Re une RP
N ARTEN RN
an a ni
VENEN

Ve ge Ra AT DR RE

Ka PR U ee

a Ce AA
a RL tya Zi ART REEL
. aa nr a an hate

Ba

Dt
ERAHNEN,
a SEE TTS

ERNST

ELVaR

ERCEL EP EZ

ee

ET TTEEN

RR ER

RENTEN TE NE
DE EN ERR IL DE Da Fol ai a ar
NR

a TE,

n ehe

ARE

Kaprun.
ne RR

arg
erahm ERIERCEEE RE
RE er, Tan MO a N
BER EZe Teer RE, a
DIE
SZ Ay
m nm wg
ERRRRSIRFSIE TWS OR UTESRENETN NER TAET EG
MO AEREER-MLÜLHRRTESTCHURSRERSSTEEE SET BEL NEE [EEE IP
EEE ENG vun»
A
rennt &
narneintrn)

DEITER SP
hr,

AN Na a

REN INNEAR
a EI EZ
anal 72 27

kN, NE.
rn

Nm ur N

ae a au a a a

rer ya

YA het Mut

u ren. Ward,
PESERERL EN IE LE ee fen

Euch taez man aA he
DEIEE 7)
PEN IAL REEL ER ar

SANS AN GGA tee uk ah SAN
ROUTE TITEL ESEL SL EL LENZE
» un u EN }

Ag gran
PER ET N TENEV

We

mtr ur rn nF
arena

wur

zul Verrat ne

Ara Au
irtege on
when
Area

IUZUED

en Mur
Pe

u arr ENDENRR

ar ap
Ant
ERIC? RER

er ee

VEN N
Da HE
EREIETETETE
DU ARE ERS A ENN
ANY iz MR
VaLDEa Er
mau haar,

ah
BL
mw wur

RA Anh
Breit
Da

Nam vor

re
Re,

BEI EEG
ehr

Ben
yerarwzut

win sur

Dre,
mei
ann mirte
ERELUEEENg CR vn
TEN E.
ine
EL VERENSTTETZE
ENT
ze
PIREEIHE TER

wa
Era VarEEr
ER
Tea Hau ET EN
BLUPITSEraV EUER DETZTEDET Kerr
N
PP ER Re!

Br
m
EEE
RT N Ae
Rehm
wm RR
PETE TEE eat:
Tee]
“rue
Nana ah

Daeur

as ENTE
Varsy. oh U Ver art ee
rer Agent tE rd wann
TDG ET u I TAZAEFFNEITI A
NN ER
Re ee
AL e 2 EEE
RRDPRETRUL SE TONER SE
LEER EP TSETI FEN
Aula

dr a ty

Ireranc er?
Bea PER LT Bella,

wann

EILTERE Te IV;
ap Areig ur N

er A

enge

Ir

PP

rue

Ronens ng
ne
PER TLN Dee ern En
az TEN "
a rast Anerıpa ung"
BEOLUBRETTN,
ver

re

ge

gen
ENT EN Tr ER

LE ERFAL
TE EN
OLZUENET EFEZET
Eee iaraazer
PRRETERTLER PERS ET

nr ange men
} ent Ne
ia Treat opera PAPER SU PO
ERDE TESET ENTER
a re ee
a ER RUE PIE

TE ERREGT NEN
re rn ea

ee De

IE

gan pr

ind
ai

en
er
FSEIEPTERWZ
ir
a
a a ETT TET

EBEPLFTZEE
eg erte

nt

73 te
TEE,

Pr

5
ee ng

PR Rn ee ee
Eee sfr ESF et ZEN DE

ESRaET
Da Ihe Abe Be
.. Fran en rn N
NER TEN SE En SB ka ee
nr PR AReTREE EB 4
a de et

EZ

ne ae art

ee nd
ne ie
a ee
wi

ac