Skip to main content

Full text of "Zeitschrift für Säugetierkunde : im Auftrage der Deutschen Gesellschaft für Säugetierkunde e.V"

See other formats


man? 


Erg 


DL 


M 


S Fe Are 


er 


= 
“= 
zu 


rn 


Pan 


% 


F 


\ 
Kr 
Y. 


BL 


nl 


7 
2. 


N ln A 


RR 


Ben 


Pe 
nz tergihese 


as ruple ale 
am are 


EE 
Erle s 


we 

were 

u ee 
eig 


SOEBEN 
ARE 
rer 

ie & 


Zr ie = 
PERMANENTE 


TITUTION NYILHLILSNI _NVINVSHLIND >310V081ı LIBRARIES SMITIMSONIAN 


S 
NI 


N 
LIBRARIES SMITHSONIAN 


NOILNLILSNI 
NOILNLILSNI 


UvVY4g17 INSTITUTION NOILNLILSNI NVINOSHLINS 


s31uvug 


SIJINVYIII LIBRARIES 
SJIUYYI1I LIBRARIES 


INSTITUTION 
INSTITUTION 


INSTITUTION NOILNLILS 


NOILNLILSNI NVINOSHLINS, S31UVY 917 


LIBRARIES _SMITHSONIAN 


NVINOSHLINS S314V4817_LIBRARIE 


SMITHSONIAN 
NYINOSHLIWS 


SMITHSONIAN 
DRG 


LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLUNS S31YV48 


S 


NOILNLILSNI NYVINOSHLINS 


TITUTION_ NOILNALILSNI s31uVy811 


II LIBRARIES SMITHSONIAN 


A — EN 
NVINOSHLINS LIBRARIES_ SMITHSONIAN 


INSTITUTION NOILNAILSNI 
S31I4UYY8171 LIBRARIES 


S3IUVYSIT_LIBRARIE 


INSTITUTION 


INSTITUTION NOILNLIASNI 
S3IyVug 


UVUglT _ LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI _NVINOSHLIWS S314Y48 


. Sn N N 
Sy 
SHIUVYAN 


NYINOSHLINS 
SMITHSONIAN 
NYINOSHLINS 
NVINOSHLINS 


TITUTION NOILNLILSNI _NVINOSHLINS SMITHSONIAN 


ARIES 


LIBRARIES SMITHSONIAN 
LIBRARIES, SMITHSONIAN 


NOILNLILSNI 
NOILNLILSNI 


ISYV49g17 LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS 


INSTITUTION 
S3INVYI17 LIBR 
INSTITUTION 


NVINOSHLINS S31I9VY9171 LIBRARIES SMITHSONIAN 


TITUTION 
N 


NVINOSHLINS S3I4V4911 
NVINOSHLINS S31UV4 8917 


SMITHSONIAN 
NVINOSHLIWS 
SMITHSONIAN 
SMITHSONIAN 


27 ‚ i “ 
I8v3gı7 LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31UVUE 


NOILNLILSNI 


LIBRARIES 
NOILNLILSNI 
LIBRARIES 


NOILNLILSNI 
u" 
LIBRARIES 


& N 
SONIA 


NOILNLILSNI 

| 
LIBRARIES 
NOILNLILSNI 


RARIES ne a u U NDIIILLSEG 
() Ro © 
= z = 
I. > 
m = E 
E - 5 
z = z 
INLILSNI S3INYY911 LIBRARIES 
z (7) z 2 
= = „zZ NR 
z = z SS 
E = EI, 
2 P> 2 ö = N 
i [77] 7 [73] A 
'RARIES SMITHSONIAN INSTITUTION NOILNLILSNI 
(77) = (77) 
m Fr Au 
2 —i — 
4 = [a4 
< y = < 
03 er a 
2 ° = 
= zZ id 


INLILSNI _NYINO 


ARIES 


Ss _S314vygaıı LIBRARIES 


un 7, 
° = 
Be X 
Al < 
m. S a 
rn = = 
NVINOSHLINS S31lyvuagı7 LIBRAR 
= z BE 
2 = 2 
B) 
> = > 
= = = 
0 .„ 2 (97) 
„„SMITHSONIAN INSTITUTION, NOILNAM 
= RE z 
102) 3 NN en”. 
: N > 
z = 


NVINOSHLINS S31IYVYI11_ 


A 
SMITHSONIAN _ INSTITUTION 


S3IMVNAIT LIBRARIES SMITHSONIAN 


INSTITUTION NOILNLILSN 


INSTITUTION NOILNLILSNI 


& = = 
= = > 
> > > 
= > = 
> A m 
RARIES „SMITHSONIAN_ INSTITUTION NOILNLILSNI NVINOSHLINS, S31UVualT LIBRAR 
| = Se = < z uw 
en | z _ M-- a) = 
R = M- iD o DE 
° T © ir ) Er NG 
< = = 2 = = N 
> = N >> = > = BIN 
7 Dogyıyy = z = ® Nol1nLI 
_NVINOSHLINS S314VY4IT_LIBRARIES SMITHSONIAN_ INSTITUTION „, 
= us z w D r= wi 
= 5 = = Win: = 
4 9 JE 
S a8: = 2 dTHE 3 E 
= a [60] = cr HE —_ 0} 
[e) Ä = | [@) za [e) _ 
= 3 = a — — 
‚RARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS Sala vyagı7 LIBRAR 
= m = Br oO ev) 
= > = En 
3. fa 2 
E& : E ER = - 
e = co = — > = 
[47] FÜ m ON. DC. (77) m “un m 
= = u Ben 
JLNLILSNI NVINOSHLIWNS S31UYY917 SMITHSONIAN INSTITUTION, NOIINLI. 
zZ 1 z 2) z : oO 
= = = = < SS - 
z — zZ = z AN = 2 0 
2 2 y 2 2 2 | \ 5 T 4 
3RARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31uvagı7 LIBRAR 
2 2 2 e x % = 
DE r SS : = 
<'/, MM = < {4 N Sitz = 
0 3 s EN: 3 
[s0) 7, —_ = — S & fe) 
an %, a er _ fe) = 


aa 9 a ha | 


Bl A 
FRE 


ER 


ZEITSCHRIFT FÜR 
SÄUGETIERKUNDE 


INTERNATIONAL JOURNAL 
OF MAMMALIAN BIOLOGY 


Organ der Deutschen Gesellschaft für Säugetierkunde 


Volume 47, 1982 ISSN 0044-3468 


Herausgeber / Editors 


P. J. H. van Bree, Amsterdam —- W. Fiedler, Wien - H. Frick, München - W. Herre, 
Kiel — H.-G. Klös, Berlin — H.-J. Kuhn, Göttingen - B. Lanza, Florenz - T. C. S. 
Morrison-Scott, London — J. Niethammer, Bonn - H. Reichstein, Kiel — M. Röhrs, 
Hannover — D. Starck, Frankfurt a.M. - F. Strauß, Bern -— E. Thenius, Wien — 
W. Verheyen, Antwerpen 


Schriftleitung/Editorial Office 
H. Schliemann, Hamburg —- D. Kruska, Kiel 


Mit 210 Abbildungen 


Verlag Paul Parey Hamburg und Berlin 


2 


NER TEE 


wi 


en 


Ai onen EA Kr 


Wissenschaftliche Autoren und Kopierrecht 


Seit Jahren wird darüber diskutiert, ob die Entwick- 
lung des Kopierens urheberrechtlich geschützter 
Texte eine Änderung der Kopierbestimmungen er- 
forderlich macht. Auf Grund von Initiativen des 
Börsenvereins des Deutschen Buchhandels hat das 
Bundesjustizministerium im Oktober 1980 einen 
Referentenentwurf veröffentlicht, der die Zahlung 
angemessener Vergütungen für das Kopieren ur- 
heberrechtlich geschützter Werke durch die Betrei- 
ber von Vervielfältigungsgeräten vorsieht. Dagegen 
hat sich u. a. der Deutsche Bibliotheksverband ge- 
wandt, der die wissenschaftlichen Autoren in einem 
Offenen Brief aufgefordert hat, gegen die geplante 
Neuregelung zu protestieren. Im einzelnen wird dar- 
in behauptet, eine Gesetzesänderung sei rechtlich 
nicht geboten, das Kopieren beeinträchtige nicht 
den Absatz wissenschaftlicher Literatur und die 
Vergütungspflicht werde eine Behinderung der wis- 
senschaftlichen Arbeit zur Folge haben. Darum liege 
eine Gesetzesänderung nicht im Interesse der wis- 
senschaftlichen Autoren. 


Der Börsenverein des Deutschen Buchhandels sieht 
sich veranlaßt, diese irreführenden Behauptungen 
des Deutschen Bibliotheksverbandes richtigzustel- 
len und deutlich zu machen, daß die Gesetzesände- 
rung gerade auch im Interesse der wissenschaftlichen 
Autoren liegt, für die die Erhaltung der wissen- 
schaftlichen Primärliteratur unverzichtbar ist. 


1. Die Gesetzesänderung ist verfassungs- 
rechtlich geboten 


Nach ständiger Rechtsprechung des Bundesver- 
fassungsgerichts und des Bundesgerichtshofs um- 
faßt die Eigentumsgarantie des Art. 14 GG auch 
das geistige Eigentum, also das Urheberrecht. 
$ 15 UrhG räumt dem Autor das alleinige Ver- 
vielfältigungsrecht ein. Einschränkungen dieses 
Rechts sind nur zum Wohle der Allgemeinheit 
und auch dann nur gegen Entschädigung zulässig. 
Die erwähnten obersten Gerichte haben wieder- 
holt festgestellt, daß Gesichtspunkte des Ge- 
meinwohls (Informationsfreiheit) Vervielfälti- 
gungen zum persönlichen und eigenen Gebrauch 
auch ohne Zustimmung des Rechtsinhabers er- 
forderlich machen können, daß dies aber nicht, 
wie bisher, vergütungsfrei erfolgen darf. Wenn 
nach Schätzungen und Hochrechnungen schon 
1978 in der Bundesrepublik Deutschland ca. 
5 Milliarden Kopien nach urheberrechtlich ge- 
schützten Texten hergestellt wurden, die sich an- 
gesichts der enormen Wachstumsraten in diesem 


Bereich bis heute weiter erheblich erhöht haben, 
spricht der Referentenentwurf mit Recht von 
einer neuen Nutzungsart der Werke, an denen der 
Rechtsinhaber auf Grund der verfassungsrecht- 
lichen Eigentumsgarantie angemessen zu beteili- 
gen ist. Der Gesetzgeber will und muß mit seinem 
Entwurf diesen verfassungswidrigen Zustand be- 
seitigen. 


Die wissenschaftliche Literatur 
ist gefährdet 


Die Auflagen wissenschaftlicher Zeitschriften 
und Monographien sind durch das Kopieren ent- 
scheidend zurückgegangen, wobei die öffentliche 
Hand diese Entwicklung durch Kürzung der Bib- 
liotheksetats noch fördert. Zwar würde nicht 
jedes Buch und jede Zeitschrift, aus denen kopiert 
wird, gekauft, aber daß diese technischen Mög- 
lichkeiten den Absatz wissenschaftlicher Litera- 
tur beeinflussen, lag für den Bundesgerichtshof 
schon 1955(!) „so klar zutage, daß dafür kein 
Beweis erbracht zu werden brauchte‘ (BGHZ 
18/44). Im übrigen gingen zwischen 1965 und 
1976 bei mehr als der Hälfte der wissenschaftli- 
chen Zeitschriften und Fachzeitschriften die Auf- 
lagen um 31% zurück, und das trotz der Verviel- 
fachung des Leserpotentials in dieser Zeit. Der 
Auflagenrückgang führt zwangsläufig zu höheren 
Preisen, wodurch wiederum der Anreiz zum Ko- 
pieren noch größer wird. 


Wenn diese Entwicklung nicht aufgehalten wird, 
haben viele wissenschaftliche Zeitschriften und 
Monographien keine Überlebenschance mehr. 
Die Wissenschaftler würden damit Publikations- 
möglichkeiten, der wissenschaftliche Nachwuchs 
Informationsmöglichkeiten verlieren, auf die sie 
dringend angewiesen sind. Die nach dem Entwurf 
von den Gerätebetreibern zu zahlenden Vergü- 
tungen sollen darum den Rechtsinhabern einen 
gewissen Ausgleich für die Nutzung ihres geisti- 
gen Eigentums geben und dadurch die Voraus- 
setzungen für die Erhaltung der wissenschatftli- 
chen Primärliteratur schaffen. 


Die Gesetzesänderung behindert nicht die 
wissenschaftliche Tätigkeit 


Der Gesetzentwurf behält die bestehende Ver- 
vielfältigungsfreiheit im wesentlichen bei. Darum 
wird die wissenschaftliche Tätigkeit in keiner 


Weise behindert. Soweit die Benutzer von den 
Gerätebetreibern mit geringfügigen Vergütun- 
gen belastet würden, müßte das auch für die 
wissenschaftlichen Autoren akzeptabel sein - in 
jedem Fall wäre dies das kleinere Übel gegenüber 
der jetzt bestehenden Gefährdung der wissen- 
schaftlichen Literatur. Für das praktische Ein- 
zugsverfahren können und müssen zwischen den 
Verwertungsgesellschaften einerseits und den 
Bibliotheken und wissenschaftlichen Instituten 
andererseits Modalitäten vereinbart werden, die 
den Verwaltungsaufwand auf ein Minimum be- 
schränken. 


Die vom Deutschen Bibliotheksverband behauptete 
Divergenz der Interessen zwischen Verlegern und 
Autoren besteht nicht: Tatsächlich liegt die Novel- 
lierung des Urheberrechtsgesetzes genauso im In- 
teresse der wissenschaftlichen Autoren wie der Ver- 
leger. Die Erhaltung der Primärliteratur ist für jeden 
Wissenschaftler und den wissenschaftlichen Nach- 
wuchs unvergleichlich wichtiger als die Belastung 
mit minimalen Zahlungen für das Anfertigen der 
Kopien. Es wird ja auch widerspruchslos für die 


technische Herstellung der Kopien sowie für alle 
anderen Arbeitsmittel, deren sich der Wissenschaft- 
ler bedient, bezahlt. Deshalb ist es nicht einzusehen, 
daß die Leistungen der Autoren und Verleger ge- 
ringer bewertet werden sollen. Wie Wissenschaftler, 
die als Autoren und Herausgeber wissenschaftlicher 
Zeitschriften mit der angesprochenen Problematik 
vertraut sind, über diese Fragen denken, geht aus 
einem an den Deutschen Bibliotheksverband gerich- 
teten Brief eines Lehrstuhlinhabers hervor, der sich 
u.a. wie folgt äußert: 


„Als wissenschaftlicher Autor und als Herausge- 
ber einer wissenschaftlichen Zeitschrift in Sorge 
um den Fortbestand unseres schriftlichen wissen- 
schaftlichen Kommunikationswesens, lege ich 
größten Wert auf eine Einschränkung des Xero- 
kopierens, lege ich größten Wert auf Erhöhung der 
Kosten all der Kopierverfahren. Ich lege ferner 
Wert darauf, daß — ähnlich wie bei Tonbändern 
und Tonbandgeräten — Verlage und Auteren aus 
den Kopierkosten Einnahmen erzielen.‘ 


Dem ist nichts hinzuzufügen. 


Frankfurt am Main, im März 1982 


BÖRSENVEREIN DES DEUTSCHEN BUCHHANDELS E.V. 


Wissenschaftliche Originalarbeiten 


ADvanı, R.: Feeding ecology of Tadarida aegyptiaca thomasi in the Indian desert. - Nahrungs- 
Ökologie von Tadarida aegyptiaca thoması aus der indischen Wüste ... 2... 2222... 


BERNARD, R. T. F.: Female reproductive cycle of Nycteris thebaica (Microchiroptera) from 
Natal, South Africa. - Weiblicher Fortpflanzungszyklus von Nycteris thebaica (Microchi- 
roBEErR)) aus Neal, Steinen den ee N er N 


BETHGE, E.: Zyklische Bestandswechsel (Gradationen) bei der Feldmaus (Microtus arvalıs), 
festgestellt durch Analyse von Eulen-Gewöllen. - Cyclic population changes of the common 
voleglgerotusarvahs)Iswudied by, analysısotowlpellets 2... 2... 


Born, E. W.; CLAUsEn, B.; HENRIKSEN, Sv. Aa.: Trichinella spiralis in walruses from the Thule 
district, North Greenland, and possible routes of transmission. — Trichinella spiralis in 
Walrossen aus dem Thule-Gebiet, Nordgrönland, und mögliche Wege der Übertragung 


BROEKHUIZEN, 9.; MAASKAMP, F.: Movement, home range and clustering in the European hare 
(Zepus europaeus Pallas) ın the Netherlands. — Ortsveränderungen, Aktionsräume und 
Gruppierungen beim Feldhasen (Zepus europaens Pallas) in den Niederlanden ........ 


BÜTTNER, D.; WoLLNnIK, FRANZISKA: Untersuchungen zur Kurzzeitperiodik beim Meer- 
schweinchen (Cavıa aperea f. porcellus). Teil I. - Investigations on ultradian rhythms in the 
euımeaspiea (@rorAapeneaitsporcellusyaBatetn 2... 0. 0u.unnenn: 


DE Jonc, W. W.; GooDMAN, M.: Mammalıan phylogeny studied by sequence analysıs of the 
eye lens protein a-crystallin. - Untersuchung der Stammesgeschichte der Säugetiere durch 
Sequenzanalusentdes’Ausgenlinsenproteinsa Kristallın „2... ....2..2..n..n 


FEER, F.: Quelques observations ethologiques sur ’Hydropote de Chine, Aydropotes inermis 
(Swinhoe, 1870), en captivite. — Some ethological observations on Aydropotes inermis 
(Swinhoe, 1870) in captivity. — Einige ethologische Beobachtungen am chinesischen Wasser- 
zehW1yaropotesinermis(Swinhoe, 1870), ın Gefangenschaft .. ..........unn.. 


FELDHAMER, G. A.; Dixon, K. R.; CHAaPMman, J. A.: Home range and movement of sıka deer 
(Cervus nıppon) in Maryland. — Reviergröße und Ortswechsel beim Sika-Hirsch (Cervus 
ERoR) a Marylarel er re ee 


GAMPERL, ROSWITHA: Die Chromosomen von Microtus arvalıs (Rodentia, Microtinae). — 
Chromosomal studies in Microtus arvalıs (Rodentia, Microtinae) .. 2... 22222200. 


GEMMEKE, H.; NIETHAMMER, J.: Zur Charakterisierung der Waldmäuse (Apodemus) Nepals. - 
©nirhercharacterisation ot fieldmice (Apodemus)from Nepal ... .............. 


GIAGIA, Eva; SavıG, I.; SOLDATOVIG, B.: Chromosomal forms of the mole rat Microspalax from 
Greece and Turkey. - Chromosomenformen von Blindmäusen (Microspalax) aus Griechen- 


sine! uma| eler Türken er en a a ER Re 2 REES ERGO 


JOHANNESSON-GROSS, KRISTINA; Gross, H.: Lernversuche mit Maulwürfen (Talpa enuropaea 
L.) unter Anwendung einer speziellen Labyrinthmethode. — Learning experiments with 
molesi@izlpzeuropaeal:.)applyingaspecialmazemethod . ... 2... zy.n.nom.. 


Kos, A.: Putzen und Putzverhalten bei Rhinolophus ferrumegquinum. — Cleaning and the 
conductioreleaninsun"Rhinolophusferrumequinum‘ „2... .. nun u..er.u...e 


KÜRTEN, L.; SCHMIDT, U.: Die Nasengruben der Vampirfledermaus Desmodus rotundus: 
Sinnesorgane zur Wahrnehmung von Wärmestrahlung. -— The nose pits of the vampire bat 
IDesızodusmorumdussisenseorganstorinfrared recepuon ... 2... .nouo.onenune: 


MAIER, W.; ALONSO, CARMEN; LANGGUTH, A.: Field observations on Callıthrix jacchus jacchns 
ae eldbeobachtunsen/an @allithnixjacchusjacchuslE. >. 2... 2. 2..n.2uu.e:n 


MARrTYys, M.: Gehegebeobachtungen zur Geburts- und Reproduktionsbiologie des Europäischen 
Wildschweines (Sus scrofa L.). - Observations on parturition and reproductive biology in 
KaptyesBuropeanwild boarsi(Sasserofal.) 22. 2.2.2.0 mn Sa... 

NADLER, C. F.; HorrMann, R. S.; VoRoNTsov, N. N.; Korprpr, J. W.; DEUTSCH, L.; 
SUKERNIK, R. 1.: Evolution in ground squirrels. Part II. - Evolution der Erdhörnchen. Teil II 

Near, B. R.: Reproductive ecology of the Rufous elephant-shrew, Elephantulus rufescens 
(Macroscelididae), in Kenya. - Die Fortpflanzungsökologie der Elefantenspitzmaus, Zle- 
IPreizealusıuugescens,Macroseelididae), ınKenia 0020... 

Near, B. R.: Reproductive biology of three species of gerbils (Genus Tatera) in East Africa. — 
Fortpflanzungsbiologie von drei Gerbilliden-Arten (Gattung Tatera) in Ostafrika .. ..... 


Nerson, B. B.; Chapman, J. A.: Age determination and population characteristics of red foxes 
from Maryland. - Altersbestimmung und Populationsmerkmale beim Rotfuchs in Maryland 


18 


12 


2415 


246 


22. 


370 


257 


175 


231 


ZI, 


72 


193 


934 


100 


198 


65 


287 


296 


PETERS, G.: A note on the vocal behaviour of the Giant panda, Arluropoda melanolenca (David, 
1869). - Beobachtungen zum Lautgebungsverhalten des Großen Panda, Arluropoda melano- 
leuca (Davıd; 1869)... 32%, oa re u ee 


PETERSEN, SUSANNE; BORN, E. W.: Age determination of the Atlantic walrus. Odobenus 
rosmarus rosmarus (Linnaeus), by means of mandibular growth layers. — Altersbestimmung 
beim Atlantischen Walroß, Odobenus rosmarus rosmarns (Linnaeus) durch Zuwachslinien 
der. Kiefern. u, scene la ae ne ee re eh ar Se N 


PULLIAINEN, E.: Scent-marking ın the pine marten (Martes martesL.) ın Finnish Forest Lapland 
in winter. — Duftmarkierung des Baummarders (Martes martes L.) im Wald-Lappland 
Finnlandsim Winter. „uns. 2 0 2 ee u N 


PULLIAINEN, E.: Habitat selection and fluctuations in numbers in a population of the arctic hare 
(Lepus timidus) on a subarctic fell in Finnish Forest Lapland. - Zur Selektion des Habitats 
und zur Fluktuation des Bestandes von Schneehasen (Lepus timıidus) auf einem subarktischen 
Ejällim Waldlappland Einnlands 77 2.2.2 


RITTNER, M.; SCHMIDT, U.: The influence of the sexual cycle on the olfactory sensitivity of wild 
female house mice (Mus musculus domesticus). - Die Beeinflussung des Riechvermögens 
durch den Sexualzyklus beı weiblichen Hausmäusen (Mus musculus domesticus) ....... 


ROTHER, G.; SCHMIDT, U.: Der Einfluß visueller Information auf die Echoortung bei Phyllosto- 
mus discolor (Chiroptera). -— The influence of visual information on echolocation in Phyllo- 
stomus.dıscolor (Chiroptera) . :. 2... 0.000 0 


RowE-ROWE, D. T.; MEESTER, J.: Population dynamics of small mammals in the Drakensberg of 
Natal, South Afrıca. - Populationsdynamik kleiner Säugetiere in den Drakensbergen von 


Natal, Südafrika ......-.. 0. 00 


SOROKER, VICTORIA; HELLWING, $.; TERKEL, J.: Parental behaviour in male and virgin white- 
toothed shrews Crocıdura russula monacha (Soricidae, Insectivora). — Elterliches Verhalten 
von Männchen und Jungweibchen bei der Spitzmaus Crocidura russula monacha (Soricidae, 
Insectivora). .. ee. la see a Le 


SCHÄFER, M. W.: Analyse des Explorationsverhaltens verschiedener Stämme und Generationen 
der Hausmaus Mus musculus domesticus Rutty. Teil I. - Analysis of exploratory behaviour 
in different strains and generations of the house mouse (Mus musculus domesticus Rutty). 
Part I’. 22029 Su N ON ee 


SCHÄFER, M. W.: Analyse des Explorationsverhaltens verschiedener Stämme und Generationen 
der Hausmaus (Mus musculus domesticus Rutty). Teil II. - Analysıs of exploratory beha- 
vıour in different strains and generations of the house mouse (Mus musculus domesticns 
Rutty). Part. I 2... a8 et a2 ee 


SCHLIEMANN, H.: Nachweis wilder Meerschweinchen für Ecuador und Beschreibung von Cavıa 
aperea patzelti subsp. nov. — A first record of Wild cavies for Ecuador and description of 
Caviasapereapatzeltisubsp:nov.'... Dun ana... ae Se 

SCHMIDT, CHRISTEL: Behavioural and neurophysiological studies of the olfactory sensitivity in 
the albino mouse. — Verhaltenspsychologische und elektrophysiologische Untersuchungen 
zur Riechfähigkeitvon Albinomäusen 2... rn u... a 


SCHMIDT, CHRISTEL; SCHMIDT, U.: Olfaction in pregnant und lactating albino mice. - Untersu- 
chungen zur Riechfähigkeit trächtiger und laktierender Albinomäuse ..... 2... ..... 


SCHMIDT, U.; JOERMAN, G.; SCHMIDT, CHRISTEL: Struktur und Variabilität der Verlassenheits- 
laute juveniler Vampirfledermäuse (Desmodus rotundus). - Structure and variability of the 
isolation calls of juvenile vampire bats (Desmodus rotundus) ...... 2... 22222200. 

STARCK, D.; PODUSCHKA, W.: Über die Ventraldrüse von Solenodon paradoxus Brandt, 1833 
(Mammalia: Insectivora). - On the ventral gland field of Solenodon paradoxus Brandt, 1833 
(Mammalia: Insectivora) .... 2.2.2000. ee re a 


STEPHAN, H.; DIETERLEN, F.: Relative brain size in Muridae with special reference to Colomys 
gosling!. — Relative Hirngröße bei Muriden mit besonderer Berücksichtigung von Colomys 


goslingt. = u: N. Sean ee ana eo 


STEPHAN, H.; Kuhn, H.-J.: The brain of Micropotamogale lamottei Heim de Balsac, 1954. - Das 
Gehirn von Micropotamogale lamotteiHeim de Balsac, 1954 ..... 2.2.2.2... 0.. 
Voss, R. S.; SıLva L., J. L.; VaLdes L., J. A.: Feeding behavior and diets of Neotropical water 
rats, genus /chthyomys Thomas, 1893. — Ernährungsweise und Nahrung der neotropischen 
Wasserratte, Genus Ichthyomys Thomas, 1893 77. nn re 


WENDLAND, V.: Häufigkeitsschwankungen der Rötelmaus (Cl/ethrionomys glareolus Schreber, 
1780). — Fluctuations of Bank vole (Clethrionomys glareolusSchreber, 1780) ......... 


236 


53 


il 


168 


47 


324 


347 


92 


149 


220 


79 


162 


50 


143 


Wissenschaftliche Kurzmitteilungen 


BARTOS, L.: A note on the sexual behaviour in red deer hind. — Bemerkungen über das 
Seatalverhaltentemer-Rothirsschkub@at ee. 


GAMPERL, ROSWITHA: Tandem-Fusion als chromosomaler Evolutionsmechanismus bei Microtus 
agrestis (Rodentia, Microtinae). -— Tandem fusion in chromosomal evolution of Microtus 
Raise WNodentiaViierouinae)e - me. ee ee ne: 


GERAETS, ADELHEID: Magendrüsenlänge und -dichte bei Crocidura russula und Suncus etruscus 
(Mammalıa: Soricidae). - Length and density of stomach glands in Crocidura russula and 
Smergeripisenhs(lammalıarSorieidae) er... ne, 

Kock, D.; KınzELBacH, R.: Der Dachs, Meles meles (Linnaeus, 1758), in NW-Syrien. — The 
BadseaMelessmeles (Eınnaeus, 1758), ım NW-Syria ..... un. 2.un wu sslun.n. 

MAcEDo, H. DE: Note on Vıicuna X Alpaca hybrids. -— Bemerkung über Hybriden von 
Yıeuma X Alselka el as or a SE 

NIETHAMMER, ]J.: Mausschläfer (Myomimus Ognev, 1924) in Afghanistan. — Mouse-tailed 
dommeusei@Myomimus©gnev, 1924)ıın Afghanistan . .. 2... 22 .me. 2 20n ek 

ROOKMAARER, L. C.: The type localıty of the Javan rhinoceros (Rhinoceros sondaıcus Desma- 
rest, 1822). - Die Terra typica des Java-Nashorns (Rhinoceros sondaıcus Desmarest, 1822) 

SPITZENBERGER, FRIEDERIKE: Der Riesenabendsegler (Nyctalus lasiopterus) ın Afrıka nachgewie- 
sen. - Giant Noctule (Nyctalus lasiopterus) found in Africa . .... 2 22222. 

SCHLOSSER-STURM, EVELYN: Zur Funktion und Bedeutung des sekundären Schultergelenks der 
Microchiropteren. - On the function and significance of the secondary articulation in the 
shonldesgolngonN\ıerochiropteran.n. un nn. a ee ee 

SCHMIDT, CHRISTEL; SCHMIDT, U.: Der Einfluß der Domestikation auf die Riechleistung der 
Hausmaus (Mus musculus). — The influence of domestication on the olfactory performance 
OISEHEIDOHSEHMOUSENLMISHTUSCHlUS) - . © nee ne 


SCHMIDT, U.; MAnskE, U.: Thermopräferenz bei der Gemeinen Vampirfledermaus (Desmodus 
rotundus). - Thermopreference in the common vampire bat (Desmodus rotundus) ....... 


Bekanntmachungen 


SEE 5 05 a ee N er a Ee 120, 


SCHEN >» 0 00 0.0.0, vi a a EEE re ee er 6202119052365 


185 


517 


113 


316 


11172 


187 


381 


145 


253 


251 


383 


384 


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 
Vervielfältigung auf photomechanischem oder ähnlichem Wege oder im Magnettonverfahren sowie der Speicherung in Datenverarbei- 
tungsanlagen, bleiben, auch bei nur auszugsweiser Verwertung, vorbehalten. Werden von einzelnen Beiträgen oder Beitragsteilen 
einzelne Vervielfältigungsstücke in dem nach $ 54 Abs. 1 UrhG zulässigen Umfang für gewerbliche Zwecke hergestellt, ıst dafür eine 
Vergütung gemäß den gleichlautenden Gesamtverträgen zwischen der Verwertungsgesellschaft Wort, vereinigt mit der Verwertungsge- 
sellschaft Wissenschaft GmbH, rechtsfähiger Verein kraft Verleihung, Goethestr. 49, D-8000 München 2, und dem Bundesverband der 
Deutschen Industrie e.V., dem Gesamtverband der Versicherungswirtschaft e.V., dem Bundesverband Deutscher Banken e.V., dem 
Deutschen Sparkassen- und Giroverband und dem Verband der Privaten Bausparkassen e. V., an die Verwertungsgesellschaft zu 
entrichten. Erfolgt die Entrichtung der Gebühren durch Wertmarken der Verwertungsgesellschaft, so ist für jedes vervielfältigte Blatt 
eine Marke im Wert von 0,40 DM zu verwenden. Die Vervielfältigungen sind mit einem Vermerk über die Quelle und den Vervielfältiger 
zu versehen. 


© 1982 Verlag Paul Parey, Hamburg und Berlin. Anschriften: Spitalerstraße 12, D-2000 Hamburg 1; Lindenstraße 4447, 
D-1000 Berlin 61. Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., Heide/Holstein 


| ISSN 0044-3468 / InterCode: ZSAEA7 47 (1-6) 1-384 (1982) | 


C21274F 


ol. 47 (1), 1-64, Februar 1982 ISSN 0044-3468 
Mommd$ 247 


ZEITSCHRIFT FÜR 


rgan der Deutschen Gesellschaft für Säugetierkunde 


ırck, D.; Poduschka, W.: Über die Ventraldrüse von Solenodon paradoxus Brandt, 1833 (Mammalia: 


Insectivora). - On the ventral gland field of Solenodon paradoxus Brandt, 1833 (Mammalia: Insectivora) 1 
rnard, R. T. F.: Female reproductive cycle of Nycteris thebaica (Microchiroptera) from Natal, South 
Africa. — Weiblicher Fortpflanzungszyklus von Nycteris thebaica (Microchiroptera) aus Natal, Südafrika 112 
vani, R.: Feeding ecology of Tadarida aegyptiaca thomasi in the Indian desert. - Nahrungsökologie von 
Tadarida aegyptiaca thomasi aus der indischen Wüste 18 


yekhuizen, S.; Maaskamp, F.: Movement, home range and clustering in the European hare (Lepus 
suropaeus Pallas) in the Netherlands. — Ortsveränderungen, Aktionsräume und Gruppierungen beim 


=eldhasen (Lepus europaeus Pallas) in den Niederlanden 22 
mmeke, H.; Niethammer, J.: Zur Charakterisierung der Waldmäuse (Apodemus) Nepals. — On the 
>haracterisation of field mice (Apodemus) from Nepal 33 
ıphan, H.; Dieterlen, F.: Relative brain size in Muridae with special reference to Colomys goslingi. — 
2elative Hirngröße bei Muriden mit besonderer Berücksichtigung von Colomys goslingi 38 


ner, M.; Schmidt, U.: The influence of the sexual cycle on the olfactory sensitivity of wild female house 
nice (Mus musculus domesticus). —- Die Beeinflussung des Riechvermögens durch den Sexualzyklus bei 


weiblichen Hausmäusen (Mus musculus domesticus) 47 
ımidt, Christel; Schmidt, U.: Olfaction in pregnant and lactating albino mice. — Untersuchungen zur 
2iechfähigkeit trächtiger und laktierender Albinomäuse 50 


tersen, Susanne; Born, E. W.: Age determination of the Atlantic walrus, Odobenus rosmarus rosmarus 
Linnaeus), by means of mandibular growth layers. — Altersbestimmung beim Atlantischen Walroß, 


Odobenus rosmarus rosmarus (Linnaeus) durch Zuwachslinien der Kiefern 55 
Shbesprechungen 62 
\ f 


'erlag Paul Parey Hamburg und Berlin 


HERAUSGEBER/EDITORS 


P. J. H. van BreeE, Amsterdam - W. FIEDLER, Wien -— H. Frick, München - W. HERRE, 

Kiel - H.-G. Krös, Berlin - H.-J. Kuhn, Göttingen - B. Lanza, Florenz - T. C. S. 

MoRrRISON-SCOTT, London - J. NIETHAMMER, Bonn - H. REIcHSTEIN, Kiel - M. Rönrs, 

Hannover - D. STARcK, Frankfurt a. M. — F. Strauss, Bern - E. ThenIus, Wien - 
W.VERHEYEN, Antwerpen 


SCHRIFTLEITUNG/EDITORIAL OFFICE 


H. SCHLIEMANN, Hamburg - D. Kruska, Hannover 


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. Verantwortli- 
cher Schriftleiter im Sinne des Hamburgischen Pressegesetzes ist Prof. Dr. Harald Schliemann. 


Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z.Hd. Priv.-Doz. Dr. 
Dieter Kruska, Institut für Zoologie, Tierärztliche Hochschule Hannover, Bünteweg 17, D-3000 
Hannover 71. Für die Publikation vorgesehene Manuskripte sollen gemäß den ‚„Redaktionellen 
Richtlinien‘ abgefaßt werden. Diese Richtlinien sind in deutscher Sprache Bd. 43, H. 1 und in 
englischer Sprache Bd. 43, H. 2 beigefügt; in ihnen finden sich weitere Hinweise zur Annahme von 
Manuskripten, Bedingungen für die Veröffentlichung 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 
Vervielfältigung auf photomechanischem oder ähnlichem Wege oder im Magnettonverfahren 
sowie der Speicherung in Datenverarbeitungsanlagen, bleiben, auch bei nur auszugsweiser Ver- 
wertung, vorbehalten. Werden von einzelnen Beiträgen oder Beitragsteilen einzelne Vervielfälti- 
gungsstücke in dem nach $54 Absatz 1 UrhG zulässigen Umfang für gewerbliche Zwecke 
hergestellt, ist dafür eine Vergütung gemäß den gleichlautenden Gesamtverträgen zwischen der 
Verwertungsgesellschaft Wort, vereinigt mit der Verwertungsgesellschaft Wissenschaft GmbH, 
rechtsfähiger Verein kraft Verleihung, Goethestr. 49, D-8000 München 2, und dem Bundesver- 
band der Deutschen Industrie e.V., dem Gesamtverband der Versicherungswirtschaft e.V., dem 
Bundesverband deutscher Banken e.V., dem Deutschen Sparkassen- und Giroverband und dem 
Verband der Privaten Bausparkassen e.V., an die Verwertungsgesellschaft zu entrichten. Erfolgt 
die Entrichtung der Gebühren durch Wertmarken der Verwertungsgesellschaft, so ist für jedes 
vervielfältigte Blatt eine Marke im Werte von 0,40 DM zu verwenden. Die Vervielfältigungen sind 
mit einem Vermerk über die Quelle und den Vervielfältiger zu versehen. 


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 internal use 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 copyıng 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 


© 1982 Verlag Paul Parey, Hamburg und Berlin. — Printed in Germany by Westholsteinische Verlagsdruckerei 
Boyens & Co., Heide/Holst. 


Z. Säugetierkunde 47 (1982) 1, 1-64 


InterCode: ZSAEA 7 47 (1), 1-64 (1982) 


Über die Ventraldrüse von Solenodon paradoxus Brandt, 1833 
(Mammalia: Insectivora)! 


Von D. STARcCK und W. PODUSCHKA 


Eingang des Ms. 15. 10. 1981 
Abstract 


On the ventral gland field of Solenodon paradoxus Brandt, 1833 (Mammalia: Insectivora) 


Studied was the ventral gland field in Solenodon paradoxus. The investigation yielded the following 
results: 

A distinct sex dimorphism is evident. The gland field of the dd forms an extended unpaired ventral 
gland which cannot be homologized to the lateral glands of the Soricidae. The extension of the gland 
field shows individual variations in form and dimension. Females have a small gland field located 
cranially. 

The histological investigation of an active ventral gland of a male showed tubular glands, the 
inidividual tubules are separated from each other by adipous tissue septa. These glands are typical 
apocrine glands with myoepithelium and plasmatic cell processes. In the gland field mentioned there is 
no increase in sebaceous glands. Contrary to MOHR’s opinion, the secretion is not continous, but 
indicates two activity phases during the annual rhythm (January to March and May to October). 
Earlier statements about colour and oily character of the secretion cannot be verified: The secretion is 
always serous, weak and nearly always colourless, of rather weak odour. 

Continued observation of captive specimens indicated the involvement of the ventral gland field in 
contact and communication behaviour. Possible relations to reproductive behaviour need further 
investigation. 


Einleitung 


Hautdrüsen sind ein echtes Schlüsselmerkmal der Säugetiere. Sie haben bereits bei basalen 
Gruppen eine bedeutende Mannigfaltigkeit erfahren und spielen in verschiedenen Funk- 
tionskreisen (Territorial-, Sozial-, Sexual-, Abwehrverhalten, Kommunikation) eine erheb- 
liche Rolle. Als Milchdrüsen gehören sie essentiell zur Organisationsform „Säugetier“. Die 
Ausbildung von Hautdrüsen ıst an bestimmte Eigenschaften des Integuments gebunden. 
Die Haut ist weich, behaart, verschieblich und in der Regel mäßig verhornt. Offensichtlich 
sind die Hautdrüsen der Mammalia, die sich bei aller Vielgestaltigkeit auf wenige Grund- 
formen zurückführen lassen, erst im Säugerstamm oder bei den unmittelbaren Ahnenfor- 
men entstanden. Eine Ableitung der Säuger-Hautdrüsen von ganz bestimmten Drüsenfor- 
men bei rezenten niederen Vertebraten kann heute nicht mehr ernstlich diskutiert werden. 

Vorkommen und Differenzierung der Hautdrüsen bei rezenten Formen läßt darauf 
schließen, daß die Funktion als olfaktorische Signalgeber primär eine dominierende Rolle 
spielt. Ancestrale Säuger waren klein, mikrophthalm und lebten nocturn. Als Makros- 
maten besitzen basale Säugetiere eine kennzeichnende Ausbildung der Nasenhöhle, die 
gegenüber Nichtsäugern progressiv ausgestaltet ist, und eine mächtige Entfaltung der dem 
Riechsinn zugeordneten zentralnervösen Strukturen (Riechhirn). Die stammesgeschichtli- 
che Entfaltung von Hautdrüsenorganen (Duftdrüsen) muß ım Zusammenhang mit der 
Differenzierung der Rezeptoren und des nervalen Zentrums gesehen werden (STARCK 
1978, 1982). 


! Herrn Direktor Dr. RıcHharnD Faust als Dank für seinen Einsatz für Solenodon paradoxus herzlich 
zugeeignet. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4701-0001 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 1-12 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


2 D. Starck und W. Poduschka 


Unter den rezenten Eutheria verdienen die Insectivora wegen ihrer basalen Stellung und 
ihrer Lebensweise als Makrosmaten mit Dominanz des Riechhirns besonderes Interesse. 
Leider sind aber gerade in dieser Ordnung im Vergleich mit anderen Eutheria unsere 
Kenntnisse noch sehr lückenhaft (Zusammenfassung siehe BRINKMANN 1912; SCHAFFER 
1940). Erwähnt werden die Seitendrüsen der Soricidae, Analdrüsen bei Potamogale, 
Solenodon, Desmaninae, Echinosoricinae (WEBER), das Subcaudalorgan von Desmana 
(MEcKEL, BrAnDrT), die Caudaldrüse von Macroscelididae (WAGNER-SCHREBER) und 
ventrale Drüsen bei Suncus (DRYDEN und ConawAyY 1967) und Blarına (EADıE 1938). 
Angaben über Sekretion von Hautdrüsen bei Solenodon liegen seit über 100 Jahren vor 
(Pory 1861). Die entsprechenden Angaben der Autoren aus der Folgezeit sind jedoch 
ebenso unpräzise und vielfach widersprechend. Insbesondere fehlen Feststellungen über 
die genaue Lokalisation der Drüsen und über ihre Struktur. Ziel dieser Arbeit ist es, diese 
offenen Fragen zu klären und über jahrelange Beobachtungen an lebenden Schlitzrüsslern 
zu berichten. 


Material und Methode 


Langfristige Beobachtungen über das Verhalten von Solenodon konnten an Zootieren durchgeführt 

werden (1.-3., W. P:). Es handelt sich um folgende Individuen: 

1. zwei Individuen ım Zoo Frankfurt a. M., 1971-1973 (PopuscHka 1975). 

2. 18 im National Zoological Park Washington, das dort fast 11 Jahre lebte, wurde mehrfach 
untersucht. 

3. Tiere im Parque Zoolögico Nacional in Santo Domingo auf zwei Reisen nach Hispaniola 1976 und 
1978°. 

4. Ab 30. 11. 1978 untersuchte PopuscHkA im Rahmen eines Zuchtprogramms für bedrohte Insekti- 
voren zwei Pärchen von Solenodon paradoxus. Es handelt sich um Wildfänge (Oktober 1978) aus 
der Dominikanischen Republik, von denen ein Paar als adult, das andere als subadult anzusehen 
war. Nach der Zitzengröße hatte das adulte Weibchen bereits mindestens einmal geworfen. Diese 
Tiere stehen seither laufend unter protokollierter Beobachtung. Die Ergebnisse werden zu gegebe- 
ner Zeit veröffentlicht. 

Zur Haltung der Tiere sei hier vorerst folgendes berichtet: Die Tiere werden in Anbetracht ihrer 

überaus empfindlichen Sohlenhaut (MoHRr 1936-38; EISENBERG 1975; eigene Beobachtungen) paar- 

weise auf weichem, nicht staubendem Torfmull gehalten und haben reichliche Möglichkeit zum 

Klettern und zum Scharren, wie sie es auch im natürlichen Habitat tun. Außerdem stehen ihnen 

Schlafkästen mit einer hochklappbaren Seitenwand über einer Glasplatte zur Verfügung. MoHRs 

Mitteilung (1937), daß die Tiere bei Haltung in Torfmull sehr schnell verschmutzen, kann absolut 

nicht bestätigt werden. Unsere Exemplare sind sehr sauber, das Fell ist glänzend, die Haut ist trocken 

und rosig. 
Für ne anatomische und histologische Untersuchung standen Hautstücke von drei Individuen zur 

Verfügung: 

a. adultes männliches Tier, das bis zum Tod lebhafte Sekretion der ventralen Körperhaut gezeigt hatte 
(PopuscHka: d C., gestorben 6. 12. 1979 an Darmverschluß). Fic.: Formol-Alkohol-Eisessig. 

b. adultes Männchen (KR: 270, Schw: 170) Museum Hamburg (T 26), Alkohol. Loc.: Sabana de la 
mar, Hispaniola, 1939. 

c. adultes Weibchen. Altes Sammlungsmaterial. Alkohol. 

Die mikroskopische Untersuchung der Hautdrüsen wurde im wesentlichen an Tier a durchgeführt. 

Ein ventrales Hautfeld von 10 cm cranıo-caudaler Länge und 8 cm Breite wurde in der Längsrichtung 

halbiert. Die rechte Hälfte wurde in 6 Blöcke zerlegt, die alternierend in Celloidin und Paraffın 

eingebettet wurden. Die Schnittdicke betrug 10 und 15 um. Färbung: Haematoxylın-Fosin, Azan, 

Resorcinfuchsin. 


* Die beiden Reisen nach Hispaniola dienten der Untersuchung der noch lebenden Bestände dieser 
äußerst bedrohten Tierart. Über die Ergebnisse wochenlanger Freilandbeobachtungen (PopDuscHKA 
1976) ist eine weitere, ausführliche Publikation in Vorbereitung. 


Über die Ventraldrüse von Solenodon paradoxus Brandt, 1833 3 


Bisherige Angaben über Hautdrüsensekretion bei Solenodon 


Pory (1861) erwähnt in seiner Arbeit über Solenodon cubanus (= Atopogale cubana) die 
Mitteilung eines Teniente D. RAFAEL ZENEA, der einige Solenodones kurze Zeit („‚poco 
tiempo“) lebend hielt, daß diese „aus dem Fell häufig eine ölige, rötliche, stark riechende 
Flüssigkeit abgesondert hätten“. Angaben über Geschlecht, Alter und Lokalisation der 
Sekretbildung fehlen. BREHM übernimmt diese Aussage in der 3. (1895) und 4. (1912) 
Auflage. ALı£n (1908) hat die vierfach vergrößerte Photographie einer „glandular surface 
of the left thigh“ veröffentlicht, diese aber im Text nicht erwähnt. 

MonHr (1936) schreibt von „Seitendrüsen‘“ bei Solenodon paradoxus. Nach ihren 
Angaben wären Achselhöhlen und Bauchseiten der Tiere stets feucht und die Haare 
teilweise miteinander verbacken. Die Farbe des Sekretes soll olivgrün und von der des 
Urins (gelblich) sehr verschieden sein. Die ganze Bauchseite des Männchens sei ständig 
grün, schmiere und färbe stark ab. Bei den Weibchen sei die Sekretbildung deutlich 
schwächer als bei den Männchen, das Sekret zeige eine mehr bräunliche Farbe. 

HERTER (1967) gibt für Schlitzrüssler diagnostisch an: „Seitendrüsen in Achsel- und 
Leistengegend.‘“ Im übrigen werden die Angaben von MORHR zitiert. 

Tısskens (1967) berichtet über die Solenodones, die 1966-1968 im Zoo Antwerpen 
gehalten wurden und erwähnt „sur les hanches et aux aiselles apparaissent des glandes 
epidermiques dont la secretion repand une odeur musc“. 

EISENBERG (1975) erwähnt ausgedehnte Drüsenfelder am Bauch, in den Achselhöhlen 
und an den Flanken, gibt aber keine weiteren eigenen Beobachtungen an. 

Es ist erstaunlich, daß bisher keinerlei morphologische und histologische Untersuchun- 
gen der Hautdrüsen bei Schlitzrüsslern durchgeführt wurden, obgleich auf die intensive 
Sekretbildung mehrfach hingewiesen wurde und diese basalen Reliktformen mit Recht 
biologisch und morphologisch großes Interesse beanspruchen dürfen. Aus der Zusammen- 
stellung der Angaben im Schrifttum ergeben sich Unstimmigkeiten und Widersprüche. 
Diese betreffen zum ersten die Färbung des Sekretes. Nach Pory (1961, Beobachtung von 
ZENEA) soll das Sekret rötlich (rojizo) sein. MoHR (1936) beschreibt das Sekret auf Grund 
eigener Beobachtungen als „olivengrün“ (RıngEway 1912 „yellowish Oil Green‘). Diese 
Unterschiede können artspezifisch sein, denn Pory berichtet über kubanısche, MOHRr über 
dominikanische Schlitzrüssler. Nachprüfung an kubanischen Tieren ıst kaum zu erwarten, 
da die Art, die lange Zeit überhaupt als erloschen galt, kaum zur Beobachtung kommt. 

Zum zweiten bestehen Unstimmigkeiten über die Lokalisation der Drüsen. PoEy 
schreibt nur generell von Durchfeuchtung des Felles („de su piel“). MOoHR spricht 
einerseits von „Seitendrüsen“, andrerseits wenige Zeilen weiter von Sekret an Achselhöh- 
len und Bauchseite. Da anatomische Untersuchungen nicht durchgeführt wurden, dürfte 
die Autorin in Hinblick auf die mehrfach beschriebenen Seitendrüsen bei Soriciden 
terminologisch einen Analogieschluß gezogen haben. 

Sehr bemerkenswert sind indessen MoHrs Verhaltensbeobachtungen, nach denen das 
Männchen bei der Werbung, seltener auch das Weibchen (!), sich mit den Vorderfüßen 
weiterstemmend, auf dem Bauch um den Partner herumrutschte oder neben ihm entlang- 
rutschte. Beide Tiere zeigten bei der Kontaktaufnahme mit dem Partner durch Unter- oder 
Überkriechen ein Verhalten, das auch wir häufig beobachten konnten. GouLD (1969) 
beschreibt das gleiche von Blarina brevicanda. EISENBERG (1975) berichtet von Solenodon: 
„An adult d will generally explore and scent mark by dragging the anogential area (!) over 
protuberances in the cage.““ Ob es sich hier um eine Markierung gehandelt hat, ist schwer 
zu entscheiden. An eine Beteiligung der Ventraldrüsen ist jedenfalls in diesem Zusammen- 
hang zu denken. Ein intensives Gleiten mit der Pectoralregion — möglicherweise auch mit 
der Bauchregion - haben wir bei Tenrecinae (Setifer setosus und Echinops telfairi) unter 
bestimmten Voraussetzungen beobachten können (PODuscHkA, unveröffentlicht). 


4 D. Starck und W. Poduschka 


Ergebnisse 


Eigene Beobachtungen über die Sekretion an der Bauchseite bei dd von 
Solenodon paradoxus 


Ab Anfang Januar 1979 zeigten sich beı beiden dd scharf abgegrenzte, etwa oval geformte 
Sekretionsfelder (10 zu 8cm) ım Bauch (Abb. 1a, b), die dadurch besonders deutlich 
sichtbar sind, daß die Tiere auf sehr weichem Torfmull gehalten werden, der sich mit der 
reichlich austretenden farblosen Flüssigkeit vermengt und an den Bauchhaaren haften 
bleibt. Dadurch wird beim hochgehobenen Tier die Ausdehnung des Drüsenfeldes gut 
sichtbar. Form und Ausdehnung des Drüsenfeldes zeigen gewisse individuelle Unter- 
schiede (Abb. 1a, b). Bei Berührung mit dem Finger spürt man eine dünnflüssige, 
keineswegs ölige Nässe. Die abgestrichene Flüssigkeit ist wasserhell. Ein einziges Mal (am 
17./18. 1. 1980) erschien das Sekret leicht grünlich. Die Prüfung auf den pH-Wert mit 
Universal-Indikatorpapier (MERcK) ergab einen Wert von 5-6 (dieser Wert kann durch 
Säure des Torfmulls beeinflußt sein). 


Abb. 1. Ausdehnung des sezernierenden, ventralen Drüsenfeldes bei 2 männlichen Solenodon parado- 
xus. Links: Männchen „Domingo“, rechts: Männchen „Chico“. Beachte die individuelle Variabilität 
ın Form und Ausdehnung des Drüsenfeldes 


In den ersten Januartagen war die Sekretion bei dem jüngeren d C (= Chico) besonders 
intensiv, doch schwankte auch bei diesem Tier sowohl die Intensität als auch die zeitliche 
Verteilung der Sekretion. Im Laufe der Monate wurde von dem älteren d D die gleiche 
Intensität erreicht. Abb. 2 demonstriert das zeitliche Auftreten der Sekretion, deren 
Höhepunkte durch ein schwarzes Hoch-Rechteck markiert sind. | 

dd wirken während der Phase starker Sekretion auffallend aktiver und lebhafter als in 
den Zwischenphasen. Außerdem trinken sie in dieser Zeit besonders reichlich, wobei das 
Wasser in ungewöhnlicher Weise um die Trinkschüssel verspritzt wird. 

Umständehalber war es leider nicht möglich, die tägliche Feststellung der Sekretion und 
ihre Intensität während des gesamten Beobachtungszeitraumes von ihrem ersten Auftreten 
Anfang Januar 1979 bis Sommer 1981 ganz lückenlos durchzuführen. Es wird aber 


Über die Ventraldrüse von Solenodon paradoxus Brandt, 1833 


JANUAR 1979 FEBRUAR MARZ APRIL 
‘pE JH FE Dinner 
Sch sa FT 3 rd DT 
on EEE TEE Bemm en 
oz Be I ı 
1980 
CO an Dee n ern Dee 
ccm Br CR (ee 
om Fre Bm DE 
on EEE mama Dane DrrTz 
MAI 1979 JUNI JULI AUGUST 
oT DT I [Deu 
she CE I He [1 
om TEN Free Ve Fr 
on BEmmıem aA nn 
1980 
KH mm IE u am PER 
Ce CZ a 5 ı 8 ] 
on Va Tr Tr 
a a | ji EEE | ze] 
SEPTEMBER 1979 OKTOBER NOVEMBER DEZEMBER 
ou Dee 04 re Werner 
Po) m mE Ce m m Tre En Be ER: = 
vn 2 | TE EEE 
oz rn Bm J 
1980 
co Ca EEE | DE | | 
oz Wi; | 
oe MI IIT De © 2 
eo  [> 10 1. KmER a 


Abb. 2. Saisonale Schwankungen der Sekretionsaktivität bei zwei Männchen und zwei Weibchen von 
Solenodon paradoxus im Jahresablauf. Phasen hoher Aktivität sind durch hohe, schwarze Rechtecke 
markiert 


deutlich, daß die Tiere keineswegs immer feucht sind, wie dies MOHR (1936) angibt. Weiter 
ist bedauerlich, daß durch den Tod des jüngeren d C die vergleichende Gegenüberstellung 
beider Tiere nur bis Anfang September 1979 möglich war. 

Abbildung2 ergibt aber deutlich, daß bei beiden Tieren zwei Sekretions-Zyklen vorka- 
men, die für beide Individuen annähernd zeitlich zusammenfallen, wenn auch geringfügige 
individuelle Verschiebungen und Intensitätsunterschiede festzustellen sind. Die eine stär- 
kere Sekretionsphase fällt in den Zeitraum von Mai bis Oktober, die zweite Phase reicht 
von Januar bis März. Ruhephasen lagen im November-Dezember und im April. Höhe- 
punkte der Sekretionsaktivität scheinen jeweils gegen Ende der Sekretionsperiode, also im 
Oktober und März vorzukommen. 

Wenn auch diese Arbeit sich zunächst mit den Hautdrüsen der männlichen Schlitzrüss- 
ler befaßt, sei doch vermerkt, daß jeweils im Frühherbst auch bei beiden ?? (M = Madre- 
cita und N = Nina) eine, wenn auch sehr viel schwächere Sekretion als bei den dd 
festzustellen war. Dies bestätigt die Angabe von MoHr (1936). Bei beiden 22 ist das 
Sekretionsareal auf ein kleines, dreieckiges Feld an der Vorderbrust konzentriert, die dort - 
wahrscheinlich durch die Sekretabsonderung - etwas rotbraun verfärbt erscheint. Saison- 
bedingte Schwankungen in der Ausbildung von Hautdrüsen und deren Sekretionsaktivität 


6 D. Starck und W. Poduschka 


sind bei Säugern weit verbreitet und deuten auf eine Beziehung zum Sexualzyklus hin. Da 
bisher keine zuverlässigen Beobachtungen zur Fortpflanzungsbiologie, besonders über 
Paarungszeiten und Graviditätsperiode von Solenodon vorliegen, sollen Aussagen zur 
biologischen Bedeutung der Sekretion zunächst vermieden werden. 


Morphologische Untersuchung der Hautdrüsen von Solenodon 


Die folgenden Ausführungen beziehen sich ausschließlich auf das Tier a(=C), da der 
gesamte Körper zur Verfügung stand und sich als ausgezeichnet fixiert erwies. Proben aus 
der Haut der Tiere b und c ergaben keine eindeutigen Resultate, Hautdrüsenkomplexe 
konnten im ventralen Rumpfbereich bei b und c nicht aufgefunden werden (Inaktivität, 
Kachexie?). Hingegen ergab die Untersuchung der ventralen Körperhaut von Tier a, das 
kurz vor seinem Tod noch Sekretionsaktivität gezeigt hatte, einen eindeutigen Befund, der 
die Aufklärung der Frage nach der Lokalisation, nach der Struktur und der morphologi- 
schen Wertung der Drüsen und damit nach der Herkunft der seit langem bekannten 
Sekretionserscheinungen von Solenodon erlaubte. 

Die Dermis der Bauchregion ist außerordentlich derb und zäh. Beim Abpräparieren 
liegt ıhr im Bereich des großen Sekretionsfeldes, das am lebenden Tier zu erkennen ist, eine 
kompakte, etwa 3-5 mm dicke gelbliche Schicht an, die eine lappıge Struktur zeigt und am 
fixierten Material wie eine relativ feste Fettgewerbsschicht aussieht. Gegen die lockere 
Subcutis mit dem Panniculus carnosus kann die Schicht leicht abgetrennt werden. Sie 
besteht (Abb. 3) aus zahlreichen, vielfach gewunden verlaufenden Schläuchen von eindeuti- 


Abb. 3. Übersichtsbild, Ventraldrüse. 
Am linken Bildrand kleine Talgdrüse 
unter dem Haarbalg 


gen a-Drüsen, die voneinander durch Fettgewebe getrennt bleiben. Diese Zwischenschicht 
zwischen den Drüsenschläuchen hat eine Dicke von durchschnittlich 5 großen Fettzellen. 
Nach den Randbezirken zu treten die Drüsenschläuche allmählich zurück, bis schließlich 
nur cutanes Fettgewebe übrigbleibt. Das Drüsenlager wird gegen die Subcutis durch eine 
sehr dünne, fascienartige Membran abgegrenzt, geht aber an der Peripherie ohne scharfe 
Grenzen ins cutane Fettgewebe über. Der ganze Drüsenkomplex besitzt also keine 
deutliche Kapsel, wie sie bei Soricidae an der Seitendrüse vorkommt (eigene Beobachtun- 
gen an Crocidura russula) und hat einen mehr diffusen Charakter. Die Seitendrüse der 
Spitzmäuse ist ein kompaktes, organartiges Gebilde, das zumindest in der Aktivitätsphase 
leicht makroskopisch präpariert werden kann. Der Drüsenkomplex von Solenodon muß als 
Ventraldrüse bezeichnet werden und ist keinesfalls der Seitendrüse beı Spitzmäusen oder 
Microtinae homolog. Beide Drüsentypen können auch genetisch kaum voneinander abge- 
leitet werden, sondern sind als gruppenspezifische Differenzierungen in verschiedenen 
Stammeslinien aufzufassen. Die Drüsen von Solenodon können nicht als Argument für die 


Über die Ventraldrüse von Solenodon paradoxus Brandt, 1833 7 


Abb. 4. Apokrine Zellkuppen in den Schläuchen der Ventraldrüse 


engere Verwandtschaft der Solenodontidae mit den Soricoidea, wie sie von MCDowELL 
(1958) angenommen wurde, angeführt werden. Seitendrüsen von Soricidae - und parallele 
Strukturen bei anderen Eutheria, z.B. Microtinae - sind abgekapselte, paarige, an den 
Flanken gelegene Organe. Der Drüsenkomplex von Solenodon ist unpaar und diffus. Das 
schließt nicht aus, daß strukturell identische Elemente ın verschiedenen lokalisierten 
Drüsenorganen vorkommen, denn alle Hautdrüsen bei Mammalıa lassen sich auf wenige 
Grundtypen (poly- und monoptyche Drüsen) zurückführen. 

Während Hautdrüsenorgane bei Säugetieren sehr häufig aus einer Kombination ver- 
schiedener Drüsentypen bestehen (Flankendrüsen der Soricidae, Antorbitaldrüsen der 
Antilopen, Temporaldrüse der Elefanten), besteht das eigentliche Drüsenorgan der Soleno- 
dontidae aus einer einzigen Drüsenart, und zwar aus a-Drüsen. Deren kennzeichnende 
Merkmale sind im Bereich des ganzen Drüsenfeldes erkennbar. Es handelt sich um 
monotypche (einschichtige) Drüsen, die weitlumige, etwas gewundene Schläuche bilden 
und deutliche Kennzeichen (Zellkuppen, Abb. 4) apokriner Sekretabgabe zeigen. 


Die Wand des ganzen 
Drüsenschlauches ist, mit 
Ausnahme der letzten 
Strecke vor der Ausmün- 
dung, mit typischen Myo- 
epithelzellen (Abb.5) be- 
setzt. Bemerkenswert ist 
das Vorkommen von zwei- 
kernigen Epithelzellen ım 
Bereich des sezernierenden 
Abschnittes der a-Drüsen 
(Abb.6). Die Ausmündung 
erfolgt stets in den Haar- 
balg, und zwar distal von 
der Mündung der ent- 
Ä sprechenden Talgdrüsen 
Abb. 5. Myoepithelien am Tangentialschnitt durch Drüsenschlauch (Abb.7). Während bei vie- 


8 D. Starck und W. Poduschka 


gs 


Abb. 6. Zweikernige Epithelzelle in a-Drüse 


einen Haarbalg 


len Säugetieren Talgdrüsen in Form dicker Pakete am Aufbau kombinierter Hautdrüsen 
beteiligt sind, finden sich bei Solenodon nur sehr spärliche Talgdrüsen (Abb.3), die 
unmittelbar an den Haarbälgen liegen und in deren tiefen Abschnitt einmünden. Sie 
nehmen am Aufbau des Drüsenkomplexes selbst kaum teil. Ihr Anteil an der Sekretbildung 
kann nur minimal sein. Melanocytenansammlungen in der Drüse, wie sie bei Antorbital- 
drüsen von Antilopen, in der Schnauzendrüse der Lagomorpha und in Scrotaldrüsen von 
Cebuella vorkommen, fehlen bei Solenodon. 


Diskussion 


Die Bedeutung der Hautdrüsen in Biologie und Stammesgeschichte der Säuger war 
eingangs hervorgehoben worden. Sie kommen in allen Säugerordnungen vor (sekundäre 
Rückbildung bei Cetacea, Sirenia, Pholidota, Bradypodidae). Ihr hohes phylogenetisches 
Alter ergibt sich aus der Tatsache, daß Drüsen gleichen Typs bei Proto-, Meta- und 
Eutheria gefunden werden. Sie erfüllen zweifellos sehr verschiedene Aufgaben in differen- 
ten Funktionskreisen und Lebensräumen. Keinesfalls ıst es zulässig, ihre Bedeutung nur im 
Dienste einer Spezialaufgabe zu sehen (Territoriumsmarkierung oder Aggression, RALLS 
1971). Die Bedeutung olfaktorischer Kommunikationssysteme (Pheromone), die zu diffe- 
renzierten Leistungen befähigt sind, ist vielfach hervorgehoben worden (SCHULTZE-WEST- 
RUM 1965; EISENBERG und KLEIMAN 1975). 

Eine systematische Ordnung der Drüsenorgane aufgrund der Feinstruktur (Form der 
sezernierenden Anteile, Art der Sekretausstoßung) reicht für eine morphologische und 
funktionelle Wertung allein nicht aus, da die Drüsenorgane oft verschiedene Grundbauele- 
mente enthalten. Histochemie der Drüsen und Biochemie der Sekrete sind noch unzurei- 


Abb. 7 (links). Endstück einer a-Drüse mit indifferent 
Epithelauskleidung unmittelbar vor der Mündung 


Über die Ventraldrüse von Solenodon paradoxus Brandt, 1833 9 


chend erforscht. Eine grobe Vergleichsbasis ergibt sich nach dem Ort der Entstehung, nach 
der Lokalisation. Für die biologische Wertung ist die ethologische Untersuchung entschei- 
dend. 

Die folgende Übersicht beschränkt sich auf Angaben über das Vorkommen von 
Ventraldrüsen im pectoralen und abdominalen Bereich bei Insectivora und einigen basalen 
Gruppen. Viele Angaben im Schrifttum beruhen auf Lebendbeobachtungen, die im 
Analogieschluß bestimmten Drüsen zugeordnet werden, ohne daß deren anatomischer 
Nachweis geführt wurde. Natürlich ergibt die Beobachtung des Verhaltens Hinweise auf 
das Vorkommen von Drüsen, doch sollte auf den präzisen, morphologischen Nachweis 
und auf die Aufklärung der Struktur nicht verzichtet werden. 

Marsupialia: Petaurus breviceps papnanus: Nachweis, auch histologisch, von Ster- 
naldrüsen neben Frontal- und Analdrüsen. Differenzierung von Individual-, Sippen- und 
Artgeruch. Die Ventraldrüse von Trichosurus (juv. Männchen) wird durch Testosteron 
aktiviert und nach Kastration rückgebildet (BoLLIGER 1944). Vorkommen von Ventraldrü- 
sen wird für Myrmecobius, Didelphis, Dendrolagus und Macropus (GAansLoSsSER 1979b) 
angegeben. 

Insectivora: Flankendrüsen (Seitendrüsen) sind bei Soricidae histologisch gut 
untersucht: Neomys fodiens, Sorex araneus, Crocidura russula (SCHAFFER), Suncus murinus 
(DryDEn, ConawaAy). Über Funktion und hormonale Abhängigkeit ist wenig bekannt. 
Der intensive Moschusduft von Suncus bleibt nach Exstirpation der Seitendrüsen erhalten 
und wird von a-Drüsen der Retroauricular- und Kehlregion verursacht. Blarina brevi- 
canda besitzt neben Seitendrüsen (a-Drüsen und Talgdrüsen) ein unpaares ventrales 
Drüsenfeld (midventral gland), das vorwiegend aus Talgdrüsen bestehen soll (EADIE 1938; 
PEARSoN 1946). Dieses ıst bei Weibchen erheblich kleiner als bei Männchen. 

Im Verhalten der Tenrecinae konnten von uns (W.P.) Situationen beobachtet 
werden, die auf das Vorkommen sternal-abdominaler Drüsenfelder hindeuten (histolog. 
Untersuchung noch nicht abgeschlossen). Setifer setosus (ähnlich auch Echinops telfaırı) 
zeigen folgendes Verhalten: Beim Versetzen in ein von Artgenossen bewohntes Areal oder 
bei der Attacke gegen einen solchen werden gelegentlich die Vorderbeine weggestreckt, so 
daß Hals- und Brustregion dem Substrat aufliegen. Gleichzeitig schiebt sich das Tier durch 
kräftige Bewegungen der Hinterbeine auf den Gegner zu. Beide Arten sind ım Freiland 
nicht territorial. Das Verhalten dürfte der Selbstbehauptung dienen. Möglicherweise liegt 
in diesem „Rutschen“ eine Parallele zu dem von MonHr und EIsEnBeErg bei Solenodon 
beschriebenen Verhalten vor. 

Über ein ventrales Drüsenfeld bei Tupaiidae, das von der Kinn- bis zur Sternal- 
gegend reicht, liegen morphologische und ethologische Untersuchungen ın größerer Zahl 
vor, auf die hier verwiesen sei (SPRANKEL 1961; v. Horst und BUERGEL-GooDwin 1975; 
v. Horst und Lesk 1975; RICHARZ und SPRANKEL 1978; SORENSON und CoNAWwAY 1966). 
Eine Zusammenstellung über das Vorkommen von Kehldrüsen beı Halbaffen bringt 
SCHILLING (1979). Der anatomische Nachweis gelang erst bei wenigen Arten. 

Unter den Simiae erwähnt Erpr£ (1967) Drüsenfelder für zahlreiche Callıtrichidae 
und Callimico, doch liegen bisher keine anatomischen Nachprüfungen vor. Das Integu- 
ment der Affen gilt im allgemeinen, von den hier nicht zu erörternden e-Drüsen abgesehen, 
als drüsenarm. Komplexe Duftdrüsenorgane kommen gelegentlich vor (Circumgenitaldrü- 
sen, Axillarorgan bei Pan, Gorilla, Homo). Hier sei auf das Vorkommen von Sternaldrüsen 
bei Ateles (WısLockı und Schutz 1925; Schwarz 1937; Krein 1971; Krein und KrEın 
1972; EisENBERG 1976), Hylobates und Pongo (A. H. SCHULTZ) verwiesen. Vermerkt sei, 
daß die Sternaldrüse von Ateles paarig ist. Der Versuch von SCHWARZ, das Organ als 
Ergebnis eines Konzentrationsprozesses in die Nähe der Milchdrüse zu rücken, bedarf 
keiner Erörterung, da es sich um ein hochspezialisiertes Organ bei evolvierten Formen 
handelt. 


Die mehrfach beschriebene Sekretion an Brust und Bauch beı Solenodon stammt aus 


10 D. Starck und W. Poduschka 


einem großen, unpaaren, ventralen Drüsenfeld. Dieses unterscheidet sich durch die 
medioventrale Lage, den Mangel einer scharfen Begrenzung und einige histologische 
Besonderheiten deutlich von den Seitendrüsen der Soricidae, Microtinae und Cricetinae. 
Nach Massenentfaltung und Intensität der Sekretion bestehen klare Sexualunterschiede, 
wie sie vielfach vorkommen (Soricidae, Tupaiidae), aber nicht allgemein für Hautdrüsen 
gelten (Retroauriculardrüse von Suncus, Kehldrüse von Ateles). Unterschiedliche Angaben 
über die Beschaffenheit des Sekretes ım Schrifttum dürften auf verschiedene Haltungsbe- 
dingungen zurückzuführen sein. Das Sekret ıst nach unseren Beobachtungen an Tieren in 
verschiedenen Zoos, an den selbst gehaltenen Tieren und in freier Wildbahn praktisch 
geruchlos. Der Beobachter (W.P.) verbrachte mehrfach die Nächte inmitten größerer 
Gruppen (6-10 Tiere) im natürlichen Lebensraum. Die Angabe von E. MoHR, daß von den 
Tieren ein übler Geruch ausgehe, bedarf der Korrektur. 

Spekulative Deutungen über Funktion und Aufgabe im Verhaltensinventar sollen 
vorerst vermieden werden, da die Anzahl der untersuchten Individuen zu gering ist und 
experimentelle Studien wegen der Seltenheit und Kostbarkeit der Tiere unmöglich sınd. 
Zur Klärung noch offener Fragen bleiben Untersuchungen zu den Beziehungen zwischen 
Drüsenaktivität und Fortpflanzungszyklus und umfassendere Studien zum Kommunika- 
tionsverhalten abzuwarten. Vor allem aber fehlt im morphologischen wie ım ethologischen 
Bereich eine genügend breite Vergleichsbasis, die alle Insectivoren umfassen sollte, denn 
diese basale Gruppe der Eutheria verdient besondere Beachtung. 

Das Vorkommen von Hautdrüsen in anderen Körpergegenden von Solenodon bedarf 
der Überprüfung. Entsprechende Untersuchungen sind im Gange. Für die Analregion von 
Solenodon gibt ORTMAnN (1960) das Vorkommen von Proctodealdrüsen, Talgdrüsen und 
a-Drüsen an. Qualitätsunterschiede der Sekrete verschiedener Drüsen und deren spezielle 
Bedeutung im Kommunikationsverhalten bedürfen weiterer Prüfung. Auf die phylogene- 
tisch bedeutsame Korrelation zwischen Ausbildung der Hautdrüsenorgane, des Geruchs- 
organs und des Riechhirns bei Makrosmaten wird verwiesen. In diesem Zusammenhang seı 
vermerkt, daß das Organon vomeronasale bei Solenodon nach MENZEL (1979) eine nur 


mäßige Ausbildung hat. 
Danksagung 


Beobachtungen und Untersuchungen an Solenodon waren nur möglich dank der Hilfe und Unterstüt- 
zung zahlreicher Personen und Institutionen. Wir haben Dank zu sagen Herrn Direktor Dr. RICHARD 
Faust, Zoologischer Garten Frankfurt a.M., für dauerndes Interesse an unserer Arbeit und für deren 
Unterstützung. SIR PETER ScoTT, IUCN/Survival Service Commission, hat das Projekt durch 
persönlichen Einsatz gefördert. In der Dominikanischen Republik unterstützten uns aktiv Dr. JAIME 
Vınas Roman, Direktor des Zoos Santo Domingo, und Lic. JOSE ALBERTO OTTENWALDER. Frau 
CHRISTL PODUSCHKA verdient unseren besonderen Dank für die aufopfernde Pflege der Tiere und 
deren tägliche Beobachtung. Herrn Prof. Dr. H. SCHLIEMANN, Hamburg, danken wir für Überlassung 
von Hautstücken aus dem Hamburger Museum sowie Frau Dipl.-Biol. G. KLAUER für die Anferti- 
gung der Photos der histologischen Schnitte. 


Zusammenfassung 


Die Untersuchung des ventralen Drüsenfeldes von Solenodon paradoxus ergab folgendes: Es besteht 
ein deutlicher Sexualdimorphismus. Das Drüsenfeld der 83 bildet eine große, unpaare Ventraldrüse, 
die nicht den Flankendrüsen (Seitendrüsen) der Soricidae homolog ist. Die Ausdehnung der Drüse 
zeigt individuelle Differenzen nach Form und Dimension. ?? besitzen ein kleines, cranıal gelegenes 
Drüsenfeld. Histologisch besteht die aktive Ventraldrüse der 83 aus Schlauchdrüsen, deren einzelne 
Schläuche durch Fettgewerbssepten getrennt sind. Diese Drüsen sind typische apokrine Drüsen mit 
Myoepithel und Zellkuppen. Talgdrüsen sind im Bereich des Drüsenfeldes nicht vermehrt. Die 
Sekretion erfolgt nicht - wie MoHR annahm - dauernd, sondern zeigt zwei Aktivitätsphasen im 
Jahreszyklus (Januar-März und Mai-Oktober). Angaben über Färbung und öligen Charakter des 
Sekretes können nicht bestätigt werden. Das Sekret ist in allen Phasen wasserklar, fast immer farblos 
und dünnflüssig, von sehr schwachem Geruch. Beobachtung der gehaltenen Tiere weist darauf hin, 
daß die Ventraldrüse eine Rolle bei der Kontaktaufnahme und im Kommunikationsverhalten spielt. 
Beziehungen zum Fortpflanzungsverhalten bedürfen weiterer Untersuchung. 


Über die Ventraldrüse von Solenodon paradoxus Brandt, 1833 11 


Literatur 


Auen, J. A. (1908): Notes on Solenodon paradoxus Brandt. Bull. Am. Mus. Nat. Hist. 34, 505-517. 

BOLLINGER, A. (1944): The response of the sternal integument of Trichosurus vulpecula to castration 
and to sex hormones. J. Proc. Roy. N. S. Wales 78, 234-238. 

BREHM, A. (1893, 1912): Brehms Tierleben, 3. und 4. Aufl., Leipzig und Wien. 

BRINKMANN, A. (1912): Die Hautdrüsen der Säugetiere (Bau und Sekretionsverhältnisse). Erg. Anat. 
Entwgesch. 20, 1173-1231. 

DryDen, G. L.; Conaway, H. C. (1967): The origin and hormonal control of scent production in 
Suncus murinus. J. Mammalogy 48, 420-428. 

EADIE, W. R. (1938): The dermal glands of shrews. J. Mammalogy 19, 171-174. 

EISENBERG, ]J. F. (1975): Tenrecs and solenodons in captivity. Int. Zoo Yearb. 15, 6-12. 

— (1976): Communication mechanisms and social integration in the black spider monkey, Ateles 
fusciceps robustus, and related species. Smithson. Contr. Zool. 213, 1-108. 

EISENBERG, ]J. F.; KLEıman, D. G. (1972): Olfactory Communication in Mammals. Ann. Rev. Ecol. 
System. 3, 1-32. 

EppL£, G. (1967): Vergleichende Untersuchungen über Sexual- und Sozialverhalten der Krallenaffen 
(Hapalidae). Folia primat. 7, 37-65. 

— (1973): Social communication by olfactory signals in marmosets. Int. Zoo Yearb. 13, 36-42. 

GANSLOSSER, U. (1979a): Soziale Interaktionen des Doria-Baumkänguruhs (Dendrolagus dorianus 
Ramsay, 1833). Z. Säugetierkunde 44, 1-18. 

— (1979b): Soziale Kommunikation, Gruppenleben, Spiel und Jugendverhalten des Doria-Baumkän- 
guruhs (Dendrolagus dorianus Ramsay, 1833). Z. Säugetierkunde 44, 137-153. 

Gouıp, E. (1969): Communication ın three genera of Shrews (Soricidae): Suncus, Blarina and 
Cryptotis. Comm. Behav. Biol., Part A, 3, 263-313. 

HERTER, K. (1967): Die Insektenesser. Grzimeks Tierleben. Zürich: Kindler-Verlag. 10, 183-263. 

Hotst, D. v.; BUERGEL-GooDwIn, U. (1975a): The Influence of Sex-hormones on Chinning by Male 
Tupaia belangeri. J. Comp. Physiol. 103, 123-151. 

— (1975b): Chinning by Male Tupaia belangerı: The Effects of Scent Marking of Conspecies and 
Other Species. J. Comp. Physiol. 103, 153-171. 

Howst, D. v.; Lesk, S. (1975): Über den Informationsgehalt des Sternaldrüsensekretes männlicher und 
weiblicher Tupaia belangeri. J. Comp. Physiol. 103, 173-188. 

Kıeın, L. L. (1971): Observations on Copulation and Seasonal Reproduction of two Species of Spider 
Monkeys, Ateles belzubuth and At. geoffroyi. Folia primat. 15, 233-248. 

Krei, L. L.; Krein, D. (1972): Aspects of social behaviour ın a colony of Spider Monkeys at San 
Francisco Zoo. Int. Zoo Yearb. 2, 175-181. 

McDoweLt jr., $S. B. (1958): The greater Antillean Insectivores. Bull. Amer. Mus. Nat. Hist. 115, 
113-214. 

MENZEL, K.H. (1979): Morphologische Untersuchungen an der vorderen Nasenregion von Solenodon 
paradoxus (Insectivora). Med. dent. Diss., Frankfurt M., 1-51. 

MoHR, E. (1936-38): Biologische Beobachtungen an Solenodon paradoxus ın Gefangenschaft. I-IV. 
Zool. Anz. 113, 177-188; 116, 65-76; 117, 233-241; 122, 132-143. 

ORTMANN, R. (1960): Die Analregion der Säugetiere. Handbuch der Zoologie. Ed. by J. G. HELMcKE, 
H. von LENGERKEN, D. STARCK, Berlin: De Gruyter. 8, Lfg. 26, 1-68. 

PEARsoN, ©. P. (1946): Scent glands of the Short-tailed Shrew. Anat. Rec. 94, 615-629. 

PopuscHkA, W. (1975): Solenodon Story. Wildlife, London, 17, 108-111. 

— (1976): Die Überlebenschancen des Schlitzrüsslers. Z. Theriophylaxe 5, 3-9. 

— (1977): Insectivore Communication. In: How Animals Communicate. Ed. by T. A. SEBEOX. 
Bloomington: Indiana University Press. 600-633. 

PoEv, F. (1861): Memorias sobre la historia natural de la ısla de Cuba. Havana. Vol. 1, 23-41, 433-435. 

Raııs, K. (1971): Mammalian Scent Marking. Science 171, 443-449. 

RICHARZ, K.; SPRANKEL, H. (1978): Daten zum Territorial-, Sexual- und Sozialverhalten von Tupaia 
glis Diard, 1820. Z. Säugetierkunde 43, 336-356. 

RıpgEway, R. (1912): Color Standards and Color Nomenclature. Washington: Selbstverlag. 

SAUER, E. G. F. (1974): Zur Biologie der Zwerg- und Riesengalagos. Z. Kölner Zoo 17, 233-248. 

SCHAFFER, J. (1940): Die Hautdrüsenorgane der Säugetiere. Berlin und Wien: Urban und Schwarzen- 
berg. 

ScHILLING, A. (1977): Olfactory Communication in Prosimians. In: The Study of Prosimian 
Behavior. London, New York: Academic Press 461-542. 

SCHULTZE-WESTRUM, T. (1965): Innerartliche Verständigung durch Dürfte beim Gleitbeutler Petau- 
rus breviceps papuanus Thomas (Marsupialia, Phalangeridae). Z. vgl. Physiol. 50, 151-220. 

SCHWARZ, W. (1937): Die Sternaldrüse bei den Klammeraffen, Ateles. Morph. Jb. 79, 600-633. 

SORENSoN, M. W.; Conaway, C. H. (1966): Observations on the Social Behavior of Tree Shrews in 
Captivity. Folia primat. 4, 124-145. 


12 R. T. F. Bernard 


SPRANKEL, H. (1961): Histologie und biologische Bedeutung eines jugulo-sternalen Duftdrüsenfeldes 
bei Tupaia glis (Diard, 1820). Verh. D. Zool. Ges. Saarbrücken, 198-206. 

STARCK, D. (1978): Das evolutive Plateau Säugetier. Sonderbd. Naturwiss. Verein Hamburg 3, 7-33. 

— (1982): Vergleichende Anatomie der Wirbeltiere 3. Berlin, Heidelberg, New York: Springer (im 
Druck). 

Tıjsskens, J. (1967): Solenodon paradoxus Brandt. Zoo Antwerp. Soc. Roy. de Zool. d’Anvers, 
142-144. 

WısLockı, G. B.; SCHULTZ, A. H. (1925): On the nature of modifications of the skin in the sternal 
region of certain Primates J. Mammalogy 6, 236-244. 


Anschriften der Verfasser: Prof. Dr. Dr. h.c. DIETRICH STARCK, Balduinstraße 88, D-6000 Frankfurt 
a. M. 70; Dr. WALTER PoDuscHKkA, Rettichgasse 12, A-1140 Wien, Öster- 
reich 


Female reproductive cycle of Nycteris thebaica (Microchiroptera) 
from Natal, South Africa 


By R. T. F. BERNARD 


Department of Zoology, University of Natal, Pietermaritzburg, South Africa 


Receipt of Ms. 23. 6. 1981 
Abstract 


Investigated monthly changes in the female reproductive organs and the reproductive cycle of Nycteris 
thebaica. Monthly changes in ovarıan and vaginal actıvity were characterized by two peaks. The first, 
between April and early June, culminated in estrus, while the second, occurred between July and 
August during pregnancy and is thought to be a relict from polyestrous ancestors. The reproductive 
cycle was characterized by copulation and fertilization in early June, a five month gestation with 
parturition in early November, approximately two months of lactation, and a period of anestrus 
between termination of lactation and onset of proestrus in April. 


Introduction 


The family Nycteridae includes a single genus of about ten species that occurs in Africa 
south of the Sahara and ın Egypt, Israel, Arabia, and on Madagascar. One species, N. 
javanıca, occurs ın Malayasıa (Hayman and Hırı 1971). Perhaps as a result of all but one 
species of this famıly occurring ın Africa, a continent on which study of chiropteran 
biology has been limited, it has received little attention. 

No aspect of the estrous cycle or associated changes in the female reproductive tract 
have been described although the reproductive cycles of tropical members have been 
reported. The tropical reproductive cycle is characterized by monotocy, polyestry with 
post-partum estrus between pregnancies, and a gestation of between two-and-a-half and 
three-and-a-half months. 

In the present study area N. thebaica was present at coastal localities throughout the 
year but at inland areas occurred only during winter, between April and September. 
Migration did not occur between known coastal and inland localıties and the summer 
roosts of the inland populations were not found. 

In the present study area it has been shown that hibernation occurs in Rhinolophus 
clivosus (Laycock 1976), Miniopterus schreibersi and M. fraterculus (BERNARD 1980a) and 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4701-0012 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 12-18 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Female reproductive cycle of Nycteris thebaica from Natal 18 


Hipposideros caffer (BERNARD 1980b). The reproductive cycles of Rhinolophus chvosus 
(Laycock 1976) and Myotis tricolor (BERNARD 1980b) include a period of sperm storage 
and delayed ovulation during winter, those of the two Miniopterus species include a period 
of delayed implantation during winter (BERNARD 1980a) while that of Hipposideros caffer 
includes a period of retarded embryonic development during winter (BERNARD 1980b). 
Nycteris thebaica did not hibernate in the present study area during winter and the aim 
of this study is to describe the estrous and reproductive cycles of this species and to see if 
and how they have been modified when compared with the tropical reproductive cycle. 


Materials and methods 


63 specimens of N. thebaica were collected from four roosts (caves and disused mines) in Natal (c. 
29° S) during 1977 and 1979. 

Specimens were killed by asphyxiation with carbon dioxide and the female reproductive tract 
removed under a dissecting microscope. Tissues were fixed in Bouin’s fixative and stored in 70 per cent 
alcohol. After routine embedding sections were cut at 5 um and routinely stained with Ehrlich’s 
haematoxylin and eosin. 

Description of the estrous and reproductive cycle has been based on an histological examination of 
monthly changes in the ovaries, uterine horns, and vaginas. All microscopic measurements were made 
with an optical micrometer. Follicular diameter was calculated from two measurements taken at right 
angles to each other, so that one measurement always included the greatest diameter. Mean thickness 
of the uterine horn wall, endometrium, and vaginal epithelium were calculated from about ten 
measurements per specimen. 

Where sample size was more than ten, mean values have been given plus or minus two standard 
deviations. Where sample size was less than ten, the mean alone has been given. 


Results 
Monthly changes in the ovaries 


Monthly changes in ovarian activity were quantified by measuring the diameters of all 
secondary and Graafian follicles and plotting mean monthly diameters (Fig. 1). In N. 
thebaica there was a marked difference in mean monthly follicular diameters from the two 
ovaries and for this reason they have been treated separately. In N. thebaica, which was 
monotocous and monestrous, there were two periods of Graafian follicle development. 


Follicle diameter (um) 


Fig. 1. Monthly changes in secondary and Graafian follicle diameter. Open symbols indicate mean 

monthly Graafian follicle diameter and closed symbols mean monthly secondary follicle diameter. 

Square symbols are measurements from the right ovary, circles from the left. Vertical lines represent 
X + 2SD 


14 R. T. F. Bernard 


The first, between Aprıl and June, occurred in the right overy and resulted in an ovulation 
in early June, while the second period occurred ın the left ovary between July and August 
and did not result ın an ovulation. 

The corpus luteum of N. thebaica, which was at least partially extruded from the ovary, 
was seen in one specimen collected in early June and four in July, but was absent from 
specimens collected during the remainder of the gestation. The diameter of the corpus 
luteum increased between early June and early July and decreased between early and late 
July (see Table). The corpus luteum comprised one type of luteal cell which was 
characterized by a large, faintly stained nucleus and prominent nucleolus. The mean 
diameter of the luteal cells decreased slıghtly between early June and early July, although 
their appearance remained constant, and decreased markedly between early and late July 


(see Table). 
Table 


Changes in mean corpus luteum and luteal cell diameters in micrometres 


Showing increasing corpus luteum diameter between early June and early July and decreasing 
diameter between early and late July 


Mean corpus luteum diameter Mean luteal cell diameter 


390,0 um 13,7 um 


516,3 um 13,0 um 
390,0 um 9,7 um 


There was very little interstitial gland tissue in the ovarıes of N. thebiaca and its cells, 
which were small with prominent nuclei, underwent no noticeable change in size or 
appearance during the year. 


Monthly changes in the vagina 


Monthly changes in the vagina were quantified by measuring thickness of the vagınal 
epithelium and plotting mean monthly thickness (Fig. 2). 

In N. thebaica there were two periods of epithelial development, both of which 
coincided with development of Graafıan follicles in the ovaries. Between October and 
March there were no cell divisions in the stratum germinativum and the epithelium 


60 


(him) 
[o} 


2 
[o} 


Vaginal epithelial thickness 


_ 
[e} 


Fig. 2. Monthly changes in thickness of the vaginal epithelium. Closed circles represent mean monthly 
thickness of vaginal epithelium and open circles mean monthly thickness of cornified layer. Vertical 
lines represent X + 258D 


Female reproductive cycle of Nycteris thebaica from Natal 15 


maintained a more or less constant thickness. During this period the epithelium comprised 
between five and nine layers of nucleated cells. 

Between Aprıl and early June the epithelium underwent a period of development. 
During Aprıl the outer layers of nucleated cells were fusiform in shape and during May the 
first superficial layers of cornified cells were present. In early June the cornified layer had 
reached a maximum thickness of 25,0 um. Delamination of the cornified layers and some of 
the underlying nucleated layers occurred during mid- and late June and in July the 
epithelium comprised approximately eight layers of nucleated cuboidal cells. Between 
April and late June leucocytes were present in the epithelium. 

During August and September there was a second increase in monthly epithelial 
thickness (Fig. 2). During this period there was no development of a cornified layer and the 
epithelium comprised nucleated cells only. During late September and early October there 
was a second period of epithelial delamination. 

The contents of the vaginal lumen changed with changes in the epithelium. Nucleated 
cells were present throughout the year but were abundant during the two periods of 
delamination. Cornified cells and leucocytes were present during the first period of 
delamination in June. Spermatozoa were seen in the vagina of one specimen collected on 6 
June, prior to ovulation. 


Monthly changes in the uterine horns 


Monthly changes in the uterine horns have been quantified by measuring thickness of the 
uterine horn wall and endometrium and plotting mean monthly thickness (Fig. 3). 

Between December and March the endometrium maintained a more or less constant 
thickness and the uterine glands were short, straight and few in number. 

Between April and June the thickness of the right uterine horn wall increased greatly 
while that of the left horn was less marked. These increases were the result of hypertrophy 
and hyperplasia of the endometrial cells and increased thickness of the myometrium. 
During this period the uterine glands increased in number and length. 

Between fertilization in early June, and early August, after which removal of the fetus 
and placenta prevented measurement of the endometrium, there was an increase in 
myometrial thickness. Although thickness of the myometrium was not measured an 


700 


(Hm) 


{e)] 
[eo] 
[o} 


500 


> 
© 
°© 


@ 
[eo] 
°o 


200 


Uterine horn wall and endometrial thickness 


Fig. 3. Monthly changes in thickness of uterine horn wall and endometrium. Triangular symbols 

indicate mean monthly thickness of the uterine horn wall and circular symbols mean monthly 

thickness of endometrium. Open symbols are measurements from the right uterine horn, closed 
symbols from the left 


16 R. T. F. Bernard 


estimate of its thickness can be obtained by comparing changes in the proportion of the 
uterine horn wall thickness that the endometrium comprised. 

After parturition, which occurred in early November, the lumina of the uterine horns 
were filled with cellular debris, erythrocytes, and leucocytes. 


Estrous cycle 


In N. thebaica the proestrous condition occurred during April and May and was charac- 
terized by the presence of developing Graafıan follicles in the ovaries and development of 
the vagınal epithelium and uterine endometrium. Estrus occurred in early June and was 
characterized by the presence of a single preovulatory Graafıian follicle in the right ovary, 
sexual receptivity, as indicated by spermatozoa in the vagina, and the presence of a layer of 
cornified cells on the vaginal epithelium. 

Estrus was followed by a short period of metestrus in mid- and late June when 
delamination of the vaginal epithelium occurred and the right oviduct of one specimen 
contained a morula. Pregnancy lasted for about five months, from early June to early 
November, and was followed by a two month period of lactation/anestrus. Although 
lactation had ceased by mid-January the anestrous condition lasted until the onset of 
proestrus in April. Anestrus was characterized by the absence of Graafian follicles from the 
ovarıes and undeveloped vaginal epithelium and uterine endometrium. 


Gestation period of N. thebaica 


The length of the gestation period has been estimated from specimens collected at Umdoni 
(30°24’S, 30°%42’E) during 1979. Of two specimens collected on 6 June one had a 
preovulatory Graafian follicle in the right ovary while in the other fertilization had recently 
occurred and there was a morula in the right oviduct. Of three specimens collected on 5 
November, two were lactating and one was still pregnant. This ıs a period of approximately 
152 days. 


Discussion 


With the exception of the second period of follicular development, the significance of 
which will be discussed later, monthly changes in the reproductive organs of N. thebaica 
were typically mammalıan. 

The corpus luteum of N. thebaica disappeared within two months of ovulation, a 
phenomenon that has been described in N. Iuteola (MATTHEws 1941) and Hıipposideros 
caffer (BERNARD 1980b). It is interesting to note that in both A. caffer and Nycteris 
thebaica the corpus luteum ıs at least partially extruded from the ovary, suggesting that its 
early degeneration may be associated with extrusion from the ovary. 

Members of the family Nycteridae are basıcally African in distribution with one species, 
N. javanica, occurring in Malaysıa (WALKER 1964). There has been no previous complete 
study of the reproductive cycle of a member of this family, although two authors have 
reported on some aspects of reproductive cycles of tropical nycterids. MATTHEwS (1941) 
reported that at c. 3°S ın Tanzanıa N. hispida and N. Iuteola are monotocous and 
polyestrous, with at least two gestations separated by a post-partum estrus. However, 
Matthews was unable to give an estimate of gestation length or whether polyestry was 
seasonal or continuous. 

Ancıaux DE FavEaux (1978) has summarized all available reproductive data for 
members of this family and has suggested reproductive cycles that are in many cases based 
on limited data but which reveal several trends common to tropical species. N. aethiopica 
from tropical latitudes north and south of the equator is monotocous and polyestrous, with 


Female reproductive cycle of Nycteris thebaica from Natal 17 


about three pregnancies per year and a gestation length of approximately three-and-a-half 
months. N. macrotiıs (= N. aethiopica, ADam and HUBERT 1976) at c. 1°S (Rwanda) is 
monotocous and seasonally polyestrous with at least two pregnancies per year separated by 
a post-partum estrus. The gestation period is about two-and-a-half months. N. thebaica 
from 11°S (Katanga) is monotocous and seasonally polyestrous with two or more pregnan- 
cies per year, separated by a post-partum estrus. The gestation period is approximately 
three months. 

To summarize these data, it would appear that the reproductive cycles of tropical 
members of this family are characterized by monotocy, polyestry which may be seasonal 
or continuous, the occurrence of a post-partum estrus, and a gestation period of between 
two-and-a-half and three-and-a-half months. 

The reproductive cycle of N. thebaica from c. 29°S (present study) differs from the 
tropical pattern in two significant ways. Firstly, N. thebaica ist monestrous and secondly, 
the gestation period is approximately five months. Both these changes are exactly what 
should be expected with the move from tropical into subtropical and temperate latitudes 
where the season suitable for reproduction is too short to allow polyestry. PEARsoN et al. 
(1952) and RAcey (1973) have shown that, in Plecotus townsendiüi and Pipistrellus pipis- 
trellus respectively, the gestation period can be experimentally lengthened by decreasing 
ambient temperature and/or food supply. Although Nycteris thebaica did not hibernate, it 
is believed that the relatively long gestation in the present study ıs caused by one or both of 
these factors. 

The second period of Graafian follicle development that occurred during pregnancy in 
the left ovary of N. thebaica ıst probably a relict from its polyestrous ancestors. Maximum 
Graafian follicle diameter in the left ovary was reached two-and-a-half to three months 
after ovulation, a period that corresponds with the length of the tropical gestation period. 

BERNARD (1976) described parts of the male reproductive cycle of N. thebaica from 
specimens collected in the present study area. Spermatogenesis begins in March and 
spermatozoa are released to the epididymides in late May, indicating that the male and 
female reproductive cycles are synchronized. 

The estrous cycle of N. thebaica at c. 29°S was characterized by a five month pregnancy/ 
luteal stage and a three month period of anestrus. Although the estrous cycle of a tropical 
member of this family has not been described it can be assumed that the pregnancy/luteal 
stage will be shorter and that, where polyestry is continuous, there will be no period of 
anestrus. 

Acknowledgements 


I would like to thank the following: Professor J. MEESTER for supervising this research and critically 
reading this manuscript; Mrs. N. Cook for typing this manuscript and the C.S.I.R. and University of 
Natal Research Fund from whom financial assistance was received. 


Zusammenfassung 


Weiblicher Fortpflanzungszyklus von Nycteris thebaica (Microchiroptera) aus Natal, Südafrika 


Der Fortpflanzungszyklus weiblicher Nycteris thebaica aus Südafrika wird auf Grund histologischer 
Untersuchungen an Ovar und Genitaltrakt beschrieben. Die Dicke des Vaginalepithels und die 
Durchmesser De Graafscher Follikel erreichen im Jahreslauf zwei Maxima. Das erste zwischen April 
und Anfang Juni entspricht dem Oestrus. Das zweite Maximum, bei dem es nicht zur Eibildung 
kommt, scheint ein Rest polyoestrischer Fortpflanzung zu sein, wie sie bei N. thebaica in den Tropen 
vorkommt. Das Corpus luteum ist - zum Teil durch Ausstoßung aus dem Ovar - zwei Monate nach 
der Ovulation verschwunden. Begattung und Befruchtung erfolgen Anfang Juni. Die Tragzeit beträgt 
fünf Monate, und die Jungen werden zwei Monate lang gesäugt. Der Anoestrus dauert anschließend 
bis zum Einsetzen des Prooestrus im April. 


18 R. Advani 


References 


ADAM, F.; HUBERT, B. (1976): Les Nycteridae (Chiroptera) du Senegal: Distribution, biomt£trie, et 
dimorphisme sexuel. Mammalıa 40, 597-613. 

ANCIAUX DE FAvEAux, M. (1978): Chez le Chiropteres cavernicoles du Shaba (S-E Zaire) et du 
Rwanda. Mammalıa 42, 453-490. 

BERNARD, R. T. F. (1976): Some aspects of the dynamics of a bat population in Natal, South Africa. 
Honours thesis, University of Natal, South Africa. 

— (1980a): Reproductive cycles of Miniopterus schreibersi natalensis (A. Smith, 1834) and Miniop- 
terus fraterculus (Thomas and Schwann, 1906) Ann. Transv. Mus. 32, 55-64. 

— (1980b): Female reproduction in five species of Natal cave-dwelling Microchiroptera. Ph. D. 
thesis, University of Natal, South Africa. 

Hayman, R. W.; Hırı, J. E. (1971): Chiroptera. In: The Mammals of Africa; an identification manual. 
Ed. by MEESTER, J.; SETZER, H. W. Washington D. C.: Smithsonian Institution. 

Lavycock, P. A. (1976): A study of cave-dwelling Microchiroptera in the Natal Midlands. M. Sc. 
thesis, University of Natal, Pietermaritzburg, South Africa. 

MATTHEwSs, L. H. (1941): Notes on the genitalia and reproduction of some African bats. Proc. Zool. 
Soc., Lond. B 111, 289-342. 

PEARSoN, O.; KOFORD, M.; PEARSoN, A. (1952): Reproduction in the lump-nosed bat (Corynorhinus 
rafinesqui) in California. J. Mammal. 33, 273-320. 

Racey, P. A. (1973): Environmental factors affecting the length of gestation in heterothermic bats. ]. 
Reprod. Fert. 19, 175-189. 

WALKER, E. P. (1964): Mammals of the World. Baltimore: John Hopkins. 


Author’s address: RicCHARD T. F. BERNARD, Department of Zoology and Entomology, Rhodes 
University, P. ©. Box 94, Grahamstown 6140, Republic of South Africa 


Feeding ecology ot Tadarida aegyptiaca thomasi 
in the Indian desert 


By R. Anvanı 


Central Arid Zone Research Institute, Jodhpur, India 


Receipt of Ms. 15. 6. 1981 
Abstract 


Studied the Egyptian free-tailed bat, Tadarida aegyptiaca thoması, collected from various districts of 
Rajasthan. It is primarily an insectivorous species. Coleoptera, Lepidoptera, Orthoptera, Hymenop- 
tera and Diptera are preferred in all main four seasons, while termites are consumed in all except 
winter season. Occurrence of ground dwelling insects, caterpillars, spiders, larger bodied insects and 
water beetles ın stomachs have been discussed in light of morpho-ethological adaptations of species. 
Presence of bat’s own fur in stomach coincides with the breeding season of the species. Presence of 
various polyphagous insect pest species of crops in feeding menu of bats show that this species can 
play a promising role in biological management of harmful insects. 


Introduction 


The Egyptian free-tailed bat or Wrinkle-lipped bat, Tadarida aegyptiaca thoması Wrough- 
ton (Chiroptera, Molossidae) is one of the most abundant species ın Rajasthan which is 
part of the Great Indian Thar Desert (24.5-30.5° N :60-70° E). Associated with warmer 
arıd and semi-arid regions, of which it is well adapted both ecologically as well as 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4701-0018 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 18-22 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Feeding ecology of Tadarıda aegyptiaca thoması 19 


physiologically, this species is not found in mountainous nor well-wooded regions. In its 
diurnal roost it is highly selective, favouring natural rock fissures or crevices in human 
settlement environment. 

In spite of the occurrence of this species in abundance, except some reports (ADvanı 
and VAZıRANI 1981; SınHA and Apvanı 1976), little is known about the ecology, biology 
and ethology of this species. To fill up this gap, the present studies were undertaken to 
investigate the food composition and seasonal variation in the feeding pattern of this 
species as this aspect has been dealt in detail in respect of temperate bats but little is known 
regarding tropical species and particularly the Indian ones. 


Methods 


The bats were collected during various periodical faunistic surveys conducted by Desert Regional 
Station, Zoological Survey of India, Jodhpur from 1975 to 1977. Collections were made mainly from 
nine districts - Jodhpur, Boondi, Dungarpur, Jhalawar, Sawai Madhopur, Tonk, Ajmer, Banswara 
and Kota, well distributed in the arıd and semi-arid parts of Rajasthan. In all 156 individuals were 
collected and examined. For each season, the break up of the sample size (N) is shown in the Table. 
After anaesthesia bats were dissected and their alimentary canals were cut open. The stomach contents 
were taken out with the help of brush and small forceps and then dried on filter paper at room 
temperature. After sorting, stomach items (insects etc.) were identified to the lowest taxonomic level 
feasıble (Order-Family), through aid of microscope and later weighed on the balance to calculate their 
percent occurrence ın the stomach contents, following MurTon et al. (1964). 

The seasonal fluctuations in the feeding pattern were determined by pooling data among four main 
seasons occurring in the Indian desert. 


Results 


The examination and analysıs of the stomach contents reveal that the Egyptian free-tailed 
or Wrinkle-Iipped bat, T. aegyptiaca thomasi ıs a prımarıly insectivorous species though 
some traces of plant material were also observed during monsoon and postmonsoon 
seasons (see table). Fur (tuft of hairs) of the same bat species occurred in larger proportions 
during summer and monsoon, whereas, it was present in traces during postmonsoon and 
winter. There were no remains of other anımals except insects and some spiders in the 
stomachs, to be considered as food items. 

In the winter season, December to February in the desert, when the temperature goes 
down to even -5 °C, this species mainly banks upon Coleoptera (34.7 percent) followed 
by Orthoptera, Hymenoptera and Diptera in about equal proportions (see Table). 
Araneida (spiders) also record their occurrence in stomachs in moderate numbers in 
winter. 

In the summer season, when the air temperature shoots up to even 50 °C, there ıs an 
increase in coleopteran diet by about 5.5 percent from winter to summer. However, 
percent frequency of Hymenoptera and Diptera decreases significantly. The bats devour 
upon an equal diet of Orthoptera (grasshoppers, crickets) during this season. Isoptera 
(termites) and Odonata (dragon and damselflies) also feature in the diet composition, 
though in low percent frequency. 

During monsoon rainy months, there is sudden rise in the consumption of winged 
termites (Microtermes obesü, Odontotermes obessus, Anacanthotermes sp.) about four 
times (24.5 percent) to that of preceding summer season. Proportion of beetles declines to 
about half, while that of Hymenoptera and Lepidoptera increases in varied percentages. 
Dependence of Orthopteran diet ıs also lowered during monsoon. 

The post monsoon season witnesses decrease in consumption of termites from 24.5 to 
5.2 percent of the total diet. Other orders like Odonata and Dictyoptera again record their 
occurrence in stomachs, while there is three fold increase in the dipteran diet. Preference 


20 R. Advani 


for Coleoptera almost remains at a same frequency of the monsoon season, whereas, moths 
are consumed in lower proportions. 


Discussion 


The Egyptian free tailed bat, T. a. thomasi is an anthrophilic species, being found in the 
close vicinity or in the midst of human environment at a tune of 100 percent of its total 
collection in western desert biome (Apvanı 1981a) and 91.3 percent ın the southern parts 
of Rajasthan (Anvanı 1981b). Its roosting habitat has certainly an impact on its feeding 
behaviour particularly in deciding composition and relative occurrence of various insect 


Table 


Stomach contents of Egyptian free-tailed bat during four main seasons in Rajasthan desert 


(expressed ın per cent of total mass) 


Stomach Seasons 
item Winter Summer Monsoon (rainy) Post-monsoon 
(Dec.-Feb.) (Mar.-June) (Jul.-Sept.) (Oect.-Nov.) 
N=32 N=41l N=40 N=43 


Coleoptera 
Scarabaeidae i ß Ä 6.1 
Carabidae 
Curculionidae 
Bostrichidae 
Dytiscidae 
Unidentified 


Lepidoptera 
Noctuidae 
Arctiidae 


Hymenoptera 
Formicidae 
Vespidae 


Neuroptera 
Mantispidae 


Diptera 
Chironomidae 
Culicidae 
Unidentified 


| Orthoptera 
Gryllidae 
Acrididae 


Isoptera 
Termitidae 


Odonata 
Libellulidae 


Dictyoptera 


Caterpillars of 
| Lepidoptera 
(unidentified) 


Araneıda 


Plant parts 


Bat’s own fur 
(tuft of hairs) 


Feeding ecology of Tadarida aegyptiaca thomasi 2 


orders, like Diptera, Coleoptera, Dictyoptera and Hymenoptera in its stomach contents 
(see Table) in sufficient proportions. However, it appears that feeding habits of this species 
are also probably a combination of opportunism and selective predation, varying with local 
conditions such as the relative abundance of different kinds of insect preys. Occurrence of 
traces of plant parts ın the stomachs of bats during monsoon and post monsoon is perhaps 
due to the remains of the undigested gut contents of insects eaten by bats. The presence of 
orthopterans, caterpillars of Lepidoptera, spiders and some ground beetles suggests that 
this species fed by picking these items from the ground or other surfaces rather than by 
aerial pursuit of prey. Moreover, its predation upon the larger sized insects like carabids, 
gryllids and dictyopterans is also an unique feature of the feeding habits of this species. 
These findings are in accordance with the general morphology of Tadarida bats, of having 
the wrinkled lips to facıliıtate a wider gap, broad wings and strong hind legs in comparison 
to other insectivorous bat species. 

Feeding habits of T. aegyptiaca markedly differ from those of the Indian false vampire, 
Megaderma lyra lyra which depends upon an equal proportion of insect as well as the 
vertebrate anımal diet (Apvanı 1981c) on an annual basis. 

Seasonwise, during winter when the temperature slides down to about - 5 °C in desert, 
and the bats are relatively inactive, they thrive upon insects available in their near vicinity 
or home ranges, like mosquitoes, flies, house crickets, cockroaches, small beetles etc. In 
this season bats can also subsist upon their own fat reserves which they accumulate after 
the monsoon season. During summer and monsoon months preference for termites is quite 
obvious, as this period coincides with emergence of winged, soft bodied termites after fırst 
few showers (from mid June onwards) in the desert, likewise, in postmonsoon season 
occurrence of winged ants and wasps, dipteran flies, dragon flies determine the diet 
composition of this species. 

The occurrence of bats’ own fur (tuft of haırs) ın stomachs ın highest percent frequency 
during summer and then in monsoon in the stomachs, is explainable on the grounds that 
the peak of the breeding season of T. aegyptiaca occurs from June to September (SInHA and 
Apvanı 1976) in the Rajasthan desert, when the inter and intra sexual fights and other 
interactions like allogrooming are quite expected. 


Acknowledgements 


I wish to thank the Director, Zoological Survey of India, Calcutta for providing facilities of 
undertaking faunistic surveys to very distant parts of State. Thanks are also due to Dr. T. G. 
VAZIRANI, Senior Entomologist, Commonwealth Institute of Entomology, British Museum Natural 
History, London for identification of insects. Thanks to Dr. IsHwAr PRAKASsH, Principal Anımal 
Ecologist, C.A.Z.R.I., Jodhpur for encouragement to start work on bat ecology as doctoral problem. 


Zusammenfassung 


Nahrungsökologie von Tadarıda aegyptiaca thomasi aus der indischen Wüste 


Fledermäuse der Art Tadarida aegyptiaca thoması wurden zu unterschiedlichen Jahreszeiten und ın 
verschiedenen Bezirken von Rajasthan (NW-Indien) gefangen. Untersuchungen der Mageninhalte 
ergaben, daß die Tiere sich in allen vier Jahreszeiten hauptsächlich von Coleopteren, Lepidopteren, 
Orthopteren, Hymenopteren und Dipteren ernähren. Termiten werden nur während der Wintermo- 
nate nicht aufgenommen. Der Nachweis von bodenlebenden Insekten, Raupen, Spinnen und Wasser- 
käfern in den Mägen der Fledermäuse deutet darauf hin, daß T. a. thomasi Nahrung nicht nur im 
freien Flug erbeutet. Ferner machten im Sommer und Monsum Fledermaushaare einen relativ großen 
Anteil am Mageninhalt aus. Dieses stimmt überein mit der Zeit der Paarung und Jungenaufzucht von 
Juni bis September. 

Die polyphage Ernährungsweise von Tadarida aegyptiaca thomasi unterstreicht die Bedeutung 
dieser Art für die Regulation von Schadinsektenpopulationen in landwirtschaftlichen Kulturen. 


22. S. Broekhuizen and F. Maaskamp 


References 


Apvanı, R. (1981a): Bioecological evaluation of Chiroptera fauna of desert biome of Rajasthan. Z. 
Angew. Zool. (in press). 

— (1981b): Ecological evaluation of the Chiroptera populations of south and southeastern biome of 
Rajasthan. Z. Angew. Zool. (in press). 

— (1981c): Seasonal fluctuations in the feeding ecology of the Indian false vampire. Megaderma lyra 
byra (Chiroptera: Megadermatidae) in Rajasthan. Z. Säugetierkunde 46, 90-93. 

ADvant, R.; VAZIRANI, T. G. (1981): Studies on ectoparasites of bats of Rajasthan and Gujarat. Rec. 
Zool. Surv. India, Occasional paper No. 22, 1-155. 

Murrton, R.; WESTWOooD, M.; Isaacson, A. J. (1964): Feeding habits of the Wood pigeon Columba 
palambus, Stockdove, Columba oenas and Turtle dove, Streptopelia turtur. Ibis 106, 177-188. 
SINHA, Y. P.; Apvanı, R. (1976): Notes on food and reproduction of some Rajasthan bats. Geobios 3, 

3740. 


Author’s address: Dr. RanjJan AoDvanı, Central Arıd Zone Research Institute, All India Rodent 
Research Project, Jodhpur, India 


Movement, home range and clustering in the European hare 
(Lepus europaeus Pallas) in The Netherlands 


By S. BROEKHUIZEN and F. MAASKAMP 


Research Institute for Nature Management, Arnhem, The Netherlands 


Receipt of Ms. 10. 6. 1981 
Abstract 


To examine aspects of dispersion ın hares, data about movements and home range were collected from 
recoveries of tagged individuals as well as from radio-tracked anımals. Most hares moved less than 
500 m from the capture place, even the sucklings, and the mean home range size observed was 29 ha, 
showing a sedentary way of life. Displaced hares mostly could establish themselves near the place of 
release, and a low degree of agonistic behaviour to an introduced tame conspecific was observed. 

Home ranges overlapped considerably, and at night clustering was observed. While ın the twilight 
hares grazing gregariously showed less attentive interruptions compared with hares grazing solıtarily, 
antı-predation behaviour was thought a function of clustering. 

As from literature is known that under circumstances with poor food supply, such as ın desert and 
in areas with deep snow cover, hares become more territorial during the reproduction season and more 
migratory in the winter, the results suggest that in our study areas food was not in short supply, 
presumably due to a high hunting pressure in autumn. 


Introduction 


Although in South-East Europe, after periods with heavy snow, groups of hundreds of 
roamıng hares have been observed (AnGERMANN 1973), roaming over large distances seems 
the exception in this species. Data collected by AnDersen (1951), RıEeck (1953, 1955), 
SZEDERJEI (1959), DoucLas (1970) and PıELowsk1 (1972) showed that about 80 percent of 
tagged anımals released on their capture place, were recovered within a distance of 3 km, 
and less than 10 percent were found more than 5 km away. RıEck (1953) calculated from 
his recoveries of tagged hares that the home range rarely exceeded 500 ha. PıELOwsKI 
(1972) recovered tagged individuals in areas of about 350 ha, while he concluded from 
observations of the movements of hares being pursued, that hares are familiar with an area 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4701-0022 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 22-32 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Movement, home range and clustering in the European hare 23 


of about the same size. However, he also pointed out that a distinction should be made 
between the larger area covered by the hare’s movements during its life time, and the 
several clearly defined areas, far smaller in size and occupied temporarily and periodically, 
situated within the boundaries of that total home range. By repeatedly shifting its area of 
activity, a hare would be kept informed of the situation within the total home range. The 
area in which PırLowsk1 observed a hare during one day was on average 20 ha. 

Merz (1976), following hares during a three month period of radio-tracking, and 
SCHNEIDER (1978), who collected data over periods of 0.5 to 1.5 years by direct observa- 
tion, found total areas of activity of about the same size as the areas of daily activity 
observed by Pırrowskı (1972). Their data clearly point to a non-migratory way of life, 
from which one might expect territoriality. Indeed, RımaTH£ (1977) and SCHNEIDER (1978) 
found indications that the movements observed in subadults were not only due to an urge 
to explore, but were also caused by intolerant behaviour from adults, possibly their own 
parents. On the other hand, Doucras (1970) and SzEDERJE1 (1959) found that most of the 
leverets were able to establish themselves in the area they were born in, and even within the 
home range of the parents. Also contrary to territorial animals, resident hares are often 
observed in groups, by day as well as at night. Although several authors showed an 
influence of field conditions on the distribution of hares (a. o. BrRüLL 1973; FIECHTER 
1975; FRYLESTAM 1976; JEZIERSKI 1968; MERZ 1976; PıELowskı 1966; RıMmATHE 1977), 
Merz (1973, 1976) observed clustering at night which he could not explain by a preference 
for certain vegetation, but in his opinion could have a social function. The same was 
suggested by PıELowskı (1966) and JEZIERSKIı (1968) dealing with clustering in the day- 
tıme. 

Since literature about spacing out and social clustering does not conform, we collected 
additional data about the size and changeability of the home range, the movements of adult 
and juvenile hares, the behaviour after introduction to unknown area and about clustering 
and its possible function. 


Material and methods 


Capturing, tagging and recoveries 


Size and changeability of the home range were examined from recoveries of tagged individuals, as well 
as by direct observation in the field. For both methods anımals had to be captured. Leverets, unless 
they were found by accident, were located by searching meadows about three quarters of an hour after 
sunset, using a binocular provided with a spotlight. At that time the leverets left their hiding places in 
order to gather for nursing (BROEKHUIZEN and MAASKAMP 1980). Attempts were made to capture the 
leverets by using a hoopnet after they had been nursed and the doe had withdrawn. In total 99 leverets 
were tagged with a numbered aluminium chicken wingmark. Beside the number, the name and phone- 
number of the institute were also punched on the tag, to facilitate the notification of recovery. Leverets 
were aged by means of body weight and length of the hind foot (BROEKHUIZEN and MAASKAMP 1979). 
After tagging they were released on the same spot where they were captured. 

Adult hares were captured by chasing them into vertically placed long nets. No drugs were treated 
for handling. Nearly five percent of the animals captured were lost through shock and accidents. Apart 
from the wing tag ın the ear already mentioned, a numbered plastic tag of the “Dalton-rototag” type 
was connected to the other ear. During the years 1968-1978 637 adult hares were captured and tagged 
at ten different locations, 132 of which were released at the spot they were captured, and 505 were 
released on fields 10 to 90 km away from their capture place. Capturing took place in the second half 
of December and the first half of January, just at the beginning of the reproduction season. 

Hares were considered as adult, when they were aged as six months or older from the stage of 
ossification of the distal epiphyseal cartilage knob of their ulna (BROEKHUIZEN and MAASKAMP 1979). 
Recoveries from our own field work were neglected, in view of the biased chance of notification. 


Measuring of the home range by direct observation 


Home ranges of 13 hares (8 females and 5 males) were determined by radio-tracking. Technical data on 
radio-transmitters, their connection to the anımal and the receiving units used were described earlier 


24 S. Broekhuizen and F. Maaskamp 


(BROEKHUIZEN et al. 1979). For these hares the ear tags and the radio-collar were plastered with 
reflecting “Scotchlite” tape, to facılitate observation at night using the spotlight. 

Since ıt was almost impossible to track hares during the whole night, it was done either from dusk 
till midnight or from midnight till morning. Generally one hare was tracked intensively. We attempted 
to locate it by triangulating the radio or by spotlight view every quarter of an hour. The location of 
other radıo-collared hares in the vicinity was determined at the beginning and at the end of the 
observations, and in the time in between as far as possible. 

Home range was defined according to BurT (1943): “that area which is traversed by the anımal 
(male) in all its activities as gathering food, resting, mating, caring for young. It excludes occasional 
sallies outside that area and shall be calculated for a specific period of time”. The size of the home 
range was determined according to the minimum observation area method of Opum and KÜNZLER 
(1955), connecting the outermost observations to the smallest convex figure in which all the other 
observations are located. 

The periods in which the home ranges of the various hares were determined, are given in table 1. 
The hares numbered as 1-6 and 10-13 lived in the area “Cortenoever” about 4 km south of Zutphen, a 
town ın the valley of the river Yssel (52.06 N; 06.13 E). This area was mainly used for dairy. The hares 
7-9 lived in the “Noordoostpolder” (52.45 N; 05.13 E), mainly used for arable farming. 


Behaviour after introduction in unknown areas 


Besides the recoveries of hares that had been moved, the behaviour after introduction in an unknown 
area was also studied from a female hare raised in captivity, which was released in September, at the 
end of the reproduction season, in the area ““Cortenoever”. The.anımal was radıo-collared for swift 
locating, and since ıt was familiar with human presence and activity, it could easily be observed from a 
car. Special attention was given as to whether the introduction induced agonistic behaviour in the 
established hares which finally resulted in the introduced hare being chased away. 

During the first three days after introduction, the anımal was observed continuously, and 
afterwards from 5.00-10.00 p.m. In the 14th night after introduction, the introduced hare was 
predated by a polecat. 


Determination of clustering 


As clustering can also be the result of nonuniformity of the habitat, we studied the distribution of 
hares on very uniform, cultivated meadows. During May and June 1977, ten meadows in the area 
“Cortenoever” were observed twelve times between 6.00 p.m. and 7.00 a.m. This was done from a car, 
using binoculars with a spotlight. The meadows, the location of which ıs indicated ın fig. 2, were 
selected for their appropriately low vegetation. In these meadows several points were marked with 
reflection-tape to facilitate the location of the hares observed. The spots on which they were first 
observed, were marked on a map. The distances between the marked hares observed per meadow area 
were compared with random distributions of the same number of hares on the same area. 


Influence of clustering on alertness 


The influence of clustering on the alertness of the hares was quantified by measuring the frequency of 
attentive interruptions during grazing, i.e. when hares stopped grazing to look around briefly. 
Observations were made between two hours before and one hour after sunset, and between one hour 
before and two hours after sunrise, from a car and without using a spotlight so as to prevent any 
disturbance of the hares’ behaviour. Protocols were recorded on tape, for being worked out 
afterwards. Grazing bouts shorter than four minutes were not taken into account. 

The observations were made on the same meadows used for studying clustering, as far as the grass 
was short enough to observe grazing and attentive behaviour in detail. The observations were 
restricted to hares located 50-100 m from the road, to enable making detailed observations without 
influence by road-traffic. 


Results 
Size and changeability of the home range 


The size of the home ranges of the 13 radio-tracked hares during the total periods of 
observation are shown in table 1. The mean size was found to be 29 ha. The relation 
between the length of the observation period and the size of the home range turned out to 
be insignificant (r = 0.29; n = 13; p. = 0.05), and the individual differences are distinct. 


Movement, home range and clustering in the European hare 25 


Table 1 
Observation periods and home ranges of 13 radio-collared hares 


C = Cortenoever, N = Noordoostpolder 


£ period of intensive number of observation-nights home range 
hare sex area observation period observations "intensive! 'incidental' (ha) 
1 ? C 31 July'73 - 30 Aug. '!73 1 Aug.'73 - 30 Aug.'73 15 17 25.5 
2 gs C 31 July'73 - 4 O0ct.'73 FAUST AO Et 31 

19 Jan.'7A - 15 Feb.'74 19 Jan.'7A - 15 Feb. '7A 13 US 99.5 
3 (6) [0 25 Aug.'73 - 8 Feb.'74 25 Aug.'73 - 20 Sep.'73 16 43 12 
4 2 (& 25 Aug.'73 - 4 Dec.'!73 25 Aug. '73 - 27 Nov.'73 45 20 16 

5 [6 C 18 0ct.'73 - 6 Dec.'74 18 Oct.'73 - 6 Dec.'73 25 21 13.5 

6 ? C 19 Jan.'74 - 15 Feb.'74 19 Jan.'74 - 15 Feb.'74 13 2 109 

7 ? N 15 Jan.'75 - 7 Feb.'75 20 Jan.'75 - 5 Feb.'75 4 9 13.5 
8 (6) N ISEgan Tr pres 20 Jan.'75 - 21 Mar.'75 19 12 30 
9 [6) N 15 Jan.'75 - 16 May '75 20 Jan.'75 - 15 May '75 27 9 172 
10 ? C 15 July'75 - 31 Oct.'75 22 July'75 - 10 Aug.'75 16 23 23 
11 ? C 24 Jan.'76 - 16 Apr.'76 2 Feb.'76 - 9 Apr.'76 10 14 57 
12 2 C 14 Jan.'76 - 16 Apr.'76 2 Feb.'76 - 9 Apr.'76 10 15 13 
13 ? C 14 Jan.'76 - 16 Apr.'76 2 Feb.'76 - 9 Apr.'76 11 14 30 


The cumulative increase of the home ranges of the hares numbered 1-6 in table 1 ıs shown 
in fig. 1°, while for the same animals fig. 1® shows the area size which would be used till 
the end of the observations. For reasons of comparison, for both graphs only data obtained 
from nights during which the anımals were tracked intensively were used. The figures 
show an increase of the total home range due to utilisation of areas not previously in use, 
and a decrease of the area which was observed to be used after that. This indicates shifts of 
the activity area which involves abandoning parts of the area. Such shifts may be gradual as 
well as abrupt. Not only new areas are involved, but also areas occupied earlıer can be 
taken ın use again. 

Home ranges of hares tracked simultaneously sometimes showed an extensive overlap, 
as is illustrated in fig. 2 for hares studied near “Cortenoever”. Hares with overlapping 
home ranges were repeatedly observed close together, also when they were of the same sex. 


Movements in adults and young 


Fig. 3 shows the distances between the places of release and recovery of individually tagged 
hares, in relation to the time involved. The location of the place of recovery was not always 
exactly known, especially when the hare was shot during a drive. Of the hares recovered 
within one kilometer from the place of release, those for which it was certain that the 
distance was less than 500 m, are indicated separately. 

From 132 adult hares released on the spot where they had been captured, 24 (18 %) 
were recovered (fig. 3%), of which 58 % were shot and 4 % killed by traffic. Nearly all the 
recoveries concerned animals which moved less than 1 km from their place of release, and 
for 58 % that distance was certainly less than 500 m. From the data it can be concluded that 
movement does not increase with age. 

From 99 leverets, estimated at 5 weeks old maximum age and supposedly still being 
nursed, 23 were recovered (fig. 3®), of which 56 % were shot. No traffic victims were 
reported. Nearly all the movements found were less then 1 km, and for two third certainly 
less than 500 m. 

The data indicate that also in young hares movement does not increase when they grow 
older. Movement of the young does not seem to differ from adults. 


26 5. Broekhuizen and F. Maaskamp 


28 2 


24 


(ha) 


207 mu N N ee Se RE a EN TEA ET RE ENTE 


16 


cumulative area of home range 


2 


10) 50 100 150 200 


(ha) 


range 


area of future home 


10) 50 100 150 200 


iength of observation period (days) 


Fig. 1. Areas of realized (A) and future (B) home range in the course of the observation periods of the 
radio-tracked hares 1-6 from table 1, based on data from nights of intensive observation 


Movements among introduced hares 


From 505 introduced adult hares, 90 (18 %) were recovered (fig. 3°), of which 70 % were 
shot and 19 % killed by traffic. Only 54 % ofthe recovered anımals moved over more than 
1 km and 9 % over more than 3 km. The data indicate that movement does not increase 
with time, so it is likely that the added movements in introduced hares mainly occurred just 
after releasing. The relatively high percentage of traffic victims also indicated movement 
before the anımals were settled. 


Movement, home range and clustering in the European hare 


u 


I, 
= 
=>. 
SZ, 
I 
LI 
DIN 
EI 
RAR 
EI 
m 
SZ 
Sn 
De 
I 
—a 
= 
Fr 
== 
== 
Fr 
LI 
en 
a 
a 
a 
= 
EZ 
£7 


Fig. 2. Home ranges of radio-tracked hares in the area ““Cortenoever”, numbered as in table 1. The 


pastures from which clustering data were collected (table 3) are shaded 


In fig. 4, the bearings indicate the direction and distance of migration related to the 
direction of the place of origin. A tendency of migration into the direction of the place of 
origin cannot be observed. 


Interactions between an introduced and indigenous hares 


Table 2 summarizes the observations of encounters between a released hare raised in 
captivity and indigenous hares during a fortnight just after releasing. Apart from this, we 
observed the hares 13 times passing by within 7 m of each other, without any obvious 
interaction. 


From table 2, no distinct difference in behaviour between the introduced hare and the 


indigenous ones can be concluded. No chasing away of the introduced hare occurred, and 
the hare stayed in the area in which it was released. 


Clustering 


The comparison between observed distributions of hares on the meadows in the “Cor- 
tenoever” area and simulated random distributions is summarized in table 3. Out of 81 
observations the observed distances were smaller than the simulated ones on 59 occasions, 


and they were larger 22 times. Since the difference is significant (sign test on observations 
and Monte Carlo simulations: T = 37; n 


81;p < 0.01) it may be concluded that hares at 


DT, 


28 S. Broekhuizen and F. Maaskamp 


5 
3 o 
E 1 . 
B 
— oO-1 9 —— - 0 - . —— 
= 0-05 —— oo © a ———————— a 
R (Dr 
s 16) 5 10 15 20 25 30 35 48 51 
E 
o 
0! B 
o ° 
= 7 
Q 
® 5 
[2 
° 3 
® 
° 1 
a 0-1 0 .. me ir Fe 4 ne Mal ee u RR RER BREEEEE er 
= 0-05 Se u 
® 
F u 
= [0] 5 10 15 20 25 30 35 64 
Qo2 
® 
© 9 o 
re 
a 7 s 
1 o 
5 A = 
o 
3 Oo .o o 
o o o 
nie ones oo. o 
1 2. “oo .,.- O O ® 
O-1 seo. 10. 1 u — 000 —— een en 
0O-0.5 ee — mem © sc o——— ... > 
man ma Un Fa EST DL IE ES EEE SE LU TESTEN UTRUSSUSTU TEE UT ETETUT U UT U UZUT retro 
(6) 5 10 15 20 25 30 35 45 47 


time between release and recovery (months) 


Fig. 3. Distances between the places of release and recovery of tagged hares, related to the time passed 
between release and recovery. A: tagged as adult, B: leverets tagged ın their suckling phase, C: tagged 


as adult and released in a strange area 


Fig. 4. Distance between 
release points (centre) and 
place of recovery (point of 
arrow) ın displaced hares, 
with the bearings of mig- 
ration related to the direc- 
tion of the place of origin 
(arrow right) 


Movement, home range and clustering in the European hare 29 


Table 2 
Behaviour in 19 encounters between a female hare released from captivity and conspecifics living 


in the wild 


behaviour hare released ‘wild’ hares 
from captivity 


contact sought 
sniffing at the conspecific 
following each other 


encircling the other one without approaching 
leaping 

aggressive behaviour (boxing, kicking, biting) 
breaking off the meeting by going away 
breaking off the meeting by flight 


m -NWOONoOMN \O 


Table 3 


Observed distribution of hares on pastures near Cortenoever (Fig. 4), compared with simulated 
random distributions 


number of total number mean sum of distances to the numbers of positive (f) and negative 
observations of hares nearest neighbour per observation (g) difference between (d) and (e) 


(b) simulated (d) observed (e) (f) (8) 


1 


_ 
AN 


m 
VOmenNNoo— 


a 
b 
c 
d 
e 
f 

8 
h 
i 

J 


ON TION PENWD PO 
»- ßen ND+-+DNM 


[eo] 
_ 
wi 
Ne) 
185) 
N 


night are more clustered than could be expected from a random distribution. During the 
observations, no indications were obtained that clustering was bound to a particular 
condition or usage of the meadows, such as height of the grass or the presence of cattle. 
Only within 10 m from the road relatively few hares were observed. 


Frequency of attentive interruptions during grazing 


For describing the distance between two grazing hares, three categories were distinguished, 
viz. 0-7 m: hares grazing gregariously; 8-30 m: hares grazing in open clusters; more than 
30 m: hares grazing solitarily. The observed frequencies of attentive interruptions within 
these categories are shown in fig. 5. A distribution is made between observations made 
around sunset and those made around sunrise. From the figure it can be concluded that 
solitarily grazing hares showed a higher frequency of attentive interruption than grazing 
hares in open clusters, whilst in the latter category the frequency exceeded that of hares 
grazing gregariously. Also at dawn the solitarily grazing hares showed the highest 
frequency of looking up. Differences between hares grazing in open clusters and those 
grazing gregariously were not observed, due to increased alert behaviour in the gregarious 
hares. It is likely that this increase is connected with a higher rate of sexual interactions in 
the morning, as found by F. OomEn-KausBEEk and K. NoOORDAM (unpubl. data). 


30 


S. Broekhuizen and F. Maaskamp 


evening 


solitary 
n=21 


20 


in open clusters 
n=25 


20 


morning 


solitary 
n=21 


20 


in open clusters 
n=20 


20 


frequency of observations (%) 


40 40 


gregarious 
n=30 


gregarious 


20 


16) 1 2 3 4 5 (0) 1 2 3 4 5 
number of attentive interruptions of grazing per min 


Fig. 5. Distribution of the number of attentive interruption of grazing per min in hares grazing 
solitarıly, in open clusters and gregariously. Left: observations from two hours before till one hour 
after sunset; right: observations from one hour before till two hours after sunrise 


Discussion 


The collected data clearly indicate that ın The Netherlands hares are very sedentary, even 
though their home range is not a fixed area. Occasionally the activity area is shifted. This, 
however, does not result in a continuous extension of the total home range, which is ın 
agreement with the ıdea of PıELowskı (1972), that the home range “contains several clearly 
defined areas, far smaller in size and occupied periodically, situated within the boundaries 
of the whole home range”. However, the size of the home ranges we observed was on 
average about ten times smaller than that mentioned by PıELowski (29 ha versus 300 ha), 
and more in agreement with the home ranges observed by MERZ (1976) and SCHNEIDER 
(1978). 

From fig. 1° it is likely that, where the observation periods varied from a few weeks to 
several months, the home ranges had not reached their ultimate size ın all cases. Neverthe- 
less we found no significant correlation between the total home range size and the length of 
the observation period, indicating that in most of the home ranges a large part of the 
ultimate size had been covered within a few weeks. This is in agreement with the recoveries 
of tagged hares, which also show that over longer periods and in other areas, most hares do 
not move more than 500 m and migration over more than 1 km is exceptional. This also 
holds for juveniles and sub-adults. Observations on dispersion of leverets after being 
nursed show that their radius of action can be several hundreds of meters at the end of the 
suckling phase (BROEKHUIZEN and MAASKAMP 1980). This indicates that the ultimate home 
range can already be reached during the suckling phase, and that generally juvenile hares do 
not need to migrate to become residents. 

Home ranges of neighbouring hares, also when they are of the same sex, can greatly 
overlap (fig. 2). This, and the low degree of agonistic behaviour, even towards newly 


Movement, home range and clustering in the European hare 31 


introduced hares, indicates a high degree of tolerance within the home range area. 
Therefore, the greater tendency to leave the place of release in introduced hares (fig. 3) ıs 
unlikely to be caused by antagonistic behaviour of indigenous conspecifics, or by ‘homing?’, 
as was found in some introduced hares by JEZIERKSI (1968) (fig. 4). As the greater moving 
distances in the former mainly occurred immediately after release, they are more likely to 
be a response to being caught and transported, or, as PIELOwskI (1972) expressed it, to be 
caused by shock. 

The distribution of hares on meadows in the ““Cortenoever” area showed an underdis- 
persed clustered pattern. This may reduce the risk of predation, because it enables the 
perception of signals from conspecifics during twighlight and in the dark, when spotting of 
potential predators is most difficult. The function of the clustering behaviour was not 
studied, but the lower frequency of attentive interruptions during grazing in the presence 
of conspecifics indicates a decreased need for attention. 

The combination of a non-migratory way of life, a low degree of intolerance, and a 
tendency to cluster does not always occur in hares. ANGERMANN (1973) mentioned 
roaming groups of hares ın South-East Europe during periods of heavy snowfall outside 
the reproduction period. LiNDLÖF (1978) observed aggressive dominance ranks in hares 
feeding from a hay stack when snow cover was deep. In Lepus americanus ıt was shown 
that when food was short, juveniles were expelled by adults and that immigrating adults 
more successfully settled than juveniles when the food supply was artıfıcıally increased 
(WINDBERG and KEITH 1976). These studies all indicate that the food supply determines 
behaviour and spatial distribution. 

The behaviour and spatial distribution of the hares we studied suggest that food 
shortage was not involved. This may hold for most of The Netherlands, because nearly all 
hare populations are hunted. Consequently, before winter starts and before the reproduc- 
tion season a substantial part of the populations will be eliminated. Besides, ın The 
Netherlands long periods with thick snow cover were exceptional in the last decade. 
Hence, the food supply is unlikely to be the limiting factor ın population regulation. 


Acknowledgements 


We are greatly indebted to the following students for their assistance: FRANGOISE OOMEN-KALSBEEK, 
Karın NooRDAM, Ton KnIjrr, JaaP VAN EGMOND, AUKE PETTEN, RUUD CRUL and WIM VAN 
Kraaıj. They all contributed to our knowledge of home range use in hares. 


Zusammenfassung 


Ortsveränderungen, Aktionsräume und Gruppierungen beim Feldhasen (Lepus europaeus Pallas) in 
den Niederlanden 


Zur Untersuchung verschiedener Aspekte der Verbreitung des Feldhasen wurden Ortsveränderungs- 
und Lebensraumdaten gesammelt an Hand von Rückmeldungen markierter Individuen sowie durch 
Verfolgung der Bewegung von Hasen, die mit Sendern versehen waren. 

Die meisten Hasen, auch Säuglinge, wanderten weniger als 500 m vom Markierungsplatz. Die 
mittlere beobachtete Lebensraumgröße war 29 ha. Dies deutet auf eine gewisse Standortstreue der 
Tiere hin. 

Versetzte Hasen siedelten sich meistens in der Nähe der Aussetzungsstelle an. Nur selten wurde 
aggressives Verhalten ortsansässiger Hasen diesen ausgesetzten Tieren gegenüber beobachtet. 

Die Lebensräume verschiedener Individuen überschneiden sich bedeutend. Nachts wurden Grup- 
pierungen beobachtet. Da während der Dämmerung gesellig äsende Hasen weniger Aufmerksamkeits- 
unterbrechungen zeigten als allein äsende Tiere, wird vermutet, daß diese Vergesellschaftung teilweise 
als Feindvermeidungsverhalten anzusehen ist. 

In der Literatur wird beschrieben, daß in Gebieten mit schlechten Nahrungsverhältnissen - wie in 
der Wüste oder in schneereichen Gebieten - Hasen während der Fortpflanzungszeit territorial sind 
und im Winter mehr wanderlustig werden. Die Ergebnisse dieser Untersuchung zeigen hingegen eine 
bleibende Standortstreue der Tiere. Diese läßt vermuten, daß in den von uns untersuchten Gebieten 
kein Futtermangel auftritt. Möglicherweise spielt dabei der hohe Jagddruck im Herbst eine Rolle. 


92 S. Broekhuizen and F. Maaskamp 


References 


ANDERSEN, J. (1951): Hvor meget vandrer haren? (On the movements of marked European hares 
[Lepus europaeus] in Denmark). Vildbiol. St. Kalö. medd. 5. 

ANGERMANN, R. (1973): De Haasachtigen. In: Het Leven der Dieren. Ed. by GrZIMEk, B. Utrecht: 
Het Spectrum, 486-540. 

BROEKHUIZEN, S. (1971): Gemerkte haas 70 km verder teruggemeldt. De Ned. Jager 76, 23, 824. 

— (1979): Survival in adult European hares. Acta theriol. 24, 34, 465-473. 

BROEKHUIZEN, S.; VAN ’t HOFF, C. A.; JANSEN, M. B.; NIEwOLD, F. J. J. (1980): Application of radio 
tracking in wildlife research ın The Netherlands. In: A handbook on biotelemetry and radio 
tracking. Ed. by C. J. AmLAnER and D. W. MacDonaArD. Oxford: Pergamon Press, 65-84. 

BROEKHUIZEN, $.; MAASKAMP, F. (1976): Behavior and maternal relations of young European hares 
during the nursing period. In: Ecology and management of European hare populations. Ed. by Z. 
PırLowskı and Z. PuceX. Warszawa: Pol. Hunt. Ass., 59-67. 

BROEKHUIZEN, $.; MAASKAMP, F. (1979): Age determination in the European hare (Lepus europaeus 
Pallas) in The Netherlands. Z. Säugetierkunde 44, 3, 162-175. 

— (1980): Behaviour of does and leverets of the European hare (Lepus europaeus) whilst nursing. ]J. 
Zool. (London) 191, 487-501. 

— (1981): Annual production of young in European hares ın the Netherlands. J. Zool. (London) 193, 
499-516. 

BrüLL, U. (1973): Wildfutterpflanzengesellschaften und Futterwert der von Feldhasen (Lepus 
europaeus Pallas) genutzten Pflanzen. Diss. Hamburg, 162 p. 

Burt, W. H. (1943): Territoriality and home range concepts as applied to mammals. J. Mammalogy 
24, 346-352. | 

Doucras, M. J. W. (1970): Movements of hares Lepus europaeus Pallas, in high country in New 
Zealand. N. Z. Jl. Sci. 13, 287-305. 

FIECHTER, A. (1975): Radiotelemetrie: Feldtechnik und Anwendung bei Reh und Feldhase. Zool. Inst. 
Univ. Zürich, Diplomarbeit. 39 pp. 

FRYLESTAM, B. (1976): Effects of cattle-grazing and harvesting of hay on density and distribution of an 
European hare population. In: Ecology and management of European hare population. Ed. by Z. 
PıeLowskı and Z. PuceEk. Warszawa: Pol. Hunt. Ass. 199-203. 

JEZIERsKT, W. (1967): Four cases of “home instinct” in the European hare. Acta theriol. 12, 173-175. 

— (1968): Some ecological aspects of introduction of the European hare. Acta theriol. 13, 1-30. 

KoeEnEn, F. (1956): Der Feldhase. Wittenberg-Lutherstadt: Ziemsen Verlag. 

LinDLör, B. (1978): Aggressive dominance rank in relation to feeding by European hare. Viltrevy 10, 
145-158. 

Merz, H. (1973): Die Gruppierungstendenz beim Feldhasen (Lepus europaeus Pallas) in Abhängigkeit 
der Nachtzeit im Juli/August 1973. Zool. Inst. Univ. Zürich, Semesterarbeit. 16 pp. 

— (1976): Nächtliche Standplatzwahl individuell gekennzeichneter Feldhasen (Lepus europaeus 
Pallas) innerhalb ihrer Aktionsräume. Zool. Inst. Univ. Zürich, Diplomarbeit. 39 pp. 

Opum, E. P.; KÜnZLer, E. J. (1955): Measurement of territory and home range ın birds. The Auk 72, 
128-137. 

PıELowskIı, Z. (1966): Forschungen über den Feldhasen. XII. Die Raumstruktur der Population. Acta 
theriol. 11, 449-484. 

— (1972): Home range and degree of residence of the European hare. Acta theriol. 17, 93-103. 

RıEck, W. (1953): Der Feldhase. Merkblätter des Niederwildausschusses des DJV 4. München: Mayer 
Verlag. 

— (1955): Feldhasen und Wildkaninchen. In: Jagd und Hege in aller Welt. Ed. by Deutscher 
Jagdschutzverband. Düsseldorf: Kölzig Verlag, 180-184. 

RımATHE£, R. (1977): Zur saisonalen Abundanzdynamik des Feldhasen (Lepus europaeus Pallas) im 
Schweizerischen Mittelland. Diss. Zürich, 126 pp. 

SCHNEIDER, E. (1978): Der Feldhase. München: BLV Verlag. 

SZEDERJEI, A. (1959): Beobachtungen über den Feldhasen in Ungarn. Z. Jagdwiss. 5, 81-94. 

WINDBERG, L. A.; KeıtH, L. B. (1976): Experimental analyses of dispersal in Snowshoe hare 
populations. Can. J. Zool. 12, 2061-2081. 


Authors’ address: Sım BROEKHUIZEN and FRANs MAASKAMP, Research Institute for Nature Manage- 
ment (R.I.N.), Kemperbergerweg 67, NL-6816 RM Arnhem, The Netherlands 


Zur Charakterisierung der Waldmäuse (Apodemus) Nepals 


Von H. GEMMERE und J. NIETHAMMER! 


Eingang des Ms. 10. 8. 1981 
Abstract 


On the characterisation of field mice (Apodemus) from Nepal 


Investigated the chromosomes of Apodemus sylvaticus and Alsomys gurkha from Nepal (Fig. 1 and 2) 
and some proteines of A. sylvaticus from that country (Tab. 1 and 2). The karyotypes are consistent 
with the membership of A. sylvaticus in the subgenus Sylvaemus and of A. gurkha with the subgenus 
Alsomys. The proteines of Nepalese A. sylvaticus show clear differences to those of European 
sylvaticus but are the same as those of long-tailed field mice from Iran. According to their proteines it 
is questionable whether Nepalese “sylvaticus” really represent that species. 


Einleitung 


In Nepal sind zwei Waldmausarten der Gattung Apodemus bekannt: Apodemus sylvaticus, 
der in der Westpaläarktis verbreiteten Untergattung Sylvaemus angehörend, und A. 
gurkha, eine Art der ostpaläarktischen Untergattung Alsomys (MARTENS und NIETHAM- 
MER 1972). Diese Zuordnung stützte sıch im Anschluß an ZIMMERMANN (1962) auf übliche 
morphologische Merkmale. 

Inzwischen sind für mehrere Arten aus beiden Untergattungen Karyogramme veröf- 
fentlicht worden (4 Arten von Sylvaemus: SOLDATOVIC et al. 1975; 4 Arten von Alsomys: 
VORONTSov et al. 1977), die hier gewisse Unterschiede zwischen beiden Gruppen erken- 
nen lassen. Sylvaemus besitzt stets 2n = 48 Chromosomen, die fast immer akrozentrisch 
sind. Nur A. mystacinus hat in dieser Gruppe zwei kleinste, metazentrische Autosomen- 
paare. Bei Alsomys ist 2n 48 oder 46. Meist sind bei ihnen 2 oder 4 kleine Autosomenpaare 
metazentrisch. Die Kenntnis der Karyogramme der beiden Arten aus Nepal schien daher 
zur Überprüfung der morphologisch getroffenen Zuordnung wünschenswert. Sie wurde 
uns durch die Überlassung lebender Tiere beider Arten durch Prof. Dr. MARTEns und Dr. 
BEcK, beide Mainz, ermöglicht. Wir möchten ihnen auch an dieser Stelle dafür danken. 

Die Einordnung der „A. sylvaticus“ aus Nepal erschien wieder fragwürdig, als Dr. BECK 
zwar A. sylvaticus aus Deutschland und „A. sylvaticus‘“‘ aus Nepal im Laboratorium 
züchtete, es ihm aber nicht gelang, beide zu kreuzen. Da sich die Apodemus-Arten in 
Europa anhand ihrer Isoproteine recht gut unterscheiden lassen (GEMMEkE 1980), bot sich 
die Möglichkeit, die Artzugehörigkeit von „A. sylvaticus“ aus Nepal durch Bestimmung 
ihrer Isoproteinmuster zu klären. 


Material und Methoden 


Für die Untersuchungen der Karyogramme standen 2 dd und 1 2 von Apodemus gurkha und 3 2? 
von A. sylvaticus zur Verfügung. (Nachfolgend wird angenommen, daß die Waldmäuse aus Nepal A. 
sylvaticus sind). Die Tiere hat Prof. MArTEns 1975 aus Nepal mitgebracht. Nach Haltung und Zucht 
in Mainz wurden sie uns von Dr. BEcK zur Verfügung gestellt. 


! Ergebnisse der Himalaya-Expeditionen von J. MARTENS, Nr. 83 - Nr. 82: J. Orn. 122, 403-427 
(1981). — Mit Unterstützung des Deutschen Akademischen Austauschdienstes (j.M.) und der Deut- 
schen Forschungsgemeinschaft (J.M., J.N.). 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4701-0033 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 33-38 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


34 H. Gemmeke und ]J. Niethammer 


Die Karyogramme wurden aus Mitose-Metaphasen von Knochenmarkszellen nach einem gängigen 
Verfahren gewonnen (z.B. Wıkıns und NIETHAMMER 1970). Es wurde nicht versucht, die 
Geschlechtschromosomen zu ermitteln. Für die Herstellung danken wir Frau G. SCHEBEN. Bei den 3 
schon genannten A. sylvaticus untersuchten wir elektrophoretisch die folgenden Proteine: Malat 
(NADP)-Dehydrogenase (MDH[NADP]), Malat NAD)-Dehydrogenase (MDH[NAD]), Indophe- 
nol-Oxidase (IPO), NADP-Isocitrat-Dehydrogenase, , , (IDH, ; >), a-Glycerophosphat-Dehydro- 
genase (a-GPD), Glutamat-Oxalacetat-Transaminase (GOT), Laktat-Dehydrogenase, , , (LDH, ; >), 
6-Phosphogluconat-Dehydrogenase (6-PGD), Albumin (Al), Transferrin (Tf). Dabei wurde die bei 
GEMMEKRE (1980) angegebene Methode befolgt. 


Ergebnisse 
Karyogramme 


Apodemus sylvaticus (Abb. 1): 2n = 48. Alle Chromosomen sind akrozentrisch. Damit 
entspricht das Karyogramm der Waldmaus (A. sylvaticus) aus Nepal zumindest in seiner 
groben Struktur demjenigen, das auch von Apodemus sylvaticus, A. flavicollis und A. 
microps aus Europa bekannt ist. 

Apodemus gurkha (Abb. 2): 2n = 48. Mindestens 2 kleine Paare sind metazentrisch, 1 
oder 2 weitere ebenfalls meta- oder submetazentrisch. Die 5 längsten Paare sind subtelo- 
zentrisch. Bei dem isolierten, nıcht subtelozentrischen, langen Element dürfte es sich um 
das X-Chromosom handeln. Unklar ist die Natur des Y-Chromosoms. Ferner könnten 
noch weitere nicht völlig akrozentrische Elemente vorhanden sein. 


m 6, er ı 


Abb. 1. Karyogramm eines ? von Apodemus sylvaticus aus Nepal. Geschlechtschromosomen nicht 
hervorgehoben 


Zur Charakterisierung der Waldmäuse Nepals 99 


ı 


N 
# 


Abb. 2. Karyogramm eines d von Apodemus gurkha aus Nepal. Die Identifikation der Geschlechts- 


chromosomen (rechts unten) ist unsicher 


Isoproteine 


Der Vergleich der 12 untersuchten Proteine zwischen Apodemus sylvaticus aus Europa, A. 
sylvaticus aus Nepal und A. flavicollis ergab: Die 3 folgenden Proteine stimmen in den drei 
Gruppen überein und sind monomorph: GOT, LDH, und MDH(NAD). Die 8 übrigen 
Proteine (Tab. 1) sind polymorph. Von ıhnen stimmen 3 in ihren häufigsten Allelen 
zwischen den drei Gruppen überein: a-GPD, 6-PGD und LDH,. Übereinstimmungen in 
den übrigen 5 Proteinen zeigt Tab. 2. Von ihnen gleichen 2 denen von A. sylvaticns ın 
Europa, unterscheiden sich aber von A. flavicollis: MDH(NADP) und Al. Umgekehrt sind 
zwei weitere dieselben wie bei A. flavicollis, aber andere als bei europäischen A. sylvaticns: 
IDH, und IPO. Tf schließlich liegt in allen drei Gruppen in verschiedener Form vor. 

A. sylvaticus aus Nepal stimmt also nicht völlig mit A. sylvaticus aus Europa überein, 
sondern unterscheidet sich in 3 von 12 Proteinen. Der gleiche Abstand besteht zu A. 
flavicollis. Nach diesen Proteinen ist also nicht zu entscheiden, welcher der beiden Arten in 
Europa die Waldmäuse aus Nepal näher stehen. 


Diskussion 
Die Karyogramme stützen die Zuordnung von A. gurkha zu Alsomys und von A. sylvaticus 


aus Nepal zu Sylvaemus. Wie schon erwähnt, sind 2-4 kleine, metazentrische Chromoso- 
menpaare bei Alsomys üblich, und A. gurkha paßt mit mindestens 2 kleinen, metazentri- 


’OdI 

"Hal 

nt 

+ IV 


*(dadVN)HAW 


edoıny edoıng jedaN 
snumvishu 'W syyonarf 'Y snaıvagks 'W 


(6/61) ‘Te 19 aHDIANvq] UoA url] sne 
uonepndodsnewpge/Q, J9P “(0861) INanWan uoA uDwwoumgn uauonepndog uayostedo.Ama 19P Ja] V 


uesj pun jedaN sne uonejndod 
-snewpfexy J9ur pun snworAjÄs Zunyedssgun 19p usıy € Taq ajapjy >puawwmsurAisg() 


GNS, 


(uuog) 

600 600 780 00T So SZO 00°1 00°1 90°0 EI‘O 8 sıyonarıf 'Y 
(redaN) 

00°I 00°1 00°1 00°1 00°1 00°1 snaıvajks 'Y 
(edoın) 


zo‘o Zz10 +H0°o 920 10°0 10°0 66°0 66°0 10°0 00°I 100 #6‘0 SO'O 100 660 €8‘0o ZTO 10°0 66°0 91 snangpajls 'V 


q 17 q e q e q e o q e “) q 17 Q q 17 fo} q 14 USW 
IV >PIIV O[>ILV a] V [P]IV [>]IV [IV [IV 
JL ‘HIT A9II-I IV AdI-» Odl 'Hal (dAdAVN)HAW 


srppoosaepz "y pun snameAjds "y UOA uauonepndog uauapaIy>sI9A UI 9UI9J0IA saydıoukjod g zuanbasp>][v 


NAT. 


Zur Charakterisierung der Waldmäuse Nepals 37 


schen Paaren dazu. Submetazentrische Autosomen, nämlich 1 bzw. 11 Paare wurden bei 
Alsomys bisher nur in 2 Untersuchungen über A. speciosus (= A. navigator) von Hondo 
angegeben (VORONTSOVv et al. 1977). Das Karyogramm von A. sylvaticus aus Nepal stimmt 
völlig mit dem von A. sylvaticus, A. flavicollis und A. microps aus Europa überein 
(SOLDATOVIC et al. 1975). 

Das Ergebnis der Isoproteinuntersuchungen führte zu unerwarteten Schwierigkeiten. 
Danach ist offen, ob A. sylvaticus aus Nepal zu A. sylvaticus, zu A. flavicollis oder einer 
eigenen Art gehört. Alle drei Möglichkeiten haben gleiche Wahrscheinlichkeit. Als positi- 
ver Befund ist vorerst nur hervorzuheben, daß offensichtlich der Abstand zwischen den 
Waldmäusen aus Nepal und solchen aus Europa wesentlich größer ist als zwischen allen 
bisher untersuchten Populationen von A. sylvaticus in Europa (vgl. GEMMEKE 1980). 

Für die weitere Erörterung setzen wir voraus, der morphologische Befund sei richtig, 
wonach A. sylvaticus aus Europa und A. sylvaticus aus Nepal enger verwandt sind als jeder 
von ihnen mit A. flavicollis. Dann müßte die Form von IDH, und IPO bei A. sylvaticus in 
Europa abgeleitet (apomorph), die bei A. flavicollis und A. sylvaticus in Nepal ursprünglich 
(plesiomorph) sein (Abb. 3a). Weniger wahrscheinlich ist die Möglichkeit, die beiden 
Proteine seien bei A. flavicollis und A. sylvaticus in Nepal zwar unterschiedlich, aber 
elektrophoretisch nicht unterscheidbar. 


A.sylvaticus A,flavıcollis 


Europa Nepal A.mystacinus 


A.sylvaticus A.tlavicollis 
Europa Nepal 


Abb. 3. Dendrogramm für (a) 3 und (b) 4 Sylvaemus-Formen. Durch O und @ ist die Verteilung von 
IDH, und IPO (a) und IPO (b) markiert. In (a) ist zu vermuten, in (b) sehr wahrscheinlich, daß Form 
O apomorph, @ plesiomorph ist 


Welche Form von zwei Allelen apo- bzw. plesiomorph ist, kann durch Einbeziehung 
weiterer Verwandter geklärt werden. Wir haben daher in Tab. 2 auch Apodemus (Sylvae- 
mus) mystacinus aufgenommen. Er stimmt in IPO mit A. flavicollis und A. sylvaticus aus 
Nepal überein. Danach wäre das IPO-Allel bei europäischen Waldmäusen apomorph. Die 
Albumine von A. sylvaticus aus Europa, A. sylvaticus aus Nepal und A. mystacinus sind 
gleich. Das abweichende Albumin von A. flavicollis dürfte apomorph sein (Abb. 3b). 

Von Bedeutung für die Einordnung der Nepal-Waldmäuse sind Proteinuntersuchungen 
an Waldmäusen aus dem Iran durch DarvIıcHEe et al. (1979). Diese gleichen in IDH,, IPO 
und Al den Waldmäusen aus Nepal. Tf haben DarvicHe et al. (1979) nicht untersucht und 
im MDH(NADP) keinen Unterschied zwischen A. sylvaticus und A. flavicollis in Europa 
gefunden. Dafür geben sie an, daß LDH, zwischen A. flavicollis und A. sylvaticus ım Iran 
übereinstimmt, beide aber von A. sylvaticus in Europa darin abweichen. Dagegen hat 
GEMMEKE (1980) diesen Unterschied — vielleicht methodisch bedingt — zwischen A. 
sylvaticus und A. flavicollis in Europa nicht finden können. 

Soweit sich die Befunde von DARrvIcHE et al. bewerten lassen, zeigen sıe völlige 
Übereinstimmung zwischen Waldmäusen aus dem Iran und aus Nepal sowie einen 
deutlichen Abstand zu europäischen Waldmäusen. 


38 H. Stephan and F. Dieterlen 


Die folgende Beobachtung verringert wieder den Abstand zwischen A. sylvaticus aus 
Europa und A. sylvaticus aus Asıen. Der Unterschied zwischen beiden im IPO-Allel ist 
hier nicht absolut. In einer A. sylvaticus-Population aus den Nordalpen wurde inzwischen 
einmal auch das für A. flavicollis typische Allel entdeckt (GEMMERE unpubl.). Damit ist 
gezeigt, daß die ursprüngliche Form auch bei A. sylvaticus in Europa noch nicht völlig 
eliminiert ıst. 


Zusammenfassung 


Eine Untersuchung der Karyogramme der Apodemus-Arten aus Nepal (Abb. 1 und 2) stützt die 
Annahme, daß Apodemus sylvaticus von dort in die Untergattung Sylvaemus, A. gurkha aber in die 
Untergattung Alsomys gehört. 

Ein proteinelektrophoretischer Vergleich der Waldmäuse A. sylvaticus aus Nepal ergab deutliche 
Unterschiede zu A. sylvaticus aus Europa, aber Übereinstimmung mit A. sylvaticus aus dem Iran (Tab. 
1 und 2). Ob die Waldmäuse aus Nepal wirklich zu A. sylvaticus gehören, muß danach vorerst offen 


bleiben. 
Literatur 


DARVICHE, D.; BENMEHDI, F.; BRITTON-DAVIDIAN, J.; THALER, L. (1979): Donnees preliminaires sur 
la systematique biochimique des genres Mus et Apodemus en Iran. Mammalia 43, 427430. 

GEMMEKE, H. (1980): Proteinvarıation und Taxonomie in der Gattung Apodemus (Mammalıa, 
Rodentia). Z. Säugetierkunde 45, 348-365. 

MARTENS, J.; NIETHAMMER, J. (1972): Die Waldmäuse (Apodemus) Nepals. Z. Säugetierkunde 37, 
144-154. 

SOLDATOVIG, B.; Savı6, I.; SETH, P.; REICHSTEIN, H.; TOLKSDORF, M. (1975): Comparative karyolo- 
gical study of the genus Apodemus (Kaup, 1829). Acta Veterinaria (Beograd) 25, 1-10. 

VoRoNTsov, N. N.; BEKAsova, T. $.; KrAL, B.; KoROBITSINA, K. V.; IvanıTskAYA, E. Yu. (1977): 
On specific status of Asian wood mice of the genus Apodemus (Rodentia, Muridae) from Siberia 
and Far East. Zool. Z. (Moskva) 56, 437-449 (russ. mit engl. Zusfg.). 

WINKING, H.; NIETHAMMER, ]J. (1970): Der Karyotyp der beiden kleinen, iberischen Pıtymys-Arten 
(Mammalia, Rodentia). Bonn. zool. Beitr. 21, 284-289. 

ZIMMERMANN, K. (1962): Die Untergattungen der Gattung Apodemus Kaup. Bonn. zool. Beitr. 13, 
198-208. 


Anschrift der Verfasser: Dr. HUBERT GEMMEKE und Prof. Dr. JOCHEN NIETHAMMER, Zoologisches 
Institut der Universität Bonn, Poppelsdorfer Schloß, D-5300 Bonn 


Relative brain size in Muridae with special reference 
to Colomys goslingi 


By H. Stepnan and F. DIETERLEN 


Max-Planck-Institut für Hirnforschung, Neurobiologische Abteilung, Frankfurt a. M. and 
Staatliches Museum für Naturkunde, Stuttgart 


Receipt of Ms. 10. 8. 1981 
Abstract 


Calculated the brain size in 18 species of Muridae either directly from brain weights (12 species) or 
indirectly from cranial capacity measurements (12 species incl. Colomys goslingi). In six species data 
from both sets are presented. In Colomys (a predator in limnetic ecosystems) relative brain size 
(encephalization) and foramen magnum size are larger than in terrestrial Muridae of equal body 
weight. The differences are similar to those found when comparing water adapted with terrestrial 
Insectivora. Comparative brain studies are in preparation. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4701-0038 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 38-47 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Brain size in Muridae 39 


Introduction 


Colomys goslingi, the african velvet or forest brook rat, is evidently a predator in limnetic 
ecosystems, as DIETERLEN and STATZNER (1981) have shown recently. Species with similar 
habits are found in the Insectivora families Tenrecidae (Limnogale mergulus; Potamo- 
galinae: Micropotamogale lamottei, M. ruwenzorü and Potamogale velox), Soricidae (e.g. 
Neomys fodiens) and Talpıdae (Desmana moschata, Galemys pyrenaicus). All of these 
species have large brains when compared with their purely or mainly terrestrial relatives. 
Within their brains there is a reduction in size of the structures of the olfactory system and 
an enlargement of the medulla oblongata, both of which are especially pronounced in 
Potamogale. As a consequence of the large medulla oblongata the foramen magnum also 
was found to be relatively large. 

The enlargement of the medulla oblongata in Potamogale is due to a marked develop- 
ment of the trigeminal system, especially of the nucleus of the spinal trigeminal tract. The 
trigeminal nerves, which also are extremely thick, innervate the strongly developed 
vibrissae of the muzzle. The vibrissae are considered to be vibration receptors adapted for 
detecting prey under water. In Colomys they seem to be “the only sense organ used in 
detecting the prey” in waters (DIETERLEN and STATZNER 1981). In predators in limnetic 
ecosystems the trıgeminal system ıs thought to replace the olfactory system and to become 
the main sensory system used in the search for food ın water (STEPHAN and Spatz 1962; 
STEPHAN 1967; BAUCHOT and STEPHAN 1967, 1968, 1970). 

Since we do not have well preserved brains of Colomys we cannot make exact 
measurements for comparative quantitative studies. However, an indication of the size of 
the brain can be obtained from cranıal capacıty measurements, and the size of the medulla 
oblongata can be inferred from the cross sectional area of the foramen magnum. By 
comparing these data with those of terrestrial Muridae we are able to evaluate, whether 
Colomys shows differences from other Muridae similar to those shown by the water 
adapted Insectivora from other Insectivora. 


Material and methods 


The cranıal capacities (CrC) and foramen magnum areas (FMA) were measured in 10 skulls of 
Colomys goslingi and compared with corresponding measurements in other murid species with similar 
body weights (Hybomys univittatus, Lemniscomys striatus, Praomys natalensıs). The average data and 
the standard error of the mean (SEM) are given in Table 1. In order to make well founded allometric 
comparisons similar measurements were made on skulls of especially small-bodied (Micromys 
minutus, Leggada minutoides) and large-bodied murids (Thamnomys venustus, Oenomys hypoxan- 
thus, Dasymys incomtus, Pelomys fallax, Rattus norvegicus, Malacomys longipes), and brain weight 
and body weight data were collected from 12 species (Table 2). 

For CrC and FMA measurements skulls, which came from specimens close to the average body 
weight of the given species, were selected. Average body weigths were calculated from the large 
collections of the Staatliches Museum für Naturkunde, Stuttgart. The CrC measurements were made 
by introducing fine dust shot into the cranial cavity through the foramen magnum, weighing the skull 
before and after introducing the dust shot, and by dividing the weight difference by 5.6121, which was 
found to be the weight in grams per cubic centimeter dust shot. The FMA measurements were made 
from enlarged photos of the foramen magnum by planimetry and/or by cutting out and weighing the 
photographic paper (STEPHAN et al. 1981). 


(% 0'0€-) 
191 
9:9] 
9'G] 

(% L'81-) 
L’8l 
8'81 
9:8] 

(% 9'67-) 
vgl 
vgl 
vgl 
0'E7 
677 
L’Ez 
uw 


(vwd) vaıe 


(% 0 FE-) 
»28 
878 
006 


(HT'IL-) 
6oll 
Izıl 
601 


(% 9'97-) 
U6 
TE6 
zıol 


rcel 
6rEl 
66C1 


‚u 


w 
(II) 


(% T8+) 
0'€9 
p’yS 
91 

(% r’7-) 
895 
765 
p'rG 

(% 1'8-) 
G'£g 
0'6+ 
0'8G 
z’ss 
9'159 
8'859 


3 
(mog) 


sKW0707 wOAF UONeIAap % 
ueauu 

sısuajvivu SCWODAT 
sW0J07 wOAF uonerAap %, 
ueaw 

snavmarun skwogldY 
sKW0707 wOAF UOLeIAIP %, 
ueaw 

snyvıus sKWwoISTuWaT 


UE9UI 


13u1]503 sKwu0]07 


wnu3ew uauwelo] Arısedes jerueıy ıy3ıam Apog 


Jy3ıam Apoq Aejruns jo Jepııny ueäLIfy Jo 
saıwads [e113531133 391yJ pue IFu1]so3 sÄw10J0O7 ur sesıe wınudew uswesoJ pue ‘sayıyede> [jeruess ‘sIJy3ıam Apoq uo eIeq] 


AL. 


'suonIpuo9 A1oJeloge] Japun paıq 31e speunue ‘(996]) NNVWHOSLAYN WOI} 
318 EIeP [JE „- '(SZ61) VASASY WON 9 pue “(Z/61) AaNDaaadloy wor} 77 ‘(ZZ61) Aa9Nnıaq WON} 0G “(II6I) Ivıvyy pue NOSATVNOT WOAF 918 spenprarpur 997 
Jo e1eP „- '(£96T) AAON WON 867 pue ‘(SO6I) vNorıaufy WO} 9 ‘(£061) LANVAgG Pure Yay9Tay, WO} € ‘(968[) vaaayy WON a1e spenprarpur 7 jo eıep „ - (0961) 


INWITM uUIOl] Ile sfenpIATpuI IpZ Be) eIep ea Sanpyuelg “unssnW 319quay9uag ulol] sppn2s Ze edunıg ‘apuny.ınyeN Any wın9snW say>meels wol] sn? 1 


667 — g76 <<6 788 g'0g 987 snivipiunp skwosy 8] 
L’ZI- 7'601 Z601 00z1 g'£9 g isndvyıs skwongdoT /1 
es] co — 9'971 ZUIZ 69 67 — L’871 SH »S78l 86 16 sadıduo skwosvoy 91 
r'8l LIE- 698 s'gI 191 L’zE- 0'£8 1601 906 0'€9 ‚ol sısuappivu sCwovA] SI 
ELE- 1'738 YpL7 OLE (167) 65€ 
671 1’97 — 0'r6 867 087 6LE— 618 9/7 »8P77 (167) „5 
ELE- 1'783 9H17 0177 167 chse susdaauou snıvy pl 
67€ 677 —- 1'86 9:01 vol LTE- 8'88 89€ BLOG Fol 18 sopıognunu vpv3837 £1 
0°67 — L’£&6 LoE 887 ZU g ds vpv3357 71 
PIE — 6'°€8 EIS 0Er 81 LaNS snjnssnu su 11 
691 I’8I- Zrol 6/1 281 0’SI- KEanı Szol »6PLl 8'959 ‚ol snyvımarın sCwogkH Ol 
TI - g’zol 886 +rlOl p'eg u 
291 LIT- 076 9/1 91 ST 8101 686 »ZOOI G'eg ‚ol 
061 — 8901 £86 0S0I € 7 snyvus shkwosstuwaT 6 
E0E— 616 gESI ‚Zirl RT 6 
Tel 19€ — €’ 18 077 6/1 ToEe— 1'76 LzsI »90P I ol 18 
IIE— 206 8191 09p 1 IzI I xoyppfskworag 8 
0'<I Re 8°S0l Salz 877 G8I— g'z01 P9p1 «P/Sl g'zol 18 snywmosu skwÄsvT 1 
8’SE— L’v8 IZEI +79 1 (76) 11 
891 L’87— 206 8°07 681 9’7E— 058 IZEI «9911 76 18 
PIE — 6°€8 IZEI OosIl (76) € snqiuvxodkq skwousQ 9 
BET — g’001 H8zı + 1621 (5'738) 6 
691 r’8I- g’eol 1’07 6°07 Ihr - 0'001 P8zı «P87l G'78 18 
E07 — 1'SOoI P8zı OSEI (5'z8) I snisnuaa skwouuvg] g 
991 — o’OII 9ES 065 y'6l .C87 snuavayls snwapody % 
ST S'zol 869 057 0€ ol sıyonavı[snwaopody € 
661 — 2'Sol £97 +8/7 g6'G el 
0’zE 67 - 896 06 L'8 661 — 9°c0I 197 787 1'9 „Ol 
E6l— F'901 IST 297 gg € snynunu skwowiWy 7 
rıl 0 Tue 1'81 0'E7 0 6TEI Ir01 GLEN z’8S ‚01 13u1so3 s(wojo) | 
(11) (01) (6) (8) (2) (9) (9) (#) (€) (d) (1) 
‚ws, wuı (1J) ‚ww ‚ww (19) du Ju 3 
sAwo]0J 001 - ; (vW43) (VWA) sAwu0]05 001 - (A.g3) (a.q) (nog) 
949 wIO.1} uonerAap VWJ4A wnugeu usw tale wnugeu WO.} UONEIAIP A193 ıydıam ureig aydıam ur 
VWÄI ader1ussıad VWJI -210} P23dxJ UusWEIOJ aderussıad AI p3199dx7 urelg poq u sawads 


sonmede> jerue1> P9119AuU09 pue sıydıam ureiq uUIO1} psurqwo> saneA | 
-sarıryedes [erueıs wol} pa119Au09 sıyaram ureig „ 'sI9Y>eı1q UT 195 a1e siydtram Apoq adeısae ayı “Papnjpur sıe sıydram Apogq umouyun yıım spenprarpur uayyy\ 
sÄUIOJOT UIOAJ sasıpur ayJ Jo 
uoreraap sdejussıad pue ‘sasıpur “saur uorssa1331 ayJ uo sanjeA p339adx> ‘sesse unuseu usure1or pue syy3lam urerg ‘sIydram Apog 


 „ 917207 I 


42 H. Stephan and F. Dieterlen 


Results 
Cranial capacity and/or brain weight 


When the three terrestrial species of approxiımately the same body weight as Colomys are 
compared with Colomys there ıs a deficit in the cranial capacity of — 16.2 % in Hybomys, 
— 26.6 % ın Lemniscomys and — 34.0 % in Praomys (Table 1). Since, however, the average 
body weights in Hybomys and Lemniscomys are somewhat lower (—2.4 % and —8.1 %) 
and in Praomys somewhat higher (+8.2 %) a regression lıne analysis is appropriate, and 
based on it, an allometric comparison. To get a stable reference line it ıs necessary to 
include more (and especially smaller and larger) species of Muridae into the comparison. 
Such a broader basis would not only stabilize the slope of the regression line, but would 
also give information about the relative position of the four species under consideration 
and especially of Colomys within the Muridae. Firstly, we plotted log brain weights against 
log body weights of 12 species of Muridae from which we have data (laboratory mice and 
rats excluded). The calculated regression line has the formula: 


log brain weight = 1.998 + 0.567 : log body weight 
The coefficıient of correlation ıs 0.987. 


Secondly, we plotted log cranial capacities (CrC) of the 12 species (so far measured) against 
log body weights. The calculated regression line has the formula: 


log cranial capacity = 1.971 + 0.585 : log body weight 


The coefficient of correlation ıs 0.968. 


Finally we combined both data sets by multiplying the CrC-values with 1.036 (specific 


>» 
JE 
Ö 
rm 
3 
£ 
< 
[e 4 
[- «} 
e)) 
.o 


log BODY WEIGHT 


1 1.9 2 2.9 


Fig. 1. Plot of logbody weight and log brain weight for 18 species of Muridae. The regression line has the 

formula log brain weight = 1.953 + 0.603 - log body weight. The coefficient of correlation is 0.975. 

* = species with brain weights converted from cranial capacities; ®= Colomys goslingi; + = species 

with brain weights combined from measured brain weights and converted cranıal capacities. (Num- 
bers as in Table 2) 


Brain size in Muridae 43 


INDICES 
BrW FMA 
—-1 Colomys 
130 130 
— 16 Malacomys - 
1657 
120 120 
— 10 Hybomys 
_4 Apodemus sylv. 
110 - 17 Lophuromys u 
_ 7 Dasymys 
3 Apodemus flav. 
—-2 Micromys he 
= 
— 9 Lemniscomys 
100] — 3 Thamnomys 1,00 
IS 
2 —- 
— 12 Leggada sp. iR= 
— 18 Acomys g9- 
= 8 Pelomys 
90 6 90 
— 13 Leggada minut. 
19 
_6 Oenomys 
— 11 Mus musculus 
— 15 Praomys 
14 Rattus norv. 2 
80 80 


Fig. 2. Diagrammatic representation of 
the encephalization indices (EI) of 18 
murid species (left hand scale) and fora- 
men magnum indices (FI) of 12 murid 
species (right hand scale). The indices are 
percentage deviations from the regression 
lines, all points of which represent 
100 %. (Numbers as in Table 2) 


gravity of the brain) to convert them into brain 
weights. Brain weights obtained in this way 
correspond well with the actual brain weights, as 
was found by STEPHAN etal. (1981) for 83 species of 
Chiroptera and by us in this study for 6 species of 
Muridae from which we have both brain weights 
and cranial capacities (+ in column 3 of Table 2 
and in Fig. 1). However, this correlation between 
brain weight and cranial capacity may be confined 
to the relatively small-sized animals (e.g. bats and 
murids) and may not be valid for large species 
with large brains (e.g. ungulates). The regression 
line resulting from all 18 species has the formula: 


log brain weight = 1.953 + 0.603 - log body 
weight 


The coefficient of correlation is 0.975. 


All three regression lines are relatively similar 
both in slope and y-intercept. For the following 
analyses we have used the common regression line 
of the combined material (Fig. 1). For the follow- 
ing comparisons this line ıs given a value of 100. 
The relative distances of the various species from 
this line (= encephalization indices, EI; see col- 
umn 5 ın Table 2 and Fig. 2) and their percentage 
deviation from Colomys (column 6 in Table 2) are 
given for all 18 species which were used to con- 
struct the common regression line of the Muridae. 

Only Malacomys ıs close to Colomys. All other 
species have distinctly lower EI values than Co/- 
omys and the percentage deficits (column 6 in 
Table 2) are distributed between — 15.0 % ın Hy- 
bomys and —37.3 % ın Rattus. When comparing 
the deficits of Hybomys, Lemniscomys and Prao- 
mys with those of the direct comparison of the 
cranıal capacities (see above) they are in Hybomys 
—15.0 % (versus —16.2 %), ın Lemniscomys 
— 22.2 % (versus —26.6 %), and ın Praomys 
— 37.3 % (versus —34.0 %). The differences are 
due to the differences in the body weights between 


the species and would not exist if all species had exactly the same body weights. The 
differences are largest in Praomys and Lemniscomys, where the average body weights are 
8.2 % higher and 8.1 % lower than that of Colomys (see Table 1). 


Foramen magnum area 


When the terrestrial species of about the same body weight as Colomys are compared with 
Colomys there is a deficit in the foramen magnum area (FMA) of — 18.7 % in Hybomys, 
— 29.6 % in Lemniscomys and —30.0 % in Praomys (Table 1). For the regression line 
analysis the four species were supplemented by eight species with clearly differing body 


weights (see column 7 in Table 2). 


44 H. Stephan and F. Dieterlen 


2.4 2.6 2.8 3.0 3.2 3.4 


Fig. 3. Plot of log cranial capacity (CrC in mm?) and log foramen magnum area (FMA in mm?) for 12 
species of Muridae. The regression line has the formula log FMA = —0.372 + 0.542 - log CrC. The 
coefficient of correlation is 0.976. ® = Colomys goslingi. (Numbers as in Table 2) 


The formula of the regression line of these 12 species is: 


log foramen magnum area = 0.710 + 0.310 - log body weight 
The coefficient of correlation ıs 0.921. 


The relative distance of each species from this line (= foramen magnum index, FI) is 
given ın column 9 of Table 2 and in Figure 2, and the percentage deviation from Colomys 
in column 10 of Table 2. Again only Malacomys is close to Colomys. All other species have 
lower FI values than Colomys and the percentage deficits (column 10 in Table 2) are 
distributed between — 16.8 % in Dasymys and —36.1 % in Pelomys. When comparing the 
deficits of Hybomys, Lemniscomys and Praomys with those obtained from the direct FMA 
comparison (see above), they are in Hybomys — 18.1 % (versus —18.7%), in Lemniscomys 
— 27.7 % (versus —29.6 %) and in Praomys —31.7 % (versus —30.0 %). Again the 
differences are due to differences in the body weights. 

In general, the distances of Malacomys and Colomys from the other species appear to be 
similar in the foramen magnum indices and in the encephalızation indices (compare both 
parts of Figure 2). In order to investigate whether or not Colomys has an disproportionally 
large FMA in relation to its brain size, we plotted FMA against CrC in a double 
logarıthmic scale and made a regression analysis (Fig. 3). The coefficient of correlation is 
0.952. The slope is 0.542 or, when plotting FMA against CrC?” (thus equalizing the 
dimensions of both axes) it is 0.813. According to this slope, which is distinctly smaller 
than 1, the FMA in species with small brains is relatively larger than in those with large 
brains. The same results are obtained when FMA is compared with CrC size in each of the 
investigated species (column 11 in Table 2). The two species with the lowest body and 
brain size (Micromys minutus and Leggada minutoides) have the highest values (32.0 and 
32.9 mm?/cm?) whereas the species with the largest body and brain size (Rattus norvegicus) 
has the lowest value (12.9 mm?/cm?). 

When the regression line in Figure 3 is given a value of 100 the highest positive 
deviations are found in Malacomys (110.1), Colomys (110.0), and Leggada minutoides 


Brain size in Muridae 45 


(108.1), whereas the highest negative deviations are found in Pelomys (84.4) and Lemnis- 
comys (91.6). All other species are close to 100. Thus, in Colomys, the FMA ıs relatively 
large when related to brain size. This relatively large FMA is, however, not exceptional as 
several other species such as Malacomys and Leggada also have relatively large FMA’s. 


Discussion 


The brain weight of Colomys was indirectly inferred from cranial capacities. The valıdity of 
such a procedure was checked in 83 species of bats (STEPHAN et al. 1981) and is verified in 
the present paper by data on 6 species of Muridae, from which both brain weights and 
cranıal capacities are available. In four of the sıx species the converted brain weights are 
slightly lower than the measured ones (Thamnomys, —4.9 %; Lemniscomys, —4.1 %; 
Pelomys, —3.7 %; Rattus, —1.0 %), whereas in the other two (Micromys, +5.6 %; 
Oenomys, +1.4 %) they are slightly higher. As ın the 83 bat species, the 6 murid species 
had similar values for measured brain weights and for those converted from cranıal 
capacıties. 

All these species are, however, either aerial or purely terrestrial and the question arises, 
whether or not such a direct correspondence also exists for water adapted forms. 
According to investigations on Potamogale velox the conversion factor from cranıal 
capacity to brain weight ıs 1.113 instead of 1.036 (specific gravity of the brain) as found in 
bats and terrestrial rodents. In Potamogale this difference is expected to be the consequence 
of the large size of the medulla oblongata and of the proximal parts of the spinal cord 
(caused by the extremely developed spinal trıgeminal nucleus; STEPHAN and SPATZ 1962). 
The spinal cord ıs always cut so that the part attached to the brain forms a square (seen 
from below). The extracranial caudal parts of the medulla oblongata and the proximal parts 
of the spinal cord left with the brain are in Potamogale, and may be in other water adapted 
species, of a larger percentage of the total brain than ın terrestrial and aerial species. Thus, 
cranial capacity measurements of water adapted species may underestimate brain sıze since 
relatively larger parts of the total brain (as recorded by brain weight) may be outside of the 
skull. | 

According to our investigations on Insectivora the adaptatıon to predatory life ın 
limnetic ecosystems ıs always accompanied by an enlargement of the medulla oblongata. 
Therefore, it can be expected, that Colomys also has a large medulla oblongata. The results 
of our FMA measurements are compatible with such an expectation, but not conclusive. A 
final decision only can be obtained from comparative investigations on the brain composi- 
tion. 

If we accept a relatively large medulla oblongata in Colomys, ıt can be expected, that ın 
Colomys (1) the brain weight is larger than that resulting from a conversion factor of 1.036, 
and (2) the EI is somewhat higher than given in Table 2. When the conversion factor of 
Potamogale (1.113; = 1.036 - 1.074) is used, the brain weight of Colomys would be 
1474 mg (1372 - 1.074) instead of 1372 mg as given ın Table 2, and the EI 141.7 instead of 
131.9. However, Potamogale seems to be more strongly adapted to water than ıs Colomys 
and, therefore, we expect the average brain weight of Colomys to be anywhere between 
1372 and 1474 mg, and the EI between 131.9 and 141.7. 

The sımilarly high position of Malacomys both in cranıal capacity and foramen magnum 
size needs further attention. It is of interest that Malacomys also ıs found near waters and ın 
swamps (WALKER 1964) and its diet is at least in part anımal food. ““Termites, crickets, 
slugs, snails, caterpillars, crabs and a toad are recorded anımal foods” (Kınapon 1974). 
However, according to DIETERLEN Malacomys may be found also far away from water, 
and its hunting strategies in water are clearly different from those of Colomys. Not the 
vibrissae but the hands seem to be used by Malacomys ın detecting prey ın shallow water. 


46 H. Stephan and F. Dieterlen 


With reference to these differences a comparative investigation of the composition of the 
brains (nearly equal ın relative size, see Fig. 2) would be of great interest. 

In conclusion, there are good indications that Colomys, which is a predator in limnetic 
ecosystems, has a larger brain and may have a larger medulla oblongata than do terrestrial 
species of Muridae. A final confirmation as well as an answer to the question, whether or 
not there is also a reduction of the olfactory structures, as is generally found in water 
adapted mammals, can only be obtained from investigations of brain components. Brain 
collection and quantitative investigations are in preparation. 


Acknowledgements 


We would like to thank Dr. D. Kock from the Senckenberg Museum, Frankfurt, for freely making 
available skulls of Micromys and Rattus. The valuable help of HELGA GROBECKER (photography) and 
HELMA LEHMANN (typing) is gratefully acknowledged. Thanks are due to JoHNn NELson, Melbourne, 
for checking the English text. 


Zusammenfassung 


Relative Hirngröße bei Muriden mit besonderer Berücksichtigung von Colomys goslingi 


Die Hirngröße von 18 Arten von Muriden wurde teils direkt aus Hirngewichten (12 Arten) oder 
indirekt aus Messungen der Hirnschädelkapazität (12 Arten inkl. Colomys goslingt) erschlossen. 6 
Arten waren mit Daten in beiden Gruppen vertreten. Bei Colomys (einer im Süßwasser jagenden Art) 
sind relative Hirngröße (Encephalisation) und Foramen magnum größer als bei terrestrischen Muriden 
gleichen Körpergewichts. Die Unterschiede ähneln jenen, wie man sie beim Vergleich an das 
Wasserleben angepaßter Insectivora mit terrestrischen Insectivora findet. Vergleichende Hirnuntersu- 
chungen sind in Vorbereitung. 


References 


BAuUcCHOT, R.; STEPHAN, H. (1967): Encephales et moulage endocraniens de quelques Insectivores et 
Primates actuels. In: Colloques internationaux du Centre National de la Recherche Scientifique, 
163, Problemes actuels de Paleontologie (Evolution des Vertebres), Paris, 1967, 575-587. 

— (1968): Etude des modifications encephaliques observees chez les insectivores adaptes a la 
recherche de nourriture en milieu aquatique. Mammalıa 32, 228-275. 

— (1970): Morphologie comparee de l’encephale des insectivores Tenrecidae. Mammalıa 34, 514-541. 

DIETERLEN, F.; STATZNER, B. (1981): The african rodent Colomys goslingi Thomas and Wroughton, 
1907 (Rodentia: Muridae) — a predator ın limnetic ecosystems. Z. Säugetierkunde 46, 369-383. 

Donapson, H. H.; HarTaı, $. (1911): A comparison of the norway rat with the albino rat in respect 
to body length, brain weight, spinal cord weight and the percentage of water in both the brain and 
the spinal cord. J. comp. Neurol. 21, 417-458. 

EBINGER, P. (1972): Vergleichend-quantitative Untersuchungen an Wild- und Laborratten. Z. Tier- 
züchtung u. Züchtungsbiologie 89, 34-57. 

GOLDBECKER, J. (1972): Acetylcholinesterase im Gehirn der Wanderratte und der Laborratte. 
Staatsexamensarbeit, Inst. Zoologie, Hannover, 1972. 

HRrDpLicka, A. (1905): Brain weight in vertebrates. Proc. Smithson. Misc. Coll. 48, 89-112. 

Kıncoon, J. (1974): East African Mammals. An Atlas of Evolution in Africa. Vol. II., Part B: Hares 
and Rodents. London and New York: Academic Press. 

KLEMMT, L. (1960): Quantitative Untersuchungen an Apodemus sylvaticus (Linnaeus, 1758). Zool. 
Anz. 165, 249-275. 

KRETSCHMANN, H.-J. (1966): Über die Cerebralisation eines Nestflüchters (Acomys cahırıinus 
dimidiatus [Cretzschmar, 1826]) im Vergleich mit Nesthockern (Albinomaus, Apodemus sylvaticus 
[Linnaeus, 1758] und Albinoratte), I. Teil: Morphologie und Allometrie. Morph. Jb. 109, 376410. 

Kruska, D. (1975): Vergleichend-quantitative Untersuchungen an den Gehirnen von Wander- und 
Laborratten. I. Volumenvergleich des Gesamthirns und der klassischen Hirnteile. J. Hirnforsch. 
16, 469-483. 

Norp, H. J. (1963): Quantitative Untersuchungen an Mus musculus dom. Rutty, 1772. Zool. Anz. 
170, 311-355. 

STEPHAN, H.; (1967): Zur Entwicklungshöhe der Insektivoren nach Merkmalen des Gehirns und die 
Definition der ‚„‚Basalen Insektivoren‘“. Zool. Anz. 179, 177-199. 

STEPHAN, H.; FRAHM, H.; Baron, G. (1981): New and revised data of volumes of brain structures in 
insectivores and primates. Folia primatol. 35, 1-29. 

STEPHAN, H.; NELSoN, J. E.; FRAHM, H. D. (1981): Brain size comparison in Chiroptera. Z. zool. 
Syst. Evolut.-Forsch. 19, 195-222. 

STEPHAN, H.; SPATZ, H. (1962): Vergleichend-anatomische Untersuchungen an Insektivorengehirnen. 


Olfactory sensitivity of female mice 47 


IV. Gehirne afrikanischer Insektivoren. Versuch einer Zuordnung von Hirnbau und Lebensweise. 
Morph. Jb. 103, 108-174. 

WALKER, E. P. (1964): Mammals of the world. Vol. I. Baltimore: Johns Hopkins Press. 

WEBER, M. (1896): Vorstudien über das Hirngewicht der Säugethiere. Festschrift f. Gegenbaur 3, 
105-123. 

WELCKER, H.; BRANDT, A. (1903): Gewichtswerthe der Körperorgane bei dem Menschen und den 
Thieren. Arch. Anthropol. 28, 1-89. 


Authors’ addresses: Dr. Heınz StEpHan, Max-Planck-Institut für Hirnforschung, Neurobiologische 
Abt., Deutschordenstr. 46, D-6000 Frankfurt/Main 71; Dr. Frırz DIETERLEN, 
Staatliches Museum für Naturkunde, Schloß Rosenstein, D-7000 Stuttgart 


The influence of the sexual cycle on the olfactory sensitivity 
of wild female house mice (Mus musculus domesticus) 


By M. Rırttner and U. SCHMIDT 


Zoological Institute, University of Bonn 


Receipt of Ms. 10. 8. 1981 
Abstract 


Studied was the influence of the sexual cycle of the olfactory sensitivity of wild female house mice 
(Mus musculus domesticus), using a two-choice training apparatus. Large fluctuations in olfactory 
sensitivity related to the sexual cycle were shown with the odourant geraniol (C,,H,;OH). During 
proestrus the mice recognized a concentration of 5 X 10° molecules geraniol/cm? air, while in 
metestrus they did not even respond to a concentration of 5 X 10!! molecules/cm?. 


Introduction 


The sense of smell plays a major role in social behaviour of house mice. For example, the 
sexual cycle of female mice may be accelerated when exposed to a male odour (WHITTEN 
1958). Olfactory contact with strange males can even lead to a depression of gravidity 
(Bruce 1959, 1962; Dominic 1966). 

While olfactory sensitivity in males is constant over a long period of time, the females’ 
varies ın relation to the sexual cycle (SchmiDT 1979). The results from these electrophy- 
siological experiments and behavioral studies with rats (PIETRAs and MouLTon; PHILLIPS 
1974 and VALLowE 1975) differ greatly. Thus, studies with trained wild mice need to be 
conducted to determine wheater these differences are a result of methods employed. 


Material and methods 


Nine female mice, approximately four months old, whose parents were caught in the wild, were used 
in this experiment. A two-choice training apparatus consisting of transparent plexiglass half-tubes was 
used. The gaseous odour was pumped via a nozzle into one of the tunnels, while filtered air entered the 
other tube. The location of the olfactory stimulus was randomly distributed. Three experimental 
animals were trained to reject the odour while the other six were trained to choose the olfactory 
marked side. The animals were punished with an electric shock each time they responded incorrectly. 
Each correct response was rewarded with food (for details of this method, see SchHMIDT 1979). Vaginal 
smears were done every day after the mice completed 30 trıals in order to determine the stage of the 
sexual cycle of each mouse (ALLen 1922; ZONDEcK and AscHHEIM 1926). 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4701-0047 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 47-50 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


48 M. Rittner and U. Schmidt 


The training odour used was geraniol (C,,H,,OH), a flowery smelling scent found in a number of 
flower oils. Even in high concentrations this essence has no damaging effects. 

Because extremely high concentrations of geraniol were rejected by all mice, the only reliable 
results were achieved with mice trained to avoid the olfactory stimulus. Therefore only the data of 
these three mice were analysed. 


Results 


For pretraining a geraniol concentration of 10'? molecules geraniol/cm? air was used. To 
determine the threshold the concentration was reduced by the factor 10 from 5 X 10!! to 
5 x 107 molecules geraniol/cm? air (each concentration level was maintained seven days). 
The reduced odour intensity was only applied during runs 1-20 (the first four trıals were 
not included ın the analysıs). The last ten runs were done at the highest level of 
concentration (10!? molecules/cm?). These ten runs served as a control and to retrain the 
anımals. 

The influence of the sexual cycle on the percentage of correct choices could already be 
observed at a concentration of 5 X 10!! molecules geraniol/cm? air. 80-100 % correct 
responses were recorded during proestrus, while only 50-70 % correct responses were 
achieved during the metestrus. Responses made during estrus and diestrus lie between 
these two extremes. The correct responses ın diestrus, proestrus and estrus were statisti- 
cally significant (t-test, p < 0.01) up to a concentration of 5 X 10° molecules geraniol/cm’ 
air. Test anımals ın all stages of their cycles no longer reacted to the essence when 
concentration reached 5 X 10’ molecules geraniol/cm? air. Figure 1 illustrates the per cent 
of positive responses of the three anımals in all four cyclıc stages. Ä 

Proestrus is significantly different (z-test, p < 0.01) from the other three stages; estrus 
and diestrus can also be differentiated from metestrus. The daily control- and retraining- 
runs at a geraniol concentration of 2 volume % (= 10!? molecules geraniol/cm? air) 
showed only little fluctuation and kept within the highly significant positive response level. 
This indicates that the differences in olfactory capacity were not due to chance and were 
not a result of training insecurity (fig. 2). 


100 


% CORRECT RESPONSE 


Fig. 1. Percentage of correct responses of the 
—— - three test animals during diestrus (D), proes- 
h : trus (P), estrus (E) and metestrus (M) at all 
above threshold concentrations of geraniol. 
The vertical bars indicate the standard devia- 
: : tion, the numbers under the colums the 
BEE number of the days investigated 


60 


Olfactory sensitivity of female mice 49 


u 100 (6) () o o 0 
ö P 
[a 
[05] 
[88] 
[04 
FH 
[(@) 
ce 90 
[04 
3 o 
S 
ee 
D D 


® 
E 


100 
NUMBER OF TRIALS 


Fig. 2. Training performance of a female house mouse during six consecutive days at an odour 
concentration of 5 X 10° molecules geraniol/cm? air - (filled circles) and at retraining with 5 x 10'? 
molecules geraniol/cm? air - (open circles). D = diestrus; P = proestrus; E = estrus; M = metestrus 


Discussion 


The results of this investigation correspond with those of C. ScHMiDT (1978, 1979) in 
which electrophysiological methods were used to examine the olfactory capacity of white 
laboratory mice. The odour substances used in SCHMIDT‘s experiment were geraniol, 
butyric acıd and butyric methyl ester. In all three test substances the typical fluctuations in 
olfactory ability were observed, indicating that these variations are independent of the 
odour offered. Neural thresholds also showed a maximal olfactory sensitivity in proestrus. 

Experiments studying olfactory sensitivity in female rats (PıETRAs and MourTon 1974; 
PhHiLLıps and VaALLowE 1975; MoULTON 1978) showed the highest sensitivity ın the estrus 
stage and a minimum of olfactory performance in metestrus and diestrus. Although these 
differences between mice and rats may be due to variations within these two species, it 
most probably was caused by the different methods of training used in each experiment. 
Each rat did ten trial runs per day at a medium odour concentration. Due to the relatively 
few runs, the performance level may have changed during sexual cycle. The security of 
response decreased during metestrus because of a decline in olfactory ability. During 
diestrus and proestrus the anımals might have successfully relearned the problem - which 
lead to a maximal surety of training in the following estrus. The continual pattern of 
trainıng and detraining could be eliminated from present investigation through daily 
trainıng with a high odour concentration so that the response level remained consistent. 

Cyeclic fluctuations in olfactory sensitivity which are expressed on the peripheral level as 
well as in perceptual level has not yet been attributable to any biological function. 
Behavioral studies need to be done to show whether female mice have gained an 
evolutionary advantage through it. 


Zusammenfassung 


Die Beeinflussung des Riechvermögens durch den Sexualzyklus bei weiblichen Hausmäusen 
(Mus musculus domesticus) 


Bei wilden 2? Hausmäusen (Mus musculus domesticus) wurde mit Hilfe einer Zweifach-Wahl- 
Dressur die Beeinflussung des Riechvermögens durch den Sexualzyklus untersucht. Für den Duftstoff 


50 Christel Schmidt and U. Schmidt 


Geraniol (C,,H},;OH) ließen sich starke zyklusabhängige Schwankungen der olfaktorischen Sensitivi- 
tät nachweisen. 

Während im Proöstrus noch 5 X 10° Moleküle Geraniol/cm? Luft erkannt wurden, wählten die 
Mäuse im Metöstrus selbst bei einer Konzentration von 5 X 10!! Molekülen Geraniol/cm? Luft nicht 
mehr signifikant positiv. 


References 


ALLEN, E. (1922): The oestrus cycle ın the mouse. Am. J. Anat. 30, 297-371. 

Bruce, H.M. (1959): An exteroceptive block to pregnancy in the mouse. Nature (London) 184, 105. 

Dominic, C. J. (1966): Block of pseudopregnancy in mice caused by exposure to male urine. 
Experıientia 22, 590-598. 

Mourron, T. H. (1978): Olfactory sensitivity during the menstrual cycle. Sens. Proc. 2, 90-98. 

PrırLıps, P. D.; VaLLowe, H. H. (1975): Cyclic fluctuations in odor detection by female rats and the 
temporal influences of exogenous steroids on ovarectomized rats. Proc. Penn. Acad. Sci. 49, 
160-164. 

PıETRAs, R. J.; MouLton, D. G. (1974): Hormonal influences on odor detection in rats: Changes 
associated with the estrus cycle, pseudopregnancy, ovariectomy and administration of testosterone 
propionate. Physiol. Behav. 12, 475-491. 

SCHMIDT, C. (1978): Olfactory threshold and its dependence on the sexual status in the female 
laboratory mouse. Naturwissenschaften 65, 601. 

SCHMIDT, C. (1979): Elektrophysiologische und verhaltensphysiologische Untersuchungen zum 
Riechvermögen der weißen Labormaus (Mus musculus). Diss. Bonn. 

WHITTEN, W.K. (1958): Modification of the oestrus cycle of the mouse by external stimuli associated 
with the male. Changes in the oestrus cycle determined by vaginal smears. J. Endocrinol. 17, 
307-313. 

ZONDECK, B.; AsSCHHEIM, $. (1926): Der Scheidenzyklus der weißen Maus als Testobjekt zum 
Nachweis des Ovarialhormons. Klin. Wschr. 22, 979-985. 


Authors’ address: M. RırtnEr and Prof. Dr. U. ScHMiDT, Zoologisches Institut der Universität, 
Poppelsdorfer Schloß, D-5300 Bonn 


Olfaction in pregnant and lactating albino mice 


By CHRISTEL SCHMIDT and U. SCHMIDT! 


Zoological Institute, University of Bonn 


Receipt of Ms. 2. 10. 1981 
Abstract 


Studied were the alterations of the neural olfactory threshold during pregnancy and lactation in albino 
mice (NMRI-strain) by means of evoked potential measurements from the Bulbus olfactorius. After 
copulation the threshold to geraniol is raised from the low proestrus level to a medium level that 
remains stable during most of the gestation period. Two to three days before parturition the olfactory 
acuity increases, till immediately after giving birth a sensitivity peak is reached. In the presence of a 
male this hyperosmia lasts a maximum of 4 days, in ısolated females up to 8 days. The biological 
significance of this effect is discussed. 


Introduction 


During the sexual cycle of females, the olfactory sensitivity changes to a very high degree. 
This has been found in humans (LE MAGnen 1952; Köster 1965; VIERLING and Rock 
1967; Dorty et al. 1981), as well as in rats (Pıerras and Mourton 1974; PHiLLıps and 


! The authors thank the Deutsche Forschungsgemeinschaft for financial support. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4701-0050 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 50-55 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Olfaction in pregnant and lactating mice Sl 


VALLowE 1975) and in mice (SCHMIDT and ScHMIDT 1980; RırTrner and SCHMIDT in 
press). In humans the span between lowest and highest sensitivity was up to the factor 10°, 
in laboratory mice the neural olfactory thresholds ın metestrus were up to factor 10° higher 
than in proestrus. Only very few observations indicate alterations of the olfactory acuity 
during pregnancy and in the post-partum-period in humans (Hansen and Gras 1936; 
Luvara and Mour1zI 1961; GUERRIER 1964; GOOD et al. 1976); in animals no investiga- 
tions have been found in the literature. 


Material and methods 


The experiments were carried out with 21 2? albino mice (NMRI-strain), 4 to 5 months old, all of 
them experienced mothers. An isolated tungsten electrode (50 um ©) was permanently implanted into 
one of the alfactory bulbs under anesthesia; the electrical activity of the olfactory bulb was recorded 
from the waking animals (for details see SCHMIDT 1978). The electrodes remained in function between 
5 days and 5 weeks. Each day the neural olfactory threshold for geraniol (C,,H,-;OH) was established 
by means of evoked potential measurements. The intensity of the stimulus was graduated in steps of 
factor 10. 

The mice, which were kept in individual cages, were divided into three groups: Group 1 was 
isolated before the experiments; two days after the operation a male was placed into the cage. In this 
group the days before copulation and the first days of pregnancy were investigated. In group 2 each 
female was constantly kept with a male. The operation was performed at the second half of gestation 
or shortly after parturition. The females of group 3 were isolated after copulation (olfactory contact to 
males, who lived in the same room, was maintained); all of them were operated the day after giving 


birth. 


_ 
[SER 
”m 


threshold concentration (molecules/cm') 


107 


4 2 
before VP 


2 4 6 87108 6 4 2 
after VP before birth 


2 4 6 8 10 
after birth days 


Fig. 1. Olfactory thresholds for geraniol of female mice before and after copulation (group 1, left half) 

resp. before and after parturition (group 2, right half). Each square represents the threshold value of 

one animal at a given day; in both groups the values of two mice are connected. Ordinate: threshold 

concentration (molecules geraniol/cm? air); the concentrations investigated differ at the factor 10 

(always 10” molecules/cm?). To indicate the number of females with the same threshold value, the 

squares are arranged vertically at each concentration. Abszissa: days before or after vaginal plug (VP) 
resp. birth (B) 


52 Christel Schmidt and U. Schmidt 


Results 
Olfactory thresholds of the females kept constantly with a male 


Before copulation the females of group 1 showed the normal cycling alterations in their 
olfactory sensitivity with a high threshold in metestrus (up to 5 - 10!* molecules geraniol/ 
cm?’ air) and a low threshold in proestrus (5 - 107° — 5 : 10° molecules/cm?). As soon as the 
vaginal plug was observed the threshold values remained at a medium level, in most of the 
cases 5 - 10° molecules/cm’. This level persisted during most of the time of pregnancy. 
Only two to three days before parturition the sensitivity increased slightly (factor 10). In 
fig. 1 the data of group 1 and 2 are combined. 

A considerable improvement of olfactory acuity was reached immediately after giving 
birth. The threshold values of all females decreased at a factor of 100 to 1000 compared 
with the long lasting gestation level. This sensitivity peak only lasted for up to 4 days. Then 
copulation occured and all anımals regained the constant level typical for pregnancy. 6 out 
of the 8 females of group 2 got their next litter 25 to 28 days after the first parturition. In 
two females the next birth took place not until 42 resp. 43 days. Nevertheless they showed 


—_ 

[o) 
—_ 
[o)} 


3 
threshold concentration (molecules/cm) 


days post partum 


Fig. 2. Olfactory thresholds for geraniol during the post-partum-period. A: lactating females which 
were kept together with a male (group 2); B: isolated lactating females (group 3) 


Olfaction in pregnant and lactating mice 53 


the same pattern of sensitivity changes as the other ones. It was not possible to determine 
when the first estrus occured, as after partirition the vaginal smears do not show the typical 
composition (GREENWALD 1958). 


Olfactory sensitivity during the post-partum-period of isolated females 


As the females of group 3 were operated the day after giving birth, the post-partum-day 2 
was the first one investigated. The minimal threshold values reached after birth were the 
same as in group 2, but the sensitivity peak remained for about double the time as in 
females kept together with males (fig. 2). In group 2 the decrease of sensitivity started 
between post-partum-day 3 to 5, in the isolated females the olfactory sensitivity was 
diminished in most of the cases not until day 7 or 8. The anımal in which a decline occured 
already at day 6 (fig. 2, squares), had a very small litter of only 5 pubs, whereas the other 
females had 8 to 12 juveniles. 

Latest at day 10 extremely high threshold values were found in all animals. The geraniol 
concentration needed to register an olfactory evoked potential was 10° to 10° times higher 
than after birth. The vaginal smears showed that at this time the females reached the 
metestrus phase. In the following days sensitivity increased again, consistent with normal 
cycling. 


Discussion 
In the course of the different sexual phases, female house mice experience a tremendous 


change of their olfactory ability. Whereas in males the olfactory thresholds differ maximal 
at the factor 10° (C. ScHmiDT 1979), in the freely cycling females the difference between the 


molecules/cm’ 


Fig. 3. Comparison of threshold values for geraniol in male mice (hatched column, a) and females at 

different reproductive stages (punctuated columns). b: freely cycling females (D diestrus, P proestrus, 

E estrus, M metestrus); c: gravid animals; d: post-partum-day 2 to 5 in lactating females kept with 

males; e: post-partum-day 2 to 5 in isolated lactating females. The figures ın the columns show the 

number of days a given threshold value was obtained; n gives the number of animals investigated ın 
each group 


54 Christel Schmidt and U. Schmidt 


lowest threshold values in proestrus and the highest in metestrus amounts up to the factor 
10°. During gestation olfactory sensitivity ıs most of the time slightly less than in the males 
(fig. 3, columns a and c); only shortly before parturition takes place male and female 
threshold values match. Irrespective of the presence of a male, the females reach very low 
thresholds after giving birth. At this time the olfactory acuity ın females is on the average 
10 times better than ın males. In the ısolated females this keen sense of smell remains for 
several days, the presence of a male reduces this sensitivity peak to 4 days at the maximum. 

What are the biological implications of these changes in olfactory acuity? Regarding the 
extremes in the cycling female, only the sensitivity peak during proestrus seems to be 
advantageous. During this time the female has to find a suitable partner which may be 
aided by olfactory stimuli. The extremely low sensitivity during metestrus, whereby the 
anımal must be nearly anosmatic, should be of disadvantage to the mouse. 

Of course these results were obtained with females artifıcıally retained from males. 
Under natural conditions males should always be available to the female. This situation is 
imitated in group 1 and 2 of the present study. Here ıt becomes evident that in the presence 
of a male the female does not reach the low metestrus level. The period of hyperosmia on 
the other hand falls in a time where olfactory guided behaviour seems to be most 
important. After parturition the female has to recognize her own pubs and, as receptivity 
commences quickly, is engaged in sexual behaviour. If a male is not available at once, as in 
group 3, the hyperosmia ıs considerably prolonged. The improvement of olfaction during a 
time, when the sense of smell is especially needed, should be of great value to the female 


mouse. 
Zusammenfassung 


Untersuchungen zur Riechfähigkeit trächtiger und laktierender Albinomäuse 


Mit Hilfe von evoked-potential-Messungen vom Bulbus olfactorius wurden die Veränderungen der 
neuralen Riechschwelle für Geraniol während Gestation und Laktation bei Albinomäusen untersucht. 
Nach der Kopulation stieg die Riechschwelle vom niedrigen Proöstrus-Niveau auf ein mittleres 
Niveau an, das fast die gesamte Tragzeit konstant blieb. Erst 2 bis 3 Tage vor der Geburt erhöhte sich 
die Riechschärfe und erreichte direkt nach der Geburt ein Maximum. Diese Hyperosmie dauerte bei 
Anwesenheit eines Männchens maximal 4 Tage, bei isolierten Weibchen jedoch bis zu 8 Tage. Die 
biologische Bedeutung der Riechschwellenänderungen wird diskutiert. 


References 


Dory, R. L.; SynDer, P. J.; Huccıns, G. R.; Lowry, L. D. (1981): Endocrine, cardıovascular, and 
psychological correlates of olfactory sensitivity changes during the human menstrual cycle. ]J. 
comp. Physiol. Psychol. 95, 45-60. 

Goop, P. R.; GEarRY, N.; Engen, T. (1976): The effect of estrogen on odor detection. Chem. Sens. 
Flavour 2, 45-50. ; 

GREENWALD, G. $. (1958): A histological study of the reproductive tract of the lactating mouse. ]. 
Endocrinol. 17, 17-28. 

GUERRIER, Y. (1964): Odorat et glandes genitales. Clin. otorhinolarıng. 16, 386-396. 

Hansen, R.; Grass, L.: (1936): Über den Geruchssinn in der Schwangerschaft. Klin. Wochenschrift 
25, 891-894. 

Köster, E. P. (1965): Olfactory sensitivity and menstrual cycle. Intern. Rhinol. 3, 57-64. 

LE Macnen, J. (1952): Les phenomenes olfacto-sexuels chez l’homme. Arch. Scı. Physiol. 6, 125-160. 

Luvara, A.; Mourızı, M. (1961): Richerche di olfactometria in gravidanza. Boll. d. mal. d. orrechie, 
d. gola, d. naso 79, 367-375. 

Prıruıps, P. D.; VaLLowz, H.H. (1975): Cyclic fluctuations in odor detection by female rats and the 
temporal influences of exogenous steroids on ovariectomized rats. Proc. Pennsylvanıa Acad. Scı. 
49, 160-164. 

Pıetras, R. J.; MouLron, D. G. (1974): Hormonal influences on odor detection in rats: Changes 
associated with the estrus cycle, pseudopregnancy, ovariectomy, and administration of testo- 
sterone proprionate. Physiol. and Behav. 12, 475491. 

Rıtrner, M.; SchMipT, U. (1982): The influence of the sexual cycle on the olfactory sensitivity of 
wild female house mice (Mus musculus domesticus). Z. Säugetierkunde 47, 47-50. 

ScHmipt, C. (1979): Elektrophysiologische und verhaltensphysiologische Untersuchungen zum 
Riechvermögen der weißen Labormaus (Mus musculus). Diss. Bonn. 


Age determination of the Atlantic walrus 55 


SCHMIDT, C.; SCHMIDT, U. (1980): Changes of the olfactory sensitivity during the estrus cycle in 
female laboratory mice. Chem. Sens. 5, 359-365. 

SCHMIDT, U. (1978): Evoked potential measurement of olfactory thresholds of laboratory mice (Mus 
musculus) to carboxylic acıds. Chem. Sens. Flavour 3, 177-182. 

VIERLING, J. S.; Rock, J. (1967): Variations ın olfactory sensitivity to Exaltolide during the menstrual 
cycle. J. appl. Physiol. 22, 311-315. 


Authors’ address: Dr. CHRISTEL SCHMIDT, Prof. Dr. Uwe ScHMiDT, Zoological Institute, Poppels- 
dorfer Schloß, D-5300 Bonn 


Age determination of the Atlantic walrus, Odobenus rosmarus 
rosmarus (Linnaeus), by means of mandibular growth layers 


By SUSANNE PETERSEN and E. W. BoRN 


Institute of Comparative Anatomy and Zoological Museum, University of Copenhagen, Denmark 


Receipt of Ms. 6. 7. 1981 
Abstract 


Examined the growth layers ın 112 walrus mandibles (Odobenus rosmarus Linnaeus) to determine 
their applicability in age determination. Growth layers are formed in the lateral portion of the ramus 
mandibularis in both males and females, from the first year of life until at least the 27th year. The 
number of mandibular growth layers was found to correspond with the number of annual growth 
layers in the cementum of the cheek teeth in the ratio of 1 : 1. It is concluded that readings of both 
mandıbular and cheek teeth growth layers should be made when determining the age in the walrus. 


Introduction 


Dental growth layers are wıdely used in age determination of seals (e. g. SCHEFFER 1950; 
Laws 1952, 1953, 1962; KLEvEZAL and KLEINENBERG 1967; JONSGÄRD, 1969; MORRIS 
1972). 

Monr (1952) called attention to the possibility of using the layering in the tooth 
cementum in age determination of the walrus (Odobenus rosmarus L.). Subsequently, age 
determination of the walrus has largely been based upon interpretation of cementum 
growth layers (BRooxks 1954; Fay 1955, 1982; MAnsrFIELD 1958; Burns 1965; KryLov 
1965; Born and KRrISTENSEN 1980). Although it ıs generally accepted that this method is 
usable, age determination in some individuals is often impeded by indistinctness of 
cementum growth layers. This is particularly marked in subadult walruses of both sexes 
and in females of all ages (Fay 1955; MansrIELD 1958; Born and KRISTENSEN 1980). 
Furthermore, attrıtion of the cheek teeth may impede age determination in older walruses 
(Burns 1965; Born and KRISTENSEN 1980). 

CHapskıı (1952) observed growth layers in the mandible of the walrus (Odobenus 
rosmarus) and the harp seal (Pagophilus groenlandicus). He demonstrated that in the harp 
seal the number of mandibular growth layers is correlated with estimates of age based on 
other criteria and suggested that the mandibular growth layers are formed annually. 
However, CHapsk1t (1952) neither gave a detailed description of the layering in the walrus 
mandible, nor did he use the layering in age determination of the walrus. Therefore we 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4701-0055 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 55-62 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


56 Susanne Petersen and E. W. Born 


decided to study the mandıbular layering ın the walrus and compare it with the growth 
layers in the cementum of the cheek teeth in order to determine the aplicability of the 
mandibular growth layers in routine age determination. 


Materials and methods 


Mandibles of 88 (47 males and 41 females) Atlantic walruses (Odobenus rosmarus rosmarus Linnaeus) 
from the Thule district, North Greenland, were collected in 1977 and 1978 in connection with an 
investigation of the exploitation of the walrus in Thule (Born and KrIsTEnsEn 1980). Of 79 walruses 
with information on date of kill 50 anımals were taken in October 1977 and 29 in the period 
March-June 1978 in the Thule district. Furthermore, four mandibles (two males and two females) 
were collected near Disko Island, Central West Greenland, in the period February-March 1978. The 
mandibles with some soft parts attached were kept frozen until cleaned in the laboratory in September 
1978. General cleanıng procedure was maceration in water (40 °C) until the soft parts could be 
removed by lightly rubbing under running tap water. After drying, a five to ten mm thick transverse 
section was cut on a rotating diamond saw used for preparation of rock samples through the region of 
the last cheek tooth. The surface of the section were polished with fine-grained sandpaper and studied 
in a dissection microscope in reflected light at a magnification of 10 x. 

In order to determine the formation of the initial growth layers two foetuses and 20 immature 
walruses (24 days to about four years old) from the collection of the Zoological Museum of 
Copenhagen were included in the analysıs. Four of the immature anımals were Pacific walruses 
(Odobenus rosmarus divergens lllıger) from Wrangell Island (U.S.S.R.) and sixteen were Atlantic 
walruses (Odobenus rosmarus rosmarus L.) from Greenland. To avoıd too much damage to the 
Museum specimens a drilling core, 12 mm in diameter, was taken from the lateral wall of the right 
ramus mandibularis below the last cheek tooth, instead of a transverse section (Fig. 1). For determina- 
tion of age by means of growth layers in the tooth cementum the method described by Born and 
KRIsTENSEn (1980) was followed. This method, essentially similar to that of MansrIELD (1958), is 
based on counts of growth layers in root cementum in transverse section of six cheek teeth from each 
animal. The undecalcified sections are studied under reflected light and the age in years ıs assessed as 
the mean number of growth layers in all sections. 


Age determination of the Atlantic walrus 57 


Results 


The combination of a broad zone of bone tissue succeeded laterally by a narrow, primary 
incremental line will constitute a “growth layer” in the description of mandibular growth. 
A corresponding annual growth layer in the tooth cementum comprises a broad opaque 
zone followed by a narrow translucent band. 

In the mandibles of two foetuses (age: about two months and 24 days prepartum, 
respectively) only spongy bone tissue had been formed. In seven neonates (age: from 24 
days until about four months postpartum) a 1-2 mm thick zone of homogeneous compact 
bone tissue had been formed on the lateral surface of the ramus, while the lingual portion 
of the mandible consisted of spongy tissue (Fig. 1). In three one-year-old calves (age: 10 to 
14 months) one fully-formed growth layer was found in the mandible as well as in the 
tooth cementum. One of these calves lived in captivity in the Copenhagen Zoo from 
October 1975 until 24 July 1976 and was reported to be 14 months old by death. 

When two to three growth layers have been formed the mandible consists almost 
exclusively of compact bone tissue. 


Fig. 2. Transverse sections of the mandible and the tooth 
cementum of a male walrus with eleven growth layers 
(Thule; October 1977) 

A: Transverse section of the right ramus mandibularis 
(Reflected light). 1: Laterally; a: Alveole of the last premo- 
lar; c. m.: Canalis mandibularis. - The bar indicates where 
the first growth layer was measured. (Photo: G. BROVAD. 
Z.M.) 

B: Transverse section of the cementum of the lower p. m. 
2. (Transmitted light). d: Dentine. — Note that the initial 
growth layers in the mandible and the tooth cementum are 
more indistinct than the following layers. (Photo: G. 
Brovan. Z.M.) 


auch wach, 


DO N RO 


In wo 


58 Susanne Petersen and E. W. Born 


Fig. 3. Transverse section 
of the right ramus man- 
dibularıs of a female with 
ten growth layers (Thule; 
May 1978); (Reflected 
light). 1: Laterally; c. m.: 
Canalis mandibularıs. -— 
Some of the primary in- 
cremental lines appear as 
double lines. (Photo: G. 
Brovap. Z. M.) 


In 20 subadults and five adults series of transverse section with five to ten mm intervals 
were cut along the whole length of the right ramus mandibularis. Growth layers were seen 
ın all sections but were most distinct in the region of the last cheek tooth. New growth 
layers are formed on the lateral surface of the ramus ın both males and females and are 
often particularly distinct in the portion lateral to the canalis mandibularis (Figs. 2A, 3, 4). 

It can thus be concluded that the first growth layer ıs formed after birth during the fırst 
year of lıfe, in the mandible as well as ın the tooth cementum. 

A neonatal line in the mandible of the sperm whale (Physeter catodon L.) was described 
by NisHrwakt et al. (1961). No such line or zorie separating the prenatal and the postnatal 
portion of the mandible was found in the walrus. Hence, the fırst-formed growth layer ın 


Fıg. 4. Transverse section 
of the right ramus man- 
dibularıs of a male walrus 
with about 24 growth lay- 
ers (Disko Island, May 
1978); (Reflected light). 1: 
Laterally; c.m.: Canalis 
mandibularis. — The first 
incremental line is partial- 
ly resorbed. (Photo: G. 
BrovaAD. Z. M.) 


Age determination of the Atlantic walrus 59 


eeretereren ers 
o 


Fıg. 5. Accumulated average widths 
of mandibular growth layers in the 
walrus (Odobenus rosmarus rosmarus 
L.) from the Thule district, North 
Greenland. Note that the width of 
the first growth layer includes bone 
formed prenatally (see text for expla- 
natıon) 


Width of mandibular growlh layers 


2 6 10 14 18 22 26 28 
No. of mandibular growth layers 


Y=0.982X + 0.218 


S°=1.084, N=86 


No. of tooth growth layers 


Fig. 6. The relationship between the 

numbers of growth layers ın the 

tooth cementum and the mandible of 

the walrus (Odobenus rosmarus ros- 

marus L.) from the Thule district, 
North Greenland 


(0) 5 10 15 20 25 
No. of mandibular growth layers 


the walrus mandible was measured from the lateral edge ot the canalis mandibularis to the 
lateral edge of the first incremental line and includes the prenatal tissue (Fig. 2A). 

In many specimens the ınıtıal two to four mandibular growth layers are more indistinct 
than the following layers. This we attribute to the accelerated growth in the premature life 
of the walrus as reported by Far (1955, 1982), MansrIELd (1958), KryLov (1965) and 
Born and KrIsTEnsen (1980). When about six to ten growth layers have been formed ın 
the male mandible, and about four to eight layers in the female, there is a decrease in the 
average thickness of the mandibular growth layers (Fig. 5). A similar deceleration in the 
growth rate is also observed in the layering of the tooth cementum (MAnsrFIELD 1958; 
Born and KriIsTENsEN 1980) and, presumably, reflects a decrease in overall body growth at 
attainment of sexual maturity. 

The relationship between the numbers of mandibular growth layers and growth layers 
in the tooth cementum is shown in Figs. 2 and 6. The slope of the regression line (Fig. 6) 


60 Susanne Petersen and E. W. Born 


does not differ from 1.0 (t = 0.824, P > 0.10, d. f. = 84) and the intercept of the line 
does not differ from 0.0 (95 % conf. int. for Y corresponding to X = 0.0: — 1.82 to 2.26). 
Thus, there is a syncrony in the formation of growth layers in the mandible and in the 
tooth cementum. The number of growth layers ın the two tissues corresponds on a ratio of 
1:1. Differences ın the numbers of growth layers found ın the two tissues reflect the 
uncertainty in interpretation of growth layers, rather than real differences in rate of 
formation. The 95 % confidence intervals of the regression line (Fig. 6) are + 2.04. In the 
few individuals where the difference between the number of mandibular and tooth 
cementum growth layers exceeds the 95 % confidence interval more layers were counted ın 
the tooth cementum. These anımals are an immature male and female (four mandibular 
layers) and two mature females (14 and 15 mandibular layers, respectively), Fig. 6. 
Indistinctness of the first-formed growth layers and counting of secondary lines in the 
tooth cementum explain these variations. 

The nature of the peripheral layer in the mandible and in the tooth cementum is shown 
in the Table. In the majority of individuals from all months there ıs a zone of bone tissue at 
the periphery and there ıs a tendency that this zone becomes broader in the period 
March-October. In the tooth cementum a translucent zone was found at the periphery ın 
most anımals in the period March-July, while 60 % of the specimens had a broad opaque 
zone at the periphery in October. Our presumption, that the formation of the narrow 
incremental line in the mandible is synchronous with the formation of a narrow translucent 
zone in the tooth cementum is not readily supported by the Table. However, in both 
tissues the nature of the narrow peripheral layers are difficult to perceive and we suppose, 
that in many cases the peripheral incremental line may have been removed from the 
mandible during preparation. In the minke whale (Balaenoptera acutorostrata Lacepede) 
parts of the newly-formed bone tissue are removed from the periosteum in bulla tym- 
panıcum during preparation (CHRISTENSEN 1980). 

Resorption of the initial growth layers occurs at a variable age, but ın none of our 
specimens had resorption progressed so far that the initial layers were not partially 
represented among the Haversian canals (Figs. 3 and 4). 

The oldest of our specimens (15 October 1977) had 27 growth layers in both mandible 
and tooth cementum. However, at only one site along the lateral surface of the mandibular 
transverse section were the 16 outermost growth layers present. These 16 growth layers 


Table 


The season of formation of the layering in the mandible and the tooth cementum in the walrus 
(Odobenus rosmarus rosmarus [L.]) from the Thule district, North Greenland 


Mandibles 
Nature of bone under formation 


Narrow incremental line 
Narrow zone of bone tissue 
Broad zone of bone tissue 


No. of anımals 


Tooth cementum 
Nature of zone under formation 


Narrow translucent zone 
Broad opaque zone 


No. of anımals 


Age determination of the Atlantic walrus 61 


totalled 1.5 mm in thickness. MAnsFIELD (1958) found walruses up to 28 years of age in 
Hudson Bay and Burns (1965) reported a 33-year-old walrus in an Alaskan sample. The 
finding of mandibular growth layers in a 27-year-old walrus indicates that age determina- 
tion from mandibular layering offers a useful tool in age determination even of very old 
walruses. 

Secondary lines were seen in the mandible of some animals. In the broad zone of bone 
tissue in the one-year-old calves an indistinct secondary line was observed. A similar line 
was present in the opaque zone of the initial growth layer in the tooth cementum of some 
anımals. The reason for this secondary line is obscure, but the long period of suckling in 
the walrus (approximately two years; MansrIELD 1958; Fay 1982) may influence on the 
regularity of the initial growth layers. In older walruses secondary lines may be present in 
the broad zone of the mandibular growth layer. The secondary lines appear irregularly in 
the normal growth layer sequence and are generally less distinct than the true primary 
incremental lines. Thus, we consider the chance of confusing secondary lines with primary 
incremental lines in age determination as low. Primary incremental lines may appear as 
closely-situated double or triple lines that fuse ventrally and dorsally (Fig. 3). 


Conclusions 


In the walrus one mandibular growth layer corresponds to one annual growth layer in the 
cementum of the cheek teeth. 

Mandibular growth layers are easier to distinguish than tooth cementum growth layers. 
Because mandibular layers are thicker, more sharply defined, and not influenced by 
eliminating processes they are more suitable in age determination than tooth cementum 
growth layers. 

The preparation of transverse sections of walrus mandibles is a quick and simple 
process. 

We therefore recommend that the counting of mandibular growth layers should be 
included with counting of growth layers in the tooth cementum in routine age determina- 
tion of the walrus. 

Acknowledgements 


This study was supported by funds from the Danish Natural Science Research Council and the 
Commission for Scientific Research in Greenland. 

We wish to thank Dr. T. KrISTENsEn, Zoological Museum of Copenhagen, who collected and 
prepared some of the material. We also wish to thank Professor K. G. WINGSTRAND and Dr. H. GRUE, 
Institute of Comparative Anatomy, Copenhagen, and Dr. E. Pıarts, Department of Zoology, 
Queens University, Belfast, for reviewing the draft and offering useful critism. 


Zusammenfassung 


Altersbestimmung beim Atlantischen Walroß, Odobenus rosmarus rosmarus (Linnaeus) 
durch Zuwachslinien der Kiefern 


Die Zuwachslinien wurden in 112 Walroßkiefern, Odobenus rosmarus rosmarus (Linnaeus) studiert, 
um über ihre Verwendbarkeit zur Altersbestimmung zu entscheiden. Im lateralen Teil des Ramus 
mandibularis werden bei beiden Geschlechtern Zuwachslinien vom 1. bis wenigstens zum 27. 
Lebensjahr gebildet. Die Zahl der Zuwachslinien im Kiefer stimmt mit der Zahl der Zuwachslinien im 
Zement der Backenzähne überein, im Verhältnis 1:1. Eine Zuwachslinie wird jährlich gebildet. 
Zählung von Zuwachslinien im Kiefer sollten daher zusätzlich zur Zählung von Zuwachslinien des 
Backenzahnzementes für die Altersbestimmung des Walrosses durchgeführt werden. 


References 


Born, E. W.; KrIsSTEnsEnN, T. (1980): Biology and exploitation of the Atlantic walrus, Odobenus 
rosmarus rosmarus (L.), in the Thule district, North Greenland. Meddr. Gronland (in press). 


62 Buchbesprechungen 


Brooks, J. W. (1954): A contribution to the life history and ecology of the Pacific walrus. Special 
Report No. 1, Alaska Cooperative Wildlife Research Unit, 103 pp. 

Burns, J. J. (1965): The walrus ın Alaska; its ecology and management. Alaska Dep. Fish Game, 
Juneau, Alaska, Annu. Proj. Segm. Rep., 5. Fed. Aid Wildl. Restor. Pro). W-6-R-5, Work Plan 
GA. 48 pp. 

CHapskn, K. K. (1952): (Determination of the age of some mammals from the microstructure of the 
bones.) Estestv. Nauchn. Inst. P. F. Lesgrafta, Akad. Pedagog. Nauk RSFSR., IZV. 25, 47-67. (In 
Russian.) 

CHRISTENSEN, I. (1980): Age determination of minke whales, Balaenoptera acutorostrata, from 
laminated structures in the tympanic bullae. Rep. int. Whal. Commn. Paper SC/32/Mi 5, 21 pp. 

Fay, F. H. (1955): The Pacific walrus (Odobenus rosmarus divergens), spatial ecology, life history and 
population. Ph. Thesis. University of British Columbia. 171 pp. 

Fay, F. H. (1982): Ecology and biology of the Pacific walrus Odobenus rosmarus divergens Illiger. 
U.S. Dep. Int. Fish Wildl. Serv. North American Fauna No XX; Washington D. C. (in press). 

JonsGÄrD, Ä. (1969): Age determination of marine mammals, in: The biology of marine mammals, H. 
T. Anperson (Ed.). New York and London: Acad. Press, 1-30. 

KLEVEZAL, G. A.; KLEINENBERG, $. E. (1967): Age determination of mammals from annual layers in 
teeth and bones. Israels Program for Scientific Translations, Jerusalem, 1969. 128 pp. 

Kryıov, V. I. (1965): Determination of age, rate of growth and analysis of age structure of the Pacific 
walrus catch. J. Fish. Res. Bd. Can. Transl. Ser. No 1004 (1968), 12 pp. 

Laws, R. M. (1952): A new method of age determination for mammals. Nature 169, 972-973. 

Laws, R. M. (1953): A new method of age determination in mammals with special reference to the 
elephant seal. Scient. Rep. Falkland Isl. Depend. Surv. 2, 1-11. 

Laws, R. M. (1962): Age determination of Pinnipeds with special reference to growth layers in the 
teeth. Z. Säugetierkunde 27, 129-146. 

MANnSFIELD, A. W. (1958): The biology of the Atlantic walrus (Odobenus rosmarus rosmarus L.), in 
Eastern Canadian Arctic. J. Fish. Res. Bd. Can. Manscr. Rep. Ser. Biol. 653, 145 pp. 

MoHRr, E. (1952): Die Robben der europäischen Gewässer. Monographien der Wildsäugetiere, Band 
XII. Frankfurt/Main: Dr. Paul Schöps, 284 pp. 

MoRrıs, P. (1972): A review of mammalıan age determination methods. Mammal. Review 4, 69-104. 

NIsHIwAkI, M.; OsHuMı, S.; Kasuya, T. (1961): Age characteristics ın the sperm whale mandible. 
Norsk hvalfangstt. 50, 499-507. 

SCHEFFER, V. B. (1950): Growth layers in the teeth of Pinnipedia as an indicator of age. Science 122, 
309-311. 


Authors’ addresses: SUSANNE PETERSEN, Institute of Comparative Anatomy, and E. W. Born, 
Zoological Museum, Universitetsparken 15, DK-2100 Copenhagen, Denmark 


BUCHBESPRECHUNGEN 


STEPHENSON, W. K.: Grundlagen der Zellbiologie. Struktur - Moleküle - Stoffwechsel. 
Ein Lernprogramm. Übers. u. bearb. von L. HArner. Berlin und Hamburg: Paul Parey, 
1980. 235 S., mehr als 200 Abb., Karte zum Abdecken der Musterantworten, Bal., 
DM 30,-. ISBN 3-489-60634-5. 


Das vorliegende Buch wendet sich an Teilnehmer von Einführungskursen der Zellbiologie für 
Botaniker, Zoologen und Mediziner und ist zum Selbststudium geeignet. Es unterscheidet sich von 
ähnlichen Werken durch die didaktische Konzeption, denn es vereinigt gleichzeitig knapp und 
prägnant gefaßte Einleitungstexte zu den einzelnen Kapiteln nach Art eines Lehrbuches mit einem gut 
durchdachten Lernprogramm mit Fragen und Musterantworten. Jeweils nach einigen Kapiteln ist für 
die einzelnen Lerneinheiten ein Lösungsschema eingeschaltet, an das sich Tests zur Selbstüberprüfung 
anschließen. 

Das Buch berücksichtigt die Strukturlehre, die biochemischen und physikalisch-chemischen 
Grundlagen und die Grundlagen des Zellstoffwechsels. Es dürfte als Unterrichtshilfe von großem 
Wert sein. D. STARcK, Frankfurt/M. 


Buchbesprechungen 63 


Austin, C. R.; SHORT, R. V.: Fortpflanzungsbiologie der Säugetiere. Bd. 4. Spezielle 
Aspekte der Fortpflanzung. Pareys Studientexte Nr. 9. Berlin und Hamburg: Paul Parey 
1981. 141 S., 57 Abb., 10 Tab. DM 30,-. 


Mit dem Erscheinen des vorliegenden Bandes wird die fünfbändige Serie ‚„‚Fortpflanzungsbiologie 
der Säugetiere‘ abgeschlossen. Damit liegt eine wertvolle Informationsquelle vor, die in gelungener 
Weise Ergebnisse vieler Einzeldisziplinen (Embryologie, Endokrinologie, Immunbiologie, Verhal- 
tensphysiologie etc.) in einer Synthese zusammenfaßt. Das Gesamtwerk ist hervorragend geeignet, 
auch dem Nichtspezialisten eine rasche und zuverlässige Übersicht über das praktisch und theoretisch 
gleich bedeutsame Gebiet zu vermitteln, zumal eine didaktisch geschickte Auswahl und Anordnung 
des Stoffes und eine klare und verständliche Sprache die Einführung auch in komplizierte Teilgebiete 
(Immunbiologie) leicht machen. Von besonderem Wert für den Mammalogen dürften im vorliegenden 
Bd. 4 die Beiträge von R. V. SHORT über spezialisierte Sonderfälle in der Fortpflanzungsbiologie der 
Säuger (Beuteltiere, Elefanten, Equiden) sein. D. STARcK, Frankfurt/M. 


SIEwING, R. (Hrsg.): Lehrbuch der Zoologie. Bd. 1: Allgemeine Zoologie. 3., völlig 
neubearb. Aufl. Stuttgart-New York: G. Fischer 1980. 910S., 723 Abb., 49 Tab., 
DM 98,-. 


In der vorliegenden 3. Auflage zeigt der frühere ‚„Wurmbach“ eine völlig neue Konzeption. Als 
Herausgeber stellt SıewinG ein Lehrbuch der allgemeinen Zoologie vor, welches in mancher Hinsicht 
von anderen Lehrbüchern gleicher Thematik abweicht. In diesem Zusammenhang ist auf das Anliegen 
im Vorwort hinzuweisen. Demzufolge wird dem Bestreben ausdrücklich entgegengewirkt, ein kurzes 
und gestrafftes, in ausgewählten Beispielen allzu exemplarisch darstellendes und stark schematisieren- 
des Lehrbuch zu schaffen. Statt dessen wird es vorgezogen, die große Mannigfaltigkeit tierischen 
Lebens und seiner Phänomene auf dem heutigen Wissensstand in kurzer und prägnanter Form 
aufzuzeigen. Es liegt somit in erster Linie ein studienbegleitendes Lehrbuch vor, das durch einen 
starken Umfang von 910 Seiten auffällt. 

Das Buch wird von 25 namhaften Autoren in folgenden Teilen und Abschnitten dargeboten: 1. 
Entstehung des Lebens auf der Erde (Biogenese); 2. Die Grundbausteine der Tiere: Zelle und Gewebe 
(Zelle, Besonderheiten der Protozoenzelle, Zellstoffwechsel, Bioelektrizität, Biologie der Gewebe); 3. 
Arterhaltung und Vererbung (Fortpflanzung, Embryologie, Genetik); 4. Der tierische Organısmus 
(Sinne, Bewegung, Ernährung, Respiration, Kreislaufsysteme, Exkretion, Osmo- und Ionenregula- 
tion, Steuerung); 5. Das Tier in seiner Umwelt (Verhalten, Ökologie); 6. Das Tier, ein geschichtlicher 
Organısmus (Evolution). Der größte Umfang entfällt auf die Teile 4 und 5, Sıewing selbst hat die 
Kapitel über Embryologie und Evolution bearbeitet. Die Darstellungen sind in allen Kapiteln klar und 
deutlich und auch für Studienanfänger verständlich. Besonders hervorzuheben sind weiterhin die sehr 
zahlreichen Abbildungen von hoher Qualität. 

Das Erscheinen dieses Buches ist in seinem Anliegen und in der Ausführung sehr zu begrüßen und 
kann jedem Studenten der Zoologie empfohlen werden. Aber auch die Lehrenden an Universitäten 
und Hochschulen werden auf diesen zusammenfassenden Band häufig zurückgreifen. Er stellt eine 
willkommene Bereicherung im deutschsprachigen zoologischen Lehrbuchangebot dar. 

D. KruskA, Hannover 


Joy, A.: A world like our own. Man and nature in Madagascar. New Haven, London: 
Yale University Press 1980. 272 S., zahlreiche Photos. £ 18,90. 


Die frühe Isolierung Madagaskars vom Kontineltalblock und die Sonderentwicklung seiner Flora 
und Fauna haben seit langem das Interesse der Biogeographen und Evolutionsforscher erregt und eine 
beachtliche Literatur entstehen lassen. Die gerade am Beispiel Madagaskars so überaus prekäre 
ökologische Situation findet, wenn auch spät, zunehmend Aufmerksamkeit. Dennoch füllt das 
vorliegende Buch eine Lücke aus. 

Die Autorin, durch langjährigen Aufenthalt in den so verschiedenen Regionen des Subkontinentes 
und durch ihre Forschungen an freilebenden Lemuren aufs beste ausgewiesen, hat ein Buch geschrie- 
ben, das in allgemein verständlicher Sprache und an zahlreichen Beispielen aus der persönlichen 
Erfahrung und den eigenen Erlebnissen bei voller Beachtung der wissenschaftlichen Qualität eine 
großartige Synthese aus Vegetationskunde und Fauna, Landeskunde und Volkskunde unter steter 
Einbeziehung der sozialen und ökonomischen Situation der Landesbewohner bringt. Die Gliederung 


64 Buchbesprechungen 


des Buches entspricht der für Madagaskar überaus deutlichen Aufteilung in Unterregionen, die sehr 
verschiedenen Lebensräumen entsprechen (Feuchtgebiete im Osten, Regenwald, Bergwald, Nebenin- 
seln Nosibe, Nosimangabe, depravierte Hochlandsavanne, Didiereaceenbusch des Südwestens, laub- 
abwerfender Trockenwald im Westen). Das Buch enthält eine Fülle von Details, auch zur Geschichte, 
Wirtschaft etc. Eine große Fülle von Photos, darunter ausgezeichnete Bilder von Tieren und Pflanzen 
(phot. Russ Kınne) erhöht den Wert des Buches, das jedem Säugetierkundler, darüber hinaus aber 
allen, die an Landeskunde und Fragen der Rettung der letzten Reste ursprünglicher Natur interessiert 
sind, wirklich empfohlen werden kann. Die ausgewogene Stellungnahme bei der Diskussion der 
Forderungen der Naturschützer und der Lebensnotwendigkeiten für den Menschen unter den 
derzeitigen Verhältnissen verdient besondere Anerkennung. D. STARcK, Frankfurt/M. 


MERKEL, F. W.: Orientierung im Tierreich. Grundbegriffe der modernen Biologie 15. 
Unter Mitarbeit von M. W. SCHÄFER. Stuttgart, New York: G. Fischer 1980. 279 S., 63 
Abb., 5 Tab. Kart. DM 48,-. 


Je mehr Beobachtungen und Experimente über die Orientierung der Tiere bekannt wurden, umso 
schwerer fällt die Orientierung auf diesem Gebiet. Bei den Tieren tragen zur Orientierung Sinnesor- 
gane bei, deren Leistungen der Mensch kaum nachempfinden kann, aber auch Fähigkeiten wie die der 
Zeitmessung oder der Magnetfeldperzeption, für die wir nicht einmal ein Organ kennen. Gründlich 
und vielseitig untersuchte Arten wie die Brieftaube sollen nach verschiedenen Versuchen die Sonne, 
Landmarken, das Erdmagnetfeld oder Düfte nutzen, wobei es zu den gleichen un auch 
zahlreiche negatıv verlaufene Experimente gibt. 

Das vorliegende Buch versucht, das Gebiet nach allgemeinen Gesichtspunkten wie der Richtungs- 
bestimmung, der Entfernungsbestimmung und der Lokalisation zu gliedern und schildert jeweils die 
wichtigsten Versuche. Der Autor, der vor allem über die Zugorientierung der Vögel gearbeitet und - 
zunächst gegen große Widerstände - auf eine Magnetfeldorientierung geschlossen hat, bemüht sich um 
eine allgemein verständliche Darstellung, die auch „biologisch Interessierte vom Imker, Sport- und 
Berufsfischer bis zum Reisetaubenliebhaber‘“‘ informieren möchte. Die verschiedenen Tiergruppen 
werden gleichrangig behandelt. Bei den Säugetieren stehen die Echoorientierung der Fledermäuse und 
die Heimkehrleistung im Vordergrund. Gewöhnlich, so bei Lachs und Aal, reichen die vielseitig 
ausgewählten, anschaulich wiedergegebenen und kritisch kommentierten Beispiele bis in die neueste 
Zeit. Bei den Fledermauswanderungen ist ausnahmsweise die neuere Literatur nicht mehr berücksich- 
gt. 

In der Form ist ein Zwischending zwischen einem Referat über den derzeitigen Wissensstand und 
einem Lehrbuch entstanden. Wer klare Antworten sucht, wird enttäuscht sein, was der Autor aber 
offenbar beabsichtigt. Anscheinend orientieren sich selbst Tiere derselben Art mit mehreren verschie- 
denen Mitteln, deren Anteil situationsbedingt wechselt. 

Der lebhafte und anschauliche Text wird durch eine Reihe instruktiver Abbildungen glücklich 
ergänzt. J- NIETHAMMER, Bonn 


REMANE, A.; STORCH, V.; WELSCH, U.: Kurzes Lehrbuch der Zoologie. 4., neubearb. 
Aufl. Stuttgart-New York: Gustav Fischer 1981. 537 S., 236 Abb. Gzl. DM 54,-, kart. 
DM 22,-. 

Studienhilfe Zoologie. Arbeitsbuch zu „Kurzes Lehrbuch der Zoologie‘. 4., neubearb. 
Aufl. Stuttgart-New York: Gustav Fischer 1981. 122 S., 1001 Fragen und Antworten, 
DM 16,80. 


Drei Jahre nach Erscheinen der 3. Auflage liegt das „Kurze Lehrbuch der Zoologie“ in 4. 
Neubearbeitung vor. Im Vergleich zur vorigen Auflage sind Aufbau und Gliederung im wesentlichen 
unverändert geblieben (Teil A: Allgemeine Zoologie, Teil B: Systematische Zoologie). In einer Reihe 
von Kapiteln sind Text und Abbildungen überarbeitet und dem aktuellen Kenntnisstand angepaßt 
sowie neue Abschnitte (Genmanipulation, Evolution der Eucyte) eingefügt worden, ohne daß 
dadurch der Gesamtumfang angewachsen ist. 

Die Autoren sind ihrem Vorsatz treu geblieben, eine kurze Gesamtübersicht über die Zoologie zu 
vermitteln. Dabei standen didaktische Überlegungen hinsichtlich der Verwendbarkeit des Buches für 
Studienanfänger im Vordergrund. Dieses Ziel ist in beachtenswerter Weise erreicht worden. 

Die „Studienhilfe Zoologie“, die eine Lern- und Orientierungshilfe darstellt, ist den Änderungen 
der 4. Lehrbuchausgabe angepaßt worden. R. LöHMER, Hannover 


Ökologie der Tiere 


Ein Lehr- und Handbuch in drei Teilen 
Von Professor Dr. Fritz Schwerdtfeger 


Band I: Autökologie 


Die Beziehungen zwischen Tier und Umwelt. 2., neubearbeitete Auflage. 1977. 460 Seiten mit 
268 Abbildungen und 55 Übersichten. Leinen 120,— DM 


Der erste Band behandelt die Beziehungen des Tieres als Individuum oder Repräsentant seiner 
Art zu den Gegebenheiten der Umwelt. Vorangestellt ist eine kurze Erörterung der Grundlagen 
der Ökologie, insbesondere ihrer Grundbegriffe. Der eingehenden Schilderung der Tier- 
Umwelt-Beziehungen ist die Gliederung der Umweltkomponenten in abiotische, trophische und 
biotische Faktoren zugrunde gelegt. Die Einflüsse der einzelnen Faktoren, beispielsweise des 
Lichtes, der Nahrungsmenge oder der Parasiten, auf den Bau, die Leistungen und das 
Verhalten des Tieres werden unter dem Gesichtspunkt vorgeführt, die Relationen eindeutig 
zahlen- und kurvenmäßig zu erfassen. 


Band Il: Demökologie 


Struktur und Dynamik tierischer Populationen. 2., neubearbeitete Auflage. 1979. 450 Seiten mit 
249 Abbildungen und 55 Übersichten. Leinen 120,- DM 


Der zweite Band behandelt nach einer Einführung zum Wesen der Demökologie und der 
Population in statistischer Sicht die Struktur der Populationen, ihre Zusammensetzung aus 
Elementen wie Dichte, Verteilung oder Altersaufbau sowie deren Beeinflussung durch Um- und 
Mitweltkomponenten. Die zu ihrer Erfassung tauglichen Methoden werden jeweils kurz vor- 
geführt. Gegenstand des zweiten Hauptteils ist die Dynamik der Populationen, die zeitliche 
Veränderung ihrer Strukturelemente; dabei werden die Wandlungen der beiden Grundelemente 
Verteilung und Dichte, also die Dispersions- und Abundanzdynamik, besonders eingehend 
behandelt. 


Band Ill: Synökologie 


Struktur, Funktion und Produktivität mehrartiger Tiergemeinschaften. Mit einem Anhang: 
Mensch und Tiergemeinschaft. 1975. 451 Seiten mit 118 Abbildungen und 125 Übersichten. 
Leinen 98,— DM 


Der Autor erläutert im dritten Band einführend das Wesen der Synökologie im allgemeinen 
sowie speziell das der Synökologie der Tiere. Ihr sind drei Hauptabschnitte gewidmet. Der erste 
befaßt sich mit der Struktur der Tiergemeinschaften, mit ihrer Entstehung, ihren Grundformen, 
Merkmalen und Kennzeichnungen sowie mit ihrer Dynamik. Thema des zweiten Abschnitts ist 
die Funktion der Tiergemeinschaften, ihr Inbetriebsein mit den dafür notwendigen Vorausset- 
zungen und Vorgängen, die in einem Funktionsplan analysiert und in der Schilderung des 
realen Funktionsgeschehens vorgeführt und gewertet werden. Der dritte Teil handelt von der in 
jüngster Zeit in den Vordergrund des Interesses gerückten Produktivität der Tiergemeinschaf- 
ten, namentlich von den Komponenten ihres Stoff- und Energiehaushalts, wie z. B. Konsum- 
tion, Assimilation, Produktion und Respiration; den Abschluß bildet der Versuch einer Stoff- und 
Energiebilanz der Tiergemeinschaften. 


BaERERNG7 PAUL PAREY :- HAMBURG UND BERLIN 


Erscheinungsweise und Bezugspreis: Die Zeitschrift erscheint alle 2 Monate; 6 Hefte bilden einen Band 
mit insgesamt 384 Seiten. Der Abonnementspreis beträgt je Band 232,- DM zuzüglich Porto. Das 
Abonnement verpflichtet zur Abnahme eines ganzen Bandes. Es verlängert sich stillschweigend, 
wenn nicht unmittelbar nach Erhalt des letzten Heftes eines Bandes Abbestellung erfolgt. 
Einzelbezugspreis der Hefte: 43,- DM. Die Preise verstehen sich ım Inland incl. Mehrwertsteuer. 
Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, 
Spitalerstraße 12, D-2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesell- 
schaft für Säugetierkunde“ erhalten die Zeitschrift unberechnet im Rahmen des Mitgliedsbeitrages. 


CANADIAN JOURNAL OF PLANT SCIENCE 
CANADIAN JOURNAL OF ANIMAL SCIENCE 
CANADIAN JOURNAL OF SOIL SCIENCE 


Each of the three Journals is published quarterly by the Agricultural 
Institute of Canada. The journals contain papers in English and 
French presenting original research findings relating to crops, soils 
and farm animals and their products. Papers are written by 
scientists from Canada and abroad and are reviewed for publication 
by respected members of the Canadian agricultural research 
community. The journals are distributed in more than fifty countries 
of the world. 


Canadian Journal of Plant Science 
Approximately 1500 pages annually. Published quarterly. Size 16.5 cm X 24.5 
cm. Annual subscription rate for individuals Can$19.00. Institutions Can$33.00. 


Canadian Journal of Animal Science 
Approximately 1100 pages annually. Published quarterly. Size 16.5X 24.5 cm. 
Annual subscription rate for individuals Can$19.00. Institutions Can$33.00. 


Canadian Journal of Soil Science 
Approximately 800 pages annually. Published quarterly. Size 16.5 cm X 24.5 
cm. Annual subscription rate for individuals Can$19.00. Institutions Can$33.00. 


Please enter my subscription for the journal(s) indicated below. 


D Canadian Journal of Plant Science 
DD Canadian Journal of Anımal Science 
D Canadian Journal of Soil Science 


All sübscriptions are on a full calendar year basis. You will receive all issues published ın the calendar 
year in which you subscribe. 


NAME 

ORGANIZATION 

ADDRESS 

CITY. „ Sen man un Rah at a en ei PRENNBEASITÄIE 
POSTALGCODE 7 eurer er Fr E Bee FREIE SOU NR 


Please include a check or money order in Canadian funds for the full amount and mail to: Agricultural 
Institute of Canada, Journals Subscriptions, Suite 907, 151 Slater Street, Ottawa, Ontario, KIP 5H4, 
Canada. 

bosmsesznnmsanmnnanHaHRRBEREREBAHEZABERBEESERRERABERmM 


Past issues of all three journals are available in microfilm from: University Microfilms 
International, 400 North Zeeb Road, Ann Arbor, Michigan 48106 USA. 


Vol. 47 (2), 65-128, April 1982 ISSN 0044-3468 C 21274 F 


SAUGETIERKUNDE 


In unsere Leser — To our Readers | 


sal, B. R.: Reproductive ecology of the Rufous elephant-shrew, Elephantulus rufescens (Macroscelididae), in 
| Kenya. — Die Fortpflanzungsökologie der Elefantenspitzmaus, Elephantulus rufescens (Macroscelididae), in 
Kenia 65 


lb, A.: Putzen und Putzverhalten bei Rhinolophus ferrumequinum. — Cleaning and the conduct of cleaning in 
, Rhinolophus ferrumequinum 72 


hliemann, H.: Nachweis wilder Meerschweinchen für Ecuador und Beschreibung von Cavia aperea patzelti 
,subsp. nov. - A first record of Wild cavies for Ecuador and description of Cavia aperea patzelti subsp. nov. 79 


lliainen, E.: Scent-marking in the pine marten (Martes martes) in Finnish Forest Lapland in winter. —- Duftmarkie- 
‚rung des Baummarders (Martes martes L.) im Wald-Lappland Finnlands im Winter 91 


lartys, M.: Gehegebeobachtungen zur Geburts- und Reproduktionsbiologie des Europäischen Wildschweines 
‚(Sus scrofa L.). - Observations on parturition and reproductive biology in captive European wild boars (Sus 
‚ scrofa L.) 100 


\issenschaftliche Kurzmitteilungen 


raets, Adelheid: Magendrüsenlänge und -dichte bei Crocidura russula and Suncus etruscus (Mammalia: 
\Soricidae). - Length and density of stomach glands in Crocidura russula and Suncus etruscus (Mammalia: 
‚Soricidae) 113 


hitzenberger, Friederike: Der Riesenabendsegler (Nyctalus lasiopterus) in Afrika nachgewiesen. — Giant Noctule 
(Nyctalus lasiopterus) found in Africa 115 


ıcedo, H. de: Note on Vicuna x Alpaca hybrids. - Bemerkung über Hybriden von Vicuna x Alpaka 17 


hmiat, U.; Manske, U.: Thermopräferenz bei der Gemeinen Vampirfledermaus (Desmodus rotundus). - Thermo- 
‚Preference in the common vampire bat, (Desmodus rotundus) 118 


Ekanntmachungen 120 


Echbesprechungen \ 121 


HERAUSGEBER/EDIT®OBS 


P. ]. H. van BrREE, Amsterdam — W. FIEDLER, Wien - H. Frıck, München -— W. HERRE, 

Kiel - H.-G. Kırös, Berlin - H.-J. Kuhn, Göttingen - B. Lanza, Florenz - T. C. S. 

MOoRRISON-SCOTT, London - J. NIETHAMMER, Bonn - H. ReıcHsTtein, Kiel- M. Rönss, 

Hannover - D. STARcK, Frankfurt a. M. - F. Strauss, Bern - E. THeEnIus, Wien — 
W.VERHEYEN, Antwerpen 


SCHRIFTLEITUNG/EDITOMIATIOFFIGE 


H. SCHLIEMANN, Hamburg —- D. Kruska, Hannover 


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. Verantwortli- 
cher Schriftleiter im Sinne des Hamburgischen Pressegesetzes ist Prof. Dr. Harald Schliemann. 


Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z.Hd. Priv.-Doz. Dr. 
Dieter Kruska, Institut für Zoologie, Tierärztliche Hochschule Hannover, Bünteweg 17, D-3000 
Hannover 71. Für die Publikation vorgesehene Manuskripte sollen gemäß den „Redaktionellen 
Richtlinien‘ abgefaßt werden. Diese Richtlinien sind in deutscher Sprache Bd. 43, H. 1 und in 
englischer Sprache Bd. 43, H. 2 beigefügt; ın ihnen finden sich weitere Hinweise zur Annahme von 
Manuskripten, Bedingungen für die Veröffentlichung 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 
Vervielfältigung auf photomechanischem oder ähnlichem Wege oder ım Magnettonverfahren 
sowie der Speicherung ın Datenverarbeitungsanlagen, bleiben, auch bei nur auszugsweiser Ver- 
wertung, vorbehalten. Werden von einzelnen Beiträgen oder Beitragsteilen einzelne Vervielfälti- 
gungsstücke in dem nach $54 Absatz 1 UrhG zulässigen Umfang für gewerbliche Zwecke 
hergestellt, ist dafür eine Vergütung gemäß den gleichlautenden Gesamtverträgen zwischen der 
Verwertungsgesellschaft Wort, vereinigt mit der Verwertungsgesellschaft Wissenschaft GmbH, 
rechtsfähiger Verein kraft Verleihung, Goethestr. 49, D-8000 München 2, und dem Bundesver- 
band der Deutschen Industrie e.V., dem Gesamtverband der Versicherungswirtschaft e.V., dem 
Bundesverband deutscher Banken e.V., dem Deutschen Sparkassen- und Giroverband und dem 
Verband der Privaten Bausparkassen e.V., an die Verwertungsgesellschaft zu entrichten. Erfolgt 
die Entrichtung der Gebühren durch Wertmarken der Verwertungsgesellschaft, so ist für jedes 
vervielfältigte Blatt eine Marke im Werte von 0,40 DM zu verwenden. Die Vervielfältigungen sind 
mit einem Vermerk über die Quelle und den Vervielfältiger zu versehen. 


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 internal use 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 copyıng 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 
BAE 


Fortsetzung 3. Umschlagseite 


© 1982 Verlag Paul Parey, Hamburg und Berlin. - Printed in Germany by Westholsteinische Verlagsdruckerei 
Boyens & Co., Heide/Holst. 


Z. Säugetierkunde 47 (1982) 2, 65-128 


InterCode: ZSAEA 7 47 (2), 65-128 (1982) 


Mit einer Beilage des Verlages Paul Parey 


le, 


Ä 


An die Leser der Zeitschrift für Säugetierkunde 


Aufgrund der weltweiten Rezession der Wirtschaft sind die Anschaffungs-Etats vieler 
wissenschaftlicher Bibliotheken im Jahre 1982 deutlich gekürzt worden. Die Zahl der von 
der jeweiligen Bibliothek abonnierten wissenschaftlichen Forschungszeitschriften wird 
reduziert werden. Damit sinkt die Verkaufsauflage der einzelnen Zeitschrift, die von 
Abbestellungen betroffen ist. Bei sinkender Verkaufsauflage steigen die Einzelstückkosten 
der Zeitschrift und damit die Preise - weitere Abbestellungen und damit der wirtschaftliche 
Zusammenbruch der Zeitschrift können die Folge sein. 

Der einzelne Leser hat diese Situation bisher kaum erkannt, da er in vielen Fällen schon 
heute nicht die Original-Zeitschrift benutzt, sondern von seiner Bibliothek anstelle des 
Original-Exemplars Xerokopien erhält. 

Man muß sich aber bewußt machen, daß die wissenschaftliche Forschungszeitschrift als 
das zentrale Medium für Befund-Dokumentation und internationalen Informationsaus- 
tausch weder heute noch in absehbarer Zukunft durch andere Medien ersetzbar ist. Weder 
Mikrofilm-Archive noch elektronische Non-Print-Medien sind eine Alternative. 

Wenn Sie unserer Argumentation soweit gefolgt sind, werden Sie uns zustimmen, daß 
auch die ökonomischen Voraussetzungen für ein Überleben der wissenschaftlichen For- 
schungszeitschriften geschaffen werden müssen. Dies läßt sich am einfachsten erreichen, 
wenn — wie erläutert - verhindert wird, daß Zeitschriften durch Bibliotheken abbestellt 
werden. 

Unser heutiger Aufruf an Sie ganz persönlich geht dahin, daß Sie möglichst bald 
mit Ihrem Bibliothekar sprechen und ihm deutlich machen, daß die schnelle Verfügbarkeit 
der Zeitschrift innerhalb Ihrer Bibliothek für die Effektivität Ihrer Forschungsarbeit sowie 
für die Ausbildung des wissenschaftlichen Nachwuchses von ausschlaggebender Bedeu- 
IR tung ist. 

Wir hoffen auf einen Erfolg unserer gemeinsamen Bemühungen. 


DıE HERAUSGEBER 


- To the Readers of the International Journal of Mammalian Biology 


Because of worldwide recession, the acquisition budget of many scientific libraries 
 decreases considerably in 1982. The number of scientific research journals, subscribed by 
© libraries, will be reduced, thereby also the edition-run of the individual journals affected by 
the cancellation. By decreasing edition-runs, the cost per copy increases and consequently 
the retail price - more cancellations and finally the economic failure will follow. 

The individual reader has not yet recognized the situation because in many cases, he 
does not use the original journal but gets a photostat of the original copy from the library. 
We have to realize, however, that the scientific research journal as the central medium 


I 


for documentation of research results and international exchange of information cannot be 
replaced today or in the near future by any other medium. Neither micro-film archives nor 
electronic non-print media are alternatives. 

If you have followed our line of reasoning thus far, you have to agree that in order for 
scientific research journals to survive, economic basic requirements must be organized. The 
easiest way would be, as we have explained, when libraries can be prevented from 
cancelling their subscriptions. 

Our appeal today to you personally is to the effect that you go to your librarian as soon 
as possible and explain to hım or her that the availability of the journal at your library is 
essential for the efficacy of your research work. 

We do hope that our joint efforts will be successful. 


THE EDIToRs 


Reproductive ecology of the Rufous elephant-shrew, 
Elephantulus rufescens (Macroscelididae), in Kenya 


By B.R. Near 


Department of Biology, University of Saskatchewan, Saskatoon, Canada 


Receipt of Ms. 25. 11. 1981 
Abstract 


The reproductive ecology of Elephantulus rufescens was studied in Meru National Park, Kenya during 
the period August 1974 to June 1975. 174 animals were collected and dissected. The body weights, 
condition of the reproductive organs, and extent of tooth wear provided information on seasonal 
changes in breeding activity, litter size, and age structure. 

The climate was very seasonal. Rainfall was confined to two periods, from November to December 
and mid-April to mid-May. Temperature increased during the dry seasons and decreased during the 
rains. However, no seasonal changes were detected in adult body mass, breeding activity, litter size, or 
age structure. 

Approximately three quarters of the adult females carried implanted embryos. The average interval 
between litters was estimated to be 62 d, so that adults had the capacity to produce just under 6 litters 
each year. The average litter size was 1.41 # 0.07 (range 1-2), and was positively correlated with 
maternal body mass. The potential annual production of young per adult female was estimated to be 
8.3 

The reproductive strategy of elephant-shrews is briefly discussed. 


Introduction 


The specific classification of elephant-shrews can be considered to be nearly definitive 
(CoRBET and Hanks 1968; CoRBET 1971). The famıly Macroscelididae, which includes 
four genera and fifteen species, is an extremely well defined taxon endemic to Africa. There 
is general agreement that the elephant-shrews are amonophyletic group not closely related 
to any other group of living mammals. Their affınity with the Insectivora and Primates is 
highly debatable (CorBET and Hanks 1968; RATHBUN 1979). 

The literature on the breeding biology of elephant-shrews has been reviewed by AsDELL 
(1964), Brown (1964), RATHBUN (1979), and Trırp (1971, 1972). Elephant-shrews are 
monogamous (RATHBUN 1979) and spontaneous ovulators (VAN DER Host 1954). Little is 
known about the length of the breeding season in most species. Breeding is confined to sıx 
months of the year in Elephantulus myurus in South Africa, and both males and females are 
in anoestrus during the remainder of the year (STocH 1954; Van DER HorsT 1946, 1954). 
Breeding occurs throughout the year in E. rufescens and Rhynchocyon chrysopygus in 
Kenya (RATHBUNn 1979), and is thought to be continuous in E. rupestris in Namibia 
(WITHERS 1979). However, samples were too small to determine seasonal changes in the 
breeding rate of these species. After a gestation period of 42 t0 56d (RaTHBun 1979; TRIıPP 
1972; Van DER Horst 1946) females give birth to a litter of 1-2 precocial young (RATHBUN 
1979). Further development of the young is rapid (RATHBUN 1979; RATHBUN and RepD- 
FORD 1981) and breeding can occur at an early age in some species (VAN DER HorsT 1946, 
1954; Trıpp 1972). 

This paper describes the breeding biology of a large sample of E. rufescens collected in 
the Meru National Park, Kenya. It complements the work of RATHBun (1979) which was 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4702-0065 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 65-71 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


66 B. R. Neal 


based on visual observations of a small group of E. rufescens ın a similar habitat 
approximately 250km south of my study area. I investigated seasonal changes in breeding 
activity and discuss the reproductive strategy of elephant-shrews. 


Material and methods 


A total of 174 animals were collected from a 2km? area immediately south of Rainkombe (0°07’N, 
38°12’E) during the period August 1974 to June 1975. Approximately half of the animals were 
collected using standard trap lines of Reporter break-back traps, and the remainder collected using 
Sherman live traps. 

Animals were measured, weighed and dissected soon after collection. An evicerated weight was 
obtained by removing the gut, from the lower oesophagus to the rectum, and also the uterus and 
embryos of pregnant females. 

The female reproductive tract was examined macroscopically, with the aid of a dissecting 
microscope where necessary. Anımals were classified as immature ıf no placental scars or corpora lutea 
were observed. Corpora lutea are particularly easy to observe in elephant-shrews because they are 
everted from the surface of the ovary (Trırp 1971). Mature females were classed as either 1. 
reproductively active, if visibly pregnant (i.e. with implanted embryos) or corpora lutea were 
observed; or 2. reproductively inactive, ıf no corpora lutea were visible. Embryos were counted, 
measured and weighed after fixation. A representative sample was also weighed before fixation to 
correct for the loss in weight due to the fixation process. 

A sperm smear was taken from the epididymides of each male. Individuals were classed as 
immature if no sperm were present, and as adults if sperm were abundant. The testes were weighed 
after fixation. 

The relative age of individuals was estimated using the eruption and subsequent wear ofthe upper 
molars and premolars as a criterion. 

Conception dates were estimated by ageing litters by the method of HuGGETT and Wıpnas (1951), 
assuming a birth mass equal to that of the heaviest embryo, a gestation period of 50 d (RATHBUN 1979; 
RATHBUN and REDFORD 1981), and that embryos attained a measurable body mass on the seventh day 
of gestation. 

The average interval between litters, and potential annual production of young per adult female 
were estimated by the method described by Near (1981). In this method the ratio of adult females 
with implanted embryos (N,) to adult females not visibly pregnant (N, + Nj) is assumed to equal the 
ratio of the duration of the post-ımplantation period of gestation (p,) to the sum of the pre- 
implantation period (p,) and the period between pregnancies (p,). In E. rufescens the equation ıs solved 
for p, by assuming that the length of the gestation period (pı + p,) ıs 50d (RATHBUN 1979) and the 
pre-implantation period (pj) ıs 4d. The mean interval between litters (p, + pı + p,) can be easily 
calculated, and the production of young is then estimated using the average lıtter sıze. 


Results 
Habitat and habits 


Elephant-shrews were confined to the area of deciduous Acacia/Commiphora bushland in 
the southern half of Meru National Park. The vegetation mainly consisted of a wide variety 
of tall Acacia and Commiphora bushes, and in some areas there was a thicker understory of 
Banhinia taitensis, Grewia villosa, Combretum aculeatum and other species of bushes 
(AMENT 1975). There were a few scattered Acacia spp, Terminalia app, Sterculia afrıcana, 
Delonix elata and Adansonia digitata trees. There was no herbaceous layer for most of the 
year, and much of the ground was badly eroded. During the dry seasons (i.e. for 8-9 
months of the year) the bushes were leafless and the whole area appeared dry and dead. 
However, during the rains (Fig. 1D) the bushes came into leaf, sparse patches of various 
annual grasses and dicotyledons appeared, and the whole area looked lush and green. 
The presence of E. rufescens was obvious because of their network of trails. My 
observations of their habits are consistent with those of RATHBun (1979). Adults occurred 
in pairs with either one or two young. They remained above ground and their movements, 
which were extremely rapid when disturbed, were almost entirely restricted to their trail 


Reproductive ecology of Elephantulus rufescens in Kenya 67 


system. They seemed oblivious to objects placed on their trails, running repeatedly into 
closed traps before moving off the trail to avoid the obstruction. Only occasionally would 
they take refuge in a hollow log or hole in the ground. The distribution of territories was 
very clumped. There were usually three or four families living more or less adjacent to one 
another, with large gaps of 100m or more between groups. There seemed to be no 
association between their distribution and the vegetation, so their absence from most of the 
available habitat may be due to the presence of spiny mice (Acomys spp) which are very 
abundant in the area (NEAL in press). WITHERS (1979) has shown that E. rupestris ın 
Namibia is poorly correlated in distribution with any plant, and appears to avoıid Aethomys 
namaquensis. 


Body size 


The mass of the heaviest embryo of a litter with a single embryo was 10.4 g; that of a litter 
of two embryos was 7.8g. The lightest juveniles collected had a body mass of 14-158. 
Derany (1964) records a female carrying a single embryo of 16g in Uganda, and WALKER 
(1955) records the body mass within 3 d of birth to be just less than 12 g. The birth mass of 
E. intufi ıs recorded as 10 g (range 8.5-10.5 g) and of E. myurus as 8.1 # 0.68 (Trıpp 1972). 

Both sexes matured at a body mass of 32-35 gand apparently atan early age. All anımals 
in which the last upper molar was partly or wholly erupted (age class II or older) were 
mature, as were 2 of 27 animals ın which the last upper molar had still not emerged through 
the surface of the jaw (age class I). Adult females usually attained a body mass of 45-65 g, 
and males a body mass of 35-55g. Adult body mass was similar to that recorded for E. 
intufi, E. rozeti and E. brachyrhynchus, and a little less than E. myurus (SMITHERS 1971; 
Trıpp 1972). 


Breeding season 


E. rufescens bred throughout the year (Fig. 1B), and there was no significant seasonal 
varlatıon in the testes mass of adults (Fig. 1A). Pregnancy rates in the wet and dry seasons 
were independent of season (x? = 1.11; P= 0.3) when tested by chi-squared analysis of a 
2 X 2 contingency table (SoKAL and ROoHLF 1969). 

RATHBUn (1979) also found that E. rufescens bred throughout the year ın Kenya, and 
VAN DER Host (1954) reports that HooGSTRAAL found this species breeding throughout 
the year in the southern Sudan. The suggestion by BRown (1964) and Kıncpon (1974) that 
breeding is seasonal in E. rufescens ıs based on inadequate data. E. myurus breeds during a 
sıx month period, from August to January, in the Transvaal (Van DER HorsT 1946, 1954). 
Its breeding season appears to be extended in Botswana because pregnancies have been 
recorded in September, March and April (SMITHERS 1971). SMITHERS (1971) also suggests 
that E. intufi breeds during the warmer, wetter months of the year, from August to 
February, in Botswana; and E. rupestris is thought to breed throughout the year in 
Namibia (WITHERs 1979). 


Litter sıze 


The mean number of embryos of 46 pregnancies was 1.41 # 0.07 (range 1-2). Two of these 
females also appeared to have a second resorbing embryo at an early stage of gestation. 
Variation in litter size was analysed in relation to the evicerated maternal body mass and 
season of conception by analysis of covariance (SNEDECOR and CocHRAN 1967). The 
sample was divided into those litters conceived during the rains and immediately after 
(November to early January, mid-April to early June), and those litters conceived during 
the dry seasons. Litter size was not significantly different in the wet and dry seasons 
(F} 43 = 0.0002; P> 0.75), but increased significantly with increase in maternal body mass 
(Fı4 = 7.34; P = 0.01). Females with an evicerated body mass of less than 40 all carried 


68 B. R. Neal 

> Elephantulus rufescens 

®) 

n 

(dp) 

(18) 

> 

n 

® 

B 

F 

® 

= 

tz) 

< 

& 

er) 

S 

a 

Ex 

S Fig. 1. Seasonal changes in reproduc- 

= tive activity of E. rufescens. Numbers 

® in Figs. A-C indicate sample sizes. A: 
Testes mass of males in age class V or 

af older. Data points represent means 

E 2x SE. B:Bercentaeegotssdule 

E females reproductively active 

= (O--O), and visibly pregnant 

o (e—-e). C: Percentage of anımals in 

S first age class. D: Maximum and 

X minimum daily temperature, and 


rainfall (histogram) 


single embryos (n = 9), whereas those over 40g had a mean litter size of 1.51 # 0.08 
(n= 37). 

Estimates of the mean litter size of E. rufescens, based on small samples, range from 1.0 
to 1.5 (Brown 1964; DELANY 1964; RATHBUN 1979; RATHBUN and REDFORD 1981; TRıPP 
1971), and give a combined estimate of 1.35 + 0.08 (n = 34). RATHBUN (1979) also noted 
that litter size increased as the mothers increased in age. The mean numbers of embryos in 
various species for which large samples are available are as follows: In 36 E. rozeti from 
northwest Africa, 2.42 # 0.11, range 1-4 (Trıpp 1971); in 26 E. intufi from the Transvaal, 
Botswana and Namibia, 1.62 + 0.11, range 1-3 (SHORTRIDGE 1934; SMITHERS 1971; TRIPP 
1971, 1972); in 20 E. brachyrhynchus from East Africa, Zambia and Botswana, 1.60 # 0.11, 
range 1-2 (BRown 1964; SHORTRIDGE 1934; SMITHERS 1971; Trıpp 1971); and in 27 E. 
myurus from the Transvaal and Botswana, 1.85 + 0.07, range 1-2 (SMITHERS 1971; TRIPP 
1971, 1972). VAN DER Horst (1944) reports that only 1 percent of E. myurus from the 
Transvaal carried single embryos. The maximum litter size appears to be restricted to two 
young in most species because implantation is restricted to a single implantation site ın each 
horn of the uterus (Trırp 1971). 


Breeding rates and production of young 


There was no seasonal varıation in either the breeding rate (see Breeding season) or litter 
size (see Litter size). Consequently, it is inferred that there was no seasonal variation ın the 
number of young entering the population. The percentage of anımals in age class I varied 
irregularly (Fig. 1C) and showed no seasonal trend. The proportion of young in the wet 
and dry seasons was independent of season (x? = 0.48; P = 0.49) when tested by chi- 
squared analysis of a2 X 2 contingency table (SoKAL and RoHLr 1969). 

Approximately three quarters (46 of 62) of adult females were visibly pregnant. The 


Reproductive ecology of Elephantulus rufescens in Kenya 69 


average interval between litters was calculated to be approximately 62 d, so that adults had 
the capacity to produce just under sıx litters each year. Assuming an average litter size of 
1.41 (see Litter size) the potential annual production of young per adult female was 
approximately 8.3. 

RATHBUN (1979) observed a mean interval of 61 # 1.23 d (range 56-65 d) between 7 
consecutive litters of this species in an area 250 km south of my study area. The potential 
production of young per adult female was very similar to my estimate because the average 
litter size was not significantly different in the two areas. VAN DER Host (1954) estimated 
that E. myurus gave birth to three litters during its six month breeding season in South 
Africa. This translates to an average interval of approximately 61 d between litters and a 
potential annual production of young per adult female of just under 6. Finally, RATHBUN 
(1979) observed an average interval of 81 d between 11 consecutive litters of Rhynchocyon 
chrysopygus ın Kenya. There was only one offspring per litter, and so the potential annual 
production of young was 4.5 per adult female. 


Table 


Comparison of life history characteristics of the four genera of elephant-shrews 


(Data summarized from Brown [1964], CoRBET and Hanks [1968], Rarugun [1979], 
VAN DER Host [1954] and this paper) 


Rhynchocyon Petrodromus Elephantulus Macroscelides 
Habitat Lowland/montane Forest, thickets, Open savanna Subdesert 

forest. Thick dense savanna woodland, steppe, 

riverine bush woodland subdesert 


Climate 

Environmental stability 

Body sıze (mass) c. 400 g c. 200 g 

Longevity 3-4 yr ? 

Age Ist reprod. Late Late? 

Breeding season All year All year? 6 months - all year 
Interval between litters 82 d ? 61-62 d 


Litter sıze 1-2 usually 1 1-2 
8 


Prod. young/female/year 4.5 6-8.3 


Discussion 


The key to understanding the reproductive strategy of Elephantulus, and perhaps all 
elephant-shrews, seems to lie in their ability to give birth to precocial young. This ability 
appears to be a secondary evolutionary development in mammals (Case 1978; LILLEGRA- 
vEN 1975), and so it is interesting to speculate on the reasons for this development. 
Elephant-shrews live above ground and most species do not build nests (RATHBUN 1979). 
They also avoid predation, in part, by running away (RATHBUN 1979) so that neonates 
must be well developed. These habits are probably associated with the evolution of 
precocial young. It is interesting to note that in Rhynchocyon, which is the only genus of 
elephant-shrews that builds nests, the neonates are not as advanced as those of the other 
three genera (RATHBUN 1979). 

The long gestation period of 42-56d (RatHsun 1979; Trıpp 1972; VAN DER HORST 
1946) and small litter size (usually 1-2) are probably associated with the development of 
precocial young (SACHER and STAFFELDT 1974). One consequence of this developmental 
strategy is that the rate of production of young per adult female is very low (less than 10 
per year) compared to most other small mammals (French et al. 1975). 


70 B. R. Neal 


To compensate for this low birth rate elephant-shrews generally breed continuously and 
have an extended period of reproductive life. The latter is achieved by increasing adult 
survivorship and decreasing the age of first reproduction (Table). The available evidence 
suggests there is an interaction of longevity and age of first reproduction with body size. 

The life history characteristics of the different genera (Table) agree with conventional 
ideas on r- and K-selection (PıankA 1970; STEARNS 1976), although the meaning of the 
concept used here may be more appropriately called birth and death rate selection (PARRY 
1981). Species inhabiting moister, less seasonal environments appear to be more K-selected 
(1.e. have lower birth and death rates) than those inhabiting arıd, more seasonal environ- 
ments. I have noted a similar set of relationships in the unstriped grass rat, Arvicanthis 
(NEAL 1981). 

The reproductive characteristics of precocial young, long gestation period, small litter 
size, continuous breeding, and early maturation are also exhibited by the spiny mice, 
Acomys (NEAL ın press), and the laminate-toothed or vleı rats, Otomys (Davıs and 
MEESTER 1981; PERRIN 1980). These genera either do not build nests, or their nests are very 
rudımentary (KınaDon 1974). Thus, these species and elephant-shrews may have evolved 
their reproductive strategy for similar reasons. 


Acknowledgements 


The field work in Kenya was carried out during a sabbatical leave from the University of Saskatche- 
wan. I wish to thank the former Director and Trustees of the Kenya National Parks (now a sub- 
department in the Ministry of Tourism and Wildlife of Kenya) for permission to study in Meru 
National Park, the Park staff for their help and cooperation, and Prof. F. MUTERE of Kenyatta 
University College for making my stay in Kenya possible. It is a pleasure to thank the Park Warden, 
Mr. P. JEnkıns, for providing excellent facilities and help for the field and initial laboratory work; the 
Park Rangers, JosEPpH NAMERIO and THoMAs NTOBURI, for their help with the field work; and my 
wife, Jenny, for help with the field and laboratory work and critically reading an earlier draft of this 
paper. Finally, it is a pleasure to thank Drs. C. RarpH and B. WUNDER of the Department of Zoology 
and Entomology, Colorado State University, for the use of the facilities of the Department during the 
completion of this work. 


Zusammenfassung 


Die Fortpflanzungsökologie der Elefantenspitzmans, Elephantulus rufescens (Macroscelididae), in Kenia 


Die Fortpflanzungsökologie von Elephantulus rufescens wurde im Meru-Nationalpark in Kenia im 
Zeitraum von August 1974 bis Juni 1975 untersucht. Insgesamt 174 Tiere wurden gefangen und 
seziert. Körpergewicht, Status der Fortpflanzungsorgane, und der Grad der Zahnabnutzung lieferten 
Informationen über Fortpflanzungsaktivität, Wurfgröße und Altersstruktur. 

Das Klima schwankte je nach Jahreszeit erheblich. Die Niederschläge beschränkten sich auf 2 
Perioden, von November bis Dezember und von Mitte April bis Mitte Maı. Die Umgebungstempera- 
turen stiegen während der Trockenzeiten an und fielen in den Regenzeiten. Es konnten jedoch keine 
jahreszeitlichen Veränderungen in bezug auf das Körpergewicht der adulten Tiere, die Fortpflan- 
zungsaktivität, Wurfgröße oder Altersstruktur festgestellt werden. 

Annähernd Dreiviertel der adulten Weibchen trugen implantierte Embryonen. Der Zeitraum 
zwischen den Würfen wurde auf durchschnittlich 62 Tage geschätzt, so daß die adulten Tiere eine 
Kapazität von knapp 6 Würfen pro Jahr hatten. Die durchschnittliche Wurfgröße war 1,41 # 0,07 
(Bereich 1-2) und korrelierte positiv mıt dem Körpergewicht des Muttertieres. Die mögliche jährliche 
Anzahl von Jungen pro adultem Weibchen wurde auf 8,3 geschätzt. 

Die Fortpflanzungsstrategie der Elefantenspitzmaus wird kurz diskutiert. 


References 


AMENT, J. G. (1975): The vascular plants of Meru National Park, Kenya. ]J. E. Afr. Nat. Hist. Soc. 
154, 1-33. 

ASDELL, $. A. (1964): Patterns of Mammalian Reproduction. 2nd Ed. Ithaca: Cornell University 
Press. 

Brown, J. C. (1964): Observations on the elephant-shrews (Macroscelididae) of equatorial Africa. 
Proc. zool. Soc. Lond. 143, 103-119. 

Case, T. J. (1978): Endothermy and parental care in the terrestrial vertebrates. Am. Nat. 112, 861-874. 


Reproductive ecology of Elephantulus rufescens in Kenya 71 


CoRBET, G. B. (1971): Family Macroscelididae. In: The mammals of Africa: an identification manual. 
Ed. by J. MEEsTER; H. W. SETZER. Washington: Smithsonian Institution. 

CoRBET, G. B.; Hanks, J. (1968): A revision of the elephant-shrews, family Macroscelididae. Bull. 
Brit. Mus. Nat. Hist. (Zool.) 16, 45-111. 

Davıs, R. M.; MEESTER, J. (1981): Reproduction and postnatal development in the Vlei rat, Otomys 
irroratus, on the van Riebeeck Nature Reserve, Pretoria. Mammalıa 45; 99-116. 

Derany, M. J. (1964): A study of the ecology and breeding of small mammals in Uganda. Proc. zool. 
Soc. Lond. 142, 347-370. 

FRENCH, N. R.; STODDART, D. M.; BoBEk, B. (1975): Patterns of demography in small mammal 
populations. In: Small mammals: their productivity and population dynamics. Ed. by GoLLEy, 
F. B.; PETRUSEwIcZz, K.; Ryszkowskı, L. Cambridge: Cambridge University Press. 

HoRSsT, C. J. VAN DER (1944): Remarks on the systematics of Elephantulus. J. Mammalogy 25, 77-92. 

— (1946): Some remarks on the biology of reproduction in the female of Zlephantulus, the holy 
anımal of Set. Trans. Roy. Soc. S. Afr. 31, 181-199. 

— (1954): Elephantulus going into anoestrous; menstruation and abortion. Phil. Trans. 238B, 27-61. 

HuccGeTT, A. St. G.; WınDas, W. F. (1951): The relationship between mammalıan foetal weight and 
conception age. J. Physiol. 114, 306-317. 

Kıncpon, J. (1974): East African Mammals: An Atlas of Evolution in Africa. Vol. II, part A: 
Insectivores and Bats. London: Academic Press. 

LILLEGRAVEN, J. A. (1975): Biological considerations of the marsupial-placental dichotomy. Evolu- 
tion 29, 707-722. 

NEAL, B. R. (1981): Reproductive biology of the unstriped grass rat, Arvicanthis, ın East Africa. Z. 
Säugetierkunde 46, 174-189. 

— (1982): The breeding pattern of two species of spiny mice, Acomys (Muridae: Rodentia), in central 
Kenya. Mammalıa (in press). 

PARRY, G. D. (1981): The Meanings of r- and K-Selection. Oecologia 48, 260-264. 

PERRIN, M.R. (1980): The feeding habits of two coexisting rodents, Rhabdomys pumilioand Otomys 
irroratus, ın relation to rainfall and reproduction. Acta Oecol. Oecol. Gen. 1, 71-9. 

Pıanka, E. R. (1970): On r- and K-selection. Am. Nat. 104, 592-597. 

RATHBun, G. B. (1979): The social structure and ecology of elephant-shrews. Z. Tierpsychol. Suppl. 
20, 1-76. 

RATHBUN, G. B.; REDFORD, K. (1981): Pedal scent marking in the Rufous Elephant-Shrew, Zlephan- 
tulus rufescens. J. Mammalogy 62, 635-637. 

SACHER, G. A.; STAFFELDT, E. F. (1974): Relation of gestation time to brain weight for placental 
mammals: Implications for the theory of vertebrate growth. Am. Nat. 108, 593-615. 

SHORTRIDGE, G. C. (1934): The mammals of South West Africa. London: Heinemann. 

SMITHERS, H. N. (1971): The mammals of Botswana. Salisbury: Nat. Mus. Rhodesıa. 

SNEDECOR, G. W.; CoCHRAN, W.G. (1967): Statistical Methods. 6th Ed. Ames: Iowa State University 
Press. 

SoKAL, R. R.; RoHLF, F. $. (1969): Biometry. San Francisco: W. H. Freeman & Co. 

STEARNS, $. C. (1976): Life history tactics: a review of the ideas. Q. Rev. Biol. 51, 3-47. 

StocH, Z. G. (1954): The male genital system and reproductive cycle of Zlephantulus myurus jamesoni 
(Chubb). Phil. Trans. 238B, 99-126. 

Trıpp, H.R.H. (1971): Reproduction in Elephant-Shrews (Macroscelididae) with special reference to 
ovulation and implantation. J. Reprod. Fert. 26, 149-159. 

— (1972): Capture, laboratory care and breeding of Elephant-Shrews (Macroscelididae). Lab. Anım. 
6, 213-224. 

WALKER, E. P. (1955): Afrıcan elephant shrews. Nature Magazıne 48, 295-297, 332. 

WITHERSs, P. C. (1979): Ecology of a small mammal community on a rocky outcrop in the Namib 
desert, South-West Africa. Madoqua 11, 229-246. 


Author’s address: Prof. Dr. B. R. Near, Department of Biology, University of Saskatchewan, 
Saskatoon, Saskatchewan, Canada S7N OWO 


Putzen und Putzverhalten bei Rhinolophus ferrumequinum 


Von A. Koıs 


Lehrstuhl für Biologie, Universität Bamberg 


Eingang des Ms. 10. 8. 1981 
Abstract 


Cleaning and the conduct of cleanıng in Rhinolophus ferrumequinum 


The procedure and process of cleaning were observed in two females. The mouth with tongue, teeth 
and labium, as well as the claws of the hind limbs, not the claw of the thumbs, serve as cleanıng 
instruments. 

Hanging by one leg, a bat combs the skin of the body-side opposite to that leg with the claws of the 
free leg, the claws being pulled through the mouth several times. When the bat cleans the other side of 
the body, the hanging leg ıs changed. The patagium and the delicate body parts, sometimes however 
also the pelt, are cleaned with the tongue. By rubbing the patagium against the facıal region, sebum is 
transferred onto the patagıum and if necessary, is spread over other body parts. 

The cleaning of single body parts takes different periods of time, most of the time being spent in 
turn in the cleanıng of the patagium and mouth. The act of cleaning the part of the body where the bat 
begins, its duration and the sequence of cleaning of individual body parts as well as the sequence of 
redoing them are governed by the nervous system in responds to the intensity of itching of the body 
parts and not by an inherited cleanıng scheme. 


Einleitung 


Das Putzen ist eine auffallende und häufig zu beobachtende Beschäftigung aller Säugetiere, 
insbesondere auch der Fledermäuse. Hufeisennasen führen diese Tätigkeit mit besonderer 
Intensität und Akribie durch. Hier sollen Einzelheiten der Putzhandlungen von Rhinolo- 
phus ferrumequinum näher beobachtet, der Zeitpunkt und die Häufigkeit dieser Handlun- 
gen festgestellt sowie die zum Putzen benutzten Organe aufgezeigt werden. Außerdem soll 
untersucht werden, ob ggf. eine spezielle Putzordnung nachzuweisen ist. 


Material und Methode 


In den vergangenen Jahren wurde gelegentlich anderer Untersuchungen auch das Putzverhalten an 
zwei Weibchen von Rbin. f. genauer beobachtet. Dabei wurden die Tiere in einen Glasschrank 
(2,40 X 0,75 X 0,75 m), der an einer Seite einen Gazestreifen (0,29 X 0,30 m) aufwies, untergebracht. 
Sie konnten fast täglich auch im Raum (5,60 X 4,80 m) frei fliegen und kehrten anschließend freiwillig 
in den Versuchsschrank zurück. 

Die Putzhandlungen wurden anfangs zur Übersicht erst allgemein beobachtet, dann Einzelhand- 
lungen genauer betrachtet, ihre Dauer mit einer Stoppuhr festgehalten und eine Reihe von Stellungen 
der Tiere fotografiert. 


Beobachtungen und Ergebnisse 
Beinstellung 


In Ruhe (Abb. 1), zur Nahrungs- und Wasseraufnahme, soweit diese nicht im Flug 
erfolgte, sowie zur Kot- (Abb. 2) und Urinabgabe (Abb. 3) hing Rbhin. f. meistens an 
beiden Beinen; bei Putzhandlungen dagegen immer nur an einem Bein, und dies selbst 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4702-0072 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 72-79 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Putzen und Putzverhalten bei Rhinolophus ferrumequinum 73 


Abb. 1 (links). Rhinolophus ferrumequimum in Schlaf- bzw. Ruhestellung, Rückenansicht. — Abb. 2 
(rechts). Kotabgabe 


Abb. 3 (links). Urinabgabe; ein Tropfen tritt gerade aus, ein zweiter ist in der Halsgegend des Tieres 
sichtbar. - Abb. 4 (rechts). Daumen und Handgelenk wird beleckt, das Putzbein liegt dem Flügel auf 


dann, wenn das andere, das Putzbein, für Putzhandlungen gar nicht benötigt wurde. In 
diesem Fall ließ es das unbeschäftigte Bein frei hängen (Abb. 4) oder legte es einem 
Körperteil auf bzw. krallte sich dort fest. Der Fuß wurde dabei senkrecht gehalten. In der 
Regel fand während eines längeren Putzens mehrmals ein Beinwechsel statt, wobei die 
‚ Häufigkeit und Größe der Intervalle des Wechselns sehr stark varıierten. Welches Bein mit 
dem Putzen begann, schien zufällig zu sein. 


74 A. Kolb 


Tätigkeit des Putzbeines bzw. -fußes 


Mit den Krallen des Putzfußes kratzte bzw. durchkämmte das Tier hauptsächlich das Fell 
derjenigen Körperseite, an der der Fuß inseriert war. So bearbeitete es mit dem rechten Fuß 
neben der Maul- und der Hufeisenregion der Nase fast ausschließlich die rechte Hälfte der 
Kopfregion sowie die rechte Rücken-, Flanken- und Bauchseite. Mitunter wurden diese 
Grenzen auch etwas überschritten. Die Strichrichtung beim Durchkämmen der Rumpf- 
region führte bevorzugt von kranıal nach kaudal, bisweilen auch zirkulär (Abb. 5). 
Besonders auffällig waren schnelle klopfartige Bewegungen des Beines beim Putzen des 
Kopfbereiches. Mit den Krallen des Fußes betrillerte es die betreffende Stelle mit so hoher 
Geschwindigkeit, daß man die einzelnen Bewegungen mit freiem Auge nicht mehr 
wahrnehmen konnte. Diese Bewegungen dauerten jeweils nur 2 bis 5 sec, wurden jedoch in 
der Regel mindestens 10mal, im Extremfall bis zu 43mal wiederholt. Sie verringerten 
mitunter gegen Ende ıhre Geschwindigkeit und liefen mit Kammbewegungen aus. 


Putzen verschiedener Körperteile mit der Zunge 


Mit der Zunge, als dem Universalputzorgan, reinigten die Hufeisennasen nicht nur alle 
unbehaarten und damit zarthäutigen Körperteile, sondern auch behaarte Körperpartien 
und die Krallen des Putzbeines. Diese Krallen wurden im ständigen Wechsel mit den 
Putzhandlungen durch das Maul gezogen, beknabbert, beleckt (Abb. 6) und so von 
Haaren, Schuppen, Parasiten und dgl. befreit, wobei auch die Lippen mitwirkten. Putzbein 
bzw. -fuß verhielten sich immer ruhig, wenn sie selbst geputzt wurden. Auch die 
Genitalregion wurde ausschließlich durch Belecken gereinigt. Dies nahm zwar meist nur 
kurze Zeit in Anspruch, wurde dafür aber sehr oft wiederholt. 


Pflege der Flügel und Flughäute 


Die gesamte Flughaut wurde als zarthäutiges Organ fast ausschließlich durch Belecken 
geputzt. Nur in vier von 121 Fällen konnte auch ein kurzes diffuses Kratzen mit den Zehen 
an Flügeln festgestellt werden. Außerdem wurde bei den beobachteten Fällen zu 69 % 
derjenige Flügel zuerst geputzt, der zur Körperseite des jeweiligen Putzfußes gehörte. 
Rhin. f. konzentrierte sich ın der Regel zuerst auf den einen, dann auf den anderen Flügel, 
doch war damit nicht immer ein Wechsel des Putzbeines verbunden. Besonders nahe- 
liegende Teile der Flügel, vor allem auch beide Daumen, wurden fast regelmäßig ohne 
Wechsel des Putzbeines gereinigt. 

Beim Putzen der einzelnen Flügelregionen (Abb. 7) konnte nur eine lockere Ordnung 
festgestellt werden. Doch wurden auch durch den unsystematischen Wechsel der Putz- 
region nach und nach alle Teile des Körpers und der Flügel von oben und unten erfaßt. 
Eine gewisse Ordnung lag auch dem Belecken der oberen basalen Teile der Fingerflughaut 
insofern zugrunde, als dieses sich meist in Richtung des Daumens erstreckte und mit dem 
Belecken desselben endete. Die Daumen und ihre Krallen waren bevorzugte Putzobjekte, 
die mitunter minutenlang beleckt wurden, obwohl sie selbst nie als Putzwerkzeuge 
fungierten. 

Bei der Pflege der Flughäute nahmen die Tiere z. T. recht akrobatische Stellungen ein. 
So wurde häufig ein Flügel mit einer Seite an den Kopf gelegt oder der Körper beı 
gleichzeitiger Torsion des Flügels hochgezogen, so daß er einen Winkel bis zu 60° zur 
Vertikalen bildete, und die Zunge dadurch auch entfernte Teile des Flügels erreichen 
konnte. Mitunter wurden die Flughäute stark ausgebeult, da der zu putzende Flügel hierbei 
von oben nach unten, der Kopf von unten nach oben drückte, so daß eine kräftige Beule 
entstand. 


Putzen und Putzverhalten bei Rhinolophus ferrumequinum 75 


Abb. 5 (links). Putzen des Rückens mit rechtwinkelig abgebogenem Putzbein. - Abb. 6 (rechts). 
Reinigung des Putzbeines 


Abb. 7 (links). Belecken der rechten Flughaut, Zunge weit herausgestreckt. - Abb. 8 (rechts). Zunge 
putzt die Schwanzflughaut 


Die Schwanzflughaut wurde ebenfalls durch Belecken auf beiden Seiten gereinigt. 
Hierbei bog sich das Tier hoch (Abb. 8) und bildete einen Winkel von fast 180°. Von der 
Schwanzflughaut beleckte es hauptsächlich in proximodistaler Richtung nur diejenige 
Hälfte, an der das Putzbein inseriert war, da diese ausgebreitet und gespannt, die andere 
Hälfte dagegen gefaltet war. Der Gesichtsbereich dürfte auch wegen seiner kräftigen 
Talgdrüsen bei der Pflege der Flughaut eine wichtige Rolle spielen, denn wegen des 
Beleckens allein wäre wohl ein so starkes Zusammendrücken von Kopf bzw. anderen 
Körperteilen und Flughaut nicht nötig. 


76 A. Kolb 


Nach der Reinigung der Flügel konnte fast regelmäßig ein eigenartiges Verhalten 
beobachtet werden. Der geputzte Flügel wurde an einer bestimmten Körperregion gerie- 
ben. In 13 registrierten Fällen wurde 6mal der gesamte Kopf, wobei sowohl Drehbewegun- 
gen des Kopfes als auch des Flügels zu beobachten waren, 5mal der Hinterkopf mit den 
Ohrmuscheln, Iimal der Wangenbereich und 1mal eine Flanke mit einem Flügel gerieben. 


Ablauf der Putzhandlungen ohne vorhergehende Nahrungsaufnahme 


Kurz nach dem Erwachen ging Rhin. f. in der Regel von der Schlafstellung, an beiden 
Beinen hängend, ın die Ruhestellung über und hing dabei nur an einem Bein. Etwa 20-60 
min nach dem Erwachen begannen die Tiere mit intensiven Putzhandlungen, die abzüglich 
aller Pausen im Schnitt etwa 28 min dauerten. Da die Putzhandlungen am Ende stark 
zerflatterten, wurden nur die der ersten 20 mın ım Abschnitt von 5 min registriert. In den 
Diagrammen wurde auf der Ordinate der prozentuale zeitliche Anteil des Putzens der 
betreffenden Körperpartien innerhalb von 5 min aufgetragen, auf der Abszisse die einzel- 
nen Körperpartien. Bei der Angabe der Werte konnte nicht berücksichtigt werden, welches 
Putzorgan, Zunge oder Putzfuß oder beides zusammen, die Reinigung durchführte und wo 
sie begannen. Die Diagramme (Abb. 9a-d) zeigen, daf} die Länge des Putzens der einzelnen 
Körperpartien sehr unterschiedlich ist. Sie lassen erkennen, daß die Tiere zum Putzen der 
Flügel die meiste Zeit (44-72 %) aufwenden, mitunter mehr als für alle übrigen Partien 
zusammen. Der Zeitanteil der übrigen Körperpartien schwankt ebenfalls stark, doch 
übersteigt kein Bereich den Anteil von 35 %. Die geringste Zeit (1-5 %) nahm das Putzen 
der Beine ın Anspruch. Es wurde jedoch keine Körperpartie vollkommen geputzt, 


6-10 min — 11-15 min 16-20 min 
[ 


60 


Flanke 
Rücken 


Bauch 


Beine 


rücken 
Bauch 


0-5 min - 6-10 min 5 11-15 min - 16-20 min 
I I { 


Rücken 
tücken 
Rücken 
Beine 


Bauch 


Beine 


a b [6 d 


Abb. 9 (oben). Diagramme (a-d) der Putzhandlungen ohne vorhergehende Nahrungsaufnahme. 

Zeitlich-prozentualer Anteil der einzelnen Körperregionen am Putzvorgang in den ersten 20 mın des 

Putzens. — Abb. 10 (unten). Diagramme (a-d) der Putzhandlungen nach vorhergehender Nahrungs- 

aufnahme. Zeitlich-prozentualer Anteil der einzelnen Körperregionen am Putzvorgang in den ersten 
20 min des Putzens 


Putzen und Putzverhalten bei Rhinolophus ferrumequinum N. 


vielmehr wurden innerhalb von 5 Minuten alle Partien wenigstens einmal geputzt und 
anschließend der Putzvorgang wiederholt, weshalb diese Zeiteinteilung gewählt wurde. 


Ablauf der Putzhandlungen nach vorhergehender Nahrungsaufnahme 


Die vorhergehende Nahrungsaufnahme hatte erheblichen Einfluß auf die Putzhandlungen. 
Die Putzzeit des Maules schnellte auf etwa 66 % hoch (Abb. 10a), während die Putzzeit für 
die Flügel auf 24 % sank. Der Verlauf des Putzgeschehens änderte sich insofern, als in 96 % 
der Fälle die Putzaktion mit dem Maul-Kopfbereich begann und eine längere Zeit in 
Anspruch nahm als alle anderen Körperteile zusammen. Bei einem Wechsel des Putzbeines 
wurde auch auf der anderen Seite mit dem Putzen des Maules begonnen. Im weiteren 
Verlauf des Putzens stieg zwar der prozentuale Anteil der Flügel (Abb. 10a-d) wieder an, 
doch überwog insgesamt die Putzzeit des Maules die der Flügel. Die übrigen vier 
Körperpartien bewegten sich während der Gesamtputzzeit in mäßigen Werten, wobei die 
Beine den geringsten Anteil (1-9%) aufwiesen. In Abb. 10d kommt außerdem zum 
Ausdruck, daß sich die Verhältnisse weitgehendst denen von 9a, also der Situation des 
Putzens ohne vorhergehende Nahrungsaufnahme, nähern. Im übrigen verliefen die Putz- 
handlungen wie bereits berichtet. 

Bei Jungtieren ist dagegen die Putztätigkeit viel kürzer und von geringerer Intensität. 


Allgemeine Feststellungen 


Durch Urin- und Kotabgabe, durch Flüge zum Trinken, 
durch kürzere oder längere Ruhepausen wurden die Putz- 
handlungen bisweilen auch unterbrochen. Das Tier hing 
dabei an einem Bein, drehte sich nach allen Seiten, schlug bei 
längeren Pausen auch die Flughaut um sich und ruhte. Im 
Herbst ging die Intensität und Dauer des Putzens etwas 
zurück, ebenso der Appetit. Fine Korrelation zwischen der 
Menge der verzehrten Beute und der Intensität bzw. Länge 
des Putzens war jedoch nicht festzustellen. 

Nach Beendigung des Putzens nahm Rbhin. f. bisweilen 
eine eigenartige Ruhestellung ein. Das Tier hing dabei an 
einem Bein. Während nun der eine Fuß meist ventral im Fell 
verankert, die Flughaut gefaltet und seitlich angelegt war, 
wurde mit der Flughaut des anderen Flügels die ganze 
Bauchseite und auch der etwas zur Brust geneigte bzw. 
hochgezogene Kopf teilweise bedeckt. Hierbei wurde das 
erste Glied des zweiten Fingers beinahe waagerecht, das 
zweite schräg und das dritte fast senkrecht nach oben gehal- 
ten. Pleuro- und Dactylopatagium hüllten also die ganze 
Ventralseite des Tieres ein. Bei Störungen konnte es, ohne 
die Bein- oder Flügelstellung zu ändern, durch Zurückbeu- 
gen des Kopfes in dorsaler Richtung das Gesicht freimachen 
(Abb. 11), durch Drehbewegungen um die eigene Körper- 
achse nach allen Seiten sichern und die Umgebung akustisch 
überwachen. 


RIRRRRTIRRERERR NE FI 


Abb. 11. Eigenartige Ruhe- 
stellung für eine gewisse Zeit 
nach dem Putzen 


78 A. Kolb 


Diskussion 


Über das Putzen der Fledermäuse gibt es, soweit mir bekannt ist, lediglich kleinere 
Angaben, aber keine Abhandlungen, die Putzorgane und Putzhandlungen geschlossen 
wiedergeben. Bei Rbhin. f. werden als Putzorgane lediglich die Krallen der Hinterfüße und 
das Maul (Zunge, Zähne, Lippen) verwendet. Die Zehen der Hinterfüße eignen sich hierzu 
recht gut. Sıe sind alle gleich lang und gleich geformt, sie stehen alle ın gleicher Richtung, 
ihre Krallen sind stark gekrümmt und sehr spitz, weshalb sie beim Ziehen durch das Fell 
wie eın Kamm wirken. Auffallend ist, daß die Daumenkrallen von Rhın. f. beim Putzen 
keine Verwendung finden, obwohl sıe bei den einheimischen Gymnorhinen wie EIsEN- 
TRAUT (1937) von Myotis myotis, MOHR (1927) von Plecotus aurıtus, NATUSCHRE (1960) 
allgemein und RümPpLER (1980) von Eptesicus serotinus angeben, dazu herangezogen 
werden. Die enorme seitliche Beweglichkeit des Hinterfußes gestattet, Kammbewegungen 
auch in zirkulärer Richtung auszuführen, weshalb die Putzbewegungen oft recht verwor- 
ren erscheinen. 

Beim Durchführen der Krallen durch das Maul wird, wıe bei allen einheimischen 
Fledermäusen, das Putzgut unter ständigem Beknabbern, wobei auch Ektoparasiten getö- 
tet werden, mit Hilfe der Zähne und Lippen von den Krallen abgelöst und z.T. ver- 
schluckt. 

Mit der rauhen, papillenbesetzten Zunge, dem Universalputzorgan, werden der ganze 
Körper, alle unbehaarten und z. T. auch die behaarten Körperteile geputzt, mit den Krallen 
des Putzfußes fast ausschließlich behaarte Teile des Körpers. Ein gegenseitiges Putzen, wie 
dies bei manchen anderen Säugern zu beobachten ist (KRUMBIEGEL 1953), wurde nicht 
festgestellt. 

Der Beginn des Putzens ist zeitlich meist nach dem Aufwachen bzw. nach dem Fressen 
gelegen und an keine bestimmte Körperstelle gebunden. Es ist auch nicht festgelegt, 
welches Putzorgan, Zunge oder Fuß, mit dem Putzen beginnt. Daß das Putzen der rechten 
Körperhälfte überwiegend mit dem rechten Fuß und umgekehrt erfolgt, ist wohl von 
Zweckmäßigkeitsgründen bestimmt, da es nur mit dem der gleichen Körperseite zugeord- 
neten Fuß möglich ist, alle Stellen dieser Körperhälfte zu erreichen. 

Das Putzen der Flughaut auf der Ober- und Unterseite mit der Zunge zeigt ein 
Höchstmaß von Akrobatik. Die Flughaut wird je nach Bedarf umgeschlagen, eingerollt, 
ausgebreitet bzw. partiell gefaltet, wodurch recht eigenartig aussehende Gebilde entstehen. 

Das Zusammenreiben der Flughaut, hauptsächlich mit dem Bereich des Kopfes, dürfte 
bei Rhin. f. die gleiche Aufgabe haben, die für alle übrigen einheimischen Fledermäuse 
angegeben wird, nämlich den Talg der Drüsen des Kopfbereiches auf die Flughaut zu 
übertragen und ggf. durch weiteres Reiben fein zu verteilen. Auf diese Weise wird die so 
wichtige Elastizität der Flughaut und die Notwendigkeit, Wasser abzustoßen, erhalten. 

In Abb. 10a ist klar zu erkennen, daß das Putzen an demjenigen Körperteil am längsten 
dauert, hier am Maul, der die stärkste Verunreinigung aufweist. Da in 96 % der beobachte- 
ten Fälle hier auch das Putzen beginnt, läßt sich daraus schließen, daß wohl ein von dieser 
Partie ausgehender Reiz sowohl für die Festlegung der Stelle des Beginns als auch für die 
der Dauer des Putzens verantwortlich ist. Weil zudem auch die Reihenfolge der zu 
putzenden Körperpartien sehr stark variiert, dürfte auch diese von der Intensität des 
Juckreizes gesteuert werden. Damit zeigen die uns anfangs als keine bestimmte Ordnung 
aufweisend erscheinenden Putzhandlungen, daß ihnen eine wohlfunktionierende, nervös 
gesteuerte Ordnung zugrunde liegt. Die Stelle des Putzbeginns, die Dauer des Putzens und 
die Reihenfolge der einzelnen Körperpartien bei dieser Prozedur wird durch das Nerven- 
system gesteuert und ist nicht durch ein ererbtes starres Schema festgelegt. Doch auch die 
Ermüdung eines Putzorgans hat Einfluß auf das Putzgeschehen. Das Betrillern z.B. weist 
etwa 20 Schläge pro sec auf, dauert 2-5sec und wird bis zu 43mal mit kurzen Zwischen- 


Putzen und Putzverhalten bei Rhinolophus ferrumequinum 79 


pausen wiederholt. Da dies etwa 100 Schläge innerhalb von 5sec sein können, ist die 
häufige Unterbrechung des Betrillerns wohl auf die Ermüdung des Putzbeines zurückzu- 
führen. Somit sind die morphologisch-anatomischen Verhältnisse bei Rhinolophus nicht 
nur bestimmend für die allgemeine Art und Weise des Putzens, sondern haben auch 


Einfluß auf die Gestaltung desselben. 


Zusammenfassung 


An zwei alten Weibchen von Rhinolophus ferrumegquinum wurde das Putzen beobachtet. Es wird 
meist nach dem Aufwachen bzw. nach der Nahrungsaufnahme durchgeführt. 

Als Putzorgan fungieren bei Rhin. f. das Maul (Zunge, Zähne, Lippen) und die Hinterbeine. 
Während des Putzens hängt das Tier nur an einem Bein, das andere besorgt das Putzen. Es wird 
jeweils die Körperhälfte geputzt, deren Bein frei ist, ein Beinwechsel erfolgt öfter. 

Die Putztätigkeit besteht im Kratzen, Kämmen und Betrillern behaarter Körperteile mit den 
Krallen des Putzfußes und im Belecken der Flughaut, des Genitalbereiches und behaarter Körperteile. 
Das Kämmen verläuft meist von kranial nach kaudal, bisweilen auch zirkulär. Die Krallen des 
Putzfußes werden immer wieder durch das Maul gezogen und vom Putzgut befreit. Das Putzen der 
Flughaut erfolgt fast ausschließlich mit der Zunge, verläuft von proximal nach distal, nur in der Nähe 
der Daumen zieht es zu diesen. Die Daumenkrallen fungieren nicht als Putzorgane. 

Jeder Körperteil wird während einer Putzperiode mehrmals bearbeitet. Die längste Zeit nimmt das 
Putzen der Flughaut und des Maules in Anspruch. Die Gesamtputzzeit dauert, abzüglich der 
Unterbrechungen, im Schnitt etwa 28 min. Die wohlfunktionierende Putzordnung wird nicht durch 
ein ererbtes, starres Schema, sondern vom Nervensystem gesteuert und ist deshalb sehr flexibel. Die 
Stelle des Putzbeginns, die Dauer des Putzens einer Körperpartie, deren Reihenfolge und Wieder- 
holung hängen von der Intensität des Juckreizes der einzelnen Partien ab. 


Literatur 


EIsENTRAUT, M. (1937): Die deutschen Fledermäuse. Leipzig: Verlag Dr. Paul Schöps. 

KRUMBIEGEL, I. (1953): Biologie der Säugetiere I. Krefeld: Agıs Verlag GmbH. 

Mon, E. (1927): Plecotus aurıtus in der Gefangenschaft. Z. Säugetierkunde 2, 82-92. 

NATUSCHKE, G. (1960): Heimische Fledermäuse. Die Neue Brehm-Bücherei. Wittenberg: A. Ziemsen 
Verlag. 

en G. (1980): Handaufzucht und Jugendentwicklung einer Breitflügelfledermaus (Eptesicus 
serotinus). Z. des Kölner Zoos 23, 25-30. 


Anschrift des Verfassers: Prof. Dr. Anton Koıs Lehrstuhl für Biologie, Universität Bamberg, 
An der Universität2, D-8600 Bamberg 


Nachweis wilder Meerschweinchen für Ecuador und 
Beschreibung von Cavia aperea patzelti subsp. nov. 


Von H. SCHLIEMANN 


Zoologisches Institut und Zoologisches Museum der Universität Hamburg 


Eingang des Ms. 24. 11. 1981 
Abstract 


A first record of Wild cavies for Ecuador and description of Cavia aperea patzelti subsp. nov. 


Cavia aperea patzelti subsp. nov. is described on the basis of five individuals collected during August 
1978 in Ecuador, Province Chimborazo. At the same time, this is also the first record of Cavia aperea 
for Ecuador. The material from Ecuador is compared with 86 individuals of several subspecies from 
Columbia and Peru. Remarkable charakters of the new subspecies — apart from size in general - are 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4702-0079 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 79-90 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


80 H. Schliemann 


size and breadth of the skull. C. a. patzelti ıs easily distinghuished from subspecies from Peru by the 
size of body and skull, while measurements of skull breadth are clearly above of even equally sized 
animals from Columbia. 


Einleitung 


Angehörige der Unterfamilie Caviinae Murray, 1866 kommen mit Ausnahme des mit 
Regenwald bedeckten Amazonasgebietes in nahezu ganz Südamerika vor. Von den vier 
Gattungen Cavia, Kerodon, Galea und Microcavia besitzt Cavia das größte Verbreitungs- 
areal. Es erstreckt sich ım östlichen Südamerika von Ostbrasilien (Pernambuco) nach 
Süden bis südlich von Buenos Aires. An der Westseite des Subkontinents reicht es vom 
Norden (Guayana, Venezuela) über die Andenstaaten Kolumbien, Peru, Bolivien bis nach 
Nordargentinien. 

Diesem großen Verbreitungsgebiet entsprechend wurde in der Gattung Cavia eine 
Vielzahl von Arten und Unterarten beschrieben (s. CABRERA 1953, 1961; HÜCKINGHAUS 
1961). HÜckInGHAus (1961) ist zuzustimmen, daß ein Großteil dieser Arten und eine 
Reihe von Unterarten als nicht valide anzusehen sind. Er konnte an Hand des ıhm 
verfügbaren Materials wahrscheinlich machen, daß das Genus Cavia lediglich drei Arten 
umfaßt, nämlich C. aperea (Erxleben, 1777), C. fulgida (Wagler, 1831) sowie C. stolida 
(Thomas, 1926). Nach HÜckINGHAUsS kommen in den Andenstaaten folgende Vertreter 
der Gattung Cavia vor: C. aperea guianae (Thomas, 1901) — vermutlich unter Einschluß 
von C. anolaimae — ın Guayana, Venezuela und Kolumbien; C. a. tschudii (Fitzinger, 
1867) mit einer weiten Verbreitung in Peru vom Norden (Cajamarca) bis zum Süden 
(Arequipa) und C. a. festina (Thomas, 1927) in Zentralperu (Junin); C. a. osgoodi 
(Sanborn, 1949) in Südostperu; C. a. sodalıs (Thomas, 1926) in Bolivien bis hinein nach 
Argentinien sowie Cavıa stolida (Thomas, 1926) in Nordperu. 

Die Verbreitungsangaben über Cavıa legen nahe, ein kontinuierliches Vorkommen von 
Angehörigen dieser Gattung in den Andenstaaten anzunehmen. Aus Ecuador waren 
jedoch bislang keine Funde bekannt geworden. Und soweit in Erfahrung zu bringen ist, 
gibt es auch in den Museumssammlungen keine Belege für ein Vorkommen von Cavıa 
dort. Demgemäß klaffte bisher zwischen den Vorkommen in Kolumbien und Nordperu 
eine erhebliche Fundlücke. Durch neuerliche Funde kann diese jetzt geschlossen werden. 
Bei den Funden handelt es sich um fünf Exemplare von Cavıa, die unter Mitwirkung von 
Herrn Erwin PATZELT (Oldenburg, Schleswig-Holstein) im August 1978 östlich von 
Riobamba in der Hochkordillere von Alao (Provinz Chimborazo) gesammelt wurden 
(Abb. 1). Ein Vergleich der ecuadorianischen Tiere mit Cavia-Material aus Kolumbien und 
Peru ergab hinsichtlich einiger Merkmale des Schädels gravierende Unterschiede, so daß 
diese Tiere im folgenden als eine neue Unterart von Cavia aperea beschrieben werden 
sollen. Diese Unterart wird nach E. PATZELT benannt, der mit seinen Veröffentlichungen 
und seiner Sammeltätigkeit wichtige Beiträge zur Kenntnis von Flora und Fauna Ecuadors 
geleistet hat. 


Vergleichsmaterial 


An Vergleichsmaterial standen insgesamt 81 Schädel und 47 Bälge sowie Maßangaben von 
5 weiteren Schädeln zur Verfügung. Hiervon stammen 52 Tiere aus Kolumbien, 34 aus 
Peru. 

Das kolumbianische Material wurde überwiegend in zentralen Teilen des Landes 
(Cundinamarca und Meta) und im Süden (Laguna Verde) gesammelt. Der südlichste 
Fundort in Kolumbien liegt ca. 350 km nördlich des ecuadorianischen Fundortes. Entspre- 
chend dem Vorschlag von HückInGHAus soll davon ausgegangen werden, daß es sich bei 
den Tieren aus Kolumbien um C. a. guianae (Thomas, 1901) handelt. 


Cavia aperea patzelti subsp. nov. aus Ecuador 81 


Die Tiere aus Peru stammen von weit auseinanderliegenden Fundorten. Ein erheblicher 
Anteil wurde in Südperu mit den Fundortangaben Yura (Arequipa), Sandia und Limbani 
(Puno) sowie in den Provinzen Cuzco und Ica gesammelt. Aus Zentralperu (Junin) war 
nur ein einziges Tier verfügbar, während aus Nordperu von wahrscheinlich zwei Fundor- 
ten (Cajamarca, Rio Utcubamba) Angaben über vier Tiere verwertet werden konnten. Der 


GUARANDA 


) 
( 


N 
ie 
SL 

NS ALausı 


Bn BUCAY 


r— 
0107 20km 


Abb. 1. Die Fundortgegend von Cavia aperea patzelti in der Provinz Chimborazo in Ecuador 


nördlichste Fundort in Peru ist etwas mehr als 400 km vom ecuadorianischen Fundort 
entfernt. Nach HückInGHAaus ist das nordperuanische Material als Cavia stolida (Thomas, 
1926) bzw. als C. a. tschudü (Fitzinger, 1867) anzusehen. C. a. tschudii wäre auch die 
Mehrzahl der peruanischen Tiere aus Ica, Arequipa und Cuzco, während aus dem 
Südosten (Limbani und Sandia) C. a. osgoodi hier durch möglicherweise vier Tiere 
vertreten ist. Folgt man HÜCKINGHAUS, so gehört das Tier aus Junin zur Unterart festina 
(Thomas, 1927). 


Beschreibung von Cavia aperea patzelti subsp. nov. 
Diagnose 


Cavia aperea patzelti ıst durch folgende Merkmale gekennzeichnet: Fellfarbe dunkles, 
rötliches Braun; erhebliche Körpergröße (Kopf-Rumpflänge ca. 280 mm); Schädel groß 
(größte Schädellänge bis über 69 mm) und vor allem breiter als bei den Unterarten aus 
Kolumbien und Peru sowie mit großen Zähnen; die entsprechenden Mittelwerte der vier 
Exemplare betragen: Rostrumbreite 12,3 mm, Gaumenbreite bei M? 18,3 mm, Hirnschä- 
delbreite 28,6 mm, Jugalbreite 40,3 mm, Länge maxill. Zahnreihe 17,7 mm. 


82 H. Schliemann 


Holotypus 


Adultes Weibchen; Schädel; Restkörper in Alkohol. Fundort: Hochkordillere von Alao, 
Provinz Chimborazo, Equador. Sammeldatum: August 1978. Aufbewahrungsort: Zoolo- 
gisches Institut und Zoologisches Museum der Universität Hamburg, Katalog-Nr. 7593. 


Paratypen 


Fundort, Sammeldatum und Aufbewahrungsort wie Holotypus. Katalog-Nr. 7594: adul- 
tes Männchen; Schädel; Restkörper in Alkohol; Katalog-Nr. 7595: adultes Männchen; 
Schädel und Balg; Katalog-Nr. 7596: adultes Männchen; Schädel und Balg; Katalog- 
Nr. 7597: juveniles Weibchen; Schädel und Balg. 


Körpermaße und Haarkleid 


Die ın Tab. 1 wiedergegebenen Körpermaße und -gewichte sind nur ein grober Anhalt. Sie 
wurden an den Bälgen bzw. Alkoholrestkörpern abgenommen; gewogen wurden die 
tiefgefrorenen Kadaver. 


Tabelle 1 


Körpermaße (in mm) und -gewichte (in g) von Cavia aperea patzelti 


Kat.nr. 


Geschlecht 
Kopfrumpflg. 


Hinterfußlg. 
Ohrlg. 
Körpergewicht 


Ein Vergleich des ecuadorianischen Materials mit Bälgen kolumbianischer und peruani- 
scher Tiere ergibt - auch bei sehr zurückhaltender Bewertung der Balgmaße -, daß die 
adulten Exemplare aus Ecuador diejenigen aus Peru an Körpergröße fast ausnahmslos 
deutlich übertreffen und auch größere Hinterfußmaße aufweisen. Die Tiere aus Kolumbien 
sind etwa so groß wie die ecuadorianischen oder übertreffen sie sogar um ein weniges. 

Die Fellfarbe auf dem Rücken der Tiere, den Körperseiten, dorsal und lateral am Kopf 
ist ein rötliches Braun (OstwALp nı 4-5). Bis an die Spitze schwarze Haare, besonders in 
der Rückenmitte zahlreich, verleihen dem Fell einen dunklen Gesamteindruck. Das 
kräftige Rötlich-Braun rührt von Haaren her, deren apikale helle Banden eben diese Farbe 
aufweisen und nur eine sehr kurze, dunkel gefärbte Spitze besitzen. Der Gesamteindruck 
der Rücken- und Kopffärbung ist bei den meisten Tieren aus Kolumbien und Peru eher ein 
Olivbraun. Ein etwa 4 cm breiter, hellerer Streifen auf der Bauchseite der ecuadorianischen 
Tiere besitzt eine gelblich-rötliche Farbe (OsTwALD gc 2); er ist gegen das dunklere Fell der 
Körperseiten mehr oder weniger scharf abgesetzt und erstreckt sich bis unter den Unter- 
kiefer, hier weniger deutlich abgegrenzt. Die Extremität wird ın distaler Richtung heller. 


Schädel 


An den Schädeln des ecuadorianischen Materials und des Vergleichsmaterials wurden die 
folgenden neun Mafe abgenommen: größte Schädellänge, Länge der Nasalıa, Länge der 
maxillaren Zahnreihe (zwischen den Alveolenrändern). Länge des Diastemas (zwischen 
den Alveolenrändern), Breite des Maxillare bei M? (Alveolenaußenrand am hinteren 


Abb. 2 und 3. Der Schädel von Cavia aperea patzelti (Zool. Inst. u. Zool. Museum Hamburg, 7593 9; 

gr. Schädellänge 96,5 mm), links von dorsal, rechts von ventral. - Abb. 4. Der Unterkiefer von Cavia 

aperea patzeltı (Zool. Inst. u. Zool. Museum Hamburg 7593). - Abb. 5 und 6. Der Schädel von Cavia 
„anolaimae“ aus Südkolumbien (Smithsonian Institution, 239985 2) 


(ejfausAoqure], 

“equiegnsyn org) 

E74 1’91 vgl L’ST 8°I7 179 vp1035 7 
(urunf) 

g‘97 L’91 9° ZI Y'6l 0°79 vunsaf v I 
(oung) 

L’e7 Or sm Gel 1’81 v‘9g 1p008s0 v '7 
(esseweley) 

g'E7 Se 0°S1 671 v'2l £&°SG npna3sı u) 
(05ZnJ) 

6°€7 Op 0°91 Ir Col v8 npna>sı u °‘) 
(e>7) 

Er7 pl gl L’ST 661 ‘09 RE ) 
(edınbaıy) 

Erz ZI e9l 6rl co g°69 He) 
(uarguungoxpnis) 

€97 8"gı 681 ZI g'eT 1’29 avuvınd v7 
(uaıqunjoyfenusz) 

T'sz 11 691 gr EZ 179 avuvım2 v 


x x s x X s x x 


"ıq ‘Inf “ıq 'qJo1sJuJ Igjopey9suapy gg ıqunnsoy ‚3fewasseig 'yauyez 'xew 37 2 Zfeipesen "3jppey>S'19, 


(s uodunysrmmgepsiepueg & [op Syasnawyırıe furur ur) 


ea1ade vIArZ uoA sjerLlsgewsystajdray uayssıuenisd pun usysstuergquinjoy sap Sgjewpapeyas 
3 ANGE 
E8€ 0'sI 0°87 92] ZEV c‘9] 8°9] O°IT 9°£9 9657 


8°Ir ZW 0°0€ v’sl ce s8l el OITT €°69 S6S/ 
l’0r v'sı 87 v’sl GI 0°81 sl EIT v9 Y69/ 


U'Ir L’ST €'87 L’8l 7a) es Y8l 677 g'69 C6S/ 


“qppp "yauyez Z]pop 
“Iq "anf “ıq "q40193u] -BUDSUNg eN "q “ıq “Aqunnsoy "3fewosseig] "xeUuI ‘37 ‘Zfeipesen -LUDS 2) "Aue 


(1x9 ] 'S ISOAJOM) 
1jJ2zJed ea1ode eıAeD uoA (ww ur) agjewpapeyas 


GSNOAET, 


Cavıa aperea patzelti subsp. nov. aus Ecuador 85 


Umfang von M’), Hirnschädelbreite (oberhalb des Proc. zygomaticus squam.), Interorbi- 
talbreite, Jugalbreite. 

Die Schädel des ecuadorianischen Materials (Maße s. Tab. 2) fallen sofort durch ihre 
Größe und ihre andersartigen Proportionen (Abb. 2-6) auf. Die Werte für das Maß 
„größte Schädellänge“ liegen im oberen Variationsbereich des Vergleichsmaterials (Tab. 3). 

Sie sind darüber hinaus, und das ist besonders bemerkenswert, deutlich breiter als auch 
gleich große Vergleichsschädel. Das zeigt sich an allen Breitenmaßen. Insbesondere wird 
die Breite des Rostrums selbst des kleinsten ecuadorianischen Tieres, mit einer Ausnahme, 
von keinem Vergleichstier erreicht, und die Breite des Oberkiefers bei M? liegt durchgängig 
klar über den entsprechenden Maßen aller verfügbaren Cavia-Schädel aus Kolumbien und 
Peru. Auffallend ist ferner die Zahngröße, wie die herausfallenden Werte des Maßes 
„Länge der maxillaren Zahnreihe“ widerspiegeln. Die Bulla tympanica ist in rostrocaudaler 
Richtung kürzer als bei den meisten übrigen Tieren. Sie endet rostral auf der Höhe der 
Naht zwischen Basioccipitale und Basisphenoid; ihr rostraler Umfang ist in transversaler 
Richtung kaum gewölbt. Die geringe Zahl der verfügbaren Schädel läßt allerdings keine 
Aussage über die Variabilität dieses Merkmales zu. Am Nahtverhalten der ecuadoriani- 
schen Schädel sind keine beachtenswerten Besonderheiten zu erkennen. 


Statistische Analyse 


Um die Einzelmerkmale des Schädels einer Gesamtbewertung zuführen zu können, 
wurden die entsprechenden Werte des ecuadorianischen Materials und der Vergleichstiere 
multivariaten statistischen Analysen unterzogen. Und zwar wurden für die ecuadoriani- 
schen Tiere und für jeweils das gesamte hierfür verwendbare adulte kolumbianische 
(n = 33) und das gesamte peruanische (n = 27) Material mit Hilfe der diskriminanzanalyti- 
schen Methode des verallgemeinerten Abstands Distanzwerte nach MAHALANOBIS errech- 
net. Diese wurden in einem F-Test auf Signifikanz geprüft (Tab. 4). Ferner wurde der 
approximative verallgemeinerte Abstand nach PEnrosE (Gesamt-, Größen- und Form- 
abstand) ermittelt (Knussmann 1962, 1967, 1974). Hierfür wurde das adulte Material 
(n = 26 bzw. n = 25) auf 10 Fundortgruppen aufgeteilt (Tab. 5). 

Die Ergebnisse dieser Berechnungen werden vor allem in Anbetracht der wenigen 
ecuadorianischen Tiere mit großer Vorsicht bewertet. Eine Aufgliederung des Materials 
nach Geschlechtern konnte selbstverständlich nicht vorgenommen werden. Die für die 
Errechnung der MAHALANOBIS-Abstände erfolgte Zusammenfassung des kolumbianischen 
Materials kann zwar als unbedenklich angesehen werden, weil es offenbar einer Subspezies 
zuzurechnen ist. Dasselbe Vorgehen bei den peruanischen Tieren ist trotz der Tatsache, 
daß der überwiegende Teil dieses Materials als C. a. tschudii anzusehen ist, kritischer zu 


Tabelle 4 


Mahalanobis-Abstände zwischen ecuadorianischen, kolumbianischen und peruanischen Tieren 
auf der Basis von neuen Schädelmaßen (s. Tab. 2 u. 3), fehlende Werte durch Gruppenmittel 
ergänzt 


Linker unterer Teil der Dreiecksmatrix die MAHALANoBIS-Abstände, rechter oberer Teil die dazuge- 
hörenden prozentualen Signifikanzniveaus 
Herkunftsland des Materials Kolumbien Equador 


Kolumbien 0,1 


Equador 4,06 — 
Peru 0,60 a 


86 H. Schliemann 


Tabelle 5 
Penrose-Abstände zwischen Cavia aperea-Material von zehn Fundortgruppen in Ecuador, 
Kolumbien und Peru auf der Basis von neun Schädelmaßen 


Interkorrelationskoeffizient (aus Gesamtmaterial berechnet) 0,7658. Linkere untere Dreiecksmatrix 
Formabstände, rechte obere Dreiecksmatrix Gesamtabstände 


Equador 


Kolumbien 
(Laguna Verde) 
Kolumbien 
(Cundinamarca) 
Kolumbien 
(Meta) 

Peru 
(Arequipa) 
Peru 

(Junin) 

Peru 

(Cuzco) 

Peru 
(Utcubamba) 
Peru 

(Puno) 

Peru 


(Ica) 


betrachten, da hierbei auch Tiere anderer Unterart- bzw. sogar Artzugehörigkeit hinzuge- 
nommen wurden. Diese Zusammenfassung erfolgte, da die Herausnahme solcher Gruppen 
aus der Berechnung die Beweislage hinsichtlich der systematischen Bewertung der Tiere 
aus Ecuador nicht verbessert hätte. Die Berechnung der MAHALAnoBIS-Abstände, die auf 
der Verwendung von Individualdaten beruht, hätte jedoch ungünstigerweise an z. T. sehr 
individuenarmen Gruppen vorgenommen werden müssen. Zur Errechnung der Abstände 
nach PENROSE wurde das Material dagegen nach Fundorten gruppiert. Auch hierbei - die 
PEnROoSsE-Methode basiert auf Mittelwerten - stellt sich die Frage nach der Zuverlässigkeit 
von Stichprobenmitteln materialarmer Gruppen. Allerdings erschien eine solche Gruppie- 
rung bei Anwendung dieser Methode im Vergleich zur Berechnung der MAHALANOBIS- 
Abstände sinnvoll; durch die Verwendung durchschnittlicher (über alle Gruppen berech- 
neter) Standardabweichungen werden zufallsbedingte Streuungen, die beı individuenarmen 
Gruppen in einem größeren Ausmaß auftreten, eher nivelliert (KnussmAnn 1967). 


Lebensraum von Cavia aperea patzelti 


Das ecuadorianische Cavia-Material stammt aus einem 3000-3800 m hoch gelegenen 
Gebiet im Bereich der feuchtkühlen, tropischen Päramos östlich von Riobamba. Die 
Paramos besitzen ein ausgesprochenes Tageszeitenklima mit nur unbedeutenden Unter- 
schieden zwischen den Jahreszeiten. Die Tagestemperaturen steigen selten über 15°C, 
können aber ın den frühen Morgenstunden unter den Gefrierpunkt sinken. Die Jahresnie- 
derschlagsmenge liegt, soweit bekannt, über 1000 mm. Die Luftfeuchtigkeit ist infolge der 
Nebel hoch (WEBER 1969). In dem Fundgebiet werden weite, z. T. sumpfige Hochflächen 
mit flachen und breiten grasbestandenen Bergrücken von den Steilhängen eines Trogtales 
begrenzt (Abb. 7). Die Vegetation ist durch hartblättrige Horstgräser vornehmlich der 


Cavia aperea patzelti subsp. nov. aus Ecuador 87 


Abb. 7. Die Landschaft des Fundortes von Cavia aperea patzelti in der Provinz Chimborazo in 
Ecuador (Photo PATZELT) 


Gattung Stipa gekennzeichnet, in deren Schutz u. a. Vertreter der Gattung Calceolaria, 
Gentiana und Altensteinia angetroffen werden. Strauchförmige Kompositen (Chugquira- 
gua) überragen das Gras (Abb. 8). In kleinen Taleinschnitten gedeihen Sträucher (Erica- 
ceen, Kompositen, Hypericum-Arten) und vor allem Wälder der Rosaceengattung Pol/yle- 
pis mit häufig krüppelhaftem Wuchs der Bäume. Die Horstgräser werden während der 
Trockenzeit angezündet. Die nachwachsenden jungen Blätter dienen den Meerschwein- 
chen u.a. als Nahrung. Auf den abgebrannten Flächen lassen sich die Röhren der 
Meerschweinchenbauten verhältnismäßig leicht ausmachen. Die hier vorgestellten Tiere 
wurden auf solchen abgebrannten Flächen gefangen. 


Diskussion 


Das Material aus Ecuador ist auf Grund seiner Körpergröße leicht von den kleineren 
Tieren aus Peru zu unterscheiden. Obwohl sich unter den Vergleichstieren aus Kolumbien 
auch einige recht kleine Individuen befinden, besitzt die Mehrzahl eine den ecuadoriani- 
schen Tieren ähnliche Körpergröße, soweit diese an Hand des Balgmaterials beurteilt 
werden kann. Bei Betrachtung des Schädels der Tiere aus Ecuador ist jedoch eine 
Unterscheidung von den Tieren aus den Nachbarländern klar zu treffen. Es soll zwar nicht 
vergessen werden, daß bislang nur vier erwachsene Meerschweinchen aus Ecuador vorlie- 
gen, aber diese Individuen zeigen die entscheidenden Merkmale (große Schädellänge, vor 
allem aber herausragende Breitenmaße) so einheitlich, daß man das Vorliegen eines 
Zufallsbefundes ausschließen darf. Auch vergleichbar große Schädel aus Kolumbien besit- 
zen mit ihren geringen Breitenmaßen charakteristisch andersartige Proportionen. Daher 
kommen größenbedingte Einflüsse als Ursache für die Verbreiterung nicht in Betracht. Für 
C. stolida (Thomas, 1926) sind ebenfalls herausfallende Breitenmaße angegeben (Hük- 


88 H. Schliemann 


KINGHAUS 1961) und kön- 
nen nach den mir vorlie- 
genden Daten bestätigt 
werden. Ganz offenbar 
liegen aber auch diese Ma- 
ße von C. stolida noch un- 
ter denjenigen von C. a. 
patzeltı, wie der Vergleich 
gleichgroßer Individuen 
zeigt. 

Der Versuch, die hin- 
sichtlich der Individuen- 
zahl unzureichenden Meß- 
daten der Schädel stati- 
stisch einer Gesamtbewer- 
tung zuzuführen, bestätigt 
diese Beobachtungen. Die 
MAHALANOBIS - Abstände 
trennen die ecuadoriani- 
schen Tiere von den peru- 
anıschen und kolumbianı- 
schen signifikant, obwohl 
nur verhältnismäßig wenig 
Schädelmaße abgenom- 
men wurden. Diesem Be- 
fund kommt in bezug auf 
die Abgrenzung von Ca- 
vıa aus Ecuador gegenüber 
Tieren aus Kolumbien un- 
eingeschränkt Bedeutung 
zu. Seine Aussagefähigkeit 
im Zusammenhang mit pe- 
ruanischen Tieren darf 


Abb. 8. Die Vegetation am Fundort von Cavia aperea patzelti. Im 
Vordergrund Horstgräser der Gattung Stipa, dahinter strauchfor- man wegen der gemeinsa- 
mige Kompositen der Gattung Chuguiragna (Photo PATZELT) men Berechnung der Di- 


stanzwerte für mehrere 
Unterarten dennoch nicht unterbewerten, da die Mehrzahl der peruanischen Tiere C. a. 
tschndii zuzurechnen ist. Die PENROSE-Abstände demonstrieren ebenfalls deutlich die 
relativ große Distanz des Materials aus Ecuador gegenüber den Tieren.aus den Nachbarlän- 
dern. Die vorgenommene dualsequentielle Clusterbildung für die Form- und Gesamtab- 
standswerte veranschaulicht diese Distanz (Abb. 9 und 10). Auf eine Interpretation der 
Abstände zwischen den Formen des übrigen Materials muß hier verzichtet werden. Diese 
könnte nur im Rahmen einer Revision von Cavıa aperea an Hand eines umfangreicheren 
Materials vorgenommen werden. Um Mißverständnissen vorzubeugen, soll aber auf 
methodische Mängel der durchgeführten Clusterbildung hingewiesen werden. In sehr 
schematischer Form werden hierbei jeweils zwei Gruppen zu einem Cluster vereinigt, die 
den relativ niedrigsten Abstand voneinander aufweisen. Auf dem nächsten Abstandsniveau 
erfolgt die Zusammenfassung mit einer weiteren Gruppe oder einem bereits gebildeten 
Cluster. Grundlage hierfür sind die Mittelwerte der Abstandsmaße der Einzelgruppen der 
Cluster. Dieses Vorgehen hat den Nachteil, daß die Hierarchie eines durch dualsequentielle 
Clusterbildung entstandenen Dendrogramms die Ähnlichkeitsverhältnisse aller Einzel- 
gruppen zueinander nur unvollkommen widerspiegelt. Der große Abstand hingegen, der 


Cavia aperea patzelti subsp. nov. aus Ecuador 89 


0.56 0.70 
0.52 0:65 
0.48 0.60 
0.44 0.55 
0.40 0,50 
0.36 0,45 
0.32 0.40 
0.28 0.35 
0.24 0. 30 
0.20 0,25 
0.16 0.20 
0.12 0.15 
0.08 0.10 
0.04 0.05 


1 2 3 4 h) 10 9 7 8 6 1 p) 3 4 a e/; g B 6 


Abb. 9 und 10. Penrose-Abstände zwischen Cavıa aperea-Material von zehn Fundortgruppen in 

Ecuador, Kolumbien und Peru. Dendrogramme nach dualsequentieller Clusterbildung zur Veran- 

schaulichung der Abstände. Abzisse: Fundorte in der Numerierung der Tab. 5; Ordinate: Abstands- 
maße; Abb . 9 (links): Formabstände, Abb. 10 (rechts): Gesamtabstände 


die Tiere aus Ecuador von dem übrigen Material distanziert, gibt das Bild, das man bei 
Betrachtung der Abstandsmatrices gewinnt, im wesentlichen richtig wieder. 

Obwohl der Fundort des Materials aus Ecuador von den nächsten bekannten Fundorten 
im Norden und Süden mehrere hundert Kilometer entfernt liegt, ıst das natürlich kein 
Beleg für eine vollständige Isolation der ecuadorianischen von den übrigen Populationen. 
Vielmehr muß vermutet werden, daß weiteres Sammeln in Ecuador zum Bekanntwerden 
anderer Fundorte von C. aperea führen würde. Bei dieser noch recht mangelhaften 
Kenntnis der Verbreitung von Cavıa aperea in dieser Region erscheint es angemessen zu 
sein, die beschriebenen morphologischen Unterschiede des ecuadorianischen Materials 
gegenüber dem Vergleichsmaterial taxonomisch nicht überzubewerten. Aus diesem 
Grunde und wegen der schmalen Materialbasıs wird die ecuadorianische Population, aus 
der die hier vorgelegten Tiere stammen, lediglich als eine Unterart von C. aperea aufgefaßst. 
Dabei wird auch berücksichtigt, daß das Vergleichsmaterial hinsichtlich der Körpergröße 
und der Schädellänge eine erhebliche Variabilität aufweist, wenn auch andererseits die 
Breitenmaße der Schädel der ecuadorianischen Tiere eine taxonomisch einschneidendere 
Bewertung bedenkenswert erscheinen lassen. Fine endgültige Klärung dieser Frage muß 
‚jedoch einer systematischen Revision mit modernen quantitativen Methoden vorbehalten 


bleiben. 


Danksagung 


Herrn Erwin PATZELT (Oldenburg, Holstein) bin ich für die Überlassung des Cavia-Materials aus 
Ecuador zu großem Dank verpflichtet. Zahlreiche Kollegen haben mir durch Ausleihen von Ver- 
gleichsmaterial und mit Auskünften Unterstützung angedeihen lassen; insbesondere möchte ich 


90 H. Schliemann 


folgenden Damen und Herren danken: PATRICIA FREEMAN (Field Museum of Natural History, 
Chicago), CHARLES ©. HanDLey (National Museum of Natural History, Washington), DAPHnE M. 
Hırıs (British Museum, London), Karı F. Koopman (American Museum of Natural History, New 
York) und JOHANNES Krarr (Zoologische Staatssammlung, München). Herrn R. KnussmAanNn 
(Anthropologisches Institut, Hamburg) danke ich für Rat und Diskussion im Zusammenhang mit der 
statistischen Bearbeitung des Materials, Frau LÜTTMAnn (ebenfalls Anthropologisches Institut Ham- 
burg) für die Durchführung der Rechnungen. 


Zusammenfassung 


Cavia aperea patzelti subsp. nov. wird an Hand von fünf Exemplaren aus der Provinz Chimborazo, 
Ecuador, beschrieben. Damit wird zugleich der Erstnachweis des Vorkommens von C. aperea in 
Ecuador geführt. Umfangreiches Material aus den Unterarten der benachbarten Länder Kolumbien 
und Peru dient dem Vergleich mit C. a. patzelti. Auffällige Merkmale der neu beschriebenen Unterart 
sind Körpergröße, Größe und vor allem Breite des Schädels. Körper- und Schädelgröße liegen deutlich 
über entsprechenden Maßen der Vergleichstiere aus Peru; die Schädelbreitenmaße unerscheiden C. a. 
patzeltı von dem Material aus Kolumbien. 


Literatur 


CABRERA, A. (1953): Los Roedores Agentinos de la Familia „‚Caviidae‘“. Escuela de Veterinaria; Publ. 
6, 1-93. Buenos Auıres. 

— (1961): Catalogo de los Mamiferos de America del sur. Vol. 4 (Sirenia, Perissodactyla, Artiodac- 
tyla, Rodentia, Cetacea). Buenos Aires: Imprenta y casa Editora Con. 

ERXLEBEN, J. C. P. (1777): Systema Regni Anımalis per Classes, Ordines, Genera, Species, Varietates, 
cum Synonymıa et Historia Anımalium. Classıs 1. Mammalia. Lipsiae; Impensis Weygandıanıs. 

FITZINGER, L. J. (1867): Versuch einer natürlichen Anordnung der Nagethiere (Rodentia). Sitgsber. 
Akad. Wiss. Wien 55, 453-515. 

HückıncHaus, F. (1961): Vergleichende Untersuchungen über die Formenmannigfalgigkeit der 
Unterfamilie Caviinae Murray 1886. Z. wiss. Zool. 166, 1-98. 

KnussMmAnn, R. (1962): Moderne statistische Verfahren in der Rassenkunde. In: Die neue Rassen- 
kunde. Hrsg. von ILsE SCHWIDETZKY. Stuttgart: G. Fischer. 

— (1967): Penrose-Abstand und Diskriminanzanalyse. Homo 18, 134-140. 

— (1974): Die Ähnlichkeitsverhältnisse im südwestlichen Afrika nach anthropometrischen Merkma- 
len. In: Bevölkerungsbiologie. Hrsg. von W. BERNHARD u. A. KANDLER. Stuttgart: G. Fischer. 

Murray, A. (1866): The geographic distribution of Mammals. London: Day and Son. 

MAYR, E. (1967): Artbegriff und Evolution. Hamburg u. Berlin: Paul Parey. 

— (1975): Grundlagen der zoologischen Systematik. Hamburg u. Berlin: Paul Parey. 

SANBORN, C. C. (1949): Cavies of southern Peru. Proc. Biol. Soc. Washington 68, 133-134. 

THomas, O. (1901): On a collection of Mammals from the Kanuku-Mountains, British Guiana. Ann. 
and Mag. Nat. Hist. Ser. 7, 8, 139-154. 

— (1926): The Spedan-Lewis South American Exploration. III. On Mammals collected by Sr. Bupın 
in the Province of Tucuman. Ann. and Mag. Nat. Hist. Ser. 9, 17, 602-609. 

— (1926): The Godman-Ihomas Expedition to Peru. III. On Mammals collected by Mr. R. W. 
HENDEE in the Chachapoyas region of North Peru. Ann. and Mag. Nat. Hist. Ser. 9, 18, 156-167. 

— (1927): The Godman-Thomas Expedition to Peru. VI. On Mammals from the Upper Huallaga and 
neighbouring highlands. Ann. and Mag. Nat. Hist. Ser. 9, 20, 594-608. 

WEBER, E. (1980): Grundriß der biologischen Statistik. Stuttgart, New York: G. Fischer. 

WEBER, H.: Zur natürlichen Vegetationsgliederung von Südamerika. In: Biogeography and ecology in 
South Amerika. Ed. by E. J. FitTkau, I. IrLıes, H. Kıinge, G. H. ScHwABE, H. Sıoıı. Vol. 2, 
475-518. The Hague: Dr. W. Junk N. V. Publishers. 


Anschrift des Verfassers: Prof. Dr. HarALD SCHLIEMANN, Zoologisches Institut und Zoologisches 
Museum, Universitätt Hamburg, Martin-Luther-King-Platz 3, D-2000 
Hamburg 13 


Scent-marking in the pine marten (Martes martes) in Finnish 
Forest Lapland in winter 


By E. PULLIAINEN 


Department of Zoology, University of Oulu 


Receipt of Ms. 1. 10. 1981 
Abstract 


Studied were scent-markings used by pine martens (Martes martes) in an uninhabited region of 
Finnish Forest Lapland, northern Europe, by tracking individuals of the species for a total of 2481 km 
over two successive winters. The frequency of defaecation decreased during the course of the winter 
(from 0.8 times/km in late October to 0.3/km ın March), but increases were found in urination (from 
0.9/km in late October-December to 1.6/km in January-March) and rubbing secretion from the anal 
glands (from 0.8/km to 5.4/km respectively). The last mentioned activity varıed considerably even 
from night to night. The pine martens scent-marked throughout their home ranges. Orientation 
(taking place even in conditions when vision is inoperative), familiarıty with a localıty and individual 
recognition are regarded as the primary functions of these scent-markings, whereas the „reading“ of 
each other’s scent-marks, and thereby the maintaining of martelism, ıs regarded as one of the effects of 
this practice under unpredictable northern conditions in which the home range of a pine marten may 
amount to almost 100 km?. 


Introduction 


The pine marten of northern Fennoscandia (Martes martes) is actıve at nıght in winter, only 
about 1 % of observations being achieved at day (PuLLiaınen 1981b). THIEssEN and RıcE 
(1976) and STODDART (1980c) have recently emphasized that nocturnalism provides a 
selective encouragement to the development of olfactory communication. The pine marten 
scent-marks its home range by urination, defaecation and rubbing secretion from the anal 
glands on conspicuous objects (EWER 1968; PULLIAINEN 1981b). JOHNSON (1973) proposed 
six functions for such scent- marking, namely as an alarm signal, epideictic display or sex 
attractant, or for individual recognition, orientation or familiarıty in a location, or for 
warning conspecifics away from occupied territory. Raııs (1971) suggested that it may 
also be used to indicate dominance to a conspecific (see also STODDART 1980a, b). 

The purpose of the present paper is to provide records on the scent-marking behaviour 
of the pine marten in Finnish Forest Lapland, northern Europe, ın winter. 


Material and methods 


The study area, of approx. 250 km, lies in eastern Itäkaira, Finnish Forest Lapland. The eastern part of 
the area includes two fell ridges (Pulkkatunturi-Sautunturi and Värriötunturi) running in a north- 
south direction about 5 km apart. Ravines are typical of the fell terrain. The area is ın a virgin state, 
with the following proportions of natural habıtats: spruce forests (Picea abies) 26.5 %. Scots pine 
forests (Pinus sylvestris) 18.5 %, birch forests (Betula spp.) 3.0 %, mixed forests 22.0 %, open bogs 
13.5 %, mountain birch forests (Betula pubescens ssp. tortuosa) 7.0 % and treeless fell summits 9.5 %. 
The forests are old and mature, including a large number of standing and fallen dead trunks. The 
density of trees is generally very low, and in some places there are young spruce and pine saplıngs 
among the very old trees. Some of the lowland birch forests were killed by larvae of the geometrid 
Epirrita autumnata in 1965-1966 (see PULLIAINEN 1976). 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4702-0091 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 91-99 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


92 E. Pulliainen 


Table 1 


Distances covered in tracking pine martens during each of the ten-day study periods during the 
winters of 1979/80 and 1980/81 


Period Distance skied, Period Distance skied, 
in km in km 


21-30 October 10-19 January 
1-10 November 20-29 January 
11-20 November 30 January-8 February 


21-30 November 9-18 February 
1-10 December 120.07 19-28 February 
11-20 December 158.35 1-10 March 
21-30 December 104.67 11-20 March 
31 December-9 January 65.99 21-30 March 


The field studies were carried out during two successive winters, 1979/80 and 1980/81. Four males 
and two females during the former winter and three males and two females (identified on the basis of 
the sıze of the footprints and urination patterns) during the latter were tracked by skiing, giving a total 
distance skied of 2481 km (Table 1). In 1979/80 a male and a female were tracked simultaneously for 67 
days, and in 1980/81 a male and a female in the same part of the study area were also tracked 
sımultaneously for 50 days. The anımals were not disturbed in any way when tracked through the 
snow. All events of urination, defaecation, and rubbing secretion from the anal glands were recorded. 
In the majority of the cases included in this paper the pine marten was tracked from one resting site to 
another. All cases, in which the snow had covered the tracks or had otherwise made it difficult to 
interpret the activities of the anımals were omitted. 


Results 


Pine martens employ a scent-marking technique which involves rubbing secretion from the 
scent anal glands onto various objects or urinating and defaecating at conspicuous locations 
(PuLziaınen 1981b). Faeces are usually left singly throughout the home range. Middens 
were seldom found, and only in the immediate vicinity of frequently used resting sites. The 
largest midden consisted of 6 faeces. Faeces were often found on the roots or trunks of 
fallen trees, on stumps or at other places where the surface of the snow was higher than 
elsewhere in the vicinity. 

Defaecation was the least frequent and least variable of the three scent-marking 
techniques (Fig. 1). In late October, however, when the pine martens ate large quantities of 
berries and mushrooms (PuULLIAINEN 1981a), the faeces were left at the average distances of 
1.3 km, while in March, when small rodents, carcasses, etc. comprised the majority of the 
diet (PuLLIAINEN 1981a), the corresponding figure was 3.6 km. The average defaecation 
frequency of the two males (or one male in two successive winters) was 0.36 times/km, and 
the corresponding mean for the female(s) in the two successive winters (in late January 
-March) 0.33. This difference is not statistically significant (t = 0.815, p<0.5, d.f. 12). 
When travelling, the pine martens visited faeces left earlier by themselves or other martens 
at conspicuous locations. 

The pine martens frequently urinated at places where the surface of the snow was higher 
than elsewhere in the vicinity, or on their own trails of those of other anımals (including 
empty snow cavities of tetraonids) or on ski tracks, on carcasses, in trees which they were 
climbing, etc. Urination occurred throughout the home range. Its average frequency was 
0.9 times/km before the turn of the year, but clearly higher in January-March (1.7 times/ 
km) (Fig. 1). Interestingly, this urination frequency was at its minımum (0.7 times/km) ın 
late October, when the pine martens were eating wet berries and mushrooms, although the 
faeces were very wet at this time. Urination was the second most frequent scent-marking 


Scent-marking in the pine marten in Finnish Forest Lapland in winter 93 


A 
= 
x 
NS 
) 
WJ 
= 
= 
& Fıg. 1. Frequencies (n/l km) of rubbing 
= z secretion from the anal glands (A), urina- 
& tion (B) and defaecation (C) in pine mar- 
u En een c tens in the period 21 October-30 March 
WO 24-4 11= 21- 1= 11= 21- 31 10- 20- 30- 9- 19- 1- 11- 21-30 during the winters of 1979/80 and 


OCT NOV DEC JAN FEB MAR 1980/81 


activity, with average frequencies in late January-March of 1.6 times/km in the male(s) (see 
above) and 2.1 times/km in the female(s). This difference is statistically significant 
(t= 3.508, p<0.01, d.f. 12). 

The pine martens rubbed secretion from their anal glands on the snow surface, sticks, 
branches of bushes or trees, etc. This was the most frequent and most changeable of the 
scent-marking activities (Fig. 1), varying considerably even from night to night and from 
one individual to another, although a certain trend can also be seen, in that the mean 
frequency was only 0.8 times/km before the turn of the year, while in January-March it 
was 5.4 times/km (Fig. 1). 

The male and female rubbed secretion from their anal glands with almost similar 
frequency (4.6 and 3.9 times/km, respectively) in January-March 1980, but in February- 


KYNSIVAARA 


® = resting Site 


SYVÄKURU 


Be N: > "ee" HÄRKÄTUNTUR —  \ 2 


Fig. 2. Movements of a male pine marten in the study area on 2 February to 31 March 1981 and 
frequencies of rubbing secretion from the anal glands between the resting sites. Frequencies: A = 0-10 
times/km; B = 11-20 times/km; C = over 20 times/km 


94 E. Pulliainen 


FREQUENCY (TIMES/KM) 


T —U = S—— er z 
21 25 30] 5 10 15 20 25 | 5 10 15 20 25 29 
JANUARY FEBRUARY MARCH 


Fig. 3. Frequency of scent-marking in a male and a female pine marten and changes in the thickness of 
the snow cover at a meteorological screen located in the middle of the study area on 21 January to 
29 March 1980 


FREQUENCY (TIMES/KM) 


10 15 20 25 5 10 15 20 25 30 
FEBRUARY MARCH 


Fig. 4. Frequency of scent-marking in a male and a female pine marten and changes in the thickness of 
the snow cover at a meteorological screen located in the middle of the study area on 9 February to 
30 March 1981 


March 1981 the corresponding frequency for the male was considerably higher than that of 
the female. Thus when the two sets of data are combined, the average frequency for the 
male(s) is 7.5 and that for the female(s) 3.6 times/km. This difference is statistically highly 
significant (t = 4.787, p<0.001, d.f. 12). | 

In order to understand any possible territorial behaviour in the pine marten it ıs 
important to know whether it scent-marks along the borders of its home range or | 
throughout it. Fig. 2 shows that the male marten tracked on 2 February-31 March 1981 had 
rubbed secretion from its anal glands onto various objects throughout its home range, the 
highest marking intensities being recorded both near the boundaries and in the inner parts. 
Other long trackings gave the same result. 


Scent-marking in the pine marten in Finnish Forest Lapland in winter 33 


Figure 3 shows the scent-marking frequencies (defaecation, urination and rubbing 
secretion from the anal glands calculated together) of the male and female pine marten, the 
home ranges of which (areas of 34 and 14 km? respectively) totally overlapped, during the 
period 21 January-29 March 1980. The female had peaks in this activity in the beginning 
and in the middle of the period, while the male had similar peaks at the end of the period. 
Altogether, however, there was no significant difference in scent-marking frequency 
between the sexes. 

A male and a female pine marten were tracked in the same part of the study area on 9 
February-30 March 1981, the areas of their home ranges being 93 and 38 km? respectively. 
Again these totally overlapped. It is possible that the martens were the same individuals as 
during the previous winter. In this case, however, scent-marking was much more frequent 
in the male than in the female (Fig. 4). Here again a considerable variation in scent-marking 
frequency is visible from night to night. 


Discussion 


The present pine martens did not throw their odorous secretion (in the urine and faeces and 
from the anal glands) to the winds, but directed it at conspicuous locations. PETERS and 
Mech (1975), studying scent-marking in wolves (Canıs lupus), point out that the elevated 
locations of raised leg urination 1. facilitates wind dispersal of the odour, 2. increases the 
evaporation surface as the urine trickles down, 3. minimizes the chances of the mark being 
covered by snow, and 4. produces a visual contrast against the snow which is visible from 
several metres away. The present study describes the behaviour of the pine marten, also in 
snow conditions. The first three points are probably relevant here, too, while the fourth is 
scarcely of any great importance in view of the darkness of the night in mid-winter. 
The term “territorial marking” implies that the marks serve to ıdentify a territory, that 
is, a fixed space which the individual will defend against rivals of the same species (RALLS 
1971). If scent-marks acted as a means of territorial defence they might be expected to 
occur most frequently at the territorial boundaries (JoHNnson 1973). PETERS and MECH 
(1975) found that wolves leave urine markings more frequently near their territorıal 
borders, and they are also known to mark more frequently on trails than off them, 
irrespective of the availability of suitable prominent sites. CAMENZIND (1978) describes 
similar behavior in the coyote (Canis latrans) and MacponauD (1980) in the red fox 
(Vulpes vulpes). Red foxes also leave urine marks specifically on visually conspicuous 


' objects (MAcDoNALD 1980), as also do coyotes (LINHART and KnowLTon 1975). 


Kruuk (1978) found that European badgers (Meles meles) demarcate their territory 
borders with middens, which were found to be located near certain landmarks more often 
than could be predicted by chance. MAcDoNALD and LinDsay (cit. in MACDONALD 1980) 
also found middens along territory borders, although some long sections of the boundaries 
had none. Kruuk and Hrwson (1978) discovered that the middens of a population of 
coastal otters (Lutra lutra) were more than twice as numerous within 100 m of the holts 
than elsewhere. These otters had linear territories along the coast, and so their middens 
were well away from this. ERLINGE (1968), studying otters ın inland waterways, found 
these to be territorial, their middens beeing associated with home range boundaries, 
whereas the scats of otters around certain Scottish lochs were not associated with territorial 
borders (Gorman 1980). Solitary moving mustelids such as Mustela frenata and M. nivalıs 


: (Quick 1951), and the pine marten (LockıE 1966; the present study), may also use 


| 


' middens. 


Red foxes (MAcDonALD 1980) and wolves (SETON 1909; PETERS and MEcH 1975; 
DaniıLov etal. 1978) often travel along well-used trails and leave their scent-marks on these 
routes and at junctions between them. The pine martens of the present study area avoid 


96 E. Pulliainen 


openings and clearly prefer forests where spruce is growing as their moving and resting 
habitats (PuLLıarınen 1981b). Partly due to the local habitat structure and partly to the 
location of favourite resting sites and scent-marked routes leading to them, the pine 
martens of the area also have routes (see PULLIAINEN 1981b; Fig. 2), but these are not the 
same game trails as are used by the semi-domestic reindeer (Rangıfer tarandus tarandus) in 
the same area. Martens only occasionally use the ski traıls of trackers, but regularly scent- 
mark the junctions between these and their own trails. 

Thus the present pine martens practised scent-marking throughout their home ranges 
and did not defend their ranges against rivals of the same species during the non-breeding 
season (PULLIAINEN 1981b). Moors (1980) also draws attention to the same point, and 
emphasizes that “actual conflict has not been observed between male mustelids establishing 
territories” and that these are “maintained more by deposition of scent than by direct 
aggression”. JOHNSON (1973) writes that “although scent marks do not cause avoidance, 
they may signal that an anımal is in foreign territory and predispose withdrawal in the 
presence of the resident animal”. STODDART (1980a: 160) states that “mammals very 
obviously distribute scent around their territories, but such odours do not prevent 
intruders from entering occupied territories” (see also THIEsSEn and Rice 1976). 

STODDART (1980a: 151) classıfies mammals for which there is clear evidence of territory 
marking into two categories, arranged according to whether the marks are principally 
placed at the territory boundary or scattered throughout the hinterland (in both categories 
there may be a concentration of marks in the immediate environs of the home site). The 
boundary markers’ group (STODDART 1980a) includes Vulpes vulpes, Canis famiharıs, 
Canis lupus (pack), Meles meles, Lutra lutra (riverine habitat), Crocuta crocuta, Canıs 
anreus, Felis tigris, Vombatus ursinus, Galago alleni and the domestic cat, and the category 
of hinterland markers the following species/groups: Herpestes auropunctatus, Hyaena 
brunnea, Lutra Iutra (marine habitat), Diceros bicornis, Oryctolagus cuniculus, most 
ungulates and most primates. 

STODDART (1980a) stresses that environmental conditions exert a strong influence on 
territorial demarcation, and this is particularly well seen in northern temperate species 
which must be catholic feeders in order to survive, while in predictable environments a 
rather constant pattern of territoriality occurs, and with it a constant demarcation 
procedure. The present pine martens are opportunistic feeders which utilize a wide variety 
of food sources, from berries and mushrooms to small mammals, birds and all kinds of 
carcasses, depending on availability (see PuLLiAINEN 1981a). 

On the basis of his own research and the work carried out by HawLey and NEwBY 
(1957) and MEcH and RoGers (1977), PULLIAINEN (1981b) states that the resource-space 
use pattern of Martes martes and M. americana includes all the alternatives from overlap- 
ping of the home ranges of two adult males or a male and one or more females to partial or 
no overlapping of home ranges among neighbouring males or a male and a female. Home 
ranges may or may not be extended into vacated areas (WECKWERTH and HAWLEY 1962; 
PuLLiAınEN 1981b). This pattern even allows for the possibility of three or four individuals 
feeding on the same cervid-sized carcass during one night (Nasımovıc 1948; PULLIAINEN 
1981b). NyHoLM (1970) and Power (1979) use the term “territory” in the case of the pine 
marten, whereas in his earlier paper (PuLLIAINEN 1981b) the present author was reluctant 
to do so for the reasons mentioned above. 

The author also suggested (PuLLiaınen 1981b) that scent-marking plays an important 
role in the pine marten’s-spatial orientation, so that it “reads” other individuals’ tracks, and 
soon learns whether it is worth hunting in the area or not. EATon (1970) suggests that 
cheetahs (Acinonyx jubatus) avoid each other’s hunting trails by “reading” fresh urine 
marks. MAcDonALD (1980) has correctly emphasized that there is a need to distinguish the 
functions and the effects of scent-marks, and in this respect the pine marten’s reading of 
another’s scent-marks belongs to the latter category. The author termed the resource-space 


Scent-marking in the pine marten in Finnish Forest Lapland in winter 97, 


use pattern of the pine marten marteliısm (PULLIAINEN 1981b), a feature which probably 
also occurs in the wolverine (Gulo gulo) (PuULLIAINEN 1981b; see also KOEHLER et al. 1980; 
cf. PULLIAINEN and OvAskKAINEN 1975), which also lives in the unpredictable environments 
of the northern taiga. | 

Scent-marking in the pine marten almost certainly also has self-centred primary 
functions in the life of the marker. It evidently plays a role in orientation and familıiarity 
with a locality (PuLLiaınen 1981b). Pine martens are very familiar with their home ranges, 
using the same resting sites on successive days and finding their way almost directly to their 
resting sites at night, when vision is of no avail (PuLLiaınen 1981b). The same favourite 
resting sites are known to have been used by the same and/or different individuals during 
successive winters. 

CAMENZIND (1978) found that coyotes had a seasonal pattern in urine marking, with a 
minimum during September and October. PETERS and MEcH (1975) recorded a peak ın 
the frequency of raised leg urination of wolves in February (3.5 times/km), fallıng to 1 
per km in March. On. the basıs of this data and his own observations on red foxes, 
MACDONALD (1980) states that in many cases the peak coincides with the period of pair 
formation and courtship, and also, perhaps fortuitously, with the dispersal of non- 
resident juveniles. 

The present data on the pine marten covers only the snowy season. Since copulation 
takes place in June (DAanıLov and TumAnov 1976), one interesting season, in the light of 
the above-mentioned studies, lies outside the scope of the present investigations. During 
the present work it was found that the frequency of defaecation decreased almost to one 
third during the course of the winter, while that of urination almost doubled and that of 
rubbing secretion from the anal glands multiplied many times over (Fig. 1), thus partly 
confirming the results of the preliminary study (PULLIAINEN and HEIKKINEN 1980). There 
was a considerable nıght-to-night variation in these frequencies, however, especially ın that 
of rubbing secretion from the anal glands (see also PuLLıAInEn 1981b). 

In most mammals the males use scent-marking more frequently than females (see review 
in JOHNnsoN 1973; 'THIEssEN and RıcE 1976; STODDART 1980a). At an earlıer stage in the 
present work (PuLLIAINEn 1981b) no differences were recorded between the sexes in this 
respect, but the present data suggest a difference corresponding to the usual pattern. The 
interindividual variation is so great, however, that further studies are needed before any 
final conclusions can be reached. 

At the same stage (PuLLIAINEN 1981b) the author also draws attention to some cases ın 
which the rubbing of secretion from the anal glands was especially frequent after snowfalls. 
A more detailed study nevertheless revealed that this is not always the case (see Figs. 3 and 
4). There are probably many factors which affect the scent-marking frequency of a pine 
marten. In early winter, when the snow cover is thin, odours from the summer can still be 
detected, but gradually their concentration decreases. The increase in the depth of the snow 
also acts in the same direction, but the scent-marks left by the pine martens sımilarly 
become weaker during long spells with no snowfall, and the spots must be re-marked. In 
the field ıt often seems that the faeces left on the tops of long stumps act in the orientation 
of martens like the buoys and beacons for ships on the sea. Thus the relatively numerous 
faeces left in early winter may partly explain the relative scarcity of other scent-markings. 
As already mentioned, external stimuli such as ski tracks or the trails of other mammals, 
birds, etc., may also trigger scent-marking in the pine marten. 

Field studies reveal that pine martens recognize the smells of the other martens 
individually (PuLLıaınen 1981b), and a male marten will now and then follow the tracks of 
a female living in his home range. Sımms (1979) found that Mustela erminea and M. frenata 
in North America have distinctly different smells, and notes that these species regularly 
traverse their ranges and advertise their presence by scent-marking, the degree of interfer- 
ence depending on population densities and the availability of refuges for the smaller 


98 E. Pulliainen 


species. The present author (PuLLIAINEN 1981b) has suggested that the pine marten also has 
its distinctive smell, which serves for interspecific recognition. 

MAcDoNALD (1980) also warns against attempts to fit urine of faecal marking into arole 
which stresses only their aversive functions. It ıs also possible that the roles which odours 
play vary intraspecifically. Thus the results obtained ın unpredictable northerly environ- 
ments, where the area of a home range of a pine marten may be almost 100 km/, are not 
necessarily relevant to more predictable environments. The costs of the different activities, 
such as defending a large space, must be taken into consideration. 


Acknowledgements 


The author wishes to express his sincere gratitude to the following persons: MARTTI LAITINEN, JOUKO 
lIVANAINEN, TEUVO HIETAJÄRVI, KALLE VAARALA, REINO TAsKINEN and JUHO TÖRMÄNEN, for their 
technical assistance in the field. 

This paper constitutes Report No. 120 from the Värriö Subarctic Research Station of the 
University of Helsinki. 


Zusammenfassung 


Duftmarkierung des Baummarders (Martes martes L.) im Wald-Lappland Finnlands im Winter 


Die Duftmarkierungen des Baummarders (Martes martes) wurden ım unberührten Wald-Lappland 
Finnlands, in Nord-Europa, untersucht, indem Spuren von Individuen dieser Art in zwei aufeinander 
folgenden Wintern 2481 km lang gefolgt wurde. Die Untersuchungen haben folgendes ergeben: Die 
durchschnittliche Defäkationshäufigkeit sank während des Winters (von 0,8 mal/km Ende Oktober 
bis 0,3 mal/km im März), während die Häufigkeiten von Urinabgabe (0,9 mal/km Ende Oktober bis 
Dezember gegenüber 1,6 mal/km im Januar bis März) und Frottieren der Analdrüsen (entsprechend 
0,8 mal/km gegenüber 5,4 mal/km) anstiegen. Die letztgenannte Handlung zeigte sogar von Nacht zu 
Nacht erhebliche Schwankungen. Die Marder duftmarkierten in ihrem gesamten Wohngebiet. Orien- 
tierung (sogar unter Bedingungen, bei denen das Sehvermögen versagt), Erkennen des Wohngebietes 
und individuelle Identifizierung werden als Primärfunktionen dieser Duftmarkierungen betrachtet. 
Demgegenüber wird das „Ablesen‘“ der Spuren anderer Artgenossen, und damit die Aufrechterhal- 
tung des sog. Martelismus (PuLLiAINENn 1981b) als ein praktischer Effekt dieser Handlung angesehen. 
Er ıst erforderlich, da unter den extremen Verhältnissen des Nordens die Größe des Wohngebietes 
eines Marders fast 100 km? betragen kann. 


References 


CAMENZIND, F. J. (1978): Behavioral ecology of coyotes on the National Elk Refuge, Jackson, 
Wyoming. In: Coyotes. Ed. by M. BEKoOFF. New York: Academic Press. 267-293. 

DanıLov, P. I.; IVANTER, E. V.; BELKIN, V. V.; NIKOLAEVSKIJ, A. A. (1978): Izmenenija Cislennosti 
ohotnicih zvere) Karelii po materialam zimnih marsrutnyh ucetov. In: Fauna i ekologija ptic ı 
mlekopitajuslih taeznogo Severo-Zapada SSSR. Ed. by E. V. IvanTEr. Petrozavodsk: Nauka. 
128-159. 

DanıLov, P. I.; Tumanov, J. L. (1976): Kuni severo-zapada SSSR. Leningrad: Nauka. 

EAToON, R. L. (1970): Group interactions, spacing and territoriality in cheetahs. Z. Tierpsychologie 27, 
481-491. 

ERLINGE, $. (1968): Territoriality of the otter, Zutra lutra L. Oikos 19, 81-88. 

EWER, R. F. (1968): Ethology of mammals. London: Elek Press. 

GORMANn, M.L. (1980): Sweaty Mongooses and other Smelly Carnivores. Symp. zool. Soc. Lond. 45, 
87-105. 

Hawıey, V. D.; Newgy, F. E. (1957): Marten home ranges and population fluctuations. J. Mammal- 
ogy 38, 174-184. 

JoHnson, R. P. (1973): Scent marking in anımals. Anım. Behavior 21, 521-535. _ 

KOEHLER, G. M.; HORNOCKER, M. G.; HasH, H. $. (1980): Wolverine Marking Behavıor. Can. Field- 
Nat. 94, 339-341. 

Kruuk, H. (1978): Spatial organisation and territorial behaviour of the European badger, Meles meles. 
J. Zool., London, 184, 1-19. 

Kruuk, H.; Hewson, R. (1978): Spacing and forgaging of otters (Lutra lutra) in a marine habıtat. ]. 
Zool. London, 185, 205-212. 

LiNHART, $. B.; KnowLTon, F. F. (1975): Determining the relative abundance of coyotes by scent 
station lines. Wildl. Soc. Bull. 3, 119-124. 

Lockie, J. D. (1966): Territory in small carnıvores. Symp. zool. Soc. London 18, 143-165. 


Scent-marking in the pine marten in Finnish Forest Lapland in winter 32 


MacponALD, D. W. (1980): Patterns of Scent Marking with Urine and Faeces Amongst Carnivore 
Communities. Symp. zool. Soc. London 45, 107-139. 

MecnH, L. D.; Rocers, L. L. (1977): Status, distribution, and movements of martens in northeastern 
Minnesota. USDA For. Serv. Res. Pap. NC-143, 1-7. 

Moos, P. J. (1980): Sexual dimorphism in the body size of mustelids (Carnivora): the roles of food 
habits and breeding system. Oikos 34, 147-158. 

Nasımovic, A. A. (1948): Ekologia lesnoj kunitsy. Trudy Laplandskogo Gos. Zapovednika 3, 81-105. 

NYHoLM, E. (1970): Näädän elintavoista, saalıstuksesta ja ravinnosta. Suomen Riista 22, 105-118. 

PETERS, R. P.; MecH, L. D. (1975): Scent marking in wolves. Am. Scient. 63, 628-637. 

PoweEıı, R. A. (1979): Mustelid spacing patterns: variations on a theme by Mustela. Z. Tier- 
psychologie 30, 153-165. 

PULLIAINEN, E. (1976): Birch forest damage caused by Oporinia autumnata Bkh. (Lep., Geometridae) 
in 1965-1966 in eastern Itäkaira, northeastern Lapland. Ann. Ent. Fenn. 42, 166-170. 

PULLIAINEN, E. (1981a): Food and feeding habits of the pine marten in Finnish Forest Lapland in 
winter. In: Proceedings of the First Worldwide Furbearer Conference, August 3-11 1980, 
Frostburg, Maryland, USA. Ed. by J. A. CHapman, D. Pursre£y. Frostburg: Worldwide Furbearer 
Conf. Inc. 

PULLIAINEN, E. (1981b): Winter habitat selection, home range, and movements of the pine marten in 
Finnish Forest Lapland. In: Proceedings of the First Worldwide Furbearer Conference, August 
3-11 1980, Frostburg, Maryland, USA. Ed. by J. A. CHapman, D. Purstey. Frostburg: World- 
wide Furbearer Conf. Inc. 

PULLIAINEN, E.; HEIKKINEN, H. (1980): Behaviour of the pine marten (Martes martes) in E-Finnish 
Forest Lapland ın winter. Suomen Riıısta 28, 30-36. 

PULLIAINEN, E.; OVASKAINEN, P. (1975): Territory marking by a wolverine (Gulo gulo) in northeast- 
ern Lapland. Ann. Zool. Fennicı 12, 268-270. 

Quick, H.F. (1951): Notes on the ecology of weasels in Gunnison County, Colorado. J. Mammalogy 
32, 281-290. 

Rauıs, K. (1971): Mammalıan scent marking. Science, Wash. 171, 443-449. 

SETON, E. T. (1909): Life histories of northern anımals 2. New York: Charles Scribner and Son. 

STODDART, D.M. (1980a): The ecology of vertebrate olfaction. London and New York: Chapman and 
Hall. 

STODDART, D. M. (ed.) (1980b): Olfaction in Mammals. Symp. zool. Soc. London 45, 1-363. 

STODDART, D. M. (1980c): Aspects of the Evolutionary Biology of Mammalıan Olfaction. Symp. zool. 
Soc. London 45, 1-13. 

THIESSsEN, D.; Rıce, M. (1976): Mammalian Scent Gland Marking and Social Behavior. Psychol. Bull. 
83.505339. 

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


Author’s address: Prof. Dr. Erkkı PurLiaınen, Department of Zoology, University of Oulu, and 
Värriö Subarctic Research Station, University of Helsinki, Kasarmintie 8, SF-90100 
Oulu 10, Finland 


Gehegebeobachtungen zur Geburts- und Reproduktionsbiologie 
des Europäischen Wildschweines (Sus scrofa L.) 


Von M. MARTYysS 


Zoologisches Institut der Universität Salzburg und Institut für Vergleichende Verhaltensforschung der 
Österreichischen Akademie der Wissenschaften (Abt. Tiersoziologie), Herzog-von-Cumberland- 
Stiftung Grünanu ı. Almtal 


Eingang des Ms. 5. 12. 1981 
Abstract 


Observations on parturition and reproductive biology in captive European wild boars (Sus scrofa L.) 


Studied the farrowing behaviour in the European wild boar Sus scrofa L. in captivity (20 litters, 119 
piglets). In addition data on certain aspects of reproductive biology are given. 

During the prepartum period the gravid females exhibited conspicuous signs of the approach of 
parturition, such as changes ın behavioural patterns, nestbuilding, the onset of milk secretion etc. 

The average time required for farrowing was 141 # 63 min. (72-319 min.), or 22.4 min. per piglet. 
Intervals between individual births varied from 1 to 124 min. 78 % of the piglets appeared with torn 
foetal membranes and 6 % had a divided umbilical cord in the moment of birth. The majority of the 
foetal membranes were expelled after all piglets had been born. The average time taken to expel the 
membranes was 69 # 42 min. A negative correlation was found between the length of the farrowing 
phase and the duration of expelling the foetal membranes (r = -— 0.88). The time required for the 
birth of a whole litter was on an average 198.84 + 34 min. 

The main farrowing season is March (60 % of 60 observed births), a second peak is in July 
(18.3 %). The average reproduction rate is 5.8 piglets per female (1-13 piglets). During the first year of 
life the mortality rate was 23 % (40 of 173 piglets). 4 piglets were stillborn, 1 deformity was registrated 
(Brachygnathia sup.). 17 % of the piglets died during the first weeks of their life because of bad 
weather conditions. 


Einleitung 


Die Fortpflanzungbiologie des Wildschweines wurde - nicht zuletzt aus forstwirtschaftli- 
chen Überlegungen - bereits mehrfach untersucht. Dabei sind Fragen bezüglich Fortpflan- 
zungszeit, Trächtigkeitsdauer, Bestandsentwicklung, Mortalitätsrate etc. im Zusammen- 
hang mit ökologischen Faktoren erörtert worden (OLoFF 1951; SLupsk1J 1956; KozLo 
1970, Rakov 1970; BRIEDERMANN 1971; JEZIERSKI 1977; STUBBE und STUBBE 1977; Vos 
und Sassanı 1977; ANDRZEJEWSKI und JEZIERSKI 1978; HECK und RASCHkE 1980; u. a.). 
Solche Studien über die Populationsdynamik des Schwarzwildes beruhen gewöhnlich auf 
umfangreichen Analysen von Jagdstrecken und sind in der .Praxis eine wichtige Grundlage 
für gezielte Hegemaßnahmen (vgl. Könıg und Hormann 1980). Die folgenden Angaben 
zur Reproduktion des Wildschweines wurden im Rahmen von Geburtsbeobachtungen in 
Gehegen ermittelt. Bedingt durch die relativ geringe Anzahl der untersuchten Individuen 
richtet sich das Augenmerk auf bestimmte Aspekte ım Fortpflanzungsgeschehen. 
Beobachtungen zur Geburtsbiologie von Sus scrofa beziehen sich meist auf die 
Trächtigkeitsperiode und auf den Zeitraum nach der Geburt, z. B. auf die Abspaltung von 
der Rotte, auf den Nestbau und auf die Brutpflege der Bache (KızssLinG 1925; HEck 1950; 
Bosack 1957; NoHL 1957; BROMLEI 1964; GUNDLACH 1968; SNETHLAGE 1974; MEYN- 
HARDT 1978). Dagegen ist der eigentliche Geburtsvorgang — besonders im Freiland - einer 
direkten Beobachtung nur schwer zugänglich. Dies ist vor allem durch die scheue 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4702-0100 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 100-113 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Zur Geburts- und Reproduktionsbiologie von Sus scrofa 101 


Lebensweise des Schwarzwildes bedingt, ferner durch die Tatsache, daß eine Bache ım 
Geburtsnest meistens den Blicken des Beobachters verborgen bleibt oder bei empfindlicher 
Störung durchaus zu heftigen Verteidigungsreaktionen bereit ist. Wegen dieser Schwierig- 
keiten, auf die schon GuUNDLACH (1968) und MEYNHARDT (1978) hingewiesen haben, 
wurden ausschließlich Gehegebeobachtungen an zahmen Tieren durchgeführt, die durch 
die Anwesenheit des Menschen nicht ırritiert sind. 

Im Gegensatz zum Hausschwein gibt es nur wenige Angaben über den Ablauf der 
Geburt bei der Wildform: HEDIGER (1961) hat m. W. erstmalig über eine Wildschweinge- 
burt im Basler Zoo berichtet. GUNDLACH (1968) behandelt zwar ausführlich das Verhalten 
der Bache vor der Geburt, insbesondere den Nestbau, nähere Angaben über den eigentli- 
chen Geburtsvorgang fehlen jedoch. Bei einer Freilandbeobachtung konnte MEYNHARDT 
(1978) keine genauen Aussagen über den Geburtsverlauf machen, da die Sicht in den 
Geburtskessel durch das viele Nestmaterial behindert war. In einem Fanggatter hat REELFS 
(1978) von einer Bache ein vollständiges Geburtsprotokoll erstellt. 


Material und Methode 


Für die Untersuchungen standen 3 verschiedene Gehege des Cumberland-Wildparkes Grünau ı. 
Almtal (Oberösterreich) zur Verfügung: (A) eines mit rund 500 m? Fläche für die Beobachtung eines 
handaufgezogenen Jungtieres, (B) das eigentliche Beobachtungsgehege (ca. 1.5 ha) mit 3-6 Bachen und 
2 Keilern, und (C) das Schwarzwildgehege im öffentlich zugänglichen Cumberland-Wildpark mit 
einer Fläche von 2.5 ha und einem wechselnden Tierbestand von 3-8 Bachen und 2 Keilern. Alle 3 
Gehege haben natürliche Vegetation. 

Für die Reproduktionsbiologie wurden Daten aus dem Zeitraum 1974-1981 verwendet. In den 
Jahren 1976-1981 konnten 20 Wildschweingeburten (119 Frischlinge) in ihrem genauen zeitlichen 
Ablauf protokolliert werden. Bei 96 Frischlingen waren vollständige Geburtsbeobachtungen möglich, 
von den restlichen 23 Frischlingen waren aus verschiedenen Gründen keine genauen Angaben zur 
Geburt (Kopf- oder Steißendlage, Nabelschnur gerissen oder nicht, usw.) erhältlich. Sie wurden daher 
in der Auswertung nicht berücksichtigt. 

Die Beobachtungen wurden direkt auf Tonband protokolliert. Wenn erforderlich, wurde während 
der Nachtstunden der Geburtsplatz mit einer Taschenlampe diffus ausgeleuchtet. Aus eigener 
Erfahrung und nach Angaben von GUNDLACH (1968) und REELFS (1978) scheinen die Tiere dadurch 
nicht irritiert zu sein. 

Methodische Schwierigkeiten ergaben sich aus verschiedenen Gründen: Manche Bachen hatten im 
Nest eine für die Beobachtung ungünstige Lage eingenommen oder waren unter einem großen Haufen 
aus Reisig und Gras verborgen, so daß der genaue Ablauf der Ereignisse nicht festgestellt werden 
konnte. Trotz der Vertrautheit mit Menschen wirkten einige Sauen durch die Anwesenheit des 
Beobachters gestört, was sich in häufigem Sichern, Schnauben, oder in Angriffsintentionen äußerte. In 
solchen Fällen wurde das Protokoll sofort abgebrochen, oder es wurde von vornherein eine Beobach- 
tung mittels Feldstecher (8 X 30) versucht. Bei den zahmen Bachen im Beobachtungsgehege (B), die 
alle vom Verfasser selbst aufgezogen worden waren, oder zumindest Töchter von diesen Tieren sind, 
konnte die Geburt meistens direkt am Nest verfolgt werden, ohne irgendwelche Störungen zu 
provozieren. 

Mit den ersten Anzeichen einer bevorstehenden Geburt wurde den hochträchtigen Bachen als 
Material für den Bau des Geburtsnestes Heu, verschieden große Fichtenzweige, und im Sommer auch 
Buchenäste angeboten, so daß zusammen mit der natürlichen Vegetation im Gehege genügend 
Nestmaterial zur Verfügung stand. Die Wahl des Geburtsplatzes wurde von den Bachen jeweils selbst 
getroffen. Im Gehege (B) bestand zusätzlich die Möglichkeit, durch einen Trennzaun die übrigen 
Tiere von der betreffenden Bache abzusondern, so daß ein ungestörter Verlauf des Geburtsvorganges 
gewährleistet war. 


Ergebnisse 
Vorbereitungsstadium 


Bei vielen Tierarten errichten die Weibchen gegen Ende der Trächtigkeitsperiode ein 
' Geburtsnest oder wählen zumindest einen für die Geburt geeigneten Platz aus (NAAKTGE- 
BOREN und SLIJPER 1970). 


102 M. Martys 


Der Bau eines Geburtsnestes steht in engem Zusammenhang mit der Multiparie. Die 
Suiden sind zwar - im Gegensatz zu den meisten multiparen Säugern — bei der Geburt 
bereits soweit entwickelt, daß man sıe als „Fast-Nestflüchter‘“ (EwER 1976) bezeichnen 
kann, doch benötigen sie ın den ersten Lebenstagen die schützende Wärme eines Nestes, da 
sie während einer poikilothermen Phase ihre Körpertemperatur nicht selbständig regulie- 
ren bzw. stabilisieren können (NAAKTGEBOREN und SLIJPER 1970). 

Außer den Wurfnestern benützen Wildschweine das ganze Jahr hindurch Ruhe- und 
Schlafnester, doch bestehen diese nur aus einer im Boden ausgewühlten Vertiefung, die 
gelegentlich mit Zweigen und Gras ausgepolstert wird. Für das Geburtsnest hingegen 
verwendet die Bache außergewöhnlich viel Zeit und Mühe. In der näheren Umgebung 
sammelt sie große Mengen von pflanzlichem Material und ordnet es nach einem bestimm- 
ten Verhaltensschema, das von FRÄDRICH (1965) und GuNDLACH (1968) beschrieben 
wurde, zu einem großen Nest. 

Noch bevor das Geburtsnest errichtet wird, setzt ein auffälliges Orientierungs- und 
Territorialverhalten ein (GunnpLAcH 1968; REELFS 1978): Die Bachen sondern sich von 
ihrer Rotte ab, durchstreifen in der folgenden Zeit abgelegene Gebiete und wählen einen 
günstigen Platz für den Nestbau aus. Die nähere Umgebung dieser Stelle wird gegen 
Artgenossen verteidigt, aber auch artfremde Individuen (Mensch, Hund, Fuchs) werden 
bei einer Annäherung sofort angegriffen (REELFS 1978). Schon einige Zeit vor dem 
Wurftermin befinden sıch dıe Bachen ın der Nähe des von ihnen ausgewählten Geburts- 
platzes und verlassen dieses Gebiet nicht mehr, um die Sicherheit dieser Stelle und ihre 
eigene Vertrautheit mit den örtlichen Gegebenheiten zu erhöhen (KızssLıng 1925). Nach 
Srupsk1J (1956) erfolgt die Abspaltung von der Rotte und die Wahl eines Nestplatzes ın 
einem Zeitraum von 5-6 Tagen, GUNDLACH (1968) gibt dafür 1-3 Tage an. Auch das 


: Geburt am 16. 
1. Nest am 9. 


Geburt am 19. 
1. Nest am 16. 


7 

Y 

3 

3 

Geburt am 19. 3. 
1. Nest am 373781 

: Geburt am 20. 4 

1. Nest am 26. 2 

7 

7. 


: Geburt am 28. 
1. Nest am 9. 


M1:800 
0 5104220 


Abb. 1. Lage der Ruhenester während der Vorbereitungsperiode und Lage der Geburtsnester von5 
Bachen im Cumberland-Wildpark, Gehege B (Skizze) 


Zur Geburts- und Reproduktionsbiologie von Sus scrofa 103 


Hausschwein beginnt rund 1-3 Tage vor der Geburt mit der Suche nach einer geeigneten 
Stelle für ein Nest (HarEz et al. 1962). 

Nach den Beobachtungen im Gehege erstreckt sich die Suche nach dem Nestplatz über 
eine bis mehrere Wochen. Anfangs sind häufig Einzelelemente des Nestbauverhaltens zu 
sehen: Während die Bache im Gelände umherstreift, nimmt sie mit dem Maul kleine Äste 
und Grasbüschel auf, trägt sie einige Schritte mit und läßt sie dann wieder fallen (REELFS 
1978). Später werden an verschiedenen Stellen Mulden ausgewühlt und zum Teil mit 
Nestmaterial ausgepolstert, das zuvor in der Umgebung eingesammelt oder einfach vom 
Nestrand in die Vertiefung eingescharrt wurde. Auf diese Weise hatte eine Bache (W1) in 
der Zeit vom 26. 2. 81 bis zur Geburt am 20. 4. 81 insgesamt 9 verschiedene Nester gebaut, 
die sie teils zum Ruhen am Tage, teils als Schlafstelle während der Nachtstunden benützte. 
Je nach Witterung legte sie die Nester bei schlechtem Wetter mit etwas Nestmaterial aus, 
an warmen Tagen jedoch meistens gar nicht. Zusätzlich angebotene Fichtenzweige und 
Heu verwendete sie zu diesem Zeitpunkt nicht weiter für den Nestbau, sondern trug sie 
höchstens einige Schritte mit sich und ließ sie dann wieder fallen. Schließlich erfolgte die 
Geburt in einem 10. Nest. In den Nestern 1 und 2, die von Bache W1 insgesamt 3 Wochen 
lang benützt wurden, duldete sie noch die Anwesenheit ihrer 3 adulten Töchter, in den 
folgenden Nestern ruhte sıe dann alleine. Die letzten 5 der 9 Schlafstellen befanden sich alle 
in unmittelbarer Nähe des späteren Geburtsplatzes, die erste davon am 9.4. 81, also 11 
Tage vor dem Wurftermin. In einem weiteren Fall hatte eine Bache (Z2) innerhalb 1 Woche 
5 verschiedene Nester benützt, bis sie ein 6. als Geburtsnest ausbaute. Im Gebiet des 
Ussuri (UdSSR) wurde eine Bache beobachtet, die 6 Nester anlegte, bevor sie das Wurfnest 
errichtete (BROMLEI 1964). In den Gehegen des Cumberland-Wildparkes konnte dieses 
Verhalten bei fast allen Bachen festgestellt werden (Abb. 1). Aber nicht nur die Bereitschaft 
Nester anzulegen, sondern auch die Intensität des gesamten Nestbauverhaltens war 
individuell verschieden. Gewöhnlich zeigten ältere Bachen mehr Eifer als junge — ım 
Gegensatz zur Annahme von BROMLEI (1964), daß jüngere Bachen allgemein nervöser sind 
und schon früher mit der Suche nach einem geeigneten Nestplatz beginnen als ältere, 
erfahrene Tiere. 

Die in der Vorbereitungsperiode errichteten Nester entsprechen zwar im Aussehen den 
normalen Ruhe- und Schlafnestern, dennoch bestehen wesentliche Unterschiede: Während 
Schlafplätze meist über einen längeren Zeitraum Verwendung finden, legen die hochgravi- 
den Bachen auffallend häufig neue Nester an, bevorzugt in unmittelbarer Umgebung des 
späteren Geburtsnestes (vgl. Abb. 1). Sie werden - ebenso wie das nachfolgende Wurfnest 
- nur an solchen Stellen errichtet, die zuvor noch kein anderes Tier benützt hat. Außerdem 
liegen die Bachen während der Vorbereitungszeit fast immer allein in ihren Nestern, ım 
Gegensatz zu den regulären Schlafnestern, in denen normalerweise mehrere Tiere gemein- 
sam ruhen. 

Die enge Bindung zwischen Mutter und Tochter bleibt bis zur Geburt bestehen: Selbst 
hochträchtige Bachen akzeptieren die Anwesenheit ihrer eigenen Mutter und dulden sogar, 
daß diese ihnen beim Bau des Geburtsnestes hilft. Auch die vorjährigen Frischlinge dürfen 
weiterhin mit ihrer Mutter in direktem Kontakt bleiben (MEYNHARDT 1978). Gegen alle 
übrigen Artgenossen werden die Bachen im Verlauf der Trächtigkeitsperiode in steigendem 
Maße aggressiv. Vor allem ın der Nähe der Nester greifen sie jetzt entgegen den bestehen- 
den Rangverhältnissen auch dominante Rottenmitglieder an. 

Nicht nur im Verhalten, auch in körperlichen Veränderungen, die für die Vorberei- 
tungsperiode charakteristisch sind, kündigt sıch das baldige Ende der knapp viermonatigen 
Tragzeit an. Die Bachen werden schwerer und auch schwerfälliger, ihr Verhalten ist durch 
eine gewisse Gelassenheit gekennzeichnet (van PUTTEN 1978). Die Tiere liegen viel und 
bewegen sich langsamer und vorsichtiger als üblich. Auch das Gesäuge nimmt an Umfang 
zu: Als erstes werden die mittleren Zitzen länger und der Zitzenhof schwillt an. Manchmal 
läßt sich mit etwas Druck ein Tropfen trüber Flüßigkeit auspressen. Bei 1 Bache war dies 


104 M. Martys 


schon 9 Tage und bei 2 weiteren Tieren 4 Tage vor der Geburt festzustellen. Bei mehreren 
Bachen war ein paar Tage vor der Geburt ın einzelnen Zitzen bereits Milch vorhanden, 
obwohl sie in der Regel erst innerhalb der letzten 24 h einschießt. Der Zeitpunkt stimmt 
meistens mit dem Beginn der Nestbauphase überein: Ungefähr 15 h (4-40 h) vor der 
Austreibung der Frischlinge sind die Zitzen prall gefüllt, so daß mit einem leichten Druck 
ein kräftiger Strahl Milch ausgepreßt werden kann. Manchmal tropft sie sogar von selbst 
aus einzelnen Zitzen. Nach SmıpT (1937) schießt die Milch beim Hausschwein gewöhnlich 
in der Eröffnungsphase ein, durchschnittlich 3-5 h vor dem 1. Ferkel. Er fand jedoch auch 
Extreme von 48 h vor und nach (!) der Geburt. Andere Autoren (zit. bei SMIDT 1937; 
Jones 1966; STUBBE und STUBBE 1977) stellten bei ähnlichen Durchschnittswerten eben- 
falls große Schwankungen von 1 h bis zu 2 Tagen ante partum fest. Offensichtlich gibt es 
von Tier zu Tier beträchtliche individuelle Unterschiede. Aber auch die Milchproduktion 
in den einzelnen Zitzen ist nicht einheitlich. Beim Wildschwein sind es die mittleren, die als 
erstes und wohl auch am meisten Milch geben. 

Zugleich mit der Vergrößerung der Zitzen beginnt sich an der Bauchregion die dichte 
Unterwolle des Winterfelles abzulösen, bis die Zitzenregion völlig kahl ist. Der Haarausfall 
setzt bei den hinteren Zitzen ein, während die vordersten manchmal erst in den letzten 
Stunden vor der Geburt freiliegen. Hochträchtige Bachen suhlen sich besonders oft und 
scheuern dann ihre Bauchseite an Baumstrünken oder ähnlich niederen Objekten, was den 
Haarausfall an dieser Körperstelle noch zusätzlich fördert. Zahme Tiere sind in diesen 
Tagen besonders sensibel an den Zitzen und im Genitalbereich. Durch eine Berührung der 
Bauchregion mit der Hand läßt sich der Mammalreflex auslösen, der bei intensivem 
Streicheln des Gesäuges die Bache veranlaßt, sich auf die Seite zu legen und die typischen 
Lockgrunzlaute wie beim Säugen ihrer Frischlinge zu äußern. 


Geburt 


Nestban und Eröffnungsphase 


Ungefähr 12-15 h vor der Geburt des 1. Frischlings wird die Bache von einer außerge- 
wöhnlichen Unruhe befallen. Sie versucht aus dem Gehege auszubrechen und läuft unter 
ständigen Grunzlauten rastlos umher. Dazwischen geht sie ins Wasser, um sich abzuküh- 
len, und setzt dabei häufig kleinere Mengen Harn ab. Mehrmals geht sie zu einer 
bestimmten Stelle im Gehege, zeigt Nestbauverhalten wie Ausmulden und Einscharren 
und läuft dann weiter, ohne vorerst Nestmaterial aufzunehmen oder zu verbauen. Später 
wird dort das Geburtsnest errichtet. Während die Bache bis jetzt die Anwesenheit ihrer 
vorjährigen Frischlinge toleriert hat, vertreibt sie diese nun aus ihrer Nähe bzw. aus dem 
unmittelbaren Bereich ihres künftigen Nestplatzes. Das hektische Benehmen der Bache 
dauert mehrere Stunden an, dann beginnt sie mit dem eigentlichen Nestbau, der sich ohne 
Unterbrechung im Durchschnitt über 5.5 h (4.5-7 h) hinzieht. 

Die Eröffnungsphase beginnt beim Hausschwein mit den ersten Anzeichen der Unruhe 
und der Nestbauhandlungen und dauert 6-12 h (SmıDT 1937). Mehrere Autoren (Zit. bei 
SMIDT 1937) vertreten die Ansicht, daß die hektische Aktivität kurz vor der Geburt mit der 
Öffnung des Zervix und mit den verstärkten Uteruskontraktionen, die für das Tier 
unangenehm sein dürften, in Zusammenhang steht. Für das Wildschwein liegen dazu keine 
Untersuchungen vor, aber wahrscheinlich beginnt die Eröffnungsphase — wie beim Haus- 
schwein — bereits mit dem Bau des Geburtsnestes. MEYNHARDT (1978) beobachtete eine 
Bache, die während des Nestbaues einige Male innehalten mußte, weil bereits die ersten 
Wehen einsetzten. Vom Ende der Nestbauphase bis zur Austreibung des 1. Frischlings 
vergehen ungefähr 2.5 h (1-4.5 h). In 2 Fällen (im Mittelwert nicht berücksichtigt) betrug 
dieser Zeitraum 8 bzw. 9 h, in denen beide Bachen allerdings nicht die ganze Zeit im Nest 
lagen, sondern wiederholt aufstanden und das Nest verließen. Es handelte sich bei diesen 


Zur Geburts- und Reproduktionsbiologie von Sus scrofa 105 


Abb. 2. Die Bache legt sich in ihr fertiges Geburtsnest (Beachte die stark geschwollene Vulva) 


Tieren um Primiparae, was mit den Angaben von SMIDT (1937) übereinstimmt, der für 
Erstlingssauen eine längere Dauer der Eröffnungsphase (8-14 h) ermittelte. 

Sobald die Bache mit dem Geburtsnest fertig ist, läßt ıhre Aktivität deutlich nach. Von 
der anfänglichen Hektik ist nun nichts mehr zu merken. Ohne weiteres Nestmaterial 
einzutragen, ordnet die Bache das fertige Nest, indem sie sperrige Äste ins Maul nımmt 
und mit seitlichen Kopfbewegungen noch weiter in den Nesthaufen hineinsteckt. Schließ- 
lich kriecht sie unter Aufwerfbewegungen des Rüssels Kopf voran an einer Schmalseite des 
Nestes hinein und läßt sich darın nieder (Abb. 2). Manchmal ist die Bache völlig unter dem 
Nesthaufen verborgen, so daß lediglich eine kleine Öffnung den Eingang anzeigt. Des 
öfteren gelingt es ihr nur, Kopf und Rumpf zu bedecken, während das Hinterteil freiliegt. 
Sie kriecht jedoch soweit in den Haufen hinein, daß ihre Rüsselscheibe auf der anderen 
Seite unter dem Nestgewirr hervorragt. Dadurch ist sie in der Lage, rechtzeitig eine 
drohende Gefahr zu wittern. Ist das Nest nicht sehr stabil gebaut, kann es schon bei den 
ersten Hinlegeversuchen auseinanderfallen. Die Bache steht dann wieder auf, richtet das 
Nest erneut - allerdings nicht mehr so gründlich — und legt sich wieder hin. Dieses 
Aufstehen und Hinlegen kann sich mehrmals wiederholen, bis die Bache endgültig still in 
ihrem Nest liegt und die Geburt erwartet. An der geschwollenen Vulva wird blutiger 
Zervikalschleim sichtbar, der durch die Uteruskontraktionen und die vorrückenden 
Früchte ausgeschieden wurde. 


Austreibungsphase 


Nach vollständiger Eröffnung des Zervix werden die Frischlinge durch starke Wehen unter 
Mitwirkung der Bauchpresse ausgetrieben. Im allgemeinen geht die Eröffnungsphase in das 
Austreibungsstadium fließend über. Manchmal setzen schon 1 h vor der Geburt einzelne 
kräftige Wehen ein. Je näher der Zeitpunkt der Austreibung rückt, desto häufiger und 
stärker werden die Kontraktionen. Dabei geht immer wieder Fruchtwasser ab. Die Bache 
ist unruhig und verändert oft ihre Lage, manchmal läßt sie vor Schmerz und Anstrengung 
eın Stöhnen und Schmatzen hören. Zunächst ist nur alle 2-3 min eine Preßwehe zu sehen. 
Nach einer Serie von Kontraktionen kann jeweils eine Wehenpause von ca. 10 min 
einsetzen. Später nehmen die Preßwehen an Intensität und Häufigkeit zu. Sıe erfolgen 
1-3mal pro Minute ohne Zwischenpausen, bis schließlich der 1. Frischling geboren ist. Bei 
den nachfolgenden Frischlingen genügen normalerweise einige wenige Preßwehen, manch- 
mal sind vor der Austreibung gar keine Anzeichen zu sehen. Die Bache liegt ganz ruhig, 
hebt plötzlich den Schwanz, und ein neugeborener Frischling gleitet heraus. 


106 M. Martys 


Abb. 3. Zur Unterstützung des Austreibungsvor- 
ganges preßt die Bache im Stehen. Der Kopf des 
Frischlings ist bereits sichtbar 


Abb. 4. Die intakte Nabelschnur hindert den 


Frischling an die Zitzen zu gelangen 


In der Regel bleiben die Bachen wäh- 
rend der Geburt in Seitenlage. Manchmal 
wechseln sie in die Bauchlage, gehen in den 
Sitz über, oder stehen sogar auf, um kräfti- 
ger pressen zu können (Abb. 3). Im Stehen 
unterstützt die Bache die Austreibung 
durch Seitwärtsbewegungen mit dem Hin- 
terteil, so daß der Frischling regelrecht her- 
ausgeschleudert wird. Bei einer schwierigen 
Geburt kann es vorkommen, daß die Bache 
plötzlich mit einem Schmerzschrei hoch- 
springt, wobei der Frischling herausrutscht. 

Die Geburt erfolgt in Vorder- oder in 
Hinterendlage. Obwohl innerhalb eines 
Wurfes keine Gesetzmäßigkeit hinsichtlich 
der Häufigkeit beider Geburtslagen zu be- 
stehen scheint, wird bei Berücksichtigung 
eines umfangreichen Materials ein annä- 
herndes Verhältnis von 1:1 deutlich 
(SMıDT 1937; NAAKTGEBOREN und SLIJPER 
1970; STUBBE und STUBBE 1977). Bei 96 
Frischlingen aus 20 Würfen waren 50 
(=52 %) Steißgeburten und 46 (=48 %) 
Kopfgeburten. Bei insgesamt 8 Würfen 
wurde jeweils 1 Kopfendlage ın unterer 
Stellung verzeichnet. Die Haltung aller 96 
Frischlinge während der Austreibung war 
durchwegs normal. In keinem Fall traten 
Abweichungen auf, wie sie u. a. von TILL- 
MANN (1951) für das Hausschwein be- 
schrieben worden sınd. Bei Steißendlage 
werden häufig schon einige Minuten vorher 
die Hinterbeine des Frischlings ın der Vul- 
va der Bache sichtbar. Sie können noch 
mehrmals verschwinden, bis der Frischling 
endgültig geboren wird. Dabei unterstützt 
er seine Austreibung durch strampelnde 
Bewegungen der Hinterbeine. 

Normalerweise platzt die Fruchtblase ın 
den Geburtswegen, nach TILLMANnN (1951) 
in den Uterushörnern, nach SMIDT (1937) 
in der Vagina. Bei einer der 96 Frischlings- 
geburten wurde jedoch 7 min vor der Aus- 
treibung in der Vulva der Bache eine intakte 


Fruchtblase von ungefähr 5 cm Durchmesser sichtbar. Die Blase zog sich zunächst wieder 
etwas zurück, dann wurde der Frischling in Vorderendlage und mit geplatzten Eihäuten 
rasch ausgetrieben. Auch für das Hausschwein wurde m. W. eine solche Ausnahme nicht 
beschrieben. Zu 78% wurden die Frischlinge mit geplatzten Eihäuten geboren. In 21 
Fällen (= 22 %) blieben die Fruchthüllen intakt und wurden erst durch die Bewegungen 
des Neugeborenen aufgerissen. Ein Zusammenhang mit der Art der Geburtslage war nicht 
ersichtlich. Die Nabelschnur war in 67 % der Geburten bereits bei der Austreibung 
durchtrennt (Tab. 1). 32 Frischlinge (= 33 %) blieben durch die Nabelschnur vorerst mit 


Zur Geburts- und Reproduktionsbiologie von Sus scrofa 107 


Tabelle 1 


Anzahl der Frischlinge, die mit durchtrennter oder intakter Nabelschnur, mit geplatzten oder 
intakten Fruchthüllen geboren wurden, in Abhängigkeit von Kopf- und Steißendlage 


Kopf- Steiß- 
endlage endlage 


Nabelschnur durchtrennt 32 32 


Nabelschnur intakt 13 19 


Fruchthülle geplatzt 39 


Fruchthülle intakt 9 


der Mutter verbunden (Abb. 4). Die intakte Nabelschnur bedeutete für die neugeborenen 
Frischlinge eine große Behinderung bei ihrem Versuch an die Zitzen zu gelangen (vgl. 
Jones 1966). Nur bei 1 Frischling war sie so lang, daß er trotz der Verbindung mit seiner 
Mutter an den hintersten Zitzen die ersten Saugversuche machen konnte. Nach durch- 
schnittlich 3.4 mın (1-7 mın) rıß die Nabelschnur durch, meistens durch die Anstrengun- 
gen der Frischlinge selbst, manchmal, weil sich die Bache erhoben hatte oder ihre Lage 
veränderte. In 2 Fällen erfolgte die Abtrennung der Nabelschnur erst, nachdem der nächste 
Frischling geboren war. 

Die Gesamtdauer der Austreibungsphase - von den ersten Prefßwehen bis zur vollende- 
ten Geburt des letzten Frischlings — betrug bei 20 Würfen im Mittel 2.3 h (x = 141 # 63 
min), bei Extremen von 72 min (5 Frischlinge) und 319 min (7 Frischlinge). Diese 
beträchtlichen Unterschiede ın der Dauer sind weniger durch die Größe des Wurfes 
bedingt, als vielmehr durch Faktoren wie Parıtät und physiologischer Zustand des Mutter- 
tieres sowie durch Größe und Gewicht der Früchte (NAAKTGEBOREN und SLIJPER 1970). 
Ausschlaggebend für die Dauer der Austreibungsphase sind vor allem die stark varıieren- 
den Zeitabstände zwischen den Geburten der einzelnen Frischlinge. Sie betrugen durch- 
schnittlich 22.4 min, wobeı 50 % aller gezählten Intervalle nur zwischen 3 und 13 mın 
dauerten und weitere 25 % zwischen 25 und 35 min. Der Rest bezog sıch auf Einzelfälle 
zwischen 1 und 124 min. Eine lange Intervallzeit bzw. Austreibungsdauer ist oft ein 
Zeichen für Geburtsschwierigkeiten, obwohl sich die Ursachen nicht immer eindeutig 
klären lassen. Kopfgeburten in unterer Stellung z. B. haben im Durchschnitt eine längere 
Intervallzeit als normale Geburten (x = 32 min). Die mit Abstand längste aller beobachte- 
ten Geburten mit einer Austreibungsdauer von 319 min und mit Intervallzeiten bis zu 124 
min (Bache Z2, Geburt am 17. 3. 1981) war durch äußere Störungen beeinträchtigt und ist 
nicht als Normalfall zu bewerten. 


Nachgeburtsphase 


Während bei den multiparen Säugern normalerweise nach der Austreibung jeder einzelnen 
Frucht auch die Nachgeburt abgeht, ist dies beim Schwein eher die Ausnahme. Gewöhn- 
lich wird nach der Geburt des letzten Frischlings ın einem Wurf durch einige kräftige 
Preßwehen die Nachgeburt ausgeschieden. Meist sind mehrere Fruchthüllen zusammenge- 
klumpt, so daß die Nachgeburt in 2-3 Portionen austritt. Bei 6 von 20 Würfen erschien 
bereits während des Geburtsvorganges oder zugleich mit dem letzten Frischling ein Teil 
der Fruchthüllen, bei 14 Geburten gab es dagegen eine deutliche Nachgeburtsphase. Ihre 
Dauer ist sehr varıabel und hängt mit der Länge der Austreibungsphase zusammen: Je 
kürzer diese ist, umso später erscheint die Nachgeburt (r = — 0.88; p< 0.001; Abb. 5). Da 
die Suiden eine Placenta diffusa besitzen, die nur oberflächlich mit der Uteruswand 
verbunden ist, kommt es relativ leicht zu einer Ablösung der Placenta ohne gröbere 


108 M. Martys 


Verletzung des Uterusge- 


sb webes (INAAKTGEBOREN 
und SLijpER 1970). Bei 

150 überdurchschnittlich lan- 
; gen Austreibungszeiten 

y=179 -0.94 x wird meistens zugleich mit 


120 dem Frischling oder inner- 


halb weniger Minuten ein 
Teil der Nachgeburt aus- 
geschieden. Im Normalfall 
dauert es rund 25-35 min, 
bis die ersten Fruchthüllen 
abgehen. Die Nachgeburt- 
sphase — gerechnet von der 
Geburt des letzten Frisch- 
lings bis zum vollständi- 
gen Ausstoßen der Nach- 


[o)) 
(@) 


Nachgeburtsphase (in min.) 
© 
(e) 


& 
(@) 


<10 60 2 120 2 180777210 geburt - dauert durch- 
Austreibungsphase (in min.) schnittlich 69 + 42 min, 

Abb. 5. Die Dauer der Nachgeburtsphase in Abhängigkeit von der im Höchstfall 145 min (bei 
Länge der Austreibungsphase (r = — 0.88) einer Austreibungszeit 


von 72 mın). Von allen Ba- 
chen wurde die Nachgeburt entweder gleich nach Beendigung der Geburt gefressen oder 
ım Lauf der nächsten Stunden, wenn sie das Nest säuberten. 

Die Austreibungsphase und die Zeitspanne, bis die komplette Nachgeburt abgegan- 
gen ist, ergeben zusammen die Geburtsdauer. Sie beträgt durchschnittlich 3.3 h 
(x = 198.84 + 34 min), bei Werten zwischen 162 und 294 min (n = 19; der durch Störun- 
gen verzögerte Geburtsablauf bei Bache Z2 am 17. 3. 1981 mit einer Dauer von 371 min 
wurde nicht mitgerechnet). Die Geburt bei Wildschweinen erfolgt demnach rascher als bei 
Haussauen, wo sıe im Schnitt 4-6 h dauert (SmıDT 1937; JOonEs 1966; STUBBE und STUBBE 
1977). Obwohl nämlich beim Hausschwein die Zeit zwischen der Austreibung zweier 
Ferkel im Schnitt kürzer ist als beim Wildschwein, ergibt sich für Haussauen - allein wegen 
der wesentlich höheren Nachwuchsrate - eine längere Gesamtgeburtsdauer. 


Reproduktionsbiologie 


Träachtigkeitsdaner 


Bei 7 Bachen konnte eine mittlere Tragzeit von 118.5 # 1.3 Tagen errechnet werden. In 
einem weiteren Fall betrug die Tragzeit 128 Tage. Henry (1968) dagegen hat eine 
Trächtigkeitsdauer von 115.2 # 2.3 Tagen ermittelt. 

Tabelle 2 


Jahreszeitliche Verteilung von 60 Wildschweingeburten im Cumberland-Wildpark Grünau 
zwischen 1974 und 1981 


Geburten je Monatshälfte 


Monate 


Geburten 


% 


Zur Geburts- und Reproduktionsbiologie von Sus scrofa 109 


Jahres- und Tageszeit der Geburten 


Von 60 registrierten Würfen zwi- 
schen 1974 und 1981 waren 46 
(= 76.7 %) in den Frühjahrsmona- 
ten Februar-April und nur 14 
(= 23.3 %) in den Sommermonaten 
Juni-August (Tab. 2). 

Nach DOMMERHOLD (zit. bei 
SLIJpER 1960) ereignen sich bei 
Haussauen die meisten Geburten in 
den Stunden vor Mitternacht. Dies 
trıfft auch für das Wildschwein zu, 
doch zeigt sich in den Soemmermona- 
ten eine auffällige Verschiebung der 
Geburten ın die späten Vormittags- 


50% 
40% 
30% 
20% 
10% 


Monat Juni - August 
(n= 10 Geburten 


10% 
20% 


30% Monat Februar - April 
40% (n= 20 Geburten) 
50% 


Abb. 6. Tageszeitliche Verteilung von Wildschweinge- 


burten im Frühjahr und im Sommer 


stunden (Abb.6). Während im 
Frühjahr die Geburten meist zwischen 21 h und 23 h stattfinden, erfolgen sie im Sommer 
am häufigsten zwischen 11 h und 13h. 

Gewöhnlich fällt bei Wildtieren der Zeitpunkt einer Geburt ın die artspezifische 
Ruhephase (NAAKTGEBOREN und SLIJPER 1970). Bei Sus scrofa ist die Aktivität im Freiland 
vorwiegend in den Nachtstunden (BRIEDERMANN 1971 a). Im Gehege jedoch sind Wild- 
schweine hauptsächlich am Tag aktiv: Während der Wintermonate und ım Frühjahr sind 
sie tagsüber oft durchgehend aktiv und legen sich dafür schon relativ bald, wenn die 
Dunkelheit einbricht, zur Ruhe. In den heißen Sommermonaten dagegen verlassen sıe 
bereits früh am Morgen ihre Nester, legen dann in der Mittagszeit eine längere Ruhepause 
ein, um erst am späten Nachmittag bis tief in die Nacht hinein aktiv zu werden. 

Die geringen Daten erlauben noch keine sicheren Aussagen über die für eine Geburt 
bevorzugte Tageszeit. Ungeklärt ist auch, ob im Freiland eine ähnliche Beziehung zwi- 
schen Ruhephase und Geburtstermin besteht, wie sie im Gehege zu beobachten ist. 


Wurfgröße 


In der Anzahl der Frischlinge zeigte sich ein deutlicher Unterschied zwischen Erstlingsba- 
chen und Mehrgebärenden: Primiparae hatten durchschnittlich 3.8 Frischlinge pro Wurf 
(1-5 Frischlinge), Pluriparae dagegen im Mittel 6.19 (4-10 Frischlinge). In einem weiteren 
Fall hatte eine Bache insgesamt 13 Frischlinge geboren (Tab. 3). Am nächsten Morgen 
waren aber bereits 7 davon tot, und in der folgenden Zeit überlebte vom ganzen Wurf nur 
ein einziger. Der Extremfall von 13 Frischlingen wurde bisher nur von OLorr (1951) 
erwähnt. Die Überlebenschancen eines solchen Wurfes sind — wie im vorliegenden Fall — 
äußerst gering. Allein durch die Zitzenanzahl der Bache ist ein oberes Limit von 10 


Tabelle 3 


Anzahl der Frischlinge pro Wurf bei 9 primiparen und 33 pluriparen Bachen 
zwischen 1974 und 1981 


n = 245 Frischlinge 


Wurfzahl 


Primiparae 


Pluriparae 


110 M. Martys 


erfolgreich aufgezogenen Frischlingen gegeben. Doch sogar bei einer solchen Wurfgröße 
ist die Zahl der Überlebenden meistens niedriger. 


Zweimaliges Frischen 


Geht ein Wurf zeitig ım Frühjahr verloren, kann die Bache erneut das Stadıum der Brunst 
erlangen und dann im Sommer desselben Jahres ein 2. Mal werfen. Im Gehege war dies bei 
einigen Tieren der Fall, doch die Mehrzahl der Bachen kam nach Verlust des Wurfes erst 
wieder ım nächsten Jahr zur Fortpflanzung. 

Während LINDEMANN (1953) und STUBBE und STUBBE (1977) ein zweitmaliges Frischen 
im Sommer als Ersatz für den Verlust des 1. Wurfes im Frühjahr durchaus für möglich 
halten, hat GoTTscHLIcH (1975) trotz entsprechender Voraussetzungen (Verluste von 
Würfen zu Jahresbeginn, Eichelmast) in einem Revier in der DDR keinerlei Anhaltspunkte 
dafür gefunden. Nach OLorr (1951) ist in Mastjahren ein zweimaliges Frischen ohne 
Verluste zumindest in westdeutschen Schwarzwildgbieten durchaus üblich. Auch HENNIG 
(1981) teilt mehrere zuverlässige Beobachtungen von der erfolgreichen Aufzucht zweier 
Würfe innerhalb eines Jahres mit. Im Cumberland-Wildpark trat so ein Fall nur ein, wenn 
sich 2 Bachen sehr bald nach der Geburt ihrer Frischlinge zu einem Mutterfamilienverband 
(REELFS 1978) zusammenschlossen und ihre Würfe gemeinsam führten. Manchmal über- 
nahm dann eine der beiden Bachen das Säugen der ganzen Frischlingsschar, wobei 
allerdings einige schwache Frischlinge zu kurz kamen und starben. Die andere Bache aber, 
die auf diese Weise ihre Frischlinge abgegeben hatte, konnte nach einer neuerlichen 
Befruchtung im Sommer desselben Jahres ein 2. Mal Frischlinge führen. 

Zwischen 1974 und 1981 gelang es einer einzigen Bache, ihre 4 Frischlinge vom 
Frühjahr (Geburt am 23. 2. 1978 im Cumberland-Wildpark) erfolgreich aufzuziehen und 
zusätzlich im Sommer ein 2. Mal zu werfen (Geburt am 10. 7. 1978, 6 Frischlinge). Anfangs 
duldete sie noch die älteren Frischlinge im Nest. Später wurden diese von ıhr vertrieben, 
besonders, wenn sie sich zum Säugen ihrer jüngeren Frischlinge hinlegen wollte. Auch in 
diesem Fall war ein zweimaliges Frischen nur möglich, da die Frischlinge des 1. Wurfes 
häufig auch bei der Schwester der betreffenden Bache saugen durften (die ihrerseits einen 
Wurf verloren hatte). Auf diese Weise sozusagen ‚‚entlastet‘‘, wurde die Bache erneut 
fortpflanzungsbereit. Ungewöhnlich ist, daß die Bache im Sommer ıhre beiden Würfe 
gemeinsam führte. 

Das Phänomen des zweimaligen Frischens ıst nıcht unbedingt auf Degenerationser- 
scheinungen von Gattersauen zurückzuführen, wie dies KıEssLıinG (1925) und STUBBE und 
STUBBE (1977) vermuten. Einerseits tritt es auch unter Gehegebedingungen sehr selten auf, 
andererseits gibt es aus Jagdrevieren Berichte, daß eine Bache innerhalb eines Jahres zwei 
Würfe erfolgreich aufgezogen hatte (mündl. Mitteilung von FM Graf SCHAFFGOTSCH, 
Fürst-Schwarzenbergische Forstverwaltung Ödenkirchen, Niederösterreich). Weiters ist 
die Bildung von Mutterfamilienverbänden auch im Freiland nachgewiesen worden (GunD- 
LACH 1968; MEYNHARDT 1978; REELFS 1978), so daß bei einem Zusammentreffen mehrerer 
günstiger Faktoren durchaus eine ähnliche Situation eintreten kann, wie sie im konkreten 
Fall aus dem Cumberland-Wildpark mitgeteilt wurde. 


Mortalität 


Die Verluste bei 173 Frischlingen aus 29 Würfen sind in Tabelle 4 aufgeführt. 

Die 4 Totgeburten waren normal entwickelt. Es ist anzunehmen, daß die Todesursache 
direkt mit dem Geburtsvorgang zusammenhing (Asphyxie). Im Vergleich zum Haus- 
schwein, bei dem rund 5-10 % aller Ferkel totgeboren werden (STUBBE und STUBBE 1977), 
scheinen solche Ausfälle beim Wildschwein seltener zu sein. In den ersten 12 h post partum 
starben 7 Frischlinge (4 %). Ein neonater Frischling wog nur 425 g. Das durchschnittliche 


Zur Geburts- und Reproduktionsbiologie von Sus scrofa 111 


Tabelle 4 


Peri- und postnatale Mortalität bei 29 Würfen (173 Frischlinge) von Sus scrofa im Cumberland- 
Wildpark zwischen 1976 und 1981 


Zeitraum Frischlinge % (n = 173) 


Totgeburt 4 


Während der ersten 12 h postpartum gestorben 7 


Im Verlauf der Säugeperiode gestorben 29 


Summe 


Geburtsgewicht liegt jedoch zwischen 700 g und 1000 g (HEPTNER et al. 1966). Der 
Frischling verendete nach ca. 1 h ohne jemals aufgestanden und zu den Zitzen gegangen zu 
sein. Bei einem 2. Frischling wurde eine Brachygnathia sup. mit Verschluß der äußeren 
Nasenöffnungen und Deformationen im Bereich des Oberkiefers und Nasenraumes festge- 
stellt. Da für das Tier keine Überlebenschance bestand, mußte es getötet werden. 2 weitere 
Frischlinge wurden kurz nach der Austreibung von der sich hinlegenden Bache erdrückt. 
Gewöhnlich reagiert sie sofort auf das Quieken ihrer Frischlinge, steht auf und richtet das 
Nest, um sich erneut hinzulegen. Ausnahmsweise können dabei Verluste durch Erdrücken 
bzw. Ersticken auftreten. Auch Haussauen in Streßsituationen gefährden ihre Ferkel, da sıe 
ständig ihre Position wechseln und dabei leicht eines der Neugeborenen erdrücken (HAFEZ 
et al. 1962). Innerhalb weniger Stunden post partum starben 3 Frischlinge, da sie im Nest 
keine Orientierung hatten und nicht zu den Zitzen fanden. Sie verkrochen sich im Stroh, 
bis sie an Entkräftung eingingen. (Von dem Wurf mit 13 Frischlingen wurden am Morgen 
nach der Geburt 7 Frischlinge tot aufgefunden. Da Zeitpunkt und Ursache ihres Todes 
unbekannt waren, konnten sie in der Auswertung nicht berücksichtigt werden). 

Während der Säugeperiode starben 29 Frischlinge (16.7 %), wobei als Hauptursache 
ungünstige Witterungsverhältnisse (hohe Schneelage, naßkaltes Wetter) zu nennen sind. 
Die Mortalität der Frischlinge während der gesamten Aufzuchtperiode beträgt 23 %. 
Auffallend ist die Übereinstimmung der vorliegenden Angaben aus dem Gehege mit 
Freilanduntersuchungen (BROMLEI 1964; JEZIERSKI 1977). Während des ersten Lebensjah- 
res sind klimatische Einflüsse besonders entscheidend für die Ausfallsquote unter den 
Frischlingen (KozLo 1970; STUBBE und STUBBE 1977; Vos und Sassanı 1977). In naturna- 
hen Biozönosen kommen noch weitere Verluste durch Freßfeinde hinzu (SLupDsk1J 1956; 
BRoMLEI 1964; RAKov 1970). 


Danksagung 


S. K.H. Ernst Ausust, Prinz von Hannover, Vorstand der Herzog-von-Cumberland-Stiftung, bin 
ich für die großzügige Förderung der Schwarzwildforschung in Grünau i. Almtal (Oberösterreich) zu 
Dank verpflichtet. Weiters gilt mein Dank für seine große Hilfsbereitschaft Herrn OFM Dipl.-Ing. K. 
HÜTHMAYR, der mir Einrichtungen und Tiere des Cumberland-Wildparkes zur Verfügung stellte. 
Meinen Lehrern Prof. Dr. Dr. h. c. K. Lorenz und Prof. Dr. H. Anpam danke ich für die vielen 
Gespräche und wertvollen Ratschläge, die wesentlich zum Gelingen der Forschungsarbeiten beigetra- 
gen haben. Meiner Mitarbeiterin Fräulein G. Burringer danke ich für ihren Einsatz bei den 
Geburtsbeobachtungen und für die Diskussion des Manuskriptes. Herrn Dr. A. GOLDSCHMID, 
Zoologisches Institut Salzburg, danke ich für Sektion und Befund der Brachygnathie sup. Frau Chr. 
MARTys, Landeskrankenanstalten Salzburg, hat in dankenswerter Weise Röntgenaufnahmen des 
Frischlings angefertigt. Für wertvolle Angaben zur Reproduktionsrate des Schwarzwildes, insbeson- 
dere in den Jahren 1974-1976, danke ich Herrn F. SANTNER, Tierpfleger im Cumberland-Wildpark 
Grünau. Herrn A. ZOHMANN danke ich für Literaturbeschaffung. Die Untersuchungen wurden vom 
Forschungsfonds des Bundesministeriums für Wissenschaft und Forschung finanziert (Projekt Nr. 
2066). 


112 M. Martys 


Zusammenfassung 


In der Vorbereitungsperiode errichtet die gravide Bache mehrere Ruhenester ın der Umgebung des 
zukünftigen Geburtsplatzes und verteidigt diese Stelle gegen Artgenossen mit Ausnahme ihrer 
vorjährigen Frischlinge und der eigenen Mutter. Ein deutliches Zeichen einer baldigen Geburt ist das 
Anschwellen des Gesäuges und die beginnende Milchproduktion in den Zitzen. Da dieser Zeitpunkt 
individuell sehr unterschiedlich ist, läßt sich daraus kein sicherer Wurftermin bestimmen. Rund 12 bis 
15 h vor der Geburt ändert sich das Verhalten der Bache auffällig: Sie ist hyperaktiv und sehr erregt. 
Auch gegen ihre Frischlinge vom Vorjahr ist sie nun aggressiv und vertreibt sie aus ihrer Nähe. Nach 
einigen Stunden beginnt sie mit dem Bau des Geburtsnestes, der sich über rund 5.5 h erstreckt. 
Währenddessen setzt die Geburt mit der Eröffnungsphase ein. 

Die Austreibungsphase dauert durchschnittlich 141 + 63 min. Die Frischlinge werden im Schnitt in 
22.4 min geboren, doch kommen sehr große Unterschiede in den Austreibungszeiten vor (1-124 min). 
Die Mehrheit der Frischlinge wird mit geplatzten Eihäuten (78 %) und mit durchtrennter Nabel- 
schnur (67 %) geboren. 

Die Länge der Nachgeburtsphase ist abhängig von der Dauer der Austreibungsphase (r = - 0.88) 
und beträgt durchschnittlich 69 # 42 min. Bei langen Austreibungszeiten werden Teile der Fruchthül- 
len bereits während der Geburt oder zugleich mit dem letzten Frischling ausgeschieden. Nach 
Beendigung der Geburt frißt die Bache die Nachgeburt. 

Die Geburtsdauer betrug im Schnitt 3.3 h (x = 198.84 + 34min) bei Werten zwischen 162 min und 
294 mın. 

Die meisten Würfe kommen im März eines Jahres (60 %), ein Nebenmaximum liegt im Juli 
(18.3 %). Dabei handelt es sich um Geburten von Bachen, die im Frühjahr ihren Wurf verloren hatten, 
erneut befruchtet wurden und dann im Sommer einen Ersatzwurf führten, oder um Jungtiere, die 
erstmals trächtig wurden. Ein zweimaliges Frischen tritt selten und nur unter bestimmten Vorausset- 
zungen auf. Die Wurfgröße reicht von 1-13 Frischlingen bei einer durchschnittlichen Reproduktions- 
rate von 5.8 Frischlingen pro Bache. Die Mortalität während der Aufzuchtsperiode liegt bei 23 % (40 
Frischlinge aus 29 Würfen). 


Literatur 


ANDRZEJEWSKI, R.; JEZIERSKI, W. (1978): Management of a wild boar population and its effects on 
commercial land. Acta Theriol. 23, 309-339. 

BoBAck, A. W. (1957): Das Schwarzwild. Verlag Radebeul. 

BRIEDERMANN, L. (1971): Zur Reproduktion des Schwarzwildes in der DDR. Beitr. Jagd- u. 
Wildforsch. 7, 169-186. 

— (1971a): Ermittlungen zur Aktivitätsperiodik des mitteleuropäischen Wildschweines. Zool. Garten 
N. F. 40, 302-327. 

BROMLEI, G. F. (1964): Das Ussurische Wildschwein (Sus scrofa ussuricus Heude). Moskau: Verlag 
Akad. Wissensch. UdSSR (Russ.). 

EWER, R. F. (1976): Ethologie der Säugetiere. Hamburg und Berlin: Paul Parey. 

FRÄDRICH, H. (1965): Zur Biologie und Ethologie des Warzenschweines (Phacochoerus aethiopicus 
Pallas) unter Berücksichtigung des Verhaltens anderer Suiden. Z. Tierpsychol. 22, 328-393. 

GOTTSCHLICH, H. ]J. (1975): Beitrag zur Vermehrung und Entwicklung eines Schwarzwildbestandes. 
Beitr. Jagd- u. Wildforsch. 9, 204-207. 

GUNDLACH, H. (1968): Brutfürsorge, Brutpflege und Tagesperiodik beim europäischen Wildschwein 
(Sus scrofa L.). Z. Tierpsychol. 25, 955-995. 

HarEZ, E. S. E.; SumpTIon, L. J.; JaKway, J. S. (1962): The behaviour of swine. In: The behaviour of 
domestic animals. Ed. by Harzz E. $. E. London: Bailliere, Tindall & Cox. pp. 334-369. 

Heck, L. (1950): Schwarzwild — Lebensbild des Wildschweines. München: BLV. 

Heck, L.; RASCHKE, G. (1980): Die Wildsauen. Hamburg und Berlin: Paul Parey. 

HEDIGER, H. (1961): Tierpsychologie im Zoo und ım Zirkus. Basel: Reinhardt AG. 

HEnnIG, R. (1981): Schwarzwild. München, Wien, Zürich: BLV Jagdbuch. 

HENRY, V. G. (1968): Length of estrous cycle and gestation in european wild hogs. J. Wildl. Mangem. 
32, 406-408. 

HEPTNER, V. G.; Nasımovic, A. A.; BannIKov, A. G. (1966): Die Säugetiere der Sowjetunion. Bd. 1. 
Jena: VEB G. Fischer. 

JEZIERsKI, W. (1977): Longevity and mortality rate in a population of wild boar. Acta Theriol. 22, 
337-348. 

Jonzs, J. E. T. (1966): Observations on parturition in the sow. Part I, II. Brit. Vet. J. 122, 420-426; 
471-478. 

KızssLing, W. (1925): Das Schwarzwild. Neumann — Neudamm. 

König, R.; Hormann, R. R. (Hrsg.), (1980): Schwarzwildsymposion Giessen. Stuttgart: F. Enke 
Verlag. 


Zur Geburts- und Reproduktionsbiologie von Sus scrofa #13 


Kozıo, P. G. (1970): Factors determining the population density of the wild boar (Sus scrofa) in the 
Belovezhskaya Pushcha. Zool. J. 49, 422-430. (Russ.). 

LINDEMANN, W. (1953): Zu „Zweimaliges Frischen der Bachen“. Wild u. Hund 56, 120. 

MEYNHARDT, H. (1978): Schwarzwild-Report. Leipzig: Neumann. 

NAAKTGEBOREN, C.; SLIJPER, E. J. (1970): Biologie der Geburt. Hamburg und Berlin: Paul Parey. 

Non, G. (1957): Frischende Bache. Wild und Hund 60, 311. 

OLorr, H. B. (1951): Zur Biologie und Ökologie des Wildschweines. Beitr. Tierkde. Tierz. 2. 
Frankfurt a. Main: P. Schöps. 

PUTTEN, G. van (1978): Schwein. In: Nutztierethologie. SAMBRAUS (Hrsg.). pp. 168-213. Hamburg 
und Berlin: Paul Parey. 

Raxov, N. V. (1970): Todesursachen beim Wildschwein und seine Beziehungen zu Räubern im 
Amurgebiet. Zool. J. 49, 1220-1228. (Russ.). 

Reeırs, H. (1978): Beitrag zur Fortpflanzungsbiologie des Europäischen Wildschweines (Sus scrofa). 
Dipl.-Arbeit Univ. Kiel. 

SLIJPER, E. J. (1960): Die Geburt der Säugetiere. In: Handbuch der Zoologie. KÜKENTHAL (Hresg.). 
Berlin: W. de Gruyter. 8/25, 1-108. 

Srupsk1J, A. A. (1956): Das Wildschwein — Ökologie und wirtschaftliche Bedeutung. Alma Ata. 
(Russ.). 

SMIDT, R. (1937): Beobachtungen bei normalen Schweinegeburten. Diss. Vet. med. HS Hannover. 

SNETHLAGE, K. (1974): Das Schwarzwild. Hamburg und Berlin: Paul Parey. 

STUBBE, W.; STUBBE, M. (1977): Vergleichende Beiträge zur Reproduktions- und Geburtsbiologie von 
Wild- und Hausschwein — Sus scrofa L. Beitr. Jagd- u. Wildforsch. 10, 153-179. 

TILLMANN, H. (1951): Zur Geburtshilfe beim Schwein. Tierärzl. Umschau 6, 317-323. 

Vos, A. de; Sassanı A. (1977). Eine Studie der Population des Schwarzwildes (Ss scrofa) in dem 
Mohammad Reza Shah Nationalpark. Z. Jagdwiss. 23, 126-131. 


Anschrift des Verfassers: MiCHAEL MARTYs, Institut für Vergleichende Verhaltsforschung, Abt. 
Tiersoziologie, Österreichische Akademie der Wissenschaften, A-4645 
Grünau ı. Almtal 11 


NZESSENSEHAETPICHETRURZMITTETELUNGEN 


Magendrüsenlänge und -dichte bei Crocidura russula 
und Suncus etruscus (Mammalia: Soricidae) 


Von ADELHEID GERAETS 
Eingang des Ms. 13. 7. 1981 


Bei der Untersuchung des Verteilungsmusters von Cardia-, Fundus- und Pylorusepithel ım 
Magen der Etruskerspitzmaus (Abb. 1) erwies sich, daß es dem von MyrcHAa (1967) für 
andere Crocidurinenarten beschriebenen entspricht. Auffällig war dagegen die Kleinheit 
der Drüsenschläuche. Sie wurden deshalb mit denen von Hausspitzmäusen verglichen. 
5-10um dicke Paraffinlängs- und -querschnitte des Magens (Färbungen: Azan, PAS, 
Hämalaun-Eosin, Hämalaun-Mucikarmin-Aurantia) von 5 Crocidura russula und 4 Suncus 
etruscus wurden mit einem Okularmikrometer vermessen. Die Drüsenlänge wurde an 
längsgetroffenen Drüsenschläuchen gemessen; bei jedem Tier wurden nach Möglichkeit 
pro Magenregion 10 Einzelschläuche erfaßt. Dies erwies sich für die Cardiaregion wegen 
ihrer geringen Ausdehnung als nicht möglich. Zudem verlaufen die Cardiadrüsen selten 
gerade, so daß genaue Längsschnitte selten sind. Deshalb konnte die Cardiadrüsenlänge 
nur an je einem Vertreter von Haus- und Etruskerspitzmaus erfaßt werden. Die Drüsen- 
dichte wurde mit einem Okularfadennetz von 0,026 bzw. 0,078 mm? ausgezählt. Pro Tier 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4702-0113 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 113-115 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


114 Adelheid Geraets 


= Oesophagusepithel 


Cardiaepithel 


ma] Fundusepithel 


M Pylorusepithel 
Verteilung der Epithelien in Oesophagus und Magen von Suncus etruscus 


wurden für jede Drüsenregion 5 Querschnitte je zweimal ausgezählt. Zur Kontrolle wurde 
die Drüsendichte auch an Längsschnitten mit dem Okularmikrometer ermittelt: die 
Anzahl der Drüsenschläuche auf 100 Teilstrichen (= 1,4mm) wurde gezählt. Dies war 
jedoch nur in der Fundus- und Pylorusdrüsenregion möglich. 

Bei beiden untersuchten Arten finden sich ähnliche Relationen der Länge der drei 
verschiedenen Drüsentypen zueinander: Die Cardiadrüsen sind am längsten, die Fundus- 
drüsen erreichen bei Crocidura 56%, bei Suncus 57%, die Pylorusdrüsen bei Crocidura 
78% und bei Suncus 81% der Cardiadrüsenlänge. In den absoluten Längen unterscheiden 
sich die Arten deutlich voneinander. Die Drüsenlängen der Hausspitzmaus übertreffen die 
der Etruskerspitzmaus um jeweils etwa die Hälfte (Tab. 1). Die gefundenen Unterschiede 
sind für die Fundusdrüsen signifikant (p < 0,01), für die Pylorusdrüsen nicht. 

Auch die Drüsendichte ist bei beiden Arten verschieden (Tab. 2); der Dichteunterschied 
ist darauf zurückzuführen, daß der Durchmesser einer Einzeldrüse bei Suncus etruscus 
geringer ist als bei Crocidura russula. Cardia- und Fundusdrüsendichten unterscheiden sich 
signifikant (p < 0,01), der Unterschied zwischen den Pylorusdrüsendichten läßt sich nicht 
absichern. 


Tabelle 1 


Länge der Magendrüsen bei Crocidura russula und Suncus etruscus 


(Angaben ın mm, Standardabweichungen in Klammern) 


Fundus Pylorus 


Crocidura russula 0,18 (0,02) 0,25 (0,06) 
Suncus etruscus 0,12 (0,01) 0,17 (0,01) 


Tabelle 2 


Magendrüsendichte bei Crocidura russula und Suncus etruscus 


(Anzahl pro mm, Standardabweichungen ın Klammern) 


Cardia Fundus Pylorus 


Crocidura russula 727885) 1390 (484) 1786 (636) 
Suncus etruscus 1332 (138) 2451 (611) 2315 (352) 


Magendrüsenlänge und -dichte bei Crocidura und Suncus 115 


Die geringere Dicke der Drüsenschicht bei der kleineren Art wird somit kompensiert 
durch eine größere Anzahl an Einzeldrüsen pro Flächeneinheit. 
Literatur 


MyrcnHa, A. (1967): Comparative studies on the morphology of the stomach in the Insectivora. Acta 
Theriol. 12, 233-244. 


Anschrift der Verfasserin: Dr. ADELHEID GERAETSs, Via Sfossato, I-07020 Luogosanto 


Der Riesenabendsegler (Nyctalus lasiopterus) in 
Afrika nachgewiesen 


Von FRIEDERIKE SPITZENBERGER 


Naturhistorisches Museum Wien 
Eingang des Ms. 12. 11. 1981 


Anläßlich der Bestandsaufnahme der Säugetierfauna des Kuf-Nationalparks bei Beida, 
Cyrenaika, Libyen, im August 1981 wurden auch 3 Riesenabendsegler gefangen. Nyctalus 
lasiopterus war bisher nur aus Europa bis W-Asien, von Portugal bis zum Ust-Urt-Plateau 
(Borowskı und WoRONZoWw 1970) bzw. Iran (ETEMAD 1970) bekannt (Abb. 1). 

Die näheren Funddaten der 3 libyschen Exemplare sind: 9. 8.1981 Wadi Kuf (32°42'N/ 
21°34’E) 8 aus Deckenkolk einer Halbhöhle; 19. 8. 1981 (Cyrene 32°47'’N/21°49’E) 2 & 
bei abendlichem Netzfang vor Apollobrunnen. Die beiden Fundorte liegen im Jebel 
Akhdar, einem bis zu 800 m aufragenden karstigen Küstenplateau. Es weist besonders im 


a ii Sr V ss 


Die Verbreitung des Riesenabendseglers (Nyctalus lasiopterus) in Europa und Afrika (Quellen s. 
Literaturverzeichnis) 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4702-0115 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 115-116 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


116 Friederike Spitzenberger 


Bereich des Kuf-Nationalparks noch ausgedehnte und teilweise gut erhaltene Wälder von 
Juniperus excelsior und Cupressus sempervirens auf. Dieses Gebiet scheint für den mediter- 
rane bis arıde Wald- und Waldsteppengebiete bevorzugenden Riesenabendsegler eine 
ideale Lebensstätte zu bieten. Bemerkenswert ist der Fund eines Riesenabendseglers in 
einem Deckenkolk einer Halbhöhle. Bisher wurden als Tagesschlafplätze für diese Art 
ausschließlich Baumhöhlen angegeben. 

Tiergeographisch betrachtet ist der Riesenabendsegler nur eines von mehreren paläark- 
tischen Elementen (z.B. Microtus mustersi) in der Säugetierfauna der Cyrenaika. 


Danksagung 


Für die Einladung zur Erforschung der Säugetierfauna des Kuf-Nationalparks und für die Organisa- 
tion unseres Aufenthaltes dort danke ich der Regierung der Sozialistischen Libysch-Arabischen Volks 
Jamahiriya und dem Arab Center for the Studies of Arıd Zones and Dry Lands, vertreten durch Mr. 
M. Dawa und Mr. ABpuL SALAM ABDUL KADER. 


Literatur 


ABELENZEV, W. 1.; PIDOPLITSCHKO, 1. G. (1956): Fauna Ukraini. Tomi. Ak. Nauk. Ukr. SSR, Kiew. 

BAUER, K. (1956): Zur Kenntnis der Fledermausfauna Spaniens. Bonn. Zool. Beitr. 7, 296-320. 

BoBrINsSKI, N. A.; KusnEzov, B. A.; Kusjakın, A. P. (1965): Opredelitel mljekopitajuschtschich 
SSSR, Usd. Prosweschtschenije. Moskau. 

Borowskı, $. G.; WoRONZov, N. N. (1970): Nyctalus lasiopterus ın the West Aral Territory. Zool. 
Zhurn. 49, 942. 

BÜRGER, P.; CERVENY, J. (1979): Large noctule, Nyctalus lasiopterus (Schreber, 1780) in the collections 
of the Museum of South Bohemia. Lynx n.s. 20, 105-106. 

Daneko, S. (1974): Der Großabendsegler, Nyctalus lasiopterus (Schreber, 1780) in der Tschechoslowa- 
kei. Lynx n.s. 15, 5-8. 

Desrayvss, M. (1965): Les Chauves-Souris. Bull. de la Murithienne 32, 1-5. 

DumITREscu, M.; TAnasAcHI, J.; ORGHIDAN, T. (1963): Raspindirea chiropterelor R. P. Romina. 
Lucr. Inst. spel. „Emil Racovita‘“ 1-2, 509-575. 

ETEMAD, E. (1970): A note on the occurrence of the Giant Noctule, Nyctalus lasiopterus Schreber, 
1780, ın Iran (Chiroptera: Vespertilionidae). Mammalıa 34, 547. 

FaTıo, V. (1869): Faune de Vertebres de la Suisse. Vol. 1. Mammiferes. Geneve et Bäle: H. Georg. 

GuLıno, G.; Dar Pıaz, G. (1939): I Chirotteri Italianı. Boll. Mus. Zool. Anat. comp. Torino 47, 1-43. 

HanAK, V.; Josırov, M. (1959): Zur Verbreitung der Fledermäuse Bulgariens. Säugetierkundl. Mitt. 
7, 145-151. 

HEIM DE Bausac, H. (1932): Un Cheiroptere nouveau pour la faune frangaise: Nyctalus maximus 
(Fatio), represente dans les collections du Museum. Bull. Mus. Nat. Hist. nat. 4, 162-164. 

HEım DE Baısac, H.; BEAUFORT, F. DE (1968): La statut des Noctules (Nyctalus) en France. 
Mammalıa 32, 204-206. 

HEINRICH, G. (1935): Über die von mir im Jahre 1935 in Bulgarien gesammelten Säugetiere. Mitt. 
Naturw. Inst. Sofia 9, 33-48. 

KAHMmann, H. (1962): Neue Ergebnisse ın der ee in der Türkei. Säugetierkundl. Mitt. 
10, 112.116. 

MILLER, G. $. (1912): Catalogue of the Mammals of Western Europe. Trustees Brit. Mus. London. 

PALMEIRIM, J.; Ramos, M.; Dras, D. (1979): Bats from Portugal in the Collection of Museu Bocage 
(Mammalia, Chiroptera). Arqu. Mus. Bocage 7, 53-66. 

RUPRECHT, A. (1970): Nyctalus lasiopterus (Schreber, 1780) — a new species in the fauna of Poland. 
Acta Theriol. 15, 370-372. 

SCHÄFER, H. (1973): Der Großabendsegler, Nyctalus lasiopterus (Schreber, 1780) ın der Hohen Tatra. 
Erstnachweis für die Tschechoslowakei. Säugetierkundl. Mitt. 21, 78-79. 

TorAL, G. (1976): New records of Vespertilio murinus Linnaeus and of Nyctalus lasiopterus (Schreber) 
in Hungary. Vertebr. Hung. 17, 9-13. 

ToscHı, A.; Lanza, B. (1959): Fauna d’Italia. Vol. IV: Mammalia. Bologna: Ed. Calderini. 

WorLr, H. (1940): Zur Kenntnis der Säugetierfauna Bulgariens. Mitt. Naturw. Inst. Sofia 13, 153-158. 

— (1964): Der Großabendsegler - Nyctalus lasiopterus (Schreber, 1780) - in Griechenland. Säugetier- 
kundl. Mitt. 12, 183-184. 


Anschrift der Verfasserin: Dr. FRIEDERIKE SPITZENBERGER, Naturhistorisches Museum Wien, 1. Zoo- 
logische Abt., Burgring 7, A-1014 Wien 


Note on Vicuna X Alpaca hybrids 


By H. DE MAcEDoO 


Receipt of Ms. 25. 11. 1981 


The possibility of getting hybrids crossing vicuna, Vicugna vicugna, and alpaca, Lama 
guanicoe forma pacos (sensu BOHLKEN 1961), ıs an old established fact. But by no means is 
this an easy task since the named species do not breed freely even when their flocks are ın 
close proximity, a rather common situation in the Peruvian highlands. An adult male 
vicugna will never copulate with a female alpaca unless the first has been timely imprinted 
to a member of the second species. Therefore, the Peruvian native breeders use exclusively 
a vicuna previously reared from earliest age by an alpaca whenever they want to get the 
hybrid called ‘pacovicuna’. 

To obtain a vicuna male involves the hunting of a just newborn anımal. Thus, a native 
shepherd-hunter first spots a flock of pregnant vicunas at the proper season (January to 
March) and then waits daily for the instant of parturition. At this very moment, he releases 
one or better, two hunting dogs brought to help him and approaches the newborn as 
quickly as possible either by foot or on horseback, trying to lasso and then to embrace the 
young beast which lıttle time after birth, is able to run as rapıdly as its mother being then 
impossible to be reached. It was precisely by this procedure that the shepherds of my ranch 
‘Checayanı’ (Azangaro province of the Department of Puno, Peru) succeeded to seize a 
young male vicuna that was presented to me as a birthday’s gift in March 1965. We all 
agreed that this anımal should be lactated and reared by an alpaca which had lost her own 
offspring, in order that when put into a harem of young female alpacas, we could get the 


First generation of Vicuna X Alpaca hybridization, so called pacovicunas 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4702-0117 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 117-118 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


118 H. de Macedo 


corresponding hybrids. This came true in March 1967 and shortly after, I took the picture 
here offered to the readers. All shown anımals belong to that particular first generation 
(Fx,). The alpacas of the parental flock were not pure strains and this fact ıs reflected in the 
great deal of variation seen in the offspring. 

It is also of interest to remember that the Peruvian alpaca breeders distinguish two 
strains or varieties. One of these ıs called “‘huacaya’ and has coarse and rather short wool. 
The name Shuacaya’ comes from the aymara word huacaa, meaning ordinary or common. 
The other alpaca variety ıs called ‘suri” and has long sılky wool. The word suri ıs the 
aymara name of the Lesser Rhea or Peruvian ‘ostrich’ Pterocnemia pennata. No doubt the 
name of the alpaca variety refers to the sımilarity of ıts wool with the long and soft loose 
webbed plumage of the bird which inhabits the highlands between the Puno and Moque- 
gua Departments. 

My endeavours to get further pacovicunas came to an end with the loss of my property 
in 1968, due to the Peruvian agrarıan reform. 


Literature 


BoHLkEn, H. (1961): Haustiere und Zoologische Systematik. Z. Tierzüchtg. Züchtgsbiol. 76, 107-113. 


Author’s address: Prof. Dr. HERNANDO DE MACEDO, Director, Museo de Historia Natural, Univer- 
sıdad Nacional Mayor de San Marcos. Apartado 11434, Lima 14, Peru 


Thermopräferenz bei der Gemeinen Vampirfledermaus 
(Desmodus rotundus) 


Von U. ScHMIDT und U. MansKE 


Zoologisches Institut der Universität Bonn 
Eingang des Ms. 8. 12. 1981 


Der Nahrungserwerb stellt die sanguivore Vampirfledermaus (Desmodus rotundus) vor 
Orientierungsprobleme, die allein mit Hilfe der Echoortung nicht gelöst werden können. 
Besonders bei der Auswahl eines geeigneten Beutetieres und bei der Selektion der Bißstelle 
müssen andere Sinnesmodalitäten z.B. Geruchs- und/oder Thermosinn, eine Rolle spielen. 
Obwohl Desmodus unter natürlichen Bedingungen wohl ausschließlich vom Blut homöo- 
thermer Vertebraten lebt, gibt es bisher keinerlei Hinweise darauf, daß thermische Reize 
bei der Lokalisation der Nahrungsquelle von Bedeutung sind. Unsere Versuche sollten 
klären, ob die Auswahl der Bißstelle von thermischen Reizen beeinflußt werden kann. 

Für die Experimente standen 2 adulte Desmodus (1,1) und 7 Hauskaninchen (Belgische 
Riesen) als Beutetiere zur Verfügung. Die Vampirfledermäuse waren in einem Flugkäfig 
(110 x 70 x 100 cm) untergebracht, an dessen einer Stirnseite 3 Isolierkäfige 
(30 x 20 x 30cm) anschlossen. Die Vampire waren daran gewöhnt, defibriniertes Rinder- 
oder Schweineblut aus Vogeltrinkröhrchen zu fressen, die von außen in die Türen der 
Isolierkäfige eingeschoben wurden. 

Vor den Präferenztests wurde jede Fledermaus 14 Tage lang darauf dressiert, Blut aus 
den Kanınchenohren aufzunehmen. Die Kaninchen waren in passenden Käfigen festgelegt; 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4702-0118 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 118-120 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Thermopräferenz bei Desmodus rotundus 119 


Vampirbiß (Pfeil) an einem Kaninchenohr. Nur der 
Hautbezirk in der Kreisöffnung war den Vampir- 
fledermäusen für das Beißen zugänglich. 
(Balkenlänge 1 cm) 


ein Ohr wurde zwischen zwei Kunststoffstreifen (Trovidur®) geklemmt. Einer der Streifen 
besaß eine kreisförmige Öffnung (®4,5 cm), die eine Hautpartie als Bißstelle freigab (s. 
Abb.). Diese Halterung wurde an der Futterstelle plaziert; die Mitte der Öffnung befand 
sich 5,5 cm über dem Käfıgboden, in Kopfhöhe einer stehenden Vampirfledermaus. Beide 
Vampire akzeptierten sofort das Kanınchenohr und frafen regelmäßig während des jeweils 
30minütigen Versuchs. Keines der Kaninchen wurde durch den Blutverlust geschädigt; die 
Beutetiere zeigten auch keinerlei Schmerzreaktion beim Biß oder Unruhe beim Fressen des 
Vampirs (SCHMIDT und ManskE 1973; GREENHALL 1976). 

Bei den Präferenztests wurden beide Ohren simultan verwendet und die Ohrhalterun- 
gen in zwei nebeneinanderliegenden Isolierkäfigen angebracht (der dritte Käfig diente dem 
gerade nicht im Experiment stehenden Desmodus als Aufenthaltsraum). Hinter dem einen 
Ohr wurde ein Kühlelement befestigt, das andere blieb unbeeinflußt; die Position des 
kalten Ohres war zufallsgemäß festgelegt. Unmittelbar nach jedem Versuch wurde die 
Oberflächentemperatur der dem Vampir zugänglichen Hautbezirke gemessen. Jeder Des- 
modus wurde an 11 Tagen untersucht. 

Die Bißstellen befanden sich zumeist im mittleren Bereich des verfügbaren Areals; es 
ließ sich keine Korrelation zur Durchblutung der Haut aufzeigen. 12 der 50 Bisse erfolgten 
in die Arteria auricularis posterior, die stets im zentralen Teil der Öffnung verlief, die 
Mehrzahl lagen in dem Anastomosengebiet (Abb.). Wenn eine Wunde zu wenig Blut 
lieferte, wurde sie mit der Zungenspitze erweitert, bis entweder ein Gefäß angeschnitten 
oder das Ohr durchbohrt war. 

Das Wahlverhalten der Fledermäuse während der Präferenztests war sehr charakteri- 
stisch: Die Vampire liefen mehrmals zwischen beiden Käfigen hin und her, inspizierten die 
Ohren, indem sie ihre Nase auf wenige Millimeter an die Hautoberfläche heranbrachten 
und begannen erst dann an der ausgewählten Bißstelle zu lecken. 17mal bevorzugten sie 
dabei das warme Ohr, 5mal bissen sie in das kalte; diese Bevorzugung ist hochsignifikant 
(z-Test, p<0,01). Die Oberflächentemperatur des warmen Ohres schwankte zwischen 
27,4 und 33,8 °C, die des kalten zwischen 12,9 und 23,8 °C. Bei den 5 Bissen in das kalte 
Ohr lag die mittlere Temperaturdifferenz um ca. 2°C niedriger (AT: 11,5 °C # 1,6; 
Temperatur des kalten Ohres: 19,4 °C + 1,1; x +), als bei den Bissen in die warme 
Hautstelle (AT: 13,9 ° + 2,5; Temperatur kaltes Ohr: 17,7 + 2,8). 

Die Versuche zeigen, daß Desmodus bei der Auswahl der Bißstelle thermische Reize 
berücksichtigt. Dadurch wird die Lokalisation einer geeigneten Hautpartie erleichtert. 
Eine thermische Orientierung könnte es z.B. dem Vampir ermöglichen, beim Fressen an 
Rindern oder Pferden zwischen dem kalten Huf und dem durchbluteten Hufansatz, einer 
bevorzugten Bißstelle (Cresro et al. 1971), zu unterscheiden. Physiologische Untersu- 


120 U. Schmidt und U. Manske 


chungen müssen zeigen, ob bei Desmodus der Thermosinn als Anpassung an die Ernäh- 
rungsweise besser ausgebildet ist als bei anderen Fledermausspezies. 


Literatur 


CRESPO, R. F.; Burns, R. J.; LinHarT, S. B. (1971): Comportamiento del vampiro (Desmodus 
rotundus) durante su alimentacion en ganado bovino en cautiverio. Tecn. Pecuaria Mexico 18, 
40-44. 

GREENHALL, A. M. (1976): The biting and feeding habits of the vampire bat, Desmodus rotundus. ]. 
Zool. London 168, 451-461. 

SCHMIDT, U.; MAnskE, U. (1973): Die Jugendentwicklung der Vampirfledermäuse (Desmodus 
rotundus). Z. Säugetierkunde 38, 14-33. 


Anschrift der Verfasser: Prof. Dr. U. Schmipr und Dr. Dr. U. Manske, Zoologisches Institut der 
Universität, Poppelsdorfer Schloß, D-5300 Bonn 


BEKANNTMACHUNGEN 


Einladung 


Die 56. Hauptversammlung der Deutschen Gesellschaft für Säugetierkunde e. V. findet von 
Montag, den 27. September, bis Freitag, den 1. Oktober 1982, in Salzburg statt. 


Vorläufiges Programm 


Montag, 27. September: Anreisetag 
Ab 19.00 Uhr Begrüßungsabend ım Peterskeller, Peters- 
bezirk 1 - Zugang von der Franziskanergasse 


Dienstag, 28. September: Institutsgebäude Akademiestraße 26, Hörsaal 340 
9.00 Uhr: Begrüßung und Eröffnung durch den 1. Vorsit- 


zenden 
anschließend Vorträge 


17.00 Uhr: Mitgliederversammlung 
abends: Begrüßung durch Stadt und Land Salzburg. 


Kammerkonzert 
Mittwoch, 29. September: Institutsgebäude Akademiestr. 26, Hörsaal 340, Vorträge 


Donnerstag, 30. September: vormittags: Institutsgebäude Akademiestr. 26, Hörsaal 340, 
Vorträge 


nachmittags: Führung durch den Tiergarten Salzburg durch 
Herrn Oberst FRIEDRICH LACCHINI 


Freitag, 1. Oktober: Exkursion zum Steinbock-Freigehege Blühmbachtal und 
zum Institut für Vergleichende Verhaltensforschung - 
Abteilung Tiersoziologie - in Grünau ım Almtal 


Bitte melden Sie Vorträge für die Tagung - die in der Regel nicht länger als 15 Minuten 
dauern sollten - möglichst frühzeitig, spätestens aber bis zum 15. Mai 1982, bei Prof. Dr. 
Hans-JüÜRG Kuhn, Anatomie der Universität, Kreuzbergring 36, D-3400 Göttingen, an. 


Bekanntmachungen 121 


Den Tagungsort betreffende und organisatorische Fragen beantwortet Herr Dr. 
ALFRED GOLDSCHMIDT, Zoologisches Institut der Universität Salzburg, Akademiestr. 26, 
A-5020 Salzburg. 


International Symposium on Vertebrate Morphology Giessen, 
August 22nd through 26th 1983 


It is the purpose of this meeting to present an overview of the current knowledge in the 
field of vertebrate morphology, emphasizing functional aspects of the major organ 
systems. 
The symposium will consist of sessions on the major organ systems. However, special 
emphasis will be devoted to the following topics: 
— interrelations between various organ systems 
— influence of the absolute sıze upon organ systems 
— relationships between embryology and ecology 
— evolution of functional systems, including the 
— documentation by paleontology 
No parallel sessions are scheduled. Besides introductory reviews, there will be oral 
presentations as well as poster sessions. 
It would be useful if those colleagues who plan to attend the symposium would write 
to: 
Prof. Dr. Dr. H.-R. Duncker 
Dept. of Anatomy and Cell Biology 
University of Giessen 
Aulweg 123 
D-6300 Giessen/W.-Germany 


BAILEFFBIESSBIRESFEUINIGEN 


DeEBLAsE, A. F.; MARTIN, R. E.: A Manual of Mammalogy with Keys to Families of the 
World. 2. Ed. Dubuque, Iowa: Wm. C. Brown Co. Publ. 1981. 436 pp. US $ 15.15. 


Dieser „Grundriß der Säugetierkunde“ liegt nun ın 2. erweiterter Auflage vor. Er ıst bestimmt als 
Leitfaden für Kurse über Säugetiere im Rahmen des Unterrichts an den amerikanischen Universitäten. 
Das Schwergewicht liegt dabei auf Belangen der Systematik. In den einleitenden Kapiteln werden die 
taxonomisch wichtigen Merkmale — wie Schädel, Gebiß, Haut und Haar, Hörner und Geweihe - 
behandelt, dann Fortpflanzung und Populationsdynamik. Der Hauptteil ist ein Bestimmungsschlüs- 
sel, der über die Ordnungen zu den rezenten Familien der Säugetiere führt. Kapitel über Statistik, 
Altersbestimmung, Sammlungstechniken (auch für Ektoparasiten) u. a. beschließen den Band. Bei den 
einzelnen Kapiteln ist in der Regel weiterführende Literatur genannt, außerdem finden sich am Ende 
eın Literaturverzeichnis und ein Register. 

Das Buch ist für Studenten sicher eine nützliche Hilfe. An deutschen Universitäten dürften so 
ausführliche Säugetierkurse leider Ausnahmeerscheinungen sein. H. BoHLken, Kiel 


BrInK, J. M. van; VORONTSov, N.N. (eds.): Animal genetics and evolution. The Hague: 
Dr. W. Junk Publishers 1980. 393 pp., 180 fıgs., 65 tables. US $ 99.-. ISBN 90-6193-602-0 


Die erheblichen Fortschritte der Populationsgenetik und der Karyologie (Cytogenetik) in den 
letzten Jahren beruhen einerseits auf Verfeinerungen der Methodik, andererseits aber nicht zuletzt auf 


122 Buchbesprechungen 


der Verbreiterung der Faktenkenntnisse und der zunehmenden Einsicht in Zusammenhänge. Der 
Überblick über das Gewonnene wird aber erschwert durch die hohe Spezialisierung der Bearbeiter auf 
einzelne Organismengruppen. So ist es zu begrüßen, daß eine Auswahl von 40 Beiträgen auf dem XIV. 
Internationalen Genetiker-Kongreß in Moskau 1978, die speziellen Problemen der Evolutions- und 
Populationsgenetik der Tiere gewidmet waren, im vorliegenden Sammelband zusammengefaßt und in 
englischer Sprache den Interessenten zugänglich gemacht wird. Dreiviertel der Beiträge stammen von 
russischen Forschern. Der außerordentlich breit gefaßte Rahmen und die hohe Spezialisierung der 
Einzelbeiträge machen eine inhaltliche Besprechung im einzelnen auf dem zur Verfügung stehenden 
Raum unmöglich. Die Hälfte der Beiträge befaßt sich mit karyologischen Problemen an Säugetieren. 6 
Beiträge sind Studien an Insekten, 1 an Vögeln und 3 niederen Vertebraten gewidmet. Der Rest 
behandelt generelle Fragen. Der Hauptwert der Veröffentlichung dürfte darin zu sehen sein, daß eine 
derartige Synopsis des Standes der Spezialforschung den Weg zu einer allgemeinen, vergleichenden 
Populations- und Evolutionsgenetik freimacht. 
Die Ausstattung des Bandes ist ausgezeichnet, der Preis allerdings recht hoch. 
D. STARcK, Frankfurt/M. 


RENNER, E.: Mathematisch-statistische Methoden in der praktischen Anwendung. 
Pareys Studientexte Bd. 31. 2., neubearb. Auflage. Berlin u. Hamburg: Paul Parey 1981. 
112 Seiten, 11 Abb., 62 Tabellen sowie 13 tabell. Übersichten. Balacron brosch. DM 24,-. 
ISBN 3-489-76034-4 


Dieses Buch wendet sich an den großen Personenkreis von Examenskandidaten, Doktoranden, 
technischen und wissenschaftlichen Mitarbeitern, die nicht in Statistik ausgebildet sind, sich aber 
dennoch bei Arbeiten häufig statistischer Methoden bedienen müssen. Auf theoretische Grundlagen 
der Statistik wird bewußt nicht eingegangen, statt dessen erfährt der Leser als Anwender, wie Daten 
sinnvoll aufbereitet werden, um Aussagen über Verteilung, Mittelwert, Varianz, Korrelation und 
Meßfehler machen zu können. An Hand von Symbol-Urlisten werden ihm die statistischen Verfahren 
hierzu anschaulich vorgestellt und an einfachen Beispielen praktisch erläutert. Für die Durchführung 
von Rechenoperationen und bei der Lösung einfacher statistischer Probleme leistet dieses Büchlein 
ausgezeichnete Hilfestellung. P. EBINGER, Hannover 


FORTMEYER, H. P.: Thymusaplastische Maus (nu/nu), thymusaplastische Ratte (rnu/ 
rnu). Haltung, Zucht, Versuchsmodelle. Schriftenreihe „Versuchstierkunde“ H. 8. Berlin, 
Hamburg: Paul Parey 1981. 94 S., 10 Abb., 6 Tab. DM 58,-. ISBN 3-489-75818-8 


Die autosomal-rezessive Mutante nu/nu der Hausmaus wurde 1966 beschrieben (FLANAGAN). 
Während bei anderen haarlosen Mutanten der Labormaus das Haarkleid sich anfangs noch normal 
entwickelt und der Haarverlust erst nach dem ersten Haarwechsel auftritt, fehlt das Haarkleid beı nu- 
Mäusen von der Geburt an. Zwar treten auch bei dieser Mutante Haaranlagen auf, doch brechen die 
Haare während des Durchbruchs auf Grund von Verhornungsanomalien sofort ab. 

Die Haarlosigkeit ist stets kombiniert mit einer Thymusaplasie, doch ist die Haarlosigkeit offenbar 
nicht die Folge der Thymusaplasie. Die Thymusaplasie macht die nu-Maus zu einem idealen 
Versuchstier für Transplantationsversuche für Immunologie und Krebsforschung, da die Abstoßreak- 
tion fehlt. 

Das vorliegende Bändchen behandelt in knapper, aber vollständiger und klarer Weise alle 
einschlägigen Fragen, die mit den morphologischen und immunbiologischen Besonderheiten des 
neuen Versuchstieres verbunden sind, wie vor allem Besonderheiten der Haltung, Ernährung, 
Pathologie und Aufzucht. Anwendungsmöglichkeiten und Versuchstechniken (immunbiologisch, 
operativ) werden berücksichtigt. Anschließend folgt ein Kapitel über die vergleichbare Mutante rnu 
bei der Laborratte. Ein umfangreiches Schrifttumsverzeichnis beschließt den Band. Das Werk kann als 
wichtiges Hilfsmittel vor allem für Experimentalforscher warm empfohlen werden. 

D. STARcK, Frankfurt/M. 


REMMERT, H.: Arctic Animal Ecology. Berlin, Heidelberg, New York: Springer-Verlag 
1980. 250 S., 156 Abb., 28 Tab. Brosch. DM 37,-. 


Die Arbeiten über die morphologischen und physiologischen Adaptationen arktischer Tier- und 
Pflanzenarten sowie über die Populationsdynamik arktischer Wirbeltiere sind in dem einschlägigen 
Schrifttum weit verstreut. Übersichtliche Synopsen sind bisher kaum erschienen. Eine der ersten war 
die von L. IrvınG (1972, ın eben demselben Verlag wie der vorliegende Titel), wo hauptsächlich die 
physiologischen Adaptationen an die niedrigen Temperaturen diskutiert wurden. Thematisch umfang- 
reicher ist der vorliegende Band. H. REMMERT beschreibt die in der arktischen Region liegenden 
Ökosysteme, wobei er gleichbedeutend sowohl auf die abiotischen Faktoren als auch auf die 


Buchbesprechungen 123 


biotischen Faktoren besonderes Gewicht legt. Das kommt zum Ausdruck in den Fallstudien, die in 
sehr anschaulicher Weise die jeweiligen Verhältnisse vor Ort beschreiben: Spitzbergen, Abisko, Point 
Barrow, Devon Island, um nur einige zu nennen. Die Ergebnisse von Spitzbergen, wo der Autor 
hauptsächlich gearbeitet hat, werden häufig in der vergleichenden Diskussion herangezogen. Neben 
den terrestrischen Ökosystemen werden auch die limnischen und marinen berücksichtigt. 

Säugetierbiologisch von besonderem Interesse sind die Kapitel über die Populationszyklen (Micro- 
tus, Clethrionomys, Lemmus), dıe saisonalen Wanderungen (Rangifer, Pinnipedia, Cetacea) und alle 
jene Stellen, in denen über die Bedeutung einer Säugetierart im Energiehaushalt des Ökosystems 
berichtet wird (Lepus, Rangifer, Ovibos). Besonders zahlreich sind die Beispiele, die die Stellung des 
Rentiers innerhalb des Ökosystems der Tundra beschreiben. 

Literaturmäßig etwas eingehender belegt wünscht man sich die Oszillationen von Wiesel- 
(Mustela), Fuchs-(Vulpes) und den anderen infrage kommenden Carnivora-Populationen. Das gilt 
sowohl für die textliche Schilderung als auch für die bildlichen Darstellungen. 

Der Autor befaßt sich auch mit den Artbildungsprozessen in der Holarktis, ein Problemfeld, das 
sicher weiterhin seine Diskussionswürdigkeit behält, was sich in den nach wie vor unterschiedlichen 
Ansichten über den Art-Unterarten-Status ausdrückt. In diese Problematik führen die ausgewählten 
Beispiele hervorragend ein. 

Vielleicht wäre es bei dem lehrbuchartigen Charakter, den die Abhandlung hat, günstig gewesen, 
wenn in den Legenden der bildlichen Darstellungen nach den Trivialnamen auch die wissenschaftli- 
chen Namen oder nur diese aufgeführt worden wären. Das würde z. B. die Benutzung der Abbildun- 
gen erleichtern. Für eine vertiefende Arbeit ist es auch von Vorteil, wenn hinter jeder Abbildung die 
Literaturstelle angegeben ist, sofern es sich nicht um Originalabbildungen handelt. Die Schwarz-weiß- 
Fotografien lassen manchmal in ihrer Deutlichkeit etwas zu wünschen übrig, was ihre Aussagekraft 
merklich mindert. 

Die von J. Wıser durchgeführte Übersetzung läßt sich erfreulich fließend lesen. Die inhaltliche 
Darstellung ist nıcht allein beschreibender, sondern problematisierender Art, so daß der Band in dieser 
Gestaltung als Lehrbuch gut benutzbar ist, und die sehr glücklich ausgewählten Beispiele als 
Diskussionsgrundlage herangezogen werden können. Nicht zuletzt aus diesem Grunde wird „Arctic 
Anımal Ecology“ in den Hochschulen eine geneigte Aufnahme finden. R. SCHRÖPFER, Osnabrück 


CIOoCHon, R. L.; CHIARELLI, A. B. (eds.): Evolutionary Biology of the New World 
Monkeys and Continental Drift. New York, London: Plenum Press 1980. 528 pp. 
US $ 49.50, ISBN 0-306-404877 


Der inhaltsreiche Band ist die stark erweiterte und überarbeitete Publikation von Beiträgen eines 
Symposions während des Internationalen Primatologenkongresses 1979 ın Bangalore. Diese zentrie- 
ren sich auf folgende Probleme: I. Welche Aussagen können heute auf Grund palaeontologischer, 
morphologischer, biochemischer Befunde über den phylogenetischen Ursprung der Platyrrhini 
gemacht werden? II. Wie ist die stammesgeschichtliche Beziehung zwischen Platyrrhinen und 
Catarrhinen zu verstehen? III. Welche Bedeutung kommt in diesen Zusammenhängen den neuen 
Auffassungen über Kontinentaldrift zu? 

Eine umfassende und klare Einführung zur Drifttheorie, besonders zur Frage der Entstehung des 
Atlantiks, mit vielen Details durch den Geophysiker D. H. Tarıınc bietet eine hervorragende Basis 
für die biologischen Beiträge. Hervorgehoben sei die Feststellung, daß eine transatlantische Immigra- 
tion nach Südamerika auf Grund der geologischen Voraussetzungen im Eozaen möglich war (Meeres- 
strömungen ost-westlich, Vorkommen von Inselketten mit Distanzen von etwa 200 km). Hingegen 
wäre die Distanz Nordamerika-Südamerika größer als die transatlantische Strecke, und die Strö- 
mungsverhältnisse wären ungünstig für einen Nord-Süd-Transport. 

Sechs Beiträge (McKEennAa, WO0D, LAVOCAT, HOFFSTETTER, GINGERICH, ROSENBERGER/SZALAY) 
behandeln die Palaeontologie und Palaeobiogeographie südamerikanischer Mammalıa im allgemeinen, 
der Primaten im besonderen. Das Problem der Caviomorpha-Phiomorpha, das so viele Analogien zur 
Platyrrhinen-Catarrhinen-Phylogenese aufweist, wird eingehend berücksichtigt. Vergleichend-mor- 
phologische Probleme und Argumente werden unter drei Gruppen (Zähne/Gebiß, cranıale Strukturen 
und andere Organe) behandelt. Hervorgehoben seien MAIERs Beitrag über Nasenknorpel und 
Vomeronasalkomplex, CArTtMILıs Versuch, Ausbildung des knöchernen Abschlusses der Orbita 
gegen die Temporalgrube bei Tarsiifformes und Simiae mit dem Auftreten einer retinalen Fovea in 
Korrelation zu bringen, und LuckETTs umfassende Analyse der Frühentwicklung und Placentation 
bei Primaten und Hystricognathen. Der argumentative Wert morphologischer Fakten bei phylogene- 
tischen Überlegungen hängt weitgehend von dem Verständnis der Einzelfunde und ihrer Wertung ab. 
Diese Voraussetzung wird von der Mehrzahl der Autoren beachtet. Demgegenüber dürfte es für 
evolutionsbiologische Deutungen ziemlich unerheblich sein, wenn ein einheitliches System nach einer 
Vielzahl von Mikromerkmalen abgesucht wird (etwa 30 Merkmale an der Haut oder eine Vielzahl von 
Oberflächenstrukturen am Spermienkopf) und diese, ohne daß Bedeutung und konstruktive Zusam- 
menhänge berücksichtigt werden, rein numerisch gewertet werden. 


124 Buchbesprechungen 


Karyologische (CHIARELLI) und biochemische Befunde (SARICH, CRONIN, BABA, DARGA, GooD- 
MAN) werden ausführlich berücksichtigt. Im Gegensatz zu einigen morphologischen Beiträgen bemü- 
hen sich die Autoren um eine synthetische Bearbeitung unter Beachtung der palaeontologischen, 
phylogenetischen und biogeographischen Ergebnisse. Zusammenfassende Erörterungen der beiden 
Herausgeber schließen den Band ab. 

Der Leser muß, um Einblick ın die Vielfalt der Meinungen und Argumentationen zu gewinnen, auf 
das Original verwiesen werden. Zusammenfassend können wir feststellen, daß die eingangs hervorge- 
hobenen Fragen noch keineswegs einheitlich beantwortet werden können. Konsens besteht darüber, 
dafS Platyrrhinen und Catarrhinen monophyletisch sind. Hingegen bleiben Ort und Zeitpunkt der 
Dichotomie noch unter Diskussion. Eine Mehrzahl von Autoren nimmt heute die Herkunft der 
Platyrrhinen von afrıkanischen Praecatarrhinen an und akzeptiert auf Grund der geophysikalischen 
Ergebnisse eine transatlantische Immigration in Südamerika im Eozaen. Der Diskussion abweichender 
Meinungen (Diphylie der Simier; Abstammung der Platyrrhina von nordamerikanischen Omomyi- 
den; Entstehungszentren der Affen in Nordostasien, Wanderung über die Behringstraße in die 
Amerikas und südwärts über S-Asıen nach Afrika) wird breiter Raum eingeräumt. 

Im ganzen handelt es sich um eine sehr beachtenswerte Neuerscheinung, die eine Fülle von neuen 
Befunden enthält, für Evolutionsbiologen, Tiergeographen, Palaeontologen und Morphologen unent- 
behrlich ist und eine Synthese des heutigen Wissensstandes bietet, von der zweifellos Impulse 
ausgehen werden. D. STARcK, Frankfurt/M. 


Mu1ızon, CH. DE: Les Vertebres fossiles de la formation Pisco (Perou). I. Deux nouveaux 
Monachinae (Phocidae, Mammalıa) du Pliocene de Sud-Sacaco. Parıs: (Edit. A.D.P.F.) 
1981. XII + 150S., 11 Taf. Trav. Inst. Franc. Etud. Andines 22 (= Rech. grand. civil. 
Mem. 6). 


Der erste Teil einer Bearbeitung der Wirbeltierfauna aus der pliozänen Pisco-Formation in Peru 
behandelt die vielleicht interessantesten Fossilfunde aus diesen marinen Küstenablagerungen der 
jüngsten Tertiärzeit, nämlich die Seehunde (Phocidae). Die Fauna setzt sich aus Haien, Knochenfi- 
schen, Krokodilen, Vögeln und Säugetieren zusammen, unter denen auch Otariiden vertreten sind. 

Der Fauna kommt vor allem in tiergeographischer Hinsicht große Bedeutung zu, was speziell für 
die Phociden gilt. Das pliozäne Alter der Pisco-Formation ist durch die Mikroflora und radiomerri- 
sche Daten gesichert. Die Evertebratenfauna weist auf etwas wärmere Wassertemperaturen als 
gegenwärtig hın. 

Die beiden, auf ziemlich vollständigen Skeletten beruhenden Phociden-Taxa werden als neue 
Gattungen und Arten (Piscophoca pacıfıca n.g. n.sp. und Acrophoca longirostris n.g. n.sp.) 
beschrieben. Es sind Angehörige der Monachinen. Piscophoca pacifica steht zwar ım allgemeinen 
Monachus nahe, zeigt jedoch Apomorphien mit den Lobodontini (Gehörregion) und scheint ein 
Abkömmling von Monotherium aberratum zu sein, der im Jung-Miozän in den Pazifik eingewandert 
ist. 

Acrophoca longirostris ıst gleichfalls ein Angehöriger der Lobodontini, der durch den langen 
Fazialschädel und die geringen Schwimmanpassungen im postcranialen Skelett von den heutigen 
Lobodontini abweicht. Acrophoca wird vom Verfasser in die zu Aydrurga führende Linie gestellt. 
Acrophoca longirostris ist als Küstenbewohner wesentlich häufiger nachgewiesen als Piscophoca 
pacıfica als mehr pelagische Art. 

Mit dem Nachweis von Piscophoca ist der Ausbreitungsweg der Monachinen in die Antarktis längs 
der pazifischen Küste Südamerikas im Jungtertiär dokumentiert, wie ihn der Referent schon vor 
Jahren angenommen hat. E. THENIUS, Wien 


NiIckEL, R.; SCHUMMER, A; SEIFERLE, E.: The Anatomy of the Domestic Animals. Vol. 
3.: The Circulatory System, the Skin, and the Cutaneous Organs of the Domestic 
Mammals. By SCHUMMER, A.; WILKENS, H.; VOLLMERSHAUS, B.; HABERMEHL, K.-H. 
Transl. by SILLER, W. G., and WIGHT, P.A.L. Berlin u. Hamburg: Paul Parey 1981. 630 pp., 
439 figs., 172 in colour. DM 186,-; US $ 98,-. ISBN 3-489-55618-6 


In the set of five volumes of English translations of the well-known German textbook of veterinary 
anatomy „Lehrbuch der Anatomie der Haustiere“ the third part is now available. Four authors 
participated in the writing of this book: the late A. SCHUMMER as well as H. WILKEns present the 
chapters on blood, heart, and arteries as well as veins, B. VOLLMERHAUS describes the Iymphatic 
organs and the lymphatıic vessels, and K.-H. HABERMEHL deals with the integument and its specialized 
structures. Comparative aspects are emphasized and seven domestic mammalıan species are consi- 
dered: dog, cat, horse, pig, and three representatives of the Pecora: goat, sheep, and cow. Only about 
120 of the 439 ıllustrations are taken from other sources, most others are originals. All figures are of 


Buchbesprechungen 125 


very high standard and are most instructive. It is of great use to the reader that not only the structures 
considered in the text are ıllustrated, but also the neighbouring organs with their topographic 
relations. The numbers representing different structures in the figures sometimes make it difficult to 
„read“ the illustrations. Full names of the structure — or at least clear abbreviations — might have 
improved the clearness of the figures. In the chapters on blood vessels comparative tables give a clear 
overview of the circulatory system. 

The terminology used ıs in Latin whenever a term appears for the first time, but subsequently the 
anglicized version is applied. This makes reading of the text easier without losing clearness and 
accurateness. It is obvious that all four authors — as well as the two translators — endeavoured to 
present a clear, readable, and scientifically precise text. This aim has been fully accomplished. The 
connection between structure and function is emphasized and practical applications of anatomical 
findings are carefully considered, e.g., in the case of clinical diagnosis and meat inspection. This 
beautifully produced volume, which ıs a textbook as well as a source of reference (Bibliography: 40 
pages!), will certainly find its readers and friends in English-speaking countries. P. LANGER, Giessen 


BAKER, R. R.: Human navigation and the sixth sense. London, Sidney, Auckland, 
Toronto: Hodder and Stoughton 1981. 138 pp. £ 8.95. ISBN 340-26082-3 


Die allgemein verbreitete Ansicht, daß der Mensch im Gegensatz zu vielen Wirbeltieren keinen 
Richtungssinn besäße, wird vom Autor in Frage gestellt. In Analogie zu den Verfrachtungsversuchen 
mit Vögeln wurden Versuche mit Studenten angestellt (Verfrachtung über etwa 50 km mit Sicht- 
blende, Veränderungen des magnetischen Feldes), die Hinweise auf die Fähigkeit zur unbewußten 
Perzeption magnetischer Einflüsse ergaben. Die Bestätigung von anderer Seite bleibt zunächst 
abzuwarten. Argumente und Hypothesen werden mit Vorsicht vorgetragen. Das Büchlein enthält eine 
interessante Zusammenstellung von Orientierungs- und Navigationsmethoden ohne Instrumente bei 
Naturvölkern. Das Buch dürfte das Interesse von Verhaltensphysiologen und Anthropologen bean- 
spruchen. D. STARcK, Frankfurt/M. 


APFELBACH, R.; DÖHL, ]J.: Verhaltensforschung. Eine Einführung. 3. neubearb. u. erw. 
Aufl. Stuttgart und New York: Gustav Fischer 1981. 193 S., 71 Abb., 2 Tab. DM 12,80. 
ISBN 3-437-20236-7 


Die Autoren geben eine kurzgefaßte Finführung in die Verhaltensforschung als biologische 
Disziplin, welche mit anderen Teilgebieten der Biologie in vielfältigen Wechselbeziehungen steht. Sie 
wenden sich vornehmlich an den jüngeren Biologiestudenten, dem die Denkweisen der biologischen 
Verhaltensforschung nahegebracht werden sollen. Das geschieht mit sehr guten Begriffserläuterungen 
sowie detaillierten Arbeitsanleitungen. Hilfreich sind die vielen Zitate, gerade auch aus kurzen 
Publikationen und Doktorarbeiten. So werden Arbeitsweisen dieses Fachgebietes in der hier eigenen 
Sprache vorgestellt, wodurch das Interesse zum Studium der Originalliteratur angeregt wird. Weiter 
findet der Student gute Hilfen zur Erarbeitung verhaltensbiologischer Literatur. 

Neben zahlreichen aktuellen Ergänzungen sowie vielen instruktiven Zeichnungen bringt die 3. 
Auflage folgende Änderungen: Im 4. Kapitel (Zentralnervöse und hormonelle Bedingungen) wird ın 
einem neuen Abschnitt die zentralnervöse Kontrolle der Lauterzeugung abgehandelt. An vielen 
Stellen hat man weiter Befunde zur Ethologie des Menschen eingearbeitet. 

Insgesamt bietet dieses Taschenbuch dem Biologiestudenten wertvolle Hilfen zum Einstieg in die 
Verhaltensbiologie. DorIT FEDDERSEN, Kiel 


JamzEson, E. W. Jr.: Patterns of Vertebrate Biology. Berlin, Heidelberg, New York: 
Springer 1981. 477 pp., 164 figs., DM 63,-. ISBN 3-540-90520-0 


Aus einer Serie von Spezialvorlesungen zur Biologie der Vertebraten hat der Autor nach annä- 
hernd 30 Jahren Tätigkeit vorliegendes Buch zusammengestellt. Darin werden folgende Abschnitte 
behandelt: 1. Stammesgeschichte der Vertebraten; Mechanismen der Evolution; Zoogeographie; 
Migrationen; 2. Atmung; Nahrung und Ernährung; Thermoregulation und Wasserhaushalt; Aktivität 
und saisonale Lethargien; 3. Sinneswahrnehmung; Signalgebung; 4. Reproduktion; Wachstum; 
Populationsdichten. 

Es werden grundsätzliche Phänomene der Wirbeltierbiologie aufgezeigt und behandelt. Säugetiere 
finden angemessene Berücksichtigung. Die Auswahl von Beispielen für extreme Lebensbedingungen 
und besondere Adaptationen aus allen Wirbeltiergruppen und aus allen Kontinenten läßt auf einen 
fundierten Wissensstand des Autors schließen. Das Buch ist ausgewogen konzipiert, kurzgefaßt und 
verständlich geschrieben. Die Auswahl der Abbildungen ist geeignet, beschriebene Sachverhalte zu 
verdeutlichen. Als Grundriß der Vertebratenbiologie ist das Buch zu empfehlen, da es sowohl 
allgemein informiert als auch zu vertieften Studien in speziellen Fragen anregt. D. Kruska, Hannover 


126 Buchbesprechungen 


BATEsoN, P. P. G.; Kroprer, P. H. (eds.): Perspectives in Ethologie. Vol. 4: Advantages 
of Diversity. New York, London: Plenum Press 1981. VII + 249 S., $ 25.-. 


Die Folge „Perspectives in Ethologie“ setzt sich in unkonventioneller und z. T. provozierender 
Art und Weise mit der Ethologie auseinander mit dem Ziel, Anregungen zu geben und eine kritische 
Auseinandersetzung mit dem Fachgebiet zu fördern. Band 4 enthält Beiträge von Ethologen und 
Wissenschaftlern anderer bzw. benachbarter Disziplinen, die auf die Gefahren zu einseitiger For- 
schungen hinweisen und eine fachübergreifende Orientierung sowie eine größere Vielseitigkeit in der 
Verhaltensforschung fordern. Dabei werden stammesgeschichtliche Fragen sowie genetische, bioche- 
mische (neurobiologische), mathematische und juristische Aspekte in Zusammenhang mit ethologi- 
schen Forschungen und Aussagen diskutiert. 

Dieses Buch wendet sich in erster Linie an Wissenschaftler, die in der Verhaltensforschung bzw. in 
benachbarten Fachgebieten arbeiten. R. LöHMER, Hannover 


COIMBRA-FILHO, A. F.; MITTERMEIER, R. A. (Eds.) 1981: Ecology and behaviour of 
neotropical primates. Vol. 1. Rio de Janeiro: Academia Brasileira de Sciencias 1981. 
496 pp. Ca. US $ 25,-. 


Die Neuweltaffen sind gegenüber den altweltlichen katarrhinen Primaten lange Zeit vernachlässigt 
worden. In den letzten Jahren setzte eine stürmische Entwicklung ein, so daß beispielsweise die 
umfangreiche Krallenaffenmonographie von HERSHKOVITZ aus dem Jahre 1977 schon jetzt in vielen 
Punkten überholt ist. 

Der vorliegende erste Band einer auf zwei Bände angelegten Monographie der Platyrrhinen ist ın 
11 Kapitel gegliedert, von denen sich die ersten drei mit der Systematik rezenter und fossiler Primaten 
der Neuen Welt beschäftigen, während die folgenden acht jeweils das gegenwärtige Wissen über eine 
Gattung zusammenfassen (Callimico, Aotus, Callicebus, Saimiri, Cebus, Pithecia, Chiropotes, Ca- 
caJao). 

Im ersten Kapitel behandelt ROSENBERGER die Klassifikation auf der Ebene höherer Taxa und wirft 
dabei die seit Gray (1821) einheitlich akzeptierte Systematik, die Trennung in Callitrichidae und 
Cebidae, über den Haufen. Die Platyrrhinen werden in zwei Familien aufgeteilt: erstens die Cebidae 
mit Cebus, Saimiri (mit seinen fossilen Verwandten) und den Krallenaffen einschließlich Callimico, 
und zweitens die Atelidae mit allen übrigen Neuweltaffen. Es ist schade, daß ROSENBERGER derartig 
weitreichende Schlußfolgerungen zieht, ohne die zugrunde liegenden Fakten bisher hinreichend 
darzulegen. Zustimmen wird man ROSENBERGER darin, daß es sich bei den charakteristischen 
Merkmalen der Krallenaffen um im Rahmen der Platyrrhinen apomorphe und nicht um plesiomorphe 
Merkmale handelt. Die angekündigten weiteren Veröffentlichungen zu diesem Thema wird man 
gespannt erwarten. 

Im umfangreichsten Kapitel des vorliegenden Bandes geben MITTERMEIER und COIMBRA-FILHO 
eine Übersicht über das System rezenter Platyrrhinen. Dieser Beitrag ist mit 69 Photographien 
hervorragend bebildert, von denen eine Reihe freilich schon früher zum Teil mehrfach veröffentlicht 
waren, einige jedoch wichtige neue Dokumente sind. Leider sind diese nicht optimal reproduziert. In 
manchen Abschnitten bedauert man, daß auf ungeklärte Fragen der Systematik nicht hinreichend 
aufmerksam gemacht ist, so ist es bis heute unklar, ob im westlichen Amazonasbecken mehrere Arten 
von Aotus und Saimiri sympatrisch vorkommen. 

Der Beitrag von Rose und FLEAGLE über die Fossilgeschichte der Primaten in der Neuen Welt holt 
im Rahmen der vorliegenden Monographie etwas weit aus. Er gibt beispielsweise eine Übersicht über 
die Microsyopidae, Plesiadapidae und andere Familien fossiler nordamerikanischer Prosimier, handelt 
die acht Gattungen fossiler Platyrrhinen jedoch etwas kursorisch ab. Die verschiedenen Hypothesen 
über den Ursprung der Platyrrhinen (Entstehung aus nordamerikanischen Prosimiern oder gemein- 
same Wurzeln mit den Katarrhinen Afrikas) werden umrissen. 

Die folgenden acht Kapitel, die sich jeweils mit einer Gattung rezenter Neuweltaffen beschäftigen, 
sind im allgemeinen wie das Kapitel über Sarmırı von BaLpwın und Barpwin gegliedert: Beschrei- 
bung der Arten, Übersicht über frühere Untersuchungen in Gefangenschaft und Feldstudien, 
Beschreibung des Habitats, Nahrung und Nahrungsaufnahme, Populationsdynamik (Gruppenzusam- 
mensetzung, Populationsdichte, Aktivitätsrhythmen, Reviergröße, Schlafplätze, Beziehungen zu 
anderen Tieren desselben Habitats), Sozialverhalten (einschließlich Geburt und Jugendentwicklung, 
Spielverhalten, Beziehungen zu erwachsenen Tieren, Grooming, usw.), Reproduktionsbiologie (jah- 
reszeitliche Zyklen, Gravidität, Oestrus, Paarungsverhalten), Kommunikation, Lokomotionsverhal- 
ten, Literatur. Die nicht englischsprachige Literatur ist teilweise berücksichtigt. Die umfangreichen 
Literaturverzeichnisse bei jedem Kapitel erleichtern zwar dem Leser das Auffinden der zitierten 
Stellen, es hätte sich jedoch insgesamt durch ein gemeinsames Literaturverzeichnis viel Platz sparen 
lassen, da einzelne Arbeiten in jedem Kapitel zitiert werden. 

Die vom Leser der einzelnen Kapitel erwarteten Vorkenntnisse differieren so stark, daß nicht ganz 
klar ist, an wen sich diese Monographie vorwiegend wendet. Die einfache Ausstattung (Paperback, 


Buchbesprechungen 1077 


kein Hochglanzpapier) bedauert man zwar bei den Halbtonbildern, sie wird dem Band durch den 
relativ günstigen Preis jedoch eine weite Verbreitung sichern. Der zweite Band soll die restlichen 
Gattungen der Neuweltaffen und zusammenfassend die Ökologie behandeln. H.-J. Kunn, Göttingen 


NIETHAMMER, J.: Säugetiere, Biologie und Ökologie. UTB 732. Stuttgart: Eugen Ulmer 
1979. 288 S., 75 Abb., 43 Tab., DM 22,80. 


Das Taschenbuch ‚Säugetiere‘ — Biologie und Ökologie - ist aus einer gleichnamigen Vorlesung 
des Autors hervorgegangen und bietet nicht nur dem Studienanfänger einen reizvollen Einstieg in das 
Gebiet der Säugetierkunde. Die Kapitel Fortbewegung, Orientierung, Ernährung, Wärmehaushalt, 
Schutz und Tarnung, Kommunikation, Fortpflanzung, Entwicklung, Einpassung in den Jahreszyklus 
und extreme Lebensräume geben mit ihren zahlreichen Beispielen eine umfangreiche und gute 
Übersicht über die Lebensweise der Säugetiere. Vergleichende Hinweise heben die physiologischen, 
morphologischen oder verhaltensbiologischen Anpassungen der Säugetiere hervor. Die Kapitel über 
Populationsstatistik sowie über die ökologische Rolle der Säugetiere und den Einfluß des Menschen 
betrachten die Säugetiere aus ökologischer Sicht. Die kapitelweise zusammengefaßte, umfangreiche 
Literatur ermöglicht dem Interessierten leicht eine Vertiefung in die jeweils gewünschten Themen. 

Das Anliegen des Autors, die Biologie der Säugetiere nicht als ein abgeschlossenes Wissensgebiet, 
sondern als eines mit vielen offenen Fragen darzustellen, ist sehr gut gelungen. 

C.-D. STURM, Hamburg 


Day, M. H. (Ed.): Vertebrate Locomotion. Symposia of the Zoological Society of 
London 48. London, New York, Toronto, Sydney, San Francisco: Academic Press 1981. 
472 pp. US $89.50. ISBN 0-12-613348-4 


Möglichkeiten zur Fortbewegung sowie Effektivität und Modus von Motilität haben große 
evolutive und adaptıve Wertigkeit. Das Studium von Anatomie und Physiologie des Bewegungsappa- 
rates differenter Lokomotionstypen bei Vertebraten auf unterschiedlichem Evolutionsniveau und bei 
unterschiedlicher Adaptation ist daher von Interesse. Die Zoological Society of London und die 
Anatomical Society of Great Britain and Ireland setzten sich daher auf einem gemeinsamen Sympo- 
sıum zum Ziel, 300 Jahre nach Gıovannı BoRELLIs ‚De Motu Anımalium‘ (1680) Beiträge zu dieser 
Thematik auf dem heutigen Wissensstand zusammenzufassen. Im vorliegenden Buch sınd 17 Beiträge 
über Schwimmen, Fliegen, Gehen, Laufen und arborikole Fortbewegung veröffentlicht. Einzelne 
Themen sind u.a.: Analysen von Schwimmbewegungen bei Gadus morhua; Antriebsmechanismen bei 
aquatischen Wirbeltieren; Lokomotion von Plattfisch-Larven; Struktur und Funktion von Fischmus- 
kulatur; Organısation des Zentralnervensystems von Fischen ın Beziehung zur Lokomotion; Flug- 
anpassung bei Vertebraten (Gleiten bei Fischen, Amphibien, Reptilien, Säugetieren; Fliegen bei 
Vögeln und Säugern); Echolokation und Ortung beim Flug; Einsatz der Muskulatur beim Fliegen, 
Schwimmen und Laufen unter Aspekten des Energiehaushaltes; Evolution von Haltung und Gangart 
bei Tetrapoden; funktionale Adaptation bei Extremitätenknochen; Allometrie der Körperproportio- 
nen von Primaten; arborikole Lokomotion von Primaten. 

Das Buch gibt einen Überblick über die Vielseitigkeit spezieller Fragestellungen. Es ist geeignet, 
über den aktuellen Wissensstand zu informieren und zu vertiefenden weiterführenden Studien ın 
vielen Bereichen anzuregen. D. Kruska, Hannover 


CsuT1, B.: Type Specimens of Recent Mammals in the Museum of Vertebrate Zoology, 
University of California, Berkeley. University of California Publications Vol. 114. Berke- 
ley, Los Angeles, London: University of California Press 1980. VII + 75 pp. US-$ 8.50 


Das 1908 gegründete Museum of Vertebrate Zoology der University of California besitzt zur Zeit 
eine 156 000 Nummern umfassende Säugetiersammlung, die auch 334 Typus-Exemplare (Holotypen) 
ausschließlich amerikanischer Arten und Unterarten enthält. 268 von ihnen stammen aus den 
westlichen Vereinigten Staaten, die übrigen aus sonstigen Teilen Nord- und Südamerikas. Die 
vorliegende Übersicht führt in systematischer Folge alle Typusexemplare auf, nennt das Zitat der 
Erstbeschreibung und den nach der letzten Revision gültigen Namen, Umfang und Zustand des 
Sammlungsstückes, Körpermaße und Gewicht, ausführliche, nach Sammlernotizen ergänzte Fundort- 
angaben, Sammeldatum, Namen und Nummer des Sammlers. Eine regional und alphabetisch geord- 
nete Liste des Typenmaterials beschließt die informative, für den Taxonomen wichtige Übersicht. 
Eine praktische und nachahmenswerte Zusammenstellung. J. NIETHAMMER, Bonn 


128 Buchbesprechungen 


NIETHAMMER, ]J.; Krapp, F. (Hrsg.): Handbuch der Säugetiere Europas. Bd. 1: 
Rodentia. Wiesbaden: Akademische Verlagsgesellschaft 1978. 476 S., 85 Abb. Subskrip- 
tionspreis: DM 234,-, sonst: DM 328,-. ISBN 3-400-00458-8 


Das Handbuch der Säugetiere Europas, konzipiert nach dem Vorbild Handbuch der Vögel Mittel- 
europas, konnte nach mehreren vergeblichen Planungen doch noch erscheinen. Dieses ist der 
Bereitschaft der Herausgeber und vieler sachkundiger Autoren zu danken, sich einer so mühevollen 
Aufgabe wie der Erstellung eines solchen Werkes zu unterziehen. 

Von den fünf vorgesehenen Bänden liegt der erste seit geraumer Zeit vor. Dem Hauptteil dieses 
ersten Bandes sind nach einer Einführung und einer Benutzungsanleitung knappe Frläuterungen u.a. 
zur Systematik, zum Skelett, zur Ökologie und zum Verhalten der Säugetiere sowie ein Ordnungs- 
schlüssel vorangestellt. Ebenfalls mit Bestimmungsschlüsseln (nach äußeren Merkmalen, nach dem 
Schädel, nach den Zähnen) für die Nagerfamilien beginnt dieser Hauptteil. Sodann werden auf rund 
400 S. Sciuriden, Castoriden, Gliriden und Muriden abgehandelt. Nach allgemeinen Vorbemerkungen 
zu den Familien (u. a. Diagnose, Verbreitung, Verwandtschaftsbeziehungen, Gattungsschlüssel) wer- 
den die einzelnen Arten ausführlich dargestellt. Diese Darstellung umfaßt nach der Diagnose eine 
detaillierte Beschreibung inkl. der Angaben ın Tabellen zusammengefaßter zahlreicher, auf Individuen 
bezogener Maße. Zu diesen Beschreibungen gehören ferner ın der Regel mehrere Abbildungen 
wichtiger Strukturen und Organe. Es folgen Text- und Karteninformationen zur Verbreitung sowie 
Hinweise zur Merkmalsvariation, Paläontologie, Ökologie, Jugendentwicklung und zum Verhalten. 
Für jede Art ist ein gesondertes Literaturverzeichnis angelegt. 

Eine so umfassende Darstellung jeder einzelnen Spezies (z. B. 22 S. für Apodemus sylvaticus) ist 
nur möglich, wenn kompetente Sachkenner sich jeweils einer oder weniger Arten annehmen. Neben 
den Herausgebern sind folgende Wissenschaftler für den ersten Band tätig gewesen: K. BECKER, W. 
BÖHME, H.-A. FREYE, F. DIETERLEN, A. Ruzıc, H. REICHSTEIN, H. M. STEINER, G. STORCH, $. 
SuLkAva und H. WILTArsky. Zu recht merkt THEODOR HALTENORTH, der bereits vor dem zweiten 
Weltkrieg die Herausgabe des Handbuches ins Auge gefaßt hatte, in seinem Geleitwort an, daß es 
zumindest so unentbehrlich sein wird wie MıLLEr’s „Catalogue of the Mammals of Western 
Europe... .““. Wenn damit der Bedarf eines umfassenden Werkes über die europäischen Säugetiere für 
den deutschen Sprachraum festgestellt wird, so soll hier ergänzt werden, daf die wissenschaftliche 
Qualität dieses ersten Bandes und seine vorzügliche Ausstattung trotz des hohen Preises erwarten 
lassen, daß das Handbuch fester Bestandteil der Bücherei jedes Mammalogen sein wird. 

H. SCHLIEMANN, Hamburg 


STARCK, D.; FIEDLER, K.; HARTH, P.; RICHTER, J.-(Hrsg.): Biologie. Eine Vorlesungsreihe 
für Mediziner und Naturwissenschaftler. Weinheim, Deerfield Beach (Florida), Basel: 
Verlag Chemie 1981. 1118 S., 666 Abb., 80 Tab. DM 88,-. ISBN 3-527-25821-3 


Aus einer interdisziplinären Vorlesungsreihe von Medizinern und Naturwissenschaftlern an der 
Universität Frankfurt ist vorliegender Band entstanden. Er soll in erster Linie Studierende über die 
üblichen exemplarischen Darstellungen hinaus informieren, ohne allzu stark vereinfachenden 
Beschneidungen der angesprochenen Problemkreise unterworfen zu sein. Die Abhandlungen sind ın 
folgenden Themenkreisen zusammengefaßt: Übergreifende Aspekte und Evolution; Zelle; Stoffwech- 
sel und -transport; Lokomotion und Biomechanik; Information im Organısmus; Mikrobiologie, 
Virologie, Molekularbiologie und Genetik; Ökologie (Lebewesen und Umwelt); Krankheit als 
biologisches Problem; Leistungen pflanzlicher Organismen; Verhalten und seine Grundlagen. Die 
Bearbeitung der einzelnen Abhandlungen teilen sich 33 sachkompetente, vorwiegend Frankfurter 
Autoren. Alles in allem wird in jedem Abschnitt die Vielfalt der Lebenserscheinungen unter 
evolutiven Aspekten vor Augen geführt und abschließend der Mensch in den Vordergrund der 
Betrachtung gestellt. Die Palette der Themen ist sehr groß und vielfältig. Sie weicht von herkömmlı- 
chen Lehrbüchern ab, und gerade dadurch gewinnt dieses Buch an Interesse. Auf das weite Themen- 
angebot mögen folgende, stellvertretend genannte Überschriften hinweisen: Grundprobleme der 
Biologie und ihre Geschichte (STARcK); Beiträge der Biologie zum Selbstverständnis des Menschen 
(Kunn); Grundlagen der Mikroskopie (RICHTER); Sexualzyklus des Menschen (TAUBERT/KUHL); 
Genchirurgie (BRENDEL); Ökologie der Oberrheinebene (SCHÄFER); Entzündung, Schock und Krebs 
(HÜBNER); Pflanzenzüchtung (BUTTERFASS); etc. 

Alle Beiträge zeichnen sich durch einen verständlichen und deutlichen Text aus und durch 
instruktive und klare Abbildungen. Besonders zu betonen sind klare Definitionen von Fachausdrük- 
ken. Das Buch stellt eine interessante, vielseitige und anregende Informationsquelle dar, nicht nur für 
Mediziner, sondern auch für alle biologisch interessierten Laien und Fachkenner. Wegen der zahlrei- 
chen Bezüge auf den Menschen ist es sicherlich auch für Lehrer an Gymnasien von Nutzen. 

D. Kruska, Hannover 


Vergleichende Untersuchungen zur Nahrungswahl 
von Gams- und Rotwild 


unter besonderer Berücksichtigung der mikrobiellen Besiedlung und der Verdauungs- 
vorgänge im Pansen. Von Ute Drescher-Kaden. Heft 12 der „Fortschritte in der Tier- 
physiologie und Tierernährung“; Beihefte zur „Zeitschrift für Tierphysiologie, Tierernäh- 
rung und Futtermittelkunde“. 1981. 108 Seiten mit 29 Abbildungen und 30 Tabellen. 
Kartoniert 54,- DM 


Bei weiterem Rückgang der Viehhaltung im Alpengebiet gilt es zu prüfen, ob ein der trag- 
baren Wilddichte entsprechendes Vorkommen von wildlebenden Wiederkäuern auf auf- 
gegebenen Almen und Ödland zur alternativen Nutzung geeignet wäre. Um hierbei alle 
Ansprüche des in den betreffenden Biotopen lebenden Wildes zu erfassen, sind neben 
Studien über Wohn- und Aktionsbereiche, Aktivitätszyklen und Sozialverhalten speziell 
auch ernährungsphysiologische Untersuchungen erforderlich. 

Diesem Ziel dienen vergleichende, in sieben Gebieten des Alpenraumes an Gams- und 
‚Rotwild durchgeführte ökophysiologische Untersuchungen sowie Vergleichsmessungen 
am Rotwild eines Gebietes der Voralpenregion. Die in Feldversuchen gewonnenen Ergeb- 
nisse verdeutlichen die Zusammenhänge zwischen Ausbildung des Verdauungstraktes und 
artspezifischer Nahrungswahl unter wechselnden Ernährungsbedingungen, geben Einblick 
in mikrobielle Besiedlung und Fermentationsleistungen des Pansens und erlauben die 
Zuordnung von Gams und Rotwild als „intermediäre Ernährungstypen‘“. 

Angesprochen sind Tier- und Ernährungsphysiologen, ferner Ökologen, Wildbiologen, 
Forstwirte, Landschaftspfleger und Gehegetierärzte, die ihre Wildkenntnisse vertiefen 
möchten und Kriterien zur Beurteilung von Biotopkapazitäten suchen. 


Palatability and Flavor Use in Animal Feeds. 
Schmackhaftigkeit des Futters und Flavor-Anwendung 


First International Symposium on Palatability and Flavor Use in Anımal Feeds, 10.-11. 
October 1978 in Zurich. 18 Vorträge, zusammengestellt von Hans Bickel. Heft 11 der 
„Fortschritte in der Tierphysiologie und Tierernährung‘‘; Beihefte zur „Zeitschrift für 
"Tierphysiologie, Tierernährung und Futtermittelkunde“. 1980. 148 Seiten mit 27 Abbil- 
dungen und 77 Tabellen. Texte englisch, deutsch oder französisch, mit je einer englischen 
und deutschen, in einem Fall auch französischen Zusammenfassung. Kartoniert 58,- DM 


Wesentliche Voraussetzung der heutigen, auf hohe Leistung ausgerichteten Tierproduk- 
tion ist eine bedarfsgerechte Fütterung. Welche Möglichkeiten sich Wissenschaft und 
Praxis bieten, um durch Verwendung von Flavor den Geschmack des Futters zu verbes- 
sern, wird in 18 Beiträgen dieses Heftes geprüft. 


BGE  BAUL PAREY : HAMBURG UND BERLIN 


Erscheinungsweise und Bezugspreis: Die Zeitschrift erscheint alle 2 Monate; 6 Hefte bilden einen Band 
mit insgesamt 384 Seiten. Der Abonnementspreis beträgt je Band 232,- DM zuzüglich Porto. Das 
Abonnement verpflichtet zur Abnahme eines ganzen Bandes. Es verlängert sich stillschweigend, 
wenn nicht unmittelbar nach Erhalt des letzten Heftes eines Bandes Abbestellung erfolgt. 
Einzelbezugspreis der Hefte: 43,- DM. Die Preise verstehen sich im Inland incl. Mehrwertsteuer. 
Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, 
Spitalerstraße 12, D-2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesell- 
schaft für Säugetierkunde“ erhalten die Zeitschrift unberechnet im Rahmen des Mitgliedsbeitrages. 


ETHOLOGISCHESWÖRTERBUCH 
ETHOLOGICAL DICTIONARY 
VOCABULAIRE ETHOLOGIQUE 


Deutsch Englisch Französisch 


1000 Wörter Ethologie. Von Dr. Armin Heymer, Brunoy/Frankreich. Unter Mitarbeit namhafter Fach- 
wissenschaftler aus Europa und Nordamerika. 1977. 238 Seiten mit 138 Abbildungen. Balacron 
broschiert DM 28,- 


„Der Verfasser hat das Problem einer mehrsprachigen Terminologie in sehr geschickter Weise ge- 
löst. Er bringt nicht etwa hintereinander drei Lexika der berücksichtigten Sprachen, sondern geht 
von den deutschen Fachausdrücken aus, die in alphabetischer Reihenfolge das Grundgerüst 
bilden. Ausschlaggebend war die Überlegung, daß ein sehr großer Teil des Schrifttums, insbeson- 
dere in den frühen Jahren der Verhaltensforschung, in deutscher Sprache veröffentlicht wurde und 
viele Grundbegriffe zunächst deutsch formuliert wurden. Dem Stichwort wird jeweils eine eingehen- 
de sachliche Erläuterung beigegeben. Dite sprachliche Erklärung des Terminus wird, soweit nötig, 
berücksichtigt. Dem deutschen Text folgt das englische und dann das französische Stichwort 
gleichfalls mit ausführlicher Erläuterung. Zahlreiche Abbildungen ’und die Verwendung verschiede- 
ner Drucktypen sowie Indices in den drei Sprachen erleichtern das Auffinden. Es ist zu hoffen, daß 
das Werk eine weite Verbreitung findet und zur Verständigung beiträgt.“ 

Zeitschrift für Säaugetierkunde - D. Starck 


“With the increasing number of behavioural publications in the most important languages English, 


German and French, terminological confusion also grew and thus the need for an ethological dic- 
tionary giving nomenclature. It may therefore be regarded as a great merit of the author to have 
compiledsuch acompetentethological dictionary. The selection of about 1000 entriesindicatesthe 
necessary restriction of the material, individual species being treated more comprehensively than 
others because of their many years of treatment by ethologists. On the other hand, the recent 
sphere of human ethology is already well represented. Numerous entries are accompanied by in- 


structive illustrations. All in all, it is a valuable aid to all teachers and students of zoology, human 2. 


ethology and psychology which will, in particular, also facilitate access to an important Slaineh 
of science to interested persons in the rising countries of the Third World.” 
Mundus, Stuttgart - Prof. Dr. H. Mergner 


“Comme toutes les sciences neuves, l’ethologie souffre d’un certain flou terminologique, et les 
chercheurs sont parfois amenes a manipuler des notions qui ne sont pas encore bien definies. Le 
travaild’A.Heymercomble une lacune evidente ence domaine et poseles bases d’un veritable effort 
de conceptualisation dans les sciences du comportement. L’equivalence destermes en trois langues 
offre aux chercheurs un outil de travail precieux en levant l’equivoque qui pese parfois sur des 
traductions approximatives ou incompletes: la, chaque terme est suivid’une explication en allemand, 
enanglaiseten francais qui leve toute ambiguite. Troisindex completent l’ouvrage etpermettent de 
se rapporteraisement au terme recherche. C’est un travail unique qui se revelera d’une tres grande 
utilite pour les chercheurs, les enseignants et les etudiants des sciences du comportement.” 


“La Recherche 
— By 
Berlin PREV U Hamburg 


En 47 (3), 9 92, Juni 1982 ISSN 0044-3468 C 21274 FF 
Nam ma Is 


EITSCHRIFT FÜR 
SAUGE TIERKUNDE 


NTERNATIONAL JOURNAL 
DF MAMMALIAN BIOLOGY 


yrgan der Deutschen Gesellschaft für Säugetierkunde 


‚ephan, H.; Kuhn, H.-J.: The brain of Micropotamogale lamottei Heim de Balsac, 1954. — Das Gehirn von 
Micropotamogale lamottei Heim de Balsac, 1954 129 


hmidt, U.; Joerman, G.; Schmidt, Christel: Struktur und Variabilität der Verlassenheitslaute juveniler Vampirfle- 
‚dermäuse (Desmodus rotundus). — Structure and variability of the isolation calls of juvenile vampire bats 
‚(Desmodus rotundus) 143 


"häfer, M. W.: Analyse des Explorationsverhaltens verschiedener Stämme und Generationen der Hausmaus Mus 
musculus domesticus Rutty. Teil I. -— Analysis of exploratory behaviour in different strains and generations of the 
house mouse (Mus musculus domesticus Rutty). Part | 149 


hmidt, Christel: Behavioural and neurophysiological studies of the olfactory sensitivity in the albino mouse. — 
‚Verhaltensphysiologische und elektrophysiologische Untersuchungen zur Riechfähigkeit von Albinomäusen 162 


illiainen, E.: Habitat selection and fluctuations in numbers in a population of the arctic hare (Lepus timidus) on a 
‚subarctic fell in Finnish Forest Lapland. — Zur Selektion des Habitats und zur Fluktuation des Bestandes von 
'Schneehasen (Lepus timidus) auf einem subarktischen Fjäll im Waldlappland Finnlands 168 


‚er, F.: Quelques observations ethologiques sur l’Hydropote de Chine, Hydropotes inermis (Swinhoe, 1870) en 
‚captivite. — Some ethological observations on Hydropotes inermis (Swinhoe, 1870) in captivity. — Einige 
'ethologische Beobachtungen am chinesischen Wasserreh, Hydropotes inermis (Swinhoe, 1870), in Gefangen- 
‚schaft 175 


‚ssenschaftliche Kurzmitteilungen 


Ertos, L.: A note on the sexual behaviour in red deer hind. - Bemerkungen über das Sexualverhalten einer 
‚Rothirschkuh 185 


'sthammer, J.: Mausschläfer (Myomimus Ognev, 1924) in Afghanistan. -— Mouse-tailed dormouse (Myomimus 
‚Ognev, 1924) in Afghanistan A _ 3 187 


‚chbesprechungen | In 190 


'erlag Paul Parey Hamburg und Berlin 


HERAUSGEBER/EDITORS 


P. ]J. H. van BREE, Amsterdam - W. FIEDLER, Wien - H. Frick, München - W. HERRE, 

Kiel - H.-G. Krös, Berlin - H.-J. Kunn, Göttingen - B. Lanza, Florenz - T. C. S. 

MORRISON-SCOTT, London - J. NIETHAMMER, Bonn —- H. REICHSTEIN, Kiel - M. Rönrs, 

Hannover -— D. STARcK, Frankfurt a. M. - F. Strauss, Bern - E. THENIUs, Wien - 
W.VERHEYEN, Antwerpen 


SCHRIFTLEITUNG/EDITORTATL OFFICE 


H. SCHLIEMANN, Hamburg - D. KruskA, Hannover 


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. Harald Schliemann. 


Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z.Hd. Priv.-Doz. Dr. 
Dieter Kruska, Institut für Zoologie, Tierärztliche Hochschule Hannover, Bünteweg 17, D-3000 
Hannover 71. Für die Publikation vorgesehene Manuskripte sollen gemäß den „Redaktionellen 
Richtlinien“ abgefaßt werden. Diese Richtlinien sind in deutscher Sprache Bd. 43, H. 1 und ın 
englischer Sprache Bd. 43, H. 2 beigefügt; in ihnen finden sich weitere Hinweise zur Annahme von 
Manuskripten, Bedingungen für die Veröffentlichung 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 
Vervielfältigung auf photomechanischem oder ähnlichem Wege oder im Magnettonverfahren 
sowie der Speicherung in Datenverarbeitungsanlagen, bleiben, auch bei nur auszugsweiser Ver- 
wertung, vorbehalten. Werden von einzelnen Beiträgen oder Beitragsteilen einzelne Vervielfälti- 
gungsstücke in dem nach $54 Absatz 1 UrhG zulässigen Umfang für gewerbliche Zwecke 
hergestellt, ıst dafür eine Vergütung gemäß den gleichlautenden Gesamtverträgen zwischen der 
Verwertungsgesellschaft Wort, vereinigt mit der Verwertungsgesellschaft Wissenschaft GmbH, 
rechtsfähiger Verein kraft Verleihung, Goethestr. 49, D-8000 München 2, und dem Bundesver- 
band der Deutschen Industrie e.V., dem Gesamtverband der Versicherungswirtschaft e.V., dem 
Bundesverband deutscher Banken e.V., dem Deutschen Sparkassen- und Giroverband und dem 
Verband der Privaten Bausparkassen e.V., an die Verwertungsgesellschaft zu entrichten. Erfolgt 
die Entrichtung der Gebühren durch Wertmarken der Verwertungsgesellschaft, so ist für jedes 
vervielfältigte Blatt eine Marke im Werte von 0,40 DM zu verwenden. Die Vervielfältigungen sind 
mit einem Vermerk über die Quelle und den Vervielfältiger zu versehen. 


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 articie 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 
SA 


Fortsetzung 3. Umschlagseite 


© 1982 Verlag Paul Parey, Hamburg und Berlin. - Printed in Germany by Westholsteinische Verlagsdruckerei 
Boyens & Co., Heide/Holst. 


Z. Säugetierkunde 47 (1982) 3, 129-192 


InterCode: ZSAEA 7 47 (3), 129-192 (1982) 


Mit zwei Beilagen des Verlages Paul Parey 


The brain of Micropotamogale lamottei Heim de Balsac, 1954 


By H. STEpHan and H.-J. KuHnn 


Max-Planck-Institut für Hirnforschung, Neurobiologische Abteilung, Frankfurt a. M. 
and Anatomisches Institut der Universität, Göttingen 


Receipt of Ms. 8. 12. 1981 


Abstract 


Studied the brain of Micropotamogale lamottei. It differs markedly from the brains of “average 
Insectivora” by less developed olfactory structures and a larger medulla oblongata. The large size of 
the latter is caused by a marked enlargement of the nucleus of the spinal trigeminal tract. Since similar 
characteristics are present in all water-adapted Insectivora, such as Zimnogale, Potamogale, Neomys, 
Desmana, and Galemys, they are thought to be related to predatory habits in limnetic ecosystems. The 
trigeminal system, innervating the strongly developed vibrissae of the muzzle, is thought to replace the 
olfactory system in water-adapted Insectivora and to become the main sensory system involved ın 
searching for food. Within the otter-shrews, the enlargement of the medulla oblongata and the 
concomitant reduction of the olfactory structures are in M. lamottei less marked than in Potamogale 
velox. Similarities in the brain characteristics of M. lamottei are with the shrew-like tenrecs of 
Madagascar (Oryzorictinae). 


Introduction 


The african water or otter-shrews comprise two genera (Micropotamogale and Potamogale) 
and three species (M. lamottei, M. ruwenzorü, and P. velox) from which, so far, only the 
brain of the giant form (P. velox) has been investigated (STEPHAN and SPATZ 1962; 
BAucHoT and STEPHAN 1967, 1968, 1970; STEPHAN 1967; STEPHAN and Anpy 1982). The 
brain of P. velox was found to have several characteristics which set it apart from most 
other Insectivora brains investigated. These are: 1. the tremendous size of the medulla 
oblongata, caused mainly by the extraordinary development of the nucleus of the spinal 
trigeminal tract; and 2. relatively very small structures of the main olfactory system. These 
special features of P. velox are obviously related to its search for food in water, and it 
would be of interest to know if similar brain characteristics are present also in the dwarf 
otter-shrews of the genus Micropotamogale. This paper focusses on four problems: 

1. Are there brain characteristics common to Potamogale and Micropotamogale which set 
the brains of these otter-shrews apart from those of other Tenrecidae and other 
Insectivora families? 

2. Can these brain characteristics be related to the search for food in water and do they 
exist in Micropotamogale to the same degree as they do in Potamogale? 

3. On the basis of brain development, what is the general evolutionary level of otter- 
shrews with respect to other Insectivora, and what is their status with respect to each 
other? 

4. Are the results compatible with the classification of otter-shrews? 


Material and methods 


The description of the macromorphology of the brain of M. lamottei ıs based on two brains (320, 
1728). For the determination of brain size, the cranial capacities (CrC) of four skulls were measured 
and, in addition, twelve skulls of P. velox and three skulls of M. ruwenzorii were also measured. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4703-0129 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 129-142 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


130 H. Stephan and H.-J. Kuhn 


Volumes of the brain parts were determined from three frontally serial-sectioned brains (320, 1227, 
1728) of M. lamottei. Corresponding volumes of two brains of P. velox (A 240, A 367) and of 26 other 
species of Insectivora (incl. 2 species of Macroscelidea) were used for comparison. Data on these 
volumes were published by STEPHAN (1967), and STEPHAN et al. (1970, 1981). 

The brain components investigated so far are: medulla oblongata, cerebellum, mesencephalon, 
diencephalon, and telencephalon; and within the telencephalon: olfactory and accessory olfactory 
bulbs, piriform lobe (= palaeocortex + amygdala), septum, striatum, hippocampus, schizocortex, and 
neocortex. 

The methods of volume determination as well as volume (and brain weight) comparison were 
described ın detail in previous papers (mentioned above) and so only brief comments need be given 
now. Firstly, all volumes are corrected to the fresh brain, and secondly, the comparisons are based on 
the allometric method. 

By aid of the allometric method the interdependency of body weight on one hand and brain weight 
or volume of varıous brain parts on the other hand was determined by calculating regression lines 
within groups of closely related mammals, usually families or subfamilies. In the double logarıthmic 
scale the regression lines of these groups have remarkably uniform slopes, even if the groups belong to 
different orders of mammals. By a canonical averaging method, for the mathematical basıs of which we 
are indebted to Dr. U. REMPE from the University in Kiel (Germany), the common slope has been 
derived using data from 34 groups within Insectivora, Primates and Chiroptera. This slope is 0.66 for 
the total brain (Fig. 7) and between 0.53 and 0.73 for the various brain parts (Figs. 9-11). A reference 
line with the slopes thus derived can be drawn anywhere in the double logarıthmic diagrams. Since we 
are interested in this paper in comparing the otter-shrews with other Insectivora, the reference lines 
were plotted through the average log body weight and the average log brain weight or log volume of 
brain structures of the 28 species of Insectivora and Macroscelidea studied. Any value on thıs line will 
be considered in the following presentation to represent the “average Insectivora”. The distance of M. 
lamottei (as well as that of P. velox or any other species) from the reference line expresses the degree of 
deviation from the “average Insectivora”. When measured parallel to the ordinate and given ın 
antılogs, such values express immediately the size of the given structure relative to that of an “average 
Insectivora” of equal body weight. The reference line itself has of course a distance of 0, the antilog of 
which is 1 (or = 100%). Parallel lines, drawn in Figures 7 and 9-11 in multiples of 0.301 (which is the 
logarıthm of 2), show when a doubling or quadrupling or when a half or a quarter is reached. We have 
called such values “indices of comparison” (allometric indices, progression indices, enlargement 
factors). In Figures 8, and 13 to 16, the indices have been scaled. 

The structure which has undergone the greatest enlargement in primate evolution is the neocortex 
(STEPHAN 1972). Consequently neocortical sıze can be expected to represent the best cerebral criterion 
(at present available) for the evolutionary level of a given species. In Figures 8, 13, 14 and 15 the 
vertical columns (= range of varıation) of the systematic (taxonomic) groups were arranged from left 
to right according to increasing neocortical indices as shown in Figure 16. The same arrangement is 
present in all scales and so the trends in the data can be easily seen, as can the relationships of the brain 
components of otter-shrews to those of other Insectivora. 


Results 
Macromorphology 


In the brain of M. lamottei, the olfactory bulbs clearly project beyond the frontal poles of 
the hemispheres and are relatively broad. In the dorsal view (Fig. 1), a very small part of the 
mesencephalon is visible between telencephalon and cerebellum. Each cerebral hemisphere 
is wedge-shaped and its largest width is found in the caudal third. From the side (Fig. 2), 
the brain appears relatıvely flat. From below (Fig. 3), a very big trigeminal nerve as well as a 
very big medulla oblongata are conspicuous. The flocculi are well developed. 

Compared with ?. velox (Figs. 4-6), the hemispheres and the olfactory bulbs of M. 
lamottei are shorter and broader. In the dorsal view, the caudal borderline of each 
hemisphere is pointed in ?. velox (Fig. 4), but broad and rounded in M. lamottei. The angle 
between the caudal contours of the two hemispheres ıs 105° and 155°, respectively. In ?. 
velox, the gap is filled by the cerebellum, and especially by its vermis, whereas in M. 
lamottei, a small cleft separates telencephalic hemispheres and cerebellum. Apart from 
these differences, the brains of M. lamottei and P. velox are sımilar in their macromor- 
phological appearance. 


The brain of Micropotamogale lamottei Heim de Balsac, 1954 131 


Figs. 1-3. Brain of Micropotamogale lamottei from dorsal side (Fig. 1), left side (Fig. 2), and ventral 

side (Fig. 3). Linear enlargement 3,0 x. The drawings (Figs. 1 and 2) are made from brain 494 byH. ]. 

Kunn, the photograph (Fig. 3) is taken from brain 1728. The latter brain was fixed in situ and not 

prepared and photographed until two weeks later. By that means the peduncles of the cerebellar 

flocculi are stretched. — Figs. 4-6. Brain of Potamogale velox (A 240) from dorsal side (Fig. 4), left side 
(Fig. 5), and ventral side (Fig. 6). Linear enlargement ca. 2.15 x 


Brain size and encephalization 
Brain and body size 


For interspecific comparisons using the allometric method, data on body size and brain 
size are needed. Data on body weights are rare but are available for all three species of 
otter-shrews. Brain weights are available only for P. velox. For the two species of 
Micropotamogale, fresh brain volumes were determined indirectly from cranıal capacities 
(CrC) and from the volumes of serially sectioned brains (VSS). CrC measurements in 
twelve skulls of P. velox, four of M. lamottei, and three of M. ruwenzorii were made by 
introducing fine dust shot into the cranıal cavity through the foramen magnum, weighing 
the skull before and after introducing the dust shot, and by dividing the weight difference 


U9INAIIL, 'SNW ‘[NYS 
u91NnAI> L, 'SnW “[[nYS 
u91NnAIIL, 'SnW “[NYS 
u91nAII ], 'sSnW “[NYS 
U9INAIIL, 'SNW ‘[NYS 
u91NAIIL, "SAW “NS 
u91nAl> ], 'SnW “NYS 
u9INAII], "SnW “nNS 
u9ANAIIL, 'SnW ‘[INYS 
u91nAT9], 'SnW “Ins 
u9NnATII L, 'SnW “[NYS 
U9INAIIL, "SAW “INS 

IdW 02 "urerg 

IdW 1102 “ureıq 

IdW 702 “urerq 

Ta o>ueag 

IdW 'I[02 “ureıq 


"wwo9a 'sıad *TIIOA 

MOAVLS [JO “ns 

LOHDNVG EIA ALIOWVT ‘[JnYS 
NHNNY 'JJO> “ns 

NHNN [03 “Tpnys 

LOHDNVI EIA TF90X ‘IdW “ureig 
NHNNY 'JJO> “ureig 

NHNN 'JJO> “ureig 


(1961) WHYY 
(0961) WHYY 
(0961) WHYY 
(0961) WHYY 
USINAIIL, 'SnW “TINYS 
USINAIIL, "SnW “Ins 
u91nAl> ]L, "sn “[JNYS 


jersagew jo uıIO 


SMIIYS 19330 UI SIUINJOA UONI9S [eLISS pur saryısede> jerue.1s ‘sJydram ureig ‘sIydram Apog 


(zı=Uu) £91‘r 
g1z‘E 
6/0‘ 
6rL‘E 
Sr0‘% 
pr6‘E 
969° 
c8T+ 
618‘E 
rEOo+ 
6€£9% 
est 
I8€‘r 


lZ/ 
0€8 
806 
8lZ 


(GUZZI 


z90° I 

691 

1071 
Zu 


p>urunsI>p 
ıy3ıam ureig 


p>ınseaui JUINTOA p>ınseoaul 
UOND9S [e119S 


BEE € - - 
c99‘E - - 
gI9E‘E 
Peg 
prG°€ 
6IcH 
058°€ 
1Er‘E 
979€ 
891 
601’ 
9E6°€ 


E erEn Ei DES rErer 


zsr 
SG 
(GEZ) 


Seesen 


r 
= 


(e=u) 850° 


+56 
OFT‘ 
6/01 
wu uıu Zu 


€ € 
p>ansesw 3 


Aynede> jerueıd ayaıam ureig ıy3ıam Apog 


ENGINE 


x0]2% 
apBowwo] 


19170140] 
opwBowviodowsiy 


ULE9UI 


NAOZUDIMMA 
appBowurı0domsıy 


Joqunu [229904 


sawads 


The brain of Micropotamogale lamottei Heim de Balsac, 1954 133 


by 5.6121, which was found to be the weight in grams per cubic centimeter dust shot. The 
results are given ın Table 1, column5. 

Both brain weights and cranial capacities are known only for P. velox. Based on the 
average values of the two parameters (Table 1, columns 4 and 5) a conversion factor (C..c) 
was calculated. 


average cranical capacity 3,741 


The same calculation was made using the maximal brain weight given in column 4 and the 
4,700 


maximal cranial capacity given in column 5 (= ET 1.114) as well as using the mınima 
3,700 mn | | 
= 3.339 = 1.108). Thus 1.113 can be considered to be an appropriate factor for converting 


cranical capacities to brain weights in P. velox. 

This factor was also used to convert the cranıal capacities of the two species of 
Micropotamogale to brain weights. This could introduce some inaccuracy in the calcula- 
tion, but it could be checked in M. lamottei by another approach in which the volumes 
were calculated from serial-sections. In more than 300 brains of insectivores, primates and 
bats, in which both the fresh weights and the serial-section volumes were measured, we 
found that the average relationship between these two parameters (Cyss) was 1.94 : 1.036 
(STEPHAN et al. 1981). 1.036 is the specific gravity of fresh brains and so can be used to 
convert brain volumes into brain weights or vice versa. 1.94 is the average shrinkage in 
brain volume due to fixation and embedding procedures. With Cyss = 2.01 the serial- 


log brain weight 


1.0 1.5 2.0 2.5 3.0 
log body weight 


Fıg. 7. Brain weight (in milligram) plotted against body weight (in gram) in a double logarithmic scale. 

The reference base-line (thick) has a slope of 0.66 and is plotted through the average log brain weight 

and the average log body weight of 26 Insectivora and 2 Macroscelidea species. Its formula is: log brain 

weight = 1.708 + 0.66 : log body weight. The thinner parallel lines point to deviations according to 

the factor 2, i.e. two times (= 200 ın Fig. 8) or one-half time. The three otter-shrew species are marked 

by big round dots. M = Micropotamogale lamottei, Mr = Micropotamogale ruwenzorü, P = 
Potamogale velox. (For further explanations see “Material and methods’) 


134 H. Stephan and H.-J. Kuhn 


section volumes of the two brains (1227, 1728) of M. lamottei (Table 1, column 6) were 
converted to fresh brain weights (column 7). The average (813 mg; n = 2) is very close to 
the average obtained from the CrC measurements (800 mg; n=4) when using 
Cc.c = 1.113. The overall average ıs 804 mg (n = 6) (Table 1). The brain of the juvenile 
anımal 494 is clearly smaller and therefore was not taken into account. This animal still had 
all of its milk-teeth, ı.e. its first dentition (KuHn 1964). 


Encephalization 


The data ın Table 1 indicate the following provisional standards of body and brains weights 
in the otter-shrews: M. ruwenzorii 95 g/1,180 mg, M. lamottei 64.2 g/800 mg, and P. velox 
660 g/4,160 mg. These standards result in encephalization indices of 115, 101 and 112, 
respectively, the average of which is 109 (Fig. 8). Thus, the weight of the brain in otter- 


200 
0 — MACROSCELIDIDAE 


TALPIDAE 
CHRYSOCHLORIDAE 


POTAMOGAL INAE Fig. 8. Brain weight (encephali- 

Mr zatıon) indices. The indices are 

h numerical values of the dis- 

tances from the base-line ın Fig. 

7 and given in percent. The 

base-line represents the “aver- 

age Insectivora” and is 100 %. 

Vertical bars represent the range 

of varıation within each sys- 

tematic group (families or sub- 

families) with the average 

SORICIDAE marked by a dot. The bars are 

arranged from left to right ac- 

cording to increasing neocorti- 

cal progression (taken from Fig. 

ERINACEIDAE 16). The three otter-shrew 

species are marked as ın Fig. 7. 

The horizontal bars to the right 

correspond with the mean of 

TENRECINAE the individual systematic group, 

the name of which is given. (For 

0 further explanations see “Mate- 
rial and methods’) 


M ORYZORICTINAE 


SOLENODON 


indices brain weight 
r 
© 
° 


shrews ıs about the same as that in the “average Insectivora” (= 1 or 100%) or slıghtly 
larger. As far as can be ascertained from the poor material available, the encephalization of 
P. velox and M. ruwenzorii is somewhat higher than that of M. lamottei. 

As can be seen from Figure 8, otter-shrews and the Oryzorictinae (the shrew-like 
tenrecs of Madagascar) have an equal degree of encephalization. Higher indices are found 
in the Macroscelididae and Talpidae, lower ones in the Erinaceidae, Soricidae and espe- 
cıially in the hedgehog-like tenrecs of Madagascar (Tenrecinae). The lowest index in 


The brain of Micropotamogale lamottei Heim de Balsac, 1954 135 


Insectivora was found to be 60 in Tenrec ecaudatus. Allometric comparisons indicate that 
the brain of M. lamottei ıs 1.7 times larger and those of M. ruwenzori and P. velox are 1.9 
times larger than that of Tenrec. When compared with the corresponding scale for the 
neocortex (Fig. 16), the encephalization (= total brain) indices of the otter-shrews (Fig. 8) 
are low. As will be shown in the following para&raphs, this is due both to a relatively large 
neocortex and to especially small non-neocortical portions of the brain, particularly the 
olfactory structures. 


Composition of the brain 


The average volumes of the various structures of M. lamottei, and the standard errors of 
the mean (SEM) in percent of the mean, are given in Table 2 (columns 1 and 2, 
respectively). Within M. Jamottei, the greatest variatıon in size occurred in olfactory bulbs, 
cerebellum, and septum, the lowest in the diencephalon. The differences in varıability of 


Table 2 


Data on brain composition in otter shrews 


Percentage composition of the total brain Allometric indices related to Insec- 
(net tissue = 100 %) tıvora regression lines (= 100 %) 


Structural complexes Average M. P. Average Average of M. P. Average 
volumes lamottei velox ofthe 26Insecti- Jamottei velox ofthe 
(mm?) in two vora and two 
M. lamottei species two Ma- species 
croscelidea 
species 


Medulla oblongata 
Cerebellum 
Mesencephalon 
Diencephalon 
Telencephalon 
Bulbus olfactorius 
Palaeocortex 
Septum 

Striatum 
Schizocortex 
Hippocampus 
Neocortex 


the various components seem to be mainly due to true intraspecific variation rather than to 
difficulties in the delineations. Structures difficult to delineate, as for example diencephalon 
and mesencephalon, have lower SEM-values than those easy to delineate (olfactory bulbs, 
cerebellum). Similar results were obtained from data of 165 brains of Insectivora and 
Primates (STEPHAN et al. 1981). 

The percentages of the measured brain components relative to the net volume of the 
total brain (= 100%) are given in column 3 of Table 2. Compared with corresponding 
values of P. velox (column 4), the cerebellum of M. lamottei ıs relatively small, and the 
telencephalon relatively large. Within the telencephalon, the olfactory bulbs, the 
palaeocortex, and the hippocampus are larger in M. lamottei than in P. velox, whereas the 
neocortex is distinctly smaller. 

Compared with the average percentages of brain components in 26 Insectivora and 2 
Macroscelidea species (column 6), the otter-shrews, on the average (column 5), have a 
clearly larger medulla oblongata but a smaller telencephalon. Within the telencephalon, the 
percentages of olfactory bulbs and palaeocortex are clearly smaller in the otter-shrews, 
whereas the neocortex ıs distinctly larger. 

Since the percentages of the various structures are not independent of one another (e.g. 


0| ıy6b,9m Apoq Bor 
0°E Ss’ 0° ST 0°T 


(‚spoy3>ur pue jersyeW, 
pue / 31] 998 suoneuejdx9 Jaylıny 10,]) 'x073@ ap8owvJo] = d Tonowp] 
apdowrjodosıy = W 'sı0p puno1 dıq Aq paymeu ae sormads Maıys-1910 
om Ay], ‘Appanaadsaı ‘97 pure pI ‘EI 'sdıg ur uaAıd are sasıpur uoneraap UL 
ysıam Apoq 30] - £Z'0 + #Z'0 = X9110909U JuınfoA 30] 
ıysıam Apoq 30] - /G'0 + #Z'0 = ging Aıorseyjo JuınjoA 30] 
ıy3ıam Apoq 30] . gG'0 + £6'0 = EIeduoggo ejjnpauu JuınjoA 30] 
:91e SepnULIOF AIOU], 's9Wads eapı[osso1se OMI pure EIOANIISU] 97 Jo 
ıydıam Apoq adeıaae ayı pur HınIon.1S Y9e9 Jo Jun[oA ade.1aAe 9yı jo yurod ayı 
ysnoıyı 91e (NOIy3) sOUT[-seq H9UA19JOY 'X3110909U :77 "17 !snı1oer]o snginq 
:07 "317 Sereduojgo ejjnpaw .6 "17 'safeos Srwupriedof ofqnop ur (we.ı3 ur) 
sıysıam Apoq ısurede P>210Jd (‚ww ur) sononas ureig jo swnjoy 'IJ-6 'sd1] 


aıng Ruoıandyro ton Bor 


ıybram Apoq Bor 6 ıybram Apoq Bor 
| | 0’E EL 0'& ST 0'T 0’E Ss'L 0'Z Sal 0'T 


s'T 
Sal _ 
je) 
(0 
per < 
o o 
175) _ 
R » 
o 
0'232 — ., & 
0’Zc 
fe 
n = 
o P 
8 o 
6; oO 
E _ 
4 5 
. * 
5’ [de) 
[°) 
SEZEE 


The brain of Micropotamogale lamottei Heim de Balsac, 1954 187 


the high percentage for the medulla oblongata will tend to reduce the percentages for all 
other brain parts) the volumes will also be compared by the allometric index method. 
The allometric indices of the various brain parts of M. lamottei indicate the deviation of 
this species from an “average Insectivora” of equal body weight. In Figures 9 to 11, the 
sızes of the indices correspond to the dis- 
tances of the big dots marked M from the 
thickly-drawn lines. They are close to 100 
when the structure under consideration is of 
similar size to that of an “average Insectiv- 
ora”, and some distance from 100 when the 250- 
structure varies in size from an “average In- 
sectivora”. In Figure 12, we have sum- 
marized the indices of all measured struc- 
tures for both M. lamottei and P. velox. 
In both species, the highest deviation 
from the “average Insectivora” is that ofthe og 
accessory olfactory bulb, but the direction 
of deviation differs between the species. 
Whereas the accessory bulb ıs minute in M. 
lamottei (index = 28) it is especially large ın 
P. velox (index = 276), i.e. nearly ten times 
larger (body size equalized). Similar differ- 


Bulbus olf. acc. 


LAMOTTEI 


Medulla obl. 


POTAMOGALE VELOX 


= 
< 
6) 
O 
z 
< 
- 
O 
[ 
O 
o£ 
— 
z 


Neocortex 


Mesencephalon 


ences, but by far less dramatic, are found in 150 - 
the cerebellum which, in M. lamottei, is Medulla obl. 
smaller than in the “average Insectivora” 
Sehizocgriex Schizocortex 
2 — a 1 1 eocortex 
(index 87) but which, in ?. velox is larger Be Cerebellum 
(index = 125). All other structures in the Septum Diencepnalen 
e : : : : Hippocampus Hippocampus 
two species deviate in the same direction Diencephalon Ieralirein 
from the “average Insectivora”. Especially 100 Total brain 
large, and comparatively larger in ?. velox Telencephalon Telencephalon 
p 
. . Striatum 
than ın M. lamottei, are medulla oblongata erebelltm 
Striatum 


(indices = 205 versus 141) (Figs. 9 and 13), 
neocortex (167 versus 125) (Figs. 11 and 16), 
and mesencephalon (154 versus 123), where- RER ie 
as in both species similar indices are found Bulböst luck 
for schizocortex (127 and 129), septum (107 
and 119), hippocampus (114 and 112) and 
diencephalon (118 and 111) (Fig. 12). 

The olfactory bulb is especially small 
both in M. lamottei and P. velox (Figs. 10 
ac Ku) mal „acıges a1 37 Aal 2), Raspes- Fig. 12. Distances from the base-lines given in 
tively. Similarly low values are found forthe 9, of the ware Imeeivon (> 100) or ex 
olfactory cortices (palaeocortex) which, structures measured in Micropotamogale 
however, also contain the amygdala. The lamottei and Potamogale velox 
low values for the olfactory structures are 
considered to be one reason for the small size of the telencephalon as a whole, which in 
the otter-shrews is smaller than in the “average Insectivora”, though the neocortex is 
clearly larger. Another reason for the small size of the telencephalon is the low value for 
the striatum in the two species (Fig. 15). The indices are 90 for M. lamottei and 79 for P. 
velox (Table 2 and Fig. 12). 


90 - 


Palaeocortex 


Bulbus olfact. 


Bulbus olf. acc. 


138 H. Stephan and H.-J. Kuhn 


Discussion 


Brain characteristics of otter-shrews compared with those of other families 
of Insectivora 


Sımilarities in macromorphology and composition of the brains of M. lamottei and P. velox 
make it possible to specify brain characteristics which are common to the otter-shrews and 
hence facilitate comparisons with the brains of other Insectivora families and/or sub- 
families as described by STEPHAN (1967), BAUCHOT and STEPHAN (1970), and STEPHAN and 
Anpy (1982). Such brain characteristics, which either alone or in combination specify 
otter-shrew brains, are: 1. no exposure (or very little) of the mesencephalic tectum; 2. a 
gradual tapering of the medulla oblongata into the spinal cord, 3. the tremendous size of 
the medulla oblongata, and 4. the very small size of the structures of the main olfactory 
system. 

The brains of the hedgehog-like tenrecs of Madagascar (Tenrecinae), of the golden 
moles (Chrysochloridae) and of the elephant shrews (Macroscelidea) differ clearly from 
those of the otter-shrews ın having distinctly larger gaps between the cerebral hemispheres 
and the cerebellum so that a large portion of the mesencephalic tectum ıs visible. 
Furthermore the brains of all of these families, as well as those of the solenodons 
(Solenodontidae), hedgehogs (Erinaceidae) and most shrews (Soricidae), are characterized 
by relatively larger olfactory bulbs and centers. From the brains of the shrews (Soricidae) 
and moles (Talpidae), those of the otter-shrews clearly differ in the shape of the medulla 
oblongata. In the otter-shrews it tapers off gradually into the spinal cord, whereas in 
shrews and moles it narrows abruptly. 

Otter-shrew brains are most like those of the shrew-like tenrecs of Madagascar 
(Oryzorictinae). In general, the brains of the latter differ in having larger olfactory 
structures and a smaller medulla oblongata, except for Limnogale mergulus, which is also 
water-adapted. 


Water adaptation 


In the otter-shrews the marked enlargement of the medulla oblongata is the result of a 
strong development of the nucleus of the spinal trıgeminal tract. Also the trigeminal nerves 
are extremely large (Figs. 3 and 6). They innervate the strongly developed vibrissae of the 
muzzle, which are thought to be vibration receptors adapted for detecting prey (mainly 
crustaceans) under water. The trıgeminal system is thought to replace the olfactory system 
in water-adapted forms, and to become the main sensory system in the search for food. The 
main olfactory system (excluding the accessory olfactory system) seems to be unqualified 
for such a function. It ıs well developed ın all terrestrial Insectivora and has probably 
undergone reduction in the water-adapted forms. 

Both features (i.e. enlargement of the trıgeminal system and simultaneous reduction of 
the olfactory system) are found also in other species with a similar adaptation. Thus, the 
olfactory system is reduced also in ZLimnogale (Oryzorictinae), Neomys (Sorıicidae), 
Galemys and Desmana (Talpidae), (marked by - ın Fig. 14), whereas the medulla oblon- 
gata ın all of these forms is large or very large (marked — in Fig. 13). 

In contrast to the main olfactory bulb, the accessory olfactory bulb seems to be 
unaffected by the animal’s adaptation to water. This small bulb, which is well developed in 
P. velox (index 276), ıs, in M. lamottei, even more reduced than the main bulb (indices 28 
versus 52). In a similar manner, Limnogale, Neomys, and the Desmaninae which are also 
predators in limnetic ecosystems, show no consistency in the sizes of the accessory bulb 
but rather show a variation that encompasses the complete range of size found ın 
Insectivora. Marked differences in accessory bulb development between species of the 
same family are found ın various Insectivora families (STEPHAN 1965, 1975). 


Indices medulla oblongata 


Indices striatum 


300 


200 


300 


200 


100 


The brain of Micropotamogale lamottei Heim de Balsac, 1954 139 


POTAMOGALINAE 


MACROSCELIDIDAE 


TALPIDAE 
ORYZORICTINAE 
— 
ud 
SOLENODON 


+ 


E: 


SORICIDAE 
| CHRYSOCHLORIDAE 
ERINACEIDAE 


TENRECINAE 


6 


TALPIDAE 


CHRYSOCHLORIDAE 
MACROSCELIDIDAE 
SOLENODON 


ORYZORICTINAE 
SORICIDAE 


M 
ö =— POTAMOGALINAE 


ERINACEIDAE 


TENRECINAE 


19 


Indices olfactory bulb 


Indices neocortex 


300 


200 


100 


400 


300 


100 


WloE 


CHRYSOCHLORIDAE 


SOLENODON 
MACROSCELIDIDAE 
ERINACEIDAE 
4 TENRECINAE 
O 


—— ORYZORICTINAE 


? —_—— TALPIDAE 


SM a 
» SORICIDAE 


a5P 


POTAMOGALINAE 


14 


TALPIOAE 
MACROSCELIDIDAE 
POTAMOGALINAE 
CHRYSOCHLORIDAE 
ORYZORICTINAE 
SORICIDAE 
SOLENODON 
ERINACEIDAE 


TENRECINAE 


16 


Figs. 13-16. Indices of the medulla oblongata (Fig. 73), bulbus olfactorius (Fig. 14), striatum (Fig. 15), 
and neocortex (Fig. 16). The indices are numerical values for the distances from the base-lines in Fig. 
9-11, respectively, and given ın percent of the “average Insectivora” (= 100). Species adapted to water 
are marked in Figs. 13 and 14 by —. (For further explanations see Fig. 8) 


140 H. Stephan and H.-J. Kuhn 


From the papers of HEım DE Barsac (1954, 1956, 1957), HEım DE Barsac and 
BOURLIERE (1955), DEWITTE and FRECHKOP (1955), GUTH et al. (1959, 1960), Raum (1961) 
and DusosTt (1965) M. lamotteiıs clearly less adapted to water than the other otter-shrews. 
M. lamottei has no laterally compressed tail, as does ?. velox, and no webbed hands and 
feet, like M. ruwenzorii. There are, however, some indications that M. Jamottei is also able 
to catch its prey in water. This species ıs found mainly in the vicinity of watercourses or of 
swamps, and crustaceans, which never leave water, were found in its stromach (GuTH et al. 
1959). The enlarged upper lip with its strongly developed vibrissae would, according to 
GUTH et al. (1959, 1960), also suggest an adaptatıon to catching food in water. 

The brain data support the suggestion that M. lamottei is less adapted to water than 
P. velox. In M. lamottei medulla oblongata and cerebellum are clearly smaller and the 
olfactory bulb is distinctly larger (Fig. 12). These structures (and others to a lesser degree) 
are more sımilar in size to those of terrestrial Insectivora. The larger sıze of the cerebellum 
in P. velox might be correlated with an adaptation to a large three-dimensional space as is 
found in their habitat, which ıs larger bodies of water. 


The general evolutionary level of otter-shrews 


When evaluated according to its encephalization (based on total brain size), the general 
evolutionary level of M. lamottei would be the same as that in the “average Insectivora” 
(index = 101), but when evaluated on the size of the neocortex, it ıs clearly higher 
(index = 125) (Fig. 12). The degree of development of the neocortex ıs probably the better 
crıterion for estimating evolutionary levels. 

According to its brain characteristics, M. lamottei is recognizably less progressive than 
P. velox. This is suggested by the somewhat lower indices of the total brain, the 
telencephalon as a whole and the diencephalon, and the clearly lower indices of medulla 
oblongata, mesencephalon, cerebellum, and especially of the neocortex. Some other brain 
parts have, in M. lamottei, somewhat higher indices than ın ?. velox. This includes the 
limbic structures (septum, schizocortex and hippocampus) and the main olfactory struc- 
tures (main bulb and palaeocortex). However, these structures are of minor importance 
when evaluating the evolutionary level. 

The general evolutionary level of the otter-shrews is considered to be in the upper half 
of the total range found in Insectivora, with M. lamottei closer to the “average Insectivora” 
and ?. velox, and possibly also M. ruwenzorii, somewhat higher. The evolutionary level of 
M. ruwenzoril could only be estimated on the basis of its encephalization, since the brain 
itself and its composition. could not be investigated. According to its encephalization, its 
place is closer to P. velox than to M. lamottei. 

A lower position for M. lamottei and/or a higher one for M. ruwenzori and P. velox 
were also found in anatomical and behavioral investigations by HEIM DE Bausac (1954, 
1956), GUTH et al. (1960), VERHEYEN (1961) and Dugosrt (1965). M. lamottei is considered 
to be closer to the basic forms than are the other two species; similarities with the shrew- 
like tenrecs of Madagascar (Oryzorictinae) are emphasized. 


Classification of otter-shrews 


Comments on anımal classification based on brain characteristics generally must be 
regarded with care, since the brain’s great plasticity frequently results in convergent 
developments (e.g. in water-adapted Insectivora). With these restrictions in mind, the 
brains of the otter-shrews are most like those of the shrew-like tenrecs of Madagascar 
(Oryzorictinae). This is true for the macromorphology of the brain, and for encephaliza- 
tion (based on comparison of total brain weights, Fig. 8), as well as for the allomertric size 
of many structures, which have totally or partly the same range of variation ın otter-shrews 


The brain of Micropotamogale lamottei Heim de Balsac, 1954 141 


as in Oryzorictinae. Such structures are: hippocampus, septum, cerebellum, schizocortex, 
telencephalon as a whole, neocortex, diencephalon, and medulla oblongata. In many 
respects M. lamottei ıs intermediate in brain composition between P. velox and the 
Oryzorictinae. 

The results of our presentation are compatible with the view that the otter-shrews are so 
closely related to the tenrecs of Madagascar, that they should be regarded as a subfamily 
(Potamogalinae) of the Tenrecidae (CABRERA 1925, after FRECHKOP 1957; WEBER 1928). In 
support of these authors and contrary to Simpson (1945), who classified the otter-shrews 
as a separate family (Potamogalıdae), the discovery of the dwarf otter-shrews induced 
HeEım DE Baısac (1954) and HEIM DE Barsac and BoURLIERE (1955) to reclassify the 
otter-shrews as a subfamily of the tenrecs. Later authors have supported this classification 
(GUTH et al. 1959, 1960; VERHEYEN 1961; RAHM 1961) and emphasized the intermediate 
position of the dwarf otter-shrews, and especially of M. lamottei, between Oryzorictinae 
and ?. velox. 

Acknowledgements 


The authors wish to thank Drs. THys VAN DEN AUDENAERDE, Tervuren, VOGEL, Lausanne, BAUCHOT 
and LAMOTTE, Paris, and STARCK, Frankfurt, for freely making available their otter-shrew material. 
The valuable help of CLÄRE ROBERG, MONIKA MARTIN (histology), HELGA GROBECKER (photogra- 
phy), HELMA LEHMANN (typing), HEIKO FRAHM (volume measurements), GERD REHKÄMPER (cranial 
capacity measurements), and MICHAEL STEPHAN (computer works) is gratefully acknowledged. 
Thanks are due to Joun NELson, Melbourne, for checking the English text. 


Zusammenfassung 


Das Gehirn von Micropotamogale lamottei Heim de Balsac, 1954 


Das Gehirn von M. lamottei unterscheidet sich vom durchschnittlichen Insektivoren-Gehirn vor allem 
durch geringer entwickelte olfaktorische Strukturen und eine größere Medulla oblongata. In der 
letzteren sind die Nuclei spinales nervi trigemini besonders stark vergrößert. Da sich ähnliche 
Merkmale bei allen wasserangepaßten Tnsektivoren, wie Limnogale, Potamogale, Neomys, Desmana 
und Galemys finden, wird ein Zusamı .ıhang mit der Nahrungssuche im Wasser angenommen. Das 
Trigeminus-System, welches die stark entwickelten Vibrissen der Oberlippe innerviert, scheint bei den 
wasserangepaßten Insektivoren das olfaktorische System als Hauptsinnessystem für die Nahrungssu- 
che zu ersetzen. 

Innerhalb der Otterspitzmäuse ist die Vergrößerung der Medulla oblongata und die gleichzeitige 
Reduktion der olfaktorischen Strukturen bei M. lamottei weniger stark ausgeprägt als bei ?. velox. 
Dies wird mit Unterschieden in der Wasseranpassung in Zusammenhang gebracht. 

In Merkmalen des Gehirns ähnelt M. lamottei am stärksten den spitzmausähnlichen Tenreks von 
Madagaskar (Oryzorictinae). 


References 


BAUCHOT, R.; STEPHAN, H. (1967): Encephales et moulages endocraniens de quelques insectivores et 
primates actuels. In: Colloques internationaux du Centre National de la Recherche Scientifique, 
163, Problemes actuels de Paleontologie (Evolution des Vertebres), Paris, 575-587. 

— - (1968): Etude des modifications encephaliques observees chez les insectivores adaptes & la 
recherche de nourriture en milieu aquatique. Mammalia 32, 228-275. 

— - (1970): Morphologie comparee de l’encephale des insectivores Tenrecidae. Mammalıa 34, 
514541. 

DEWITTE, G.-F.; FRECHKOP, $. (1955): Sur une espece encore inconnue de mammifere africain, 
Potamogale ruwenzorii, sp. n. Bull. Inst. roy. Sci. Nat. Belgique 31, Nr. 84, 1-11. 

Dusost, G. (1965): Quelques renseignements biologiques sur Potamogale velox. Biol. gabon. 1, 
257-272. 

FRECHKOP, $. (1957): A propos de nouvelles especes de Potamogalines. Mammalıa 21, 226-234. 

GUTH, CH.; HEIM DE Baısac, H.; LAMOTTE, M. (1959): Recherches sur la morphologie de 
Micropotamogale lamottei et l’evolution des Potamogalinae. I. - Ecologie, denture, anatomie 
cränıenne. Mammalıa 23, 423-447. 

— — — (1960): Recherches sur la morphologie de Micropotamogale lamottei et l’evolution des Potamo- 
galinae. II. - Rachis, visceres, position systematique. Mammalia 24, 190-217. 

HEIM DE Baısac, H. (1954): Un genre inedit et inattendu de Mammifere (Insectivore Tenrecidae) 
d’Afrique Occidentale. Comptes rendus des seances de I’Academie des Sciences 239, 102-104. 


142 H. Stephan and H.-J. Kuhn 


— (1956): Morphologie divergente des Potamogalinae (Mammiferes Insectivores) en milieu aquati- 
que. Comptes rendus des seances de l’Academie des Sciences 242, 2257-2258. 

— (1957): Evolution des vertebres caudales chez les Potamogalinae (Mammiferes Insectivores Tenre- 
cıdae). Comptes rendus des seances de l’Acad&mie des Sciences 245, 562-564. 

HEIM DE Baısac, H.; BOURLIERE, F. (1955): Ordre des Insectivores, Systematique. In: Grass£, P.-P. 
(ed.), Traite de Zoologie 17, 1653-1697. Paris: Masson. 

Kunn, H.-J. (1964): Zur Kenntnis von Micropotamogale lamottei Heim de Balsac, 1954. Z. Säugetier- 
kunde 29, 152-173. 

— (1971): An adult female Micropotamogale lamottei. J. Mammalogy 52, 477-478. 

RAHMm, U. (1960): Note sur les specimens actuellement connus de Micropotamogale (Mesopotamogale) 
ruwenzoril et leur repartition. Mammalia 24, 511-515. 

— (1961): Beobachtungen an der ersten in Gefangenschaft gehaltenen Mesopotamogale ruwenzorü 
(Mammalia-Insectivora). Rev. Suisse Zool. 68, 73-90. 

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

STEPHAN, H. (1967): Zur Entwicklungshöhe der Insektivoren nach Merkmalen des Gehirns und die 
Definition der ‚„‚Basalen Insektivoren“. Zool. Anz. 179, 177-199. 

— (1972): Evolution of primate brains: A comparative anatomical investigation. In: The Functional 
and Evolutionary Biology of Primates. Ed. by TUTTLE, R. 155-174. Chicago: Aldine/Atherton 
Inc. 

— (1975): Allocortex. In: Handbuch der mikroskopischen Anatomie des Menschen. Hrsg. BARG- 
MANN, W.4. Bd., 9. Teil, 998 S. Berlin: Springer. 

STEPHAN, H.; Anpy, ©. J. (1982): General brain characteristics and septal areas of the insectivores. In: 
Comparative Correlative Neuroanatomy of the Vertebrate Telencephalon. Ed. by Crosgy, E. C.; 
SCHNITZLEIN, H. N. 525-564. New York: MacMillan. 

STEPHAN, H.; BAUCHOT, R.; Anpy, O. J. (1970): Data on size of the brain and of various brain parts in 
insectivores and primates. In: The Primate Brain, Advances in Primatology. Ed. by NoBack, 
C. R.; MonTacna, W. 1, 289-297. New York: Appleton Century Crofts. 

STEPHAN, H.; FRAHM, H.; Baron, G. (1981): New and revised data on volumes of brain structures in 
insectivores and primates. Folıa primatol. 35, 1-29. 

STEPHAN, H.; SPATZ, H. (1962): Vergleichend-anatomische Untersuchungen an Insektivorengehirnen. 
IV. Gehirne afrikanischer Insektivoren. Versuch einer Zuordnung von Hirnbau und Lebensweise. 
Morph. Jb. 103, 108-174. 

VERHEYEN, W.N. (1961a): Recherches anatomiques sur Micropotamogale ruwenzorü. 1. Lamorpho- 
logie externe, les visceres et l’organe genital mäle. 2. La myologie de la tete, du cou et de la patte 
anterieure. Bull. Soc. Roy. Zool. Anvers 21. 

— (1961b): Recherches anatomiques sur Micropotamogale ruwenzorü. 3. La myologie de la patte 
posterieure, de la region sacrale et de la queue. 4. Observations osteologiques et considerations 
generales. Bull. Soc. Roy. Zool. Anvers 22. 

WEBER, M. (1928): Die Säugetiere. Einführung in die Anatomie und Systematik der recenten und 
fossilen Mammalıa. Bd. II, Systematischer Teil. Jena: Fischer. 


Authors’ addresses: Dr. HEınz STEPHAN, Max-Planck-Institut für Hirnforschung, Neurobiologische 
Abt., Deutschordenstr. 46, D-6000 Frankfurt/M.-71; Prof. Dr. Hans-JÜRG 
Kunn, Anatomisches Institut der Universität Göttingen, Kreuzbergring 36, D- 
3400 Göttingen 


Struktur und Variabilität der Verlassenheitslaute juveniler 
Vampirfledermäuse (Desmodus rotundus) 


Von U. SCHMIDT, G. JOERMANN und CHRISTEL SCHMIDT 


Zoologisches Institut der Universität Bonn 
Eingang des Ms. 30. 12. 1981 


Abstract 


Structure and variabılıty of the isolation calls of juvenile vampire bats (Desmodus rotundus) 


Studied were the isolation calls of 15 juvenile vampire bats (Desmodus rotundus murinus). These calls 
consist of distinct units (1-14) and are emitted in series. The frequency of the fundamental of the 
harmonic units ranges from 5 to 30 kHz. According to the frequency modulation pattern, most of the 
units belong to two types, but individual varıations are found. The calls of the different juveniles show 
varied characteristics (number of units/call; type of units used; graduation of frequency; frequency 
range); the typical pattern remains stable throughout ontogenetic development. The significance of 
individual distinctions of the isolation calls in Desmodus ıs discussed. 


Einleitung 


Das Zusammenfinden von Mutter und Jungtier wird bei allen bisher untersuchten Fleder- 
mausarten durch soziale Laute (Verlassenheitslaute, Isolationslaute, Stimmfühlungslaute) 
der Jungen erleichtert (z. B. GouLp 1975, 1979). Aus diesen Verlassenheitslauten entwik- 
keln sich bei einigen Spezies die Orientierungslaute (MÖHRESs 1953; KONSTANTINOV 1973, 
MATSUMURA 1979), manchmal bleiben sie als spezielle Soziallaute über längere Zeit 
erhalten (BRown 1976; HÄUSSLER et al. 1981). 

Bei der Gemeinen Vampirfledermaus, Desmodus rotundus, sind die Jungtiere besonders 
lange von ihren Müttern abhängig. Vom 4. Lebensmonat an begleiten sie ihre Mütter zwar 
auf der Futtersuche, können sich jedoch erst mit ca. 10 Monaten selbständig ernähren 
(SCHMIDT und MAnNSKE 1973; SCHMIDT et al. 1980). Bis zu diesem Zeitpunkt stoßen sie bei 
Trennung von der Mutter Verlassenheitslaute aus (SCHMIDT 1972), die sehr intensiv und 
auf weite Entfernungen hörbar sind. Es wird vermutet, daß die Mutter ıhr eigenes Junges 
anhand dieser Laute erkennt. Diese Untersuchung sollte klären, ob die Variabilität der 
Verlassenheitslaute verschiedener Jungtiere und die Konstanz bei einem Individuum groß 
genug sind, ein persönliches Erkennen zu ermöglichen. 


Material und Methode 


Für die Lautaufnahmen standen 15 Desmodus rotundus murinus verschiedenen Alters zur Verfügung. 
10 der Jungtiere wurden in Mexiko untersucht (2 Desmodus in einer Höhle, H1 + 2;8 am Abend des 
Fangtages ın der Station „Las Margaritas“, M1 — 8); 5 im Zoologischen Institut der Universität Bonn 
geborene Jungtiere wurden für genauere Analysen herangezogen. Das Alter der im Freiland gefange- 
nen Fledermäuse wurde anhand der Unterarmlänge und der Zahnentwicklung geschätzt (s. Tab.). 
Die Jungtiere wurden bei den Aufnahmen in kleinen Käfigen isoliert gehalten; ihre Mutter oder ein 
anderes adultes Weibchen befand sich ca. 3 m entfernt in einem zweiten Käfig. Bei jedem Jungtier 
wurden 20-30 Minuten lang alle sozialen Laute registriert, für die Auswertung jedoch nur die 
Verlassenheitslaute verwendet (der Erkennungslaut trat in dieser Versuchssituation nur selten auf). 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4703-0143 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 143-149 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


144 U. Schmidt, G. Joermann und Christel Schmidt 


Tabelle 


Maße und Alter der untersuchten Jungtiere 


Unterarmlänge 
(mm) 


46,5 
5152 
46,3 
45,0 
54,6 
32,8 
56,4 
54,6 
48,1 
495 


Bel (I) 20 / (II) 33 / (III) 110 
38,4 / 37,2 / 53,6 (D) 22/(M) 47 / (IM) 82 
26,3 / 27,7 / 29,8 A). 17 Say 
30,8 / 31,4 (IV) 22 / (V) 29 

monatl. (1.-9. Monat) 


ö 
6) 
2. 
6) 
% 
(6) 
(6) 
S 
2 
(6) 
2? 
6) 
6) 
? 


oO, 


Die Aufnahmen erfolgten mit einem Kondensatormikrofon (Sennheiser MKH 415) auf ein „‚Uher- 
Report“-Tonbandgerät (Laufgeschwindigkeit 19 cm/s) oder, zur Analyse der hohen Frequenzanteile, 
mit Ultraschallmikrofon (B&K 4135 mit Vorverstärker B&K 2618) und Hochfrequenzbandgerät 
(Lennartz 6000/800). Die Auswertung wurde an Speicheroszillograph (Tektronix 5113), Sonagraph 
(Kay-Electric 7029) und Frequenzspektrograph (Nicolet UA 500 A) vorgenommen. 


Ergebnisse 
Allgemeine Charakterisierung der Verlassenheitslaute 


Die Verlassenheitslaute juveniler Desmodaus sind aus distinkten Einzelelementen aufgebaut 
und werden in Serien abgegeben. Eine solche Serie besteht je nach Erregungszustand des 
Jungtieres aus 2 bis 30 Lauten mit Intervallen von 35 bis 115 ms. Die Laute weisen max. 4 
Harmonische auf; der Grundton liegt im Frequenzbereich zwischen 5 und 30 kHz und 
enthält stets die Hauptenergie des Lautes. Da die bei den Freilanduntersuchungen benutzte 
Registrierapparatur nur Frequenzen bis zu 30 kHz aufnahm, konnte das Spektrum der 
Obertöne nicht bei allen Tieren analysiert werden. Der hohe Frequenzanteil besitzt jedoch 
kaum eine funktionelle Bedeutung, da eine Kommunikation über größere Entfernungen 
auf relativ niedrige Frequenzen beschränkt ist; bei der Schallausbreitung werden hohe 
Frequenzen sehr stark gedämpft. 

Die Anzahl der Elemente pro Laut schwankt zwischen 1 und max. 14 (mehr als 10 
Elemente/Laut traten jedoch nur in weniger als 1% der Fälle auf); die Lautdauer ist 
entsprechend variabel (min. 12 ms bei 1 Element / Laut bis ca. 180 ms bei 10 Elementen/ 
Laut). Die 1-Element-Laute treten vorwiegend am Anfang einer Serie auf, das verwendete 
Element ist immer relativ lang. Die Mehrzahl der Elemente lassen sich 2 Grundtypen 
zuordnen: Im einfachsten Fall handelt es sich um einen abwärts modulierten ‚‚sweep“ 
(Abb. 1: 2.-5. Element), beim 2. Grundtyp folgt auf die Abwärts- eine Aufwärtsmodula- 
tion (z. B. Abb. 2: 1. Element des 2. Lautes). Dieser Aufwärts-sweep kann sich abrupt dem 
Abwärts-sweep anschließen, der Übergang kann aber auch allmählich erfolgen (Abb. 1: 
1. Element), oder es tritt eine Pause zwischen beiden Frequenzanteilen auf (Abb. 6b: 
3. Element). Außer diesen Grundtypen können individuelle Varianten gebildet werden 
(z. B. Abb. 2: 4. Element des 5. Lautes). Die Dauer eines Elementes variiert zwischen 3 ms 


Struktur und Variabilität der Verlassenheitslaute juveniler Vampirfledermäuse 145 


Abb. 1. Sonagramm eines Ver- 

lassenheitslautes mit 5 Elemen- 

ten (L3/I). Ordinate: Frequenz 
(kHz), Abszisse: Zeit (ms) 


100 200 300 


Abb. 2. Serie von Verlassenheits- 
lauten (H1). Ordinate: Frequenz 
(kHz), Abszisse: Zeit (ms) 


400 500 ms 600 


(steiler Abwärts-sweep) und 20 ms (Grundtyp 2 mit langsamer Modulation); die Intervalle 
zwischen den Elementen sind 7 bis 20 ms lang. Ein Laut kann aus verschiedenen Element- 
Typen aufgebaut sein. 


Individuelle Konstanz der Laute 


Eine wichtige Voraussetzung dafür, daß die Mutter ıhr eigenes Junges an den Verlassen- 
heitslauten erkennen kann, ist, daß die Laute jedes Jungtiers typische Merkmale aufweisen, 
die sich über längere Zeit nicht verändern. Bei den ım Labor geborenen Desmodus (L2-L5) 
konnte untersucht werden, inwieweit 
Veränderungen im Verlauf der ontogene- 
tischen Entwicklung auftreten. 

Bei allen 4 Tieren zeigte sich eine aus- 
geprägte Konstanz. Nicht nur, daß stets 
die gleichen Elemente zum Lautaufbau 


a 


verwandt wurden, sondern auch indivi- 304 
duelle Charakteristika blieben über grö- 20 \ 
ßere Zeiträume erhalten. Abb. 3 zeigt 10 s \V \ v \v \ 


z. B., daß bei L4 am 1. Lebenstag ein 7- 
Elemente-Laut genauso aufgebaut ist wie 

am 29. Tag. Selbst der Frequenzbereich 30 
hat sich während dieses Monats nur un- 20 

wesentlich verschoben. Der einzige Un- 10 Er äh 
terschied liegt darin, daß der am 1. Tag 
registrierte Laut um ca. 20 % länger ist. 
Auch die Anzahl der pro Laut auftreten- 4,5, 3. 7er-Laut von L4 am 1. Lebenstag (a), 13. 
den Elemente bleibt über die jeweilige Tag (b) und 29. Tag (c) 
Untersuchungszeit konstant. Das Jung- 

tier L4 besitzt meist 5 bis 8 Elemente/Laut (Abb. 4a), bei L2 sind fast ausschließlich 1- und 
2-Elemente-Laute zu registrieren (22. Tag: 88 %; 47. Tag: 90 %; 82. Tag: 98 %), und bei 
L5 dominieren die 2er- und 3er-Laute (Abb. 4b). 


100 200ms 


146 U. Schmidt, G. Joermann und Christel Schmidt 


50 a Said oO r2ııı 
% 5 D 8a R 
40 A 80 L3 11 
EN v 22d Sam 
EN N 
30 7 NER“ m 29d 70 Zi 
EN 
Fe N 
20 j 60 
10 So \ 
° 40 RN 
30 \ y 
50 o en 
20 EN 
40 s = Sy 5: Be, 
r N, 7 
30 5 N ®- Ve v 
U er (To an KDD Seen en 
20 
1) 2,, 30.2 05 „oT 
10 Elemente /Laut 
5 Abb. 4. Prozentuale Verteilung der Elemente pro 


Laut. Ordinate: Häufigkeit (%), Abszisse: An- 
ee a OR zahl der Elemente/Laut (je Tier n = 100). a) L4 

Elemente/Laut (1.-29. Tag), b) L5 (1.-7. Monat), c) Vergleich 
von 4 Tieren 


Unterschiede zwischen den Jungtieren 


Am auffälligsten unterscheiden sich die Verlassenheitslaute in der Anzahl der Elemente pro 
Laut (Abb. 4). Bei einigen Desmodus ist diese Anzahl sehr konstant: So verwendet z.B. L3 
fast ausschließlich Doppellaute, M5 in über 90 % 5er-Laute. Andere Individuen besitzen 
ein breites Spektrum: z.B. M1 bildet je ca. 15 % der Laute aus 3, 4, 5, 6 und 7 Elementen, 
M8 in etwa gleichem Verhältnis aus 5 bis 10 Elementen. 

Auch die Verwendung von Element-Typen weist individuelle Charakteristika auf. In 
den meisten Fällen sind beide Grundtypen vertreten, jedoch in sehr unterschiedlicher 
Ausprägung und Häufigkeit. Bei M5 konnten nur Abwärts-sweeps registriert werden, bei 
M6 war die Aufwärtskomponente meist nur angedeutet. Im Gegensatz dazu waren bei MI 
und L1 beide Typen etwa gleich häufig vertreten, während bei L2 und L4 Typ 2 weitaus 
dominierte. Die Modulationsrate und der Frequenzumfang der 2. Komponente weisen bei 
den einzelnen Tieren beträchtliche Unterschiede auf. Dazu kommen Individuen, die sehr 
spezifische Elementvariationen ausgebildet hatten (M2, M3), so daß jedes Jungtier anhand 
des Sonagrammes eindeutig zu identifizieren ist. In Abb. 5 sind charakteristische 4er-Laute 
von 6 Jungtieren gegenübergestellt. 

Ein weiteres Unterscheidungsmerkmal bietet die Kadenz der Elemente. Im einfachsten 
Fall liegen die Frequenzen aller Elemente eines Lautes im gleichen Bereich. Häufig ist 
jedoch eine typische Frequenzabstufung der einzelnen Elemente feststellbar (Abb. 6). 
Einige Tiere weisen unterschiedliche Laute auf; so sind z. B. bei M8 einfache Laute, die aus 
Typ-1-Elementen aufgebaut sind, gemischt mit Lauten, bei denen jeweils Typ-1- und Typ- 
2-Elemente abwechseln. Die Typ-1-Elemente fallen dabei auf 9-11 kHz, die Typ-2- 
Elemente nur auf 16-18 kHz. 

Auf den Sonagrammen läßt sich schon erkennen, dafß die Laute z.T. beträchtliche 
Frequenzunterschiede aufweisen. Für jedes Tier läßt sich ein Frequenzbereich festlegen, 
der z. T. gegen andere Individuen statistisch absicherbar ist. In Abb. 7 unterscheiden sich 
z.B. die untersten Frequenzen des 1. Elementes eines 7er-Lautes von M6 und M8 um ca. 7 
kHz. Der Frequenzabstand verändert sich zwar entsprechend der individuellen Kadenz ım 
Verlauf des Lautes, M6 bleibt jedoch stets niederfrequenter als M8. Der Frequenzbereich 
ist vom Alter unabhängig. Sowohl unter den sehr jungen, als auch unter den älteren Tieren 


Struktur und Variabilität der Verlassenheitslaute juveniler Vampirfledermäuse 147 


30 
kHz 


ea. NS  \ 


10 


ENEN Na 


ur 


50 100 50 100 ms 100 


Abb. 5 (links). Vergleich der 4er-Laute von 6 Jungtieren. a) H2, b) M8, c) M4, d) MI, e) M7, f) M2. - 
Abb. 6 (rechts). Vergleich langer Laute von M8 (a), M3 (b) und M4 (c). Die Laute unterscheiden sich ın 
den Element-Typen und ım Frequenzmuster 


fanden sich Individuen, bei denen die Elemente unter 10 kHz abfielen, zum anderen waren 
relativ hochfrequente Jugendlaute bei verschiedenen Altersstufen feststellbar. 

Nimmt man alle Unterscheidungsmerkmale zusammen —- Anzahl der Elemente pro 
Laut, Elementtypen, Kadenz, Frequenzbereich - so besitzt jedes Jungtier ein charakteristi- 
sches Lautmuster, das sich von dem aller untersuchten Desmodus unterscheidet. Die 
Unterschiede sind so groß, daß die Experimentatoren beim langsamen Abspielen der 
Tonbänder ohne Schwierigkeiten die Verlassenheitslaute den einzelnen Individuen zuord- 
nen konnten. 


Diskussion 


In vielen Publikationen werden zwar die Verlassenheitslaute bei den unterschiedlichsten 
Fledermausspezies erwähnt und z.T. auch analysiert, jedoch nur wenige Untersuchungen 
geben Hinweise auf die ontogenetische Entwicklung und die Individualität dieser Laute. 
Sowohl die Konstanz der Verlassenheitslaute über eine längere Zeit als auch spezifische 
Lautcharakteristika sind eine Voraussetzung dafür, daß eine Mutter ıhr Jungtier akustisch 


erkennen kann. 
kHz 


© 
tn 
Ze 
: 
\ D } 


Abb. 7. Untere Frequenzgrenzen der Elemente 
von 7er-Lauten bei M6 (®), M8 (m) und L4 (V). 
Mittelwerte und Standardabweichung aus je 20 
Lauten 1. 2. 3. 4. 5% 6. 7. 

Element 


200 ms 


148 U. Schmidt, G. Joermann und Christel Schmidt 


In dieser Hinsicht unterscheiden sich die einzelnen Chiropterenspezies sehr deutlich. 
Die einfachsten Verhältnisse finden sich bei Rhinolophus ferrumequinum (MATSUMURA 
1979) und Myotis oxygnathus (KONSTANTINOV 1973). Bei beiden Arten sind die Verlassen- 
heitslaute wenig strukturiert und wandeln sich im Verlauf von 4 resp. 7 Wochen in die 
Ortungslaute um. Die sukzessive Veränderung und die geringe Variationsmöglichkeit 
machen es unwahrscheinlich, daß diese Laute für ein individuelles Erkennen geeignet sind. 
Für das Auffinden des eigenen Jungtieres müssen olfaktorische Einflüsse maßgebend sein 
(KuLZErR 1962; Kos 1977). Bei Myotis myotis scheinen verschiedene Lauttypen vorzu- 
kommen (Kos 1981). Sehr viel variationsreichere Verlassenheitslaute besitzen Molossus 
molossus (HÄUSSLER et al., 1981) und Antrozous pallidus (BROowNn 1976). Sie lassen sich bei 
beiden Spezies ca. 2 Monate lang registrieren. Antrozons ist die einzige Fledermausart, bei 
der individuelle Unterschiede der Verlassenheitslaute nachgewiesen wurden. Diese Laute, 
die wie bei Desmodus ın Serien abgegeben werden, besitzen eine wellenförmige Frequenz- 
modulation, ein Parameter, der eine vielfältige Variation ermöglicht. Dieses Modulations- 
muster bleibt im Verlauf der Ontogenie erhalten, obwohl sich der Frequenzbereich 
langsam verschiebt, und auch die Laute kürzer werden. Noch vıel konstanter als bei 
Antrozous ıst das individuelle Lautmuster bei Desmodus, außerdem bleiben bei den 
Vampiren die Verlassenheitslaute bis zum 10. Lebensmonat erhalten. Zum anderen weisen 
diese Laute, bedingt durch ihren komplizierteren Aufbau, weit mehr Variationsmöglich- 
keiten auf, so daß jedes Jungtier eindeutig akustisch identifizierbar ist. 

Die Ausbildung individueller Laute scheint mit einer verlängerten Laktationsphase 
korreliert zu sein. Während beı Rhinolophus und Myotis (M. velhfer, BROwN 1976; Myotis 
lucıfugus, BARCLEY et al. 1979) die Umstellung von der Milch- auf die Insektennahrung 
sehr schnell erfolgt, wird Antrozous bis zum 55. Lebenstag gesäugt, obwohl die Jungtiere 
bereits mit 30 Tagen Insekten fressen. Ungewöhnlich langsam verläuft die Nahrungsum- 
stellung bei Desmodus (SCHMIDT et al. 1980). Schon im ersten Lebensmonat wird das 
Jungtier mit geringen Mengen Blut gefüttert; mit ca. 4 Monaten begleitet es die Mutter auf 
der Nahrungssuche, nimmt jedoch nur langsam zunehmende Quantitäten Blut auf. Erst 
mit ca. 10 Monaten ist Desmodus selbständig, bis zu diesem Alter müssen die Jungtiere 
gesäugt werden. Das gemeinsame Ausfliegen von Mutter und Jungtier erfordert ein 
effektives Kommunikationssystem, das über große Entfernungen wirksam ist. Auch 
müssen die Individuen eindeutig identifizierbar sein, da ın der gleichen Region oft viele 
Vampire leben. Die Verlassenheitslaute von Desmodus erfüllen alle geforderten Vorausset- 
zungen. 

Danksagung 


Herrn Prof. Dr. H. SCHNEIDER danken wir für die Überlassung der Analysegeräte, der Deutschen 
Forschungsgemeinschaft für finanzielle Unterstützung. 


Zusammenfassung 


Die Verlassenheitslaute von 15 juvenilen Vampirfledermäusen (Desmodus rotundus murinus) wurden 
analysıert. Diese Soziallaute sind aus unterschiedlich vielen Einzelelementen (max. 14) zusammenge- 
setzt und werden in Serien abgegeben. Die Elemente sind harmonisch aufgebaut mit einem intensiven 
Grundton im Bereich von 5-30 kHz; es lassen sich 2 Grundtypen unterscheiden, außerdem kommen 
individuelle Variationen vor. Die Laute der einzelnen Tiere weisen charakteristische Unterschiede auf 
(Anzahl der Elemente/Laut; Elementtypen; Frequenzabstufung der Einzelelemente; Frequenzbe- 
reich). Die für ein Jungtier typischen Merkmale bleiben im Verlauf der ontogenetischen Entwicklung 
erhalten. Die Bedeutung solcher individualspezifischen Verlassenheitslaute wird diskutiert. 


Literatur 


BarcLAY, R. M. R.; FEnton, M. B.; Thomas, D. W. (1979): Social behavior of the little brown bat, 
Myotis lucifugus. II. Vocal communication. Behav. Ecol. Sociobiol. 6, 137-146. 

Brown, P. (1976): Vocal communication in the pallid bat, Antrozous pallidus. Z. Tierpsychol. 41, 
34-54. 


Explorationsverhalten verschiedener Stämme und Generationen der Hausmaus 149 


Goup, E. (1975): Neonatal vocalızations in bats of eight genera. J. Mammalogy 56, 15-29. 

— (1979): Neonatal vocalizations of ten species of Malaysıan bats (Megachiroptera and Microchirop- 
tera). Amer. Zool. 19, 481-491. 

HÄUSSLER, U.; MÖLLER, E.; SCHMIDT, U. (1981): Zur Haltung und Jugendentwicklung von Molossus 
molossus (Chiroptera). Z. Säugetierkunde 46, 337-351. 

Koıs, A. (1977): Wie erkennen sich Mutter und Jungtier des Mausohrs, Myotıs myotis, bei der 
Rückkehr vom Jagdflug wieder? Z. Tierpsychol. 44, 423—431. 

— (1981): Entwicklung und Funktion der Ultraschallaute bei den Jungen von Myotis myotis und 
Wiedererkennung von Mutter und Jungen. Z. Säugetierkunde 46, 12-19. 

KOoNSTANTINOV, A. J. (1973): Development of echolocation in bats ın postnatal ontogenesis. Period. 
biol. 75, 13-19. 

KuLzER, E. (1962): Über die Jugendentwicklung der Angola-Bulldogfledermaus Tadarida (Mops) 
condylura (A. Smith, 1833) (Molossidae). Säugetierkundl. Mitt. 10, 116-124. 

MATSUMURA, S. (1979): Mother-infant communication in a horseshoe bat (Rhinolophus ferume- 
quinum nippon): Development of vocalizatıon. J. Mammalogy 60, 76-84. 

MöhRrss, F. P. (1953): Jugendentwicklung des Orientierungsverhaltens bei Fledermäusen. Die 
Naturwissenschaften 40, 298-299. 

SCHMIDT, C.; SCHMIDT, U.; MANSKE, U. (1980): Observations of the behavior of orphaned juveniles 
in the common vampire bat (Desmodus rotundus). Proc. 5th Int. Bat Res. Conf. Ed. by D. E. 
WıLson, A. L. GARDNER. Texas Tech Press, Lubbock. pp. 105-111. 

SCHMIDT, U. (1972): Die sozialen Laute juveniler Vampirfledermäuse (Desmodus rotundus) und ihrer 
Mütter. Bonn. zool. Beitr. 23, 310-316. 

— MAaNnsKE, U. (1973): Die Jugendentwicklung der Vampirfledermäuse (Desmodus rotundus). Z. 
Säugetierkunde 38, 14-33. 


Anschriften der Verfasser: Prof. Dr. UwE SCHMIDT, GERHARD JOERMANN, Dr. CHRISTEL SCHMIDT, 
Zoologisches Institut der Universität, Poppelsdorfer Schloß, D-5300 
Bonn 


Analyse des Explorationsverhaltens verschiedener Stämme und 
Generationen der Hausmaus Mus musculus domesticus Rutty 


I. Methodik, Datenzuordnung und Einflüsse des sozialen Umfeldes 


Von M. W. SCHÄFER 


Fachbereich Biologie (Zoologie) der Universität Frankfurt a. M. 
Eingang des Ms. 5. 11. 1981 


Abstract 


Analysıs of exploratory behaviour in different strains and generations of the house mouse (Mus 
musculus domesticus Rutty). I. Methods, grouping of data, and social effects 


Analized were different strategies of exploratory behaviour in white laboratory mice of the NMRI- 
strain and “wild” (grey) Mus musculus of the fifth to seventh generation born in laboratory under 
similar conditions. Individuals of these two strains were crossed to (grey) F,-hybrids, the distribution 
of the sexes in the parental generations combined white females with grey males as well as vice versa. 
Further crossings of F,-hybrids resulted in white (25 %) and grey (75 %) descendents of the F,- 
generation. 

To control possible socıal effects single young anımals (1 to 5 days old) of one strain were 
introduced to breeding parents of the other stock. 

As the experimental environment the white bottom of a large plastic box served as an ““open-field” 
with a plastic cup fixed ın the centre of the field as the only threedimensional structure. To start the 
observation an anımal was placed in a corner of the open-field and the responses like sitting, running, 
jumping, exploring or hiding from view inside the cup were recorded every second for 30 minutes. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4703-0149 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 149-162 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


150 M. W. Schäfer 


The great number of data were grouped to mean latency-values of the different behavioural 
parameters, to portions of the total observation time ın percentage, to portions of the time an anımal 
spent actıvely and to mean values of time measured in seconds; to prove the significance of differences 
non-parametric tests and analysıs of data clumping at zero were used. 

Different behavioural patterns of white and grey mice could be demonstrated including very small 
latencies of running and exploring in the NMRI-anımals (opposite to grey ones), a very distinctive 
change of sitting and quick running phases ın the grey mice and nearly no jumping-tendencies in the 
albinos while the grey anımals started to jump after about 15 minutes. 

Reproduction of data was demonstrated by observations of some individuals in a second 
experiment. Because no social influences effected the exploratory behaviour the differences between 
the two strains were interpreted as expressions of different genetic bases, investigations of hybrid- 
generations appeared to be useful. 


Einleitung 


Der Nachweis einer Verhaltensweise als „angeboren‘ kann nicht ausreichend durch den 
Ausschluß von Lernprozessen geführt werden, vielmehr ist die erbliche Fixierung als 
Ausprägung art- oder rassenspezifischer Charakteristika oder als Erbgang in Kreuzungs- 
experimenten zu demonstrieren. Letztlich mag in vielen Fällen ein Beharren auf der 
Alternative ‚„angeboren‘ oder „erworben“ der realen Komplexität in der Ausprägung von 
Verhaltensweisen nicht gerecht werden. Ein Modell interagierender Faktoren in unter- 
schiedlicher Wertigkeit ist statt dessen in Betracht zu ziehen. Bezüglich dieser Wertigkeit 
sind nach wie vor unterschiedliche Diskussionsakzente feststellbar (z. B. RoTH 1974; 
Money und EHRHARDT 1975; WIESER 1976; LORENZ 1978). 

Bereits in der frühen Menschheitsgeschichte wurden Tierarten durch Haltung und 
Zucht einer an menschlichen Interessen orientierten Auslese unterworfen. Dieser als 
Domestikation bezeichnete Prozeß bedeutete die Isolation einer Gruppe von Individuen 
von ihrer natürlichen Stammpopulation unter Veränderung ihrer Umwelt und Ernährung 
(RÖHRsS 1961; HERRE und RÖHrs 1973) mit der Folge konvergenter Veränderungen in 
morphologischen, physiologischen und ethologischen Bereichen (HArEz 1969). Neben 
erblich fixierten Domestikationsmerkmalen kommen durchaus modifikationsbedingte 
hinzu, der Vergleich von Verhaltensweisen zwischen Haustieren und ihren Stammformen, 
wie auch die immer noch bestehende Möglichkeit der Kreuzung zwischen beiden, eröffnen 
jedoch die Möglichkeit, erblich fixierte Verhaltensanteile experimentell zu isolieren. 

Die vorliegende Arbeit untersucht vergleichend das Explorationsverhalten „wilder“ 
(besser grauer) Hausmäuse und deren Albinoform (beide Mus musculus domesticus Rutty). 

Die Laborhaltung der Albinomaus ist für Europa seit etwa 100 Jahren belegbar (FREYE 
und FREYE 1960), kultische Verehrung ist jedoch schon um 1400 v. Chr. nachzuweisen 
(HAGEMANN 1959). Klassıfiıziert man die Albinomaus als „domestiziert‘“ (FRICK und 
NorD 1963), so ist zu bedenken, daß ıhr die „moderne“ graue Hausmaus nicht vorbehalts- 
los als Wildtyp gegenübergestellt werden kann. Die kommensale Lebensweise als Vorrats- 
schädling des Menschen mag sie vielmehr als ‚„spurenhaft andomestiziert‘“ erscheinen 
lassen, auch können Rassen der grauen Form in ihrem Verhalten differieren (PLOMIN und 
MAanoseVvITz 1974). 

Allgemein erscheinen die Albinos plumper im Körperbau, von größerer Fluchtlatenz 
bei reduziertem Sprungvermögen, geringer Beißintention beim Einfangen und mit im 
ganzen „gemächlicherer‘“ Fortbewegung (EıBL-EiBEsFELDT 1950, 1958). Vergleichende 
Beobachtungen der Verhaltensweisen von wilden und Labormäusen sowie Filialgeneratio- 
nen führten zum Schluß, daß wenige Gene dominant ‚„Wildheit“ bedingen (CoBurn 1975; 
Dawson 1932). Als weitere Charakteristika der Wildformen werden genannt: Schnelles 
Vermeidungslernen, höhere Laufaktivität im Wechsel mit „erstarrtem‘“ Sitzen, größere 
Aggressivität gegen Artgenossen, hohe Fluchtbereitschaft u. a. (BARNETT und SMART 1975; 
Connor 1975; SMITH 1972, 1978). 

Vergleichende Beobachtungen in einem in Sektoren aufgeteilten Versuchsfeld, dem 


Explorationsverhalten verschiedener Stämme und Generationen der Hausmaus 151 


„open-field“ (WAarsH und Cummins 1976) sind besonders geeignet, Erkundungs- und 
Fluchtverhalten in weitgehend deckungsloser Umgebung über einen bestimmten Zeitraum 
und nach Maßgabe gut abgrenzbarer Verhaltensmuster zu analysieren. So versucht die 
vorliegende Arbeit in Form eines erweiteren open-field-Tests das Verhalten grauer und 
weißer Mäuse quantitativ zu erfassen und Unterschiede auf genetische Grundlagen mittels 
Hybridgenerationen zu prüfen. Im ersten Teil der Studie wird die Methodik der Verhal- 
tensmessung sowie die statistische Auswertung der Daten vorgestellt, weiterhin wird 
versucht, Einflüsse des sozialen Umfeldes (Aufzuchterfahrung) abzuschätzen. 


Material und Methode 
Herkunft, Haltungsbedingungen und Maße der Versuchtstiere 


Als Vertreter der Parentalgeneration standen zur Verfügung: 1. Weiße Mäuse des NMRI-Stammes (w) 
aus der 9. bis 11. Generation eigener Institutszucht sowie 2. graue Mäuse (g) aus der 5. bis 7. 
Generation eigener Institutszucht, deren „Stammeltern‘ ın der Umgebung Frankfurts gefangen 
wurden. Beide Zuchten gingen aus Populationen von 20 bis 30 Individuen hervor. 

Reziproke Kreuzungsansätze lieferten die folgenden Filialgenerationen: 1. Weiße Weibchen (w) 
gepaart mit grauen Männchen (g), lieferten als F, (w) bezeichnete graue Hybriden, und 2. graue 
Weibchen (g) gepaart mit weißen Männchen (w) lieferten als F,(g) bezeichnete, gleichfalls graue 
Hybriden. 

Kreuzungen der F,-Generationen untereinander wurden nach folgenden Ansätzen durchgeführt: 
1. F,(w) x F,(w) lieferten die zweite Filialgeneration F,(w) und 2. F,(g) X F,(g) lieferten die zweite 
Filialgeneration F,(g). 

Die F,Tiere waren zu etwa einem Viertel weiß, bezeichnet als wF,(w) bzw. wF,(g) und zu drei 
Vierteln grau, bezeichnet als gF,(w) bzw. gF,(g). 

Die Versuchstiere aller Gruppen waren ın einem gemeinsamen Haltungsraum in „Makrolon- 
Standardkäfigen“ (40 x 25 X 15 cm) mit Futterraufen und Wasserflaschen untergebracht. Die Böden 
der Käfige waren mit einer ca. 5 cm hohen Torfschicht bedeckt, als Nestbaumaterial diente weißes 
Toilettenpapier. Wasser und Futter (‚„altromin 1324‘ — Standarddiät) standen ständig zur Verfügung. 
Jeweils nach der wöchentlichen Reinigung wurden zusätzlich Salat, Äpfel und Mohrrüben angeboten. 
Die Besatzdichte eines Käfigs lag durchschnittlich beı 15 Tieren, konnte nach gleichzeitigen Würfen 
mehrerer Weibchen allerdings kurzfristig ansteigen. Die Tiere der F,- und F,-Generationen lebten die 
ersten drei Wochen bis zur Entwöhnung bei ıhren Eltern und wurden dann in eigene Käfige verbracht, 
wobei gegebenenfalls weiße und graue Tiere gemeinsam gehalten wurden. 

Als Kontrollexperimente wurden jeweils weiße (bzw. graue) Jungtiere im Alter von 1 bis 5 Tagen 
„transplantiert“‘, also einer grauen (bzw. weißen) Mutter unterschoben, so daß einzelne weiße Tiere 
unter grauen aufwuchsen und umgekehrt und so reziproke Aufzuchterfahrungen resultierten. Trans- 
plantierte weiße und graue Tiere werden im folgenden mit w’ bzw. g’ bezeichnet. 


Tabelle 1 


Verteilung der Versuchstiere 


Die Zahl zweifach getesteter Individuen ist jeweils in Klammern angegeben; insgesamt 233 Tiere 


Herkunft, Generation 


W 
w' 


’ 


5 
8 


F, (w) 
F, (8) 
wF, (w) 
gF, (w) 
wF; (8) 
gF, (8) 


152 M. W. Schäfer 


Tabelle 2 
Körpermaße der Gruppen w (n=10), g (10), F, (23) und F, (38) 


Angegeben sind Mittelwerte und Standardabweichungen folgender Größen: Schwanzlänge, von der 
Schwanzwurzel bis zur Spitze gemessen, Kopf-Rumpf-Länge von der Nasenspitze bis zur Schwanz- 
wurzel, Ohrlänge als Abstand zwischen proximalstem und distalstem Punkt der Ohrmuschel, 
Hinterfußlänge von der Ferse bis zur Spitze der längsten Zehe (ohne Kralle) und Bruttokörpergewicht 


©) 
7 


Hinterfuß Gewicht 
(8) 


Kopf-Rumpf 
(mm) 


E 
3 
E 
s 


9859.22 359 
89,7 2 4,4 
98598211359, 
IS 22032 


& 


oOOO 
NO UN 


4553 
28,5 
Sy, ae 
939 


“ 


NEN 
DN 
ON 


“w w 


ker ae le 


at 
ae 
ae 
Se 


o 


Insgesamt wurden 233 Tiere der Verhaltensbeobachtung im ‚„open-field‘ unterzogen, davon 32 
einem zweiten Test (im Abstand von mindestens 10 Tagen) zur Überprüfung der Reproduzierbarkeit 
ihrer Daten. Die genaue Verteilung der Versuchstiere nach Abstammung und Geschlecht gibt Tab. 1 
wieder. 

Zur näheren Charakterisierung der einzelnen Versuchsgruppen bzw. Generationen wurde die 
Bestimmung der Körpermaßse (entspr. NIETHAMMER und KrApp 1978) von 81 Tieren herangezogen 
(Rab. 2)! 

Die Größen Schwanzlänge, Kopf-Rumpf-Länge und Gewicht erreichten bei den grauen Tieren (g) 
höhere Werte als bei freilebenden (STRESEMANN 1967; NIETHAMMER und Krapp 1978), lagen jedoch 
deutlich unter den Maßen der weifßen Mäuse. Die Daten der F, und F, erwiesen sich als nahezu 
identisch, was bemerkenswerterweise auch für die verschiedenen Untergruppen der F,-Generationen 
[wF,(w), gF,(w), wF,(g) und gF,(g)] galt. Kopf-Rumpf-Länge und Schwanzlänge der len 
nen ordneten sıch hierbei der Größenordnung der weißen Parentalindividuen zu, während das 
Gewicht ‚„intermediäre‘“ Werte zwischen w und g, allerdings noch im weiteren Streubereich der 
weißen Tiere, annahm. 

Als Mindestalter für einen Versuch wurden 2 Monate festgesetzt, eine obere Altersbegrenzung gab 
es nicht, ein Versuchstier durfte lediglich keine äußeren Verletzungen aufweisen. Die Entnahme eines 
Tieres zum Versuch erfolgte jeweils dann, wenn in dessen Käfig alle Individuen aktıv angetroffen 
wurden. Es wurde dann mittels eines kleinen, mit Torf versehenen Käfigs zum Versuchsraum 
transportiert. Zwischen der Entnahme aus dem Heimkäfig und dem Einsetzen in die Versuchtsappara- 
tur lag gewöhlich eine Zeitspanne von 3 Minuten. 


Apparatur, Verhaltensweisen und Protokollierung 


Als Versuchsfeld diente der glatte Boden eines weißen, oben offenen Kunststoff-Containers von 40 cm 
Wandhöhe („BRAUCKE-Polykunststoffkasten‘“), dessen Bodenfläche mittels 5 mm breiten schwar- 
zen Klebebandstreifen in sechs Quadrate unterteilt wurde, denen die Zahlen 1 bis 6 zur Protokollie- 
rung zugeordnet wurden. Auf der Feldmitte wurde ein weißer Plastikbecher so am Boden liegend 
befestigt, daß lediglich der aufliegende Rand der Becheröffnung mit durchsichtigem Klebestreifen 
fixiert war, der Becher also durch mögliche Aktivitäten eines Versuchstieres in begrenztem Maße aus 
seiner Lage gebracht werden konnte. Dieser Becher stellte das einzige dreidimensional strukturierte 
Element des open-fields dar. Die genauen Maße des Versuchsfeldes gibt Abb. 1A wieder. Die 
Beleuchtung im sonst völlig verdunkelten Versuchsraum lieferte eine 1,70 m über dem Versuchsfeld 
zentrierte 40-Watt-Glühbirne (Beleuchtungsstärke von der Feldmitte nach oben 21 Lux, von der 
Feldmitte senkrecht auf die Wandflächen konstant 9 Lux). Vor jedem Versuch wurde die Bodenfläche 
mit 15%iger Lysoform-Lösung ausgewischt. Sobald der Kasten getrocknet war, wurde ein Versuchs- 
tier am Schwanz erfaßt und in die Ecke des Feldes 1 eingesetzt. Mit diesem Versuchsbeginn wurde eine 
Stoppuhr ausgelöst und in 1-Sekunden-Abstand der Aufenthaltsort sowie die gerade auftretende 
Verhaltensweise auf einem mit einer „Karo-Matrix“ (1 Kästchen = 1 s) versehenen Protokollvordruck 
registriert. Die Verhaltensbeobachtung eines Individuums endete nach 30 Minuten. Einen Ausschnitt 
aus einem typischen Protokollvordruck findet sich in Abb. 1B. 

Nach beendetem Versuch wurde die Maus am Schwanz erfaßt und individuell markiert. Hierzu 
wurden den weißen Tieren Punktmuster aus Pikrinsäure mit einem Pinsel aufgetragen, grauen Mäusen 
wurden mit einer Schere kleine Fellpartien geschoren, da die graubraune Fellfärbung die Erkennbar- 
keit von Farbmustern (aus unschädlichen Farben) unmöglich machte. Besondere Reaktionen auf Fang 
und Markierung wurden gesondert vermerkt. Hierbei ergaben sich bereits recht deutliche Unter- 


Explorationsverhalten verschiedener Stämme und Generationen der Hausmaus 153 


schiede. So verhielten sich die Albinos überwiegend unbeteiligt und zeigten geringe Erregung, 
während bei den grauen Tieren Fluchtversuche, Lautäußerungen (Quietschen) oder auch Beifßen 
auftraten. Die meisten Individuen der Filialgenerationen verharrten hingegen bewegungslos. Rezi- 
proke Anwendung von Schere und Pinsel stellten sicher, daß nicht die unterschiedlichen Behandlungs- 
methoden die Ursache dieser differierenden Reaktionen waren. 

Die folgenden Verhaltensweisen konnten während der 30minütigen Beobachtungszeit registriert 
werden: 

1. Schnuppern (Sch): Witterungsaufnahme mit oder ohne Ortsveränderung, an Bewegung von 
Nasenspitze und „Schnurrhaaren“ erkennbar. 

2. Aufrichten (A): Erheben der Vorderbeine, auf freier Fläche oder an der Kastenwand, dann auch 
oft Abheben eines Hinterbeines vom Boden, mit Schnuppern kombiniert auftretend. 

3. Putzen (P): Jede Art körperbezogenen Pflege-(Komfort-)Verhaltens als Ordnen des Fells mit den 
Vorderextremitäten, Belecken von Körperteilen, Wischen mit den Vorderextremitäten über Kopf 
und Gesicht, Kratzen und Reinigen des äußeren Gehörganges mit einem Hinterfuß. 

4. Nagen (N): Versuche, Wand, Boden oder Klebebandstreifen mit den Schneidezähnen zu bear- 
beiten. 

5. Schwanzrasseln (R): Meist seitlich geführte, vibrierende Schwanzausschläge. 

6. Aktionen am Becher (a. B.): Beschnuppern oder Benagen des Bechers, Berührungen mit den 
Extremitäten, Aufrichten oder Besteigen, Lageveränderungen. 

7. Aufenthalt im Becher (z. B.): Mindestens 50 % des Mäusekörpers befinden sich im Becher. 

8. Sitzen (Sit): Bewegungsloses Verharren, entspannt oder starr. 

9. Springen (Sp): Alle vier Beine ohne Bodenkontakt, häufig an einer Wand oder in einer Ecke 
ausgeführt. 

10. Bewegung (Bew): Lokomotion ohne eine der unter 1) bis 6) genannten Verhaltensweisen; 
Fortbewegung auf der Lauffläche, als Wechsel in ein anderes Quadrat registriert, wenn mehr als 
50 % des Körpers die Grenzmarkierung zweier Quadranten passiert hatten. 

11. Defäkieren (Kot): Absetzen (besser Verlieren) von Kotballen. 

12. Urinieren (Urin): Urinausscheidung als sichtbare Lache. 

Qualitative Vermerke und Beschreibungen rundeten die Protokollierung ab. So wurden vermerkt: 

Geschwindigkeit und „flüssiger“ Ablauf der Fortbewegung, eine besondere Erscheinungsform des 

Sitzens als ‚„erstarrtes Sitzen“, ein Verharren in beliebiger Stellung bei weit geöffneten Augen, 

aufgerichteten Ohren und leicht gesträubtem Fell sowie eine besondere Art der Fortbewegung, im 

folgenden als ‚„‚klebendes Laufen“ bezeichnet. Letzteres erschien kombiniert mit Schnuppern als eine 

Fortbewegungsform, bei der zunächst die Hinterbeine ihren Standort nicht verließen, während es 


108 


ale ala a ra A| a gr 
ae ee 27, 


A 
A 


60 


20 A B 


Abb. 1. A: Grundriß des Versuchsfeldes mit Quadranteneinteilung und der Position des liegenden 
Bechers (B); Maße ın cm, das Kreuz im Feld 1 kennzeichnet den Ort, wo ein Versuchstier bei 
Beobachtungsbeginn eingesetzt wurde. B: Ausschnitt aus dem verwendeten Protokollvordruck. 1 
Karo entspricht 1 s, ın einer vollständigen horizontalen Reihe sind 30 Eintragungen möglich. Die 
eingetragenen Zahlen 1 bis 6 kennzeichnen den Aufenthaltsort des Versuchstieres, die Abkürzungen 
einzelne Verhaltensweisen wie Schnuppern (Sch), Aufrichten (A) und Aufrichten an einer Wand (/A), 
Sitzen (Sıt), Putzen (P) und Nagen (N); das eckige, unten offene Symbol zeigt Aktivitäten am Becher 
an, ein senkrechter Strich im Becher bedeutet Aufenthalt in diesem 


154 M. W. Schäfer 


unter Vortreten der Vorderbeine zu einer maximalen Körperstreckung kam, dann schoben sich die 
Hinterbeine eine kurze Strecke ın Fortbewegungsrichtung vor, um erneut zu verharren, worauf 
wieder die eingangs beschriebene Streckung folgte usw. So entstand gewissermaßen der Eindruck die 
Maus klebe unterseits am Boden und vermöge sich nur mit Mühe fortzubewegen. 


Ermittlung und statistische Auswertung der Daten 


Die als standardisierter Protokollvordruck benutzte „Karo-Matrix“ lieferte eine Abfolge registrierter 

Verhaltensmuster im 1s-Abstand vom Einsetzen in die Apparatur bis zum Versuchsende nach 30 

Minuten. Als einfache Addition ergaben sich folgende Werte: 

1. Summe von Schnuppern und Aufrichten (Sch + A), 

2. Gesamtanteil der Aktivität (Akt) durch Subtraktion der Zeiten des Aufenthaltes im Becher (:. B.) 
und des Sitzens (Sit) als 30 min - ı. B. - Sıt. 

3. Gesamtanteil reiner Lokomotion (Bew) als Anteil der Aktivität minus Schnuppern (Sch), Aufrich- 

ten (A), Putzen (P), Nagen (N), Schwanzrasseln (R) und Aktionen am Becher (a. B.). 

Die Werte 1 bis 3 werden als prozentuale Anteile der Versuchszeit ausgedrückt. 

Weiterhin ist es möglich, die Elemente Sch, A, Sch+ A, P, N, R, a. B. und Bew. als Anteil der 
Gesamtaktivität x/Akt darzustellen. 

Schließlich wurden ermittelt die Anzahl von Sprüngen (Sp), Kot und Urin sowie die Zahl 
überlaufener Quadranten als Felder pro Minute (F/min), als Felder pro Minute Aktivität (F/akt. min) 
und als Felder pro Minute Lokomotion (F/bew. min). 

Die mittlere Dauer einer Verhaltensweise x ergibt sich aus ihrer Gesamtdauer t und ihrer 
Häufigkeit r» während des Beobachtungszeitraumes ım 1s-Raster, im folgenden für die einzelnen 


Verhaltensweise angegeben als t/x. Z. B. ist t/Sch = "sh , YP =? usw. 


—— ) 


Asch np 
Die mittlere Dauer der aktiven Verhaltenselemente ergibt sich als 


Re N een Ar 
Akt er Sch A P N R a.B. \ 
Nsch + n4 + np + nn + nR + Nn,.B. 


die mittlere Dauer sämtlicher Verhaltensweisen als 


30 min 


t/Ges = —  ——— . 
Nike F NiB. IF ng; 

Weiterhin wurden für sämtliche Verhaltensweisen die mittleren Latenzzeiten als Zeitintervall 
zwischen Einsetzen ın die Apparatur und erstem Auftreten ermittelt. 

Die statistische Auswertung erfolgte mit nicht-parametrischen Testverfahren. So konnte die bei 
der Datenfülle aufwendige Prüfung auf Normalverteilungen vermieden und den zum Teil beträchtli- 
chen Streuungen Rechnung getragen werden. Überdies ergaben bereits Stichproben höchst unter- 
schiedliche Datenverteilungen (Abb. 2 und 3). Generell erfolgten die Tests mit zweiseitiger Fragestel- 
lung auf einem Signifikanzniveau von p = 0,01. 

Im einzelnen kamen folgende Testverfahren zur Anwendung: 

1. Vorzeichentest nach Dixon und Moop für verbundene Stichproben (in SacHs 1978): Vergleich 
der Befunde an zweifach geprüften Tieren. 2. Vergleich zweier unabhängiger Stichproben nach dem 
U-Test von WıLcoxon, MANN und WHITNEY (in MÜHLENBERG 1976 und Sachs 1978): Vergleich der 
Gruppen w-w’, g-g’, F,(w) - F,(g), F-(w) - Fı(g) und wF, - gF). 3. Vergleich mehrerer Süchproben 
nach dem A-Test von Kruskar und WarLıis (in LiEnerT 1973 und Sachs 1978) für 7>YX° 1,9): 
Vergleich der F,-Untergruppen wF,(w) - gF,(w) - wF,(g) - gFı(g). 4. Reduktion der Rangzuteilungen 
bei Stichproben > 100 durch das Verfahren nach RAaATz (1966), eine Modifikation des WHITE-Tests 
(1952), die auf den oben genannten A-Test anwendbar ist: Vergleich der Gruppen w-g-F,-F}. 5. 
Paarweise Prüfung mittlerer Rangsummen nach NEMmEnYI (in SacHs 1978) bei signifikanten - 
Werten. 6. Verfahren nach LACHENBRUCH (1976) ım Falle einer Wertehäufung um Null als Proportio- 
nenvergleich der Null-Ergebnisse beider Meßreihen (z,) mit den Nicht-Null-Ereignissen beider 
Reihen (z,). Im Falle = 25 + 25 >xX%,,) liegt Signifikanz vor. 


Ergebnisse 


Häufigkeitsverteilungen von Verhaltensparametern in den Parental- 
und Filialgenerationen 


Ordnet man Häufigkeiten (Tieranzahlen) den „‚Klassenbreiten‘“ einer Verhaltensweise, 
gemessen als prozentualer Anteil an der Beobachtungsdauer, zu, so ergibt sich jeweils für 
alle Generationen ein ähnliches Verteilungsbild bei durchaus unterschiedlichen Mittelwer- 


OL 


Ge 


JOA ıySıu uadaı] uS3ungra219AJeuI 
-ION 'U9I9L], uaneI3 19q uassepy uayoy 
nz 3unna1NS 9Yıeıs Jur pun USUOTEIIU 
-9feiftg uop ur 9119, apuayerssne ıyauı 
uassepyy U919J013 nz y2opal “aysıfuye 
“osneM Alam ıny usssepy] usdLıparu usp 
ur usdunjnep] 9wanxa usdIaz uadungıaı 
-IIA IQ '% S JoTy arrsıquassepy ‘7 'gqgV 
-ıdsıua 3undengny %7 pun !7 “3 ‘m uad 
-dn19 usp ur „u9zy1g“ ıny usaysıned 


19p Aapfıgqsdungsuss‘ (519294) € "94V 


1391] UST], UasTaM pun uaneıd 
uayasımz Noyzıpydıaıamz 19191n2pa3ur 
au usdadury 7 aıp “uayssıdsıus u1ayg 
uaneı13 uap Zrspewsdungts219A 77 A9Pp 9491] 
aIp pusiyem ‘usuoneIsusgferusIe] UP 
usyssımz usdunddenisqn us3ıanyaurıad 
ınu Ju USdUnJI91I19AJeULION Spulsyeuue 
usdıaz 19, IQ "usay2Traı1ıa Nozsdung 
-y9egoag I9p ur 9%, (G‘T + X) SIQ %, X UOA 
„[aruessddnuyss“ usywwmsaq usurs a1p 
“Jı9L], [yezuy 19p ajnes aurs Iyarıdsıua 
0S '% ST NZ uSNSIquUaSSeIN] ur Jonep 
-sysnsI9AIuwIesaH) Jap ue „ulsddnuyag“ 
IsTOMsusıfeyls A I9p [Iaauy usjenıuszo1d 
UIP 9ssızsqyy aIp “I9PaIM USNPIAIPUJ UOA 
(/ zuanbaı,] ‘[yezuy) 1oy3ıynerg aıp ıqı3 
aeumpıg aıq “y pun /J 3 m usddnıy 
usp ur „unsddnuysg“ ıny usıay3ıyneg 


19P zoppıgsdungana (sau) z "gqV 


% 


156 M. W. Schäfer 


ten. Für einzelne Verhaltensweisen konnten neben annähernden Normalverteilungen auch 
einseitige Datenhäufungen in z. B. niedrigen Klassen festgestellt werden (Abb. 2 und 3). 


Latenzzeiten 


Pro Versuchsgruppe und Generation wurden die mittleren Latenzzeiten der einzelnen 
Verhaltensparameter aufgetragen und verglichen. Allerdings erfolgte dabei keine Aussage 
darüber, ob alle Individuen eine bestimmte Verhaltensweise überhaupt zeigten oder ob der 
angegebene ‚Mittelwert‘ nur auf das Verhalten eines Individuums zurückzuführen ist. So 
gilt etwa für „Springen“, daß die überwiegende Zahl weißer Mäuse der Parentalgeneration 
dieses Verhalten während der 30minütigen Versuchszeit nicht zeigte. Trat dieses Verhalten 
jedoch einmal auf, so erst kurz vor Versuchsende. Die Abb. 4 und 5 geben für insgesamt 
zwölf Verhaltensmuster die Latenzzeiten aller Versuchsgruppen wieder, wobei nach 
Verhaltensweisen innerhalb der ersten zwei bzw. fünf Minuten und solchen, die in der 
Regel erst nach zehn Minuten oder mehr auftraten, untergliedert wurde. Die Auftragungen 
lassen sowohl Aussagen über die Reihenfolge des Auftretens von Verhaltensweisen wie 
auch über Unterschiede zwischen Gruppen und Generationen zu. So kennzeichnet die 
Gruppen w und w’ sofortiges Schnuppern in der unbekannten Umgebung, dem Lokomo- 
tion folgt (Bew). Für die grauen Tiere (g und g’) ıst diese Abfolge genau umgekehrt und 
entspricht damit jener der F,, während bei beiden Parentalgruppen Aufrichten die dritte 
Stelle der Abfolge einnimmt. Während dann graue Tiere früher sitzen (Beginn der 2. Min.), 
beginnen die weißen bereits mit Putzen. Sitzen zeigen weifse Mäuse zwei Minuten später, 
Koten und Urinieren setzt hingegen früher ein. Der erste Aufenthalt im Becher erfolgt bei 
weißen Tieren erst nach dem doppelten Zeitintervall der grauen. Während letztere nach der 
Hälfte der Beobachtungszeit zum erstenmal springen, zeigte nur ein weißes Tier dieses 
Verhalten sehr viel später, dıe übrigen nie. 

Die Tiere der F, zeigen beı pauschaler Betrachtung mehr dem grauen Elternteil ähnliche 
Werte. In der F, sınd eine Reihe ‚‚intermediärer‘‘ Daten zwischen beiden Elternteilen zu 
beobachten, wobei die Aufgliederung in weiße und wildfarbene Individuen einige Entspre- 
chungen zur gleichgefärbten Parentalgeneration erkennen läßt. Für die transplantierten 
Individuen (w’, g’) sind neben unveränderten Größenordnungen (gegenüber w und g) 
gleichsinnige Veränderungen feststellbar: so treten fast alle Verhaltensweisen früher auf als 
bei jenen Tieren, die in genetisch gleicher Sozietät heranwuchsen. Die Verhältnisse von w’ 
zu g’ entsprechen jedoch weitgehend jenen von w zu g (Abb. 4, 5). 


Reproduzierbarkeit der Daten und Geschlechtsunterschiede 


Wie bereits in Tab. 1 angegeben, wurden zur Überprüfung der Reproduzierbarkeit der 
erhobenen Daten jeweils acht Individuen der Gruppen w, g, F, und F, einem zweiten Test 
im Abstand von mindestens zehn Tagen unterzogen. Die Resultate wurden mit dem 
Vorzeichentest auf Unterschiede geprüft. Bei 43 so verglichenen Wertepaaren ergab sich ın 
41 Fällen kein signifikanter Unterschied zwischen erstem und zweitem Test. Lediglich eine 
Verhaltensweise zeigte bezüglich ihrer Latenzzeiten wie auch der Änderung ihrer Anzahl 
mit p<0,01 zu sichernde Unterschiede: das Springen. Die Tiere sprangen während des 
zweiten Versuchs später und weniger. Da sich für die Daten der übrigen Verhaltenspara- 
meter zu über 95 % Reproduzierbarkeit ergab, wird die experimentelle Erhebung als 
zuverlässig angesehen. 

Eine Gruppierung der Daten nach Geschlechtern erwies sich als unnötig, da die 
Mittelwerte männlicher und weiblicher Tiere meist nur geringfügig voneinander abwichen 
und die wenigen Unterschiede niemals das Maß der Differenzen genetisch abgrenzbarer 
Versuchsgruppen erreichten. 

Da „Schwanzrasseln“ nur bei wenigen Individuen auftrat und weniger als 0,2 % der 


umge 0C OL 0€ 0c Ol 


21°) 


[6) 
104 Denn 
M 


Eos 07 Ol DE 4 Ol 


usınurm 0E UoA ume1Iazsdungysego 


-9g usıwesad up A9qn usdundenny 


‘„usaarunn“ pun „unoy“ ‘ usdurdg“ 
‘„Joyag wı ıfeyyuspny“ ‘,,ujassei 
-Zuemyag“ °,usden“ USSIOMSUHITeyU1II A 
J9P U9NIZZUIIET HAI (519294) S 'ggV 


ue susıpey1sA sap uurdag Usp ıqı13 

SfoquÄsfra}g I9p uey] sıyssıyuas “oyur 
aIp “usyenyne yoZsuniysegoag I9p us 
nur Fun} 'AZq I9MZ U91S19 J9p pusiyem 
aIp “uaswmasusreyts\ pums usdenasiny 
"„uszırg“ pun ‘,19y99g we USIENANNY“ 
‘„uszıng“ ‘, usaysuıyny“ “ ursddnuysg“ 
‘ 3undanag“ usstaMmsusgfeyIsA Ip Any 
usuonelausn, pun usddnıy Jouspary>s 


-I9A USNIZZUSIET ION '(SYu) p 'ggV 


uw G 


Mag 


158 M. W. Schäfer 


Beobachtungszeit einnahm, wird diese Verhaltensweise im folgenden vernachlässigt; das 
gleiche gilt für „Nagen‘ mit einem Anteil unter 1 % der Versuchszeit bei allen Gruppen. 


Aufzuchterfahrung und soziales Umfeld (Transplantationen) 


Weiße wie graue Tiere sind als genetisch abgegrenzte Gruppen definiert und zeigen 
gruppenspezifische Verhaltensunterschiede ihrer Explorationsstrategien, überdies sind die 
Gruppen jedoch auch als sozial abgegrenzt zu bezeichnen. Diese Abgrenzung bedeutet 
z.B. für ein neugeborenes weißes Individuum zwingend die eingeschränkte Aufzuchter- 
fahrung durch seine weiße Mutter, durch ihm gleichende Geschwister und adulte Indivi- 
duen. Der Anteil sozietätkonform „geprägter‘‘ Verhaltensanteile in der Explorationsten- 
denz kann also nur durch Prüfung reziprok ‚„untergeschobener“ (transplantierter) Jung- 
tiere abgeschätzt werden, wobei eine solche Prozedur gewöhnlich problemlos verlief. D. h. 
die kurz nach der Geburt transplantierten Individuen wuchsen als sozial voll integrierte 
Mitglieder der genetisch anderen Sozietät heran, ohne daß von irgendeiner Seite soziale 


Isolation beobachtet war. 
Die oben geschilderten Möglichkeiten zur Auswertung und Gruppierung der Daten 


100] %, 10019, 
60 60 
20 20 
green ngr ET = 
= > 


Abb. 6. Vergleich der Gruppen w und w’ (jeweils rechte, stärker umrandete Säule) bezüglich der 

Dauer verschiedener Verhaltensweisen, aufgetragen als mittlerer Prozentwert mit Standardabwei- 

chung der Gesamtbeobachtungszeit (linkes Diagramm) und bezüglich der Dauer als mittlerer prozen- 

tualer Anteil der Aktivitätszeit (rechtes Diagramm). Ein signifikanter Unterschied zwischen beiden 

Gruppen besteht nur für den Parameter Sch/Akt, also dem „‚Schnupperanteil“ an der aktiven Zeit, der 
bei den transplantierten Individuen höher liegt 


10) fs] 50] (s] 10]n] sol[n] 
6 ; 30 6 30 
2 10 2 10 

Sea nazıa a0 

Fa. 22 2230 5 

5; 0 2% ® oz Dre ® 

o > 3 z 

=. 3 

3 =o 

3 


Abb. 7. Vergleich der Gruppen w und w’ (jeweils rechte, stärker umrandete Säule) bezüglich der 

mittleren Dauer verschiedener Verhaltensweisen in s (linkes Diagramm) sowie der mittleren Anzahl 

der Ereignisse „Kot“, „Urin“, „überlaufene Felder‘ pro verschiedener Zeitanteile und „Springen“ 

(rechtes Diagramm); signifikante Unterschiede, die transplantationsbedingte Veränderungen anzeigen 

könnten, sind nicht nachweisbar. Zur näheren Erläuterung einiger Veränderungstendenzen s. Text 
und Tab. 3 


Tabelle 3 


Vergleich der Gruppen g und g’ bezüglich der Dauer einzelner Verhaltensweisen als Anteil der Gesamtzeit sowie der Aktivitätszeit und bezüglich 
der mittleren Dauer der Verhaltensweisen sowie der Anzahl der Ereignisse „Kot“, „Urin“, „überlaufene Felder“ und „Springen“ 


Explorationsverhalten verschiedener Stämme und Generationen der Hausmaus 159 


verringerten Werte der überlaufenen Felder für alle 


auffällig sind die jeweils bei g’ 


Unterschiede zwischen den Gruppen sind in keinem Fall sicherbar, 
drei zeitlichen Bezugsgrößen, eine Tendenz, die auch zwischen den 


” beobachtbar ist (s. Abb. 7). Abkürzungen der 


und 


weißen Tieren der Gruppen w 


Verhaltensparameter entspr. Kapitel Material und Methode 


15 =E N0,E) 
16,1 #& 6,4 


5142.-2223,0 
48,9.2202959 


13,4 + 18,5 
7,4 & 13,4 
Bew/Akt 


Voll 


19,3 4,4 
Sch + A/Akt 


5,0 


Dh ze 
10) ae 252 


a. B./Akt 


P/Akt 


A/Akt 


6,2 


10,4 
169222 1458 


97,92=41252 
38,3 220,9 


Em 
os 
No) 
+ 
en 
Ne) 
2) 


1554.:252354 


F/bew.min 


F/akt.min 


konnten für die Gruppen 
w, w’ und g, g’ (wie auch 
für die übrigen Gruppen) 
vergleichend wie folgt her- 
angezogen werden: 1. Die 
Latenzzeiten (Abb. 4, 5), 
2. die Dauer der Verhal- 
tensweisen als Anteil der 
Versuchszeit insgesamt 
(@Abb6, abe 5): die 
Dauer der Verhaltenswei- 
sen als mittlerer prozen- 
tualer Anteil der aktiv ver- 
brachten Zeit (Abb. 6, 
Tab. 3), 4. die ‚„‚absolute‘“ 
mittlere Dauer der Verhal- 
tensweisen in Sekunden 
(Abb. 7, Tab. 3) und 5. die 
„absolute‘“ mittlere An- 
zahl einzelner Verhaltens- 
weisen (Abb. 7, Tab. 3). 
Während ansonsten die 
Verhaltensanteile an der 
Gesamtzeit konstant blie- 
ben, also keine Wirkung 
der Transplantationen er- 
kennen ließen, sind aus 
den Darstellungen einige 
gleichsinnige (wiewohl 
nicht sicherbare) Tenden- 
zen erkennbar: Bei w’ und 
g’ sinkt der Anteil über- 
laufener Felder sowohl 
pro Minute wie auch pro 
akt. Min. und pro bew. 
Min. (Abb. 7, Tab. 3). Da- 
gegen wächst die Anzahl 
der Defäkationen (,,Kot“, 
Abb. 7 und Tab. 3) bei w’ 
und g’, bei w’ nimmt über- 
dies das Urinieren zu. Tat- 
sachlich urinierten nur 
60 % der w-Gruppe, je- 
doch 100 % der w’-Grup- 
pe während des Testver- 
laufes, wobei unter Be- 
rücksichtigung der Daten- 
häufung bei Null und der 
daraus abzuleitenden An- 
wendung des Signifikanz- 
tests nach LACHENBRUCH 
(1976) die Erhöhung der 


160 | M. W. Schäfer 


Urinationsanzahlen bei w’mit p < 0,01 (für Z = 11,36 > 9,21 = x? a, 0.1, Sicherbar ist. Als 
weitere tendenzielle Veränderung seı auf die verringerten Werte von t/Akt und t/Ges für 
die Gruppen w’ und g’ hingewiesen (Abb. 7, Tab. 3), d. h. daß die transplantierten 
Individuen etwas erhöhte Frequenzen aller Verhaltensweisen zeigten, also ihr Verhalten 
häufiger wechselten als w und g. 

Die Prüfung mit dem U-Test nach Wırcoxon et al. ergab lediglich im Falle eines 
Parameters, dem Anteil des Schnupperns an der aktiven Zeit (Sch/Akt), eine signifikante 
Veränderung: Dieser Anteil liegt für die Gruppe w’ höher als für die Vergleichsgruppe 
(Abb. 6). 

Eine zusammenfassende Wertung der Parameter ‚‚normal“ aufgewachsener und trans- 
plantierter Individuen weist für lediglich 3 % der Daten eine Veränderung aus, ansonsten 
bleiben die veränderten Aufzuchterfahrungen und das soziale Umfeld aus Tieren eines 
unterschiedlichen „Genpools“ ohne Einfluß auf das Explorationsverhalten. 


Diskussion 


Die vorliegende Arbeit versucht, das „open-field-Verhalten“ verschiedener Mäusestämme 
in möglichst engmaschigen Zeit- und Parameter-Rastern zu quantifizieren. Es wurde also 
bewußt nicht versucht, einen vorgefaßten Begriff von ‚„Wildheit“ anhand weniger Verhal- 
tensmuster bestätigt zu finden und dann mittels Züchtungsexperimenten auf wenige 
„dominante Wildheits-Gene“ zurückzuführen. 

Der Versuch, möglichst alle abgrenzbaren Verhaltensformen zu quantifizieren, führte 
zwangsläufig zu großer Datenfülle und über die Eigenart der Datenverteilungen zur 
notwendigen Anwendung verschiedener nicht-parametrischer Testverfahren. Gerade die 
Tatsache der von Normalverteilungen abweichenden Frequenzen der Werte spiegelt 
jedoch besonders eindeutig Unterschiede zwischen den Gruppen wider, eben dann, wenn 
die überwiegende Anzahl in einer Gruppe eine Verhaltensweise wenig oder nicht, die 
überwiegende Zahl einer zweiten Gruppe diese jedoch häufig oder ausschließlich zeigt. 
Das Auftreten ‚um Null gehäufter Ereignisse‘ erfordert dann ebenso wie die bei Verhal- 
tensmessungen häufigen großen Streuungen differenziertere Prüfverfahren. 

Der Nachweis von Verhaltensunterschieden zwischen den Parentalgruppen erscheint 
sicher genug, um Kreuzungsansätze zu entwickeln. Dabei sollen nicht von vornherein 
Unterschiede zwischen ‚„domestizierten‘ und ‚wilden‘ Tieren bzw. zwischen weißen und 
grauen Individuen aufgefunden werden, vielmehr dient die weiße Fellfarbe zunächst nur 
zur Charakterisierung eines Genpools, der bedingt durch lange menschliche Züchteraktivi- 
tät und Käfighaltung einen höheren Homozygotiegrad erreicht haben sollte als graue Tiere 
weniger Käfiggenerationen bei identischen Haltungsbedingungen. 

Eine Diskussion genetischer Wirkungen setzt die Abschätzung bzw. Abgrenzung 
postnataler Einflüsse auf das Verhalten voraus. Diese Einflüsse sind sowohl sozialer Art 
(Eltern, Geschwister, Gruppenmitglieder) wie ökologischer Art (Beschaffenheit des 
Lebensraumes). Letzterer wurde für beide Stämme identisch dargeboten, ein unterschied- 
lich richtender Einfluß, „Makrolon-Käfig“ versus „natürliche Umwelt“, ist also nicht 
überprüfbar. Hingegen belegen Befunde von SMITH und Connor (1974, 1976) und 
Connor (1975), daß Korrelationen zwischen verschiedenen Habitaten und Verhaltensmu- 
stern bei Inzuchtstämmen, Hybriden und Wildformen nicht auffindbar sind. Z. B. wird 
open-field- und anderes Verhalten nicht dadurch unterschiedlich, daß eine Gruppe Labor- 
bedingungen, die andere ‚natürlicher‘ Haltung ausgesetzt wurde. Während also Habitat- 
einflüsse gering einzuschätzen sind, ist die Überprüfung postnataler Einflüsse durch die 
Mutter und andere Sozialpartner unerläßlich. 

Der Vergleich der Tiere aus den Gruppen w, w’, g und g’ weist lediglich im Falle der 
erhöhten Schnupperrate als Anteil an der aktiven Zeit bei w’gegenüber w einen signifikan- 


Explorationsverhalten verschiedener Stämme und Generationen der Hausmaus 161 


ten Unterschied aus. Allerdings steigert dieser Unterschied die Differenz zwischen den 
transplantierten Individuen und ihrer grauen Aufzucht-Sozietät. Statt eines Angleichs 
übertrifft so der „Weiß-Grau-Unterschied“ noch jenen zwischen den Gruppen w und g. 
Weitere gleichsinnige Veränderungen bezüglich verminderter Bewegungsaktivität, erhöh- 
ter Verhaltensfrequenzen und verkürzter Latenzzeiten sind nicht auf Sozietätseinflüsse 
spezifisch rückführbar (allenfalls auf die Tatsache der Transplantation als solche!), da sie 
beide Gruppen (w’und g’) gleichermaßen betreffen und somit keine Angleichung zwischen 
den transplantierten Individuen und ihren Sozialpartnern aufzeigen. Damit sind dominie- 
rende Einflüsse des sozialen Umfeldes auf die Ausprägung des analysierten Verhaltens 
auszuschließen. Die Untersuchung von Verhaltensmustern verschiedener Generationen 


erscheint sinnvoll. 
Danksagung 


Mein besonderer Dank gilt Frau CrauDıa RÖTER für die im Rahmen ihrer Diplom-Schrift im 
Fachbereich Biologie, Frankfurt, geleistete Mitarbeit sowie Frau MARGOT NAUMANN für die tierpfle- 
gerische Betreuung. 

Zusammenfassung 


Untersucht wurden die Explorationstendenzen einschließlich Flucht- und kryptischem Verhalten bei 
Hausmäusen verschiedener Zuchtstämme und Generationen ım „open-field“, einer in Quadranten 
unterteilten, weitgehend deckungslosen Versuchsfläche. Als Material dienten neben einem Albino- 
Stamm graue Mäuse aus gleichen Aufzucht- und Haltungsbedingungen, reziprok „transplantierte‘“ 
Jungtiere sowie Hybrid-Filialgenerationen reziproker Zuchtansätze. 

Sowohl die Häufigkeitsverteilungen der Individuen in Relation zu einzelnen Verhaltensmustern 
wie auch die mittleren Latenzzeiten ließen Unterschiede zwischen den Tiergruppen und Generationen 
erkennen. 

Die Prüfung der Explorations-Strategien transplantierter Individuen wies das open-field-Verhalten 
als weitgehend unabhängig von Aufzuchterfahrungen aus und ließ eine genetische Grundlage ver- 
muten. 

Literatur 


BARNETT $. A.; SMART, J. L. (1975): The Movements of Wild and Domestic House Mice in an 
Artificial Enviroment. Behav. Biol. 15, 85-94. 

CoBuRN, C. A. (1922): Heredity of Wildness and Savageness in Mice. Behav. Monog. 4, 1-71. 

CONNOR, J. L. (1975): Genetic Mechanisms Controlling the Domestication of a Wild House Mouse 
Population (Mus musculus L.). J. Comp. Physiol. Psychol. 89, 118-130. 

Dawson, W. M. (1932): Inheritance of Wildness and Tameness ın Mice. Genetics 17, 296-314. 

EıgL-EißBEsreLpT, I. (1950): Beiträge zur Biologie der Haus- und der Ährenmaus nebst einigen 
Beobachtungen an anderen Nagern. Z. Tierpsychol. 7, 558-587. 

— (1958): Das Verhalten der Nagetiere. In: KüKENTHAL, Hb. Zool. 8 (10), 1-88. Berlin: de Gruyter. 

FREYE, H. A.; FREYE, H. (1960): Die Hausmaus. Wittenberg: Neue Brehm Bücherei. 

Frick, H.; Norp, H. ]J. (1963): Domestikation und Hirngewicht. Anat. Anz. 113, 307-316. 

Harez, E. S. E. (1969): The Behaviour of Domestic Animals. London: Bailliere, Tindall & Cassell. 

HAGEMANnNn, E. (1959): Über die Haustierwerdung der weißen Maus. Desinfektion und Gesundheits- 
wesen 11, 154-156. 

HERRE, W.; RöHrs, M. (1973): Haustiere — zoologisch gesehen. Stuttgart: G. Fischer. 

LACHENBRUCH, P. A. (1976): Analysıs of Data with Clumping at Zero. Biometr. Zeitschr. 18, 
351-356. 

LIENERT, G. A. (1973): Verteilungsfreie Methoden ın der Biostatistik I und II. Meisenheim am Glan: 
Hain. 

LoRENnzZ, K. (1978): Vergleichende Verhaltensforschung, Grundlagen der Ethologie. Wien, New 
York: Springer. 

Money, J.; EHRHARDT, A. (1975): Männlich-weiblich. Die Entstehung der Geschlechtsunterschiede. 
rororo-Sexologie. Reinbek bei Hamburg: Rowohlt Taschenbuch-Verlag. 

MÜHLENBERG, M. (1976): Freilandökologie. Heidelberg: Quelle & Meyer. 

NIETHAMMER, J.; Krapp, F. (1978): Handbuch der Säugetiere Europas Bd. I, Rodentia I. Wiesbaden: 
Akadem. Verlagsanstalt. 

Promin, R.; MAnosevItz, M. (1974): Behavioral Polytypism in Wild Mus musculus. Behav. Genetics 
4, 145-157. 

RAATZ, U. (1966): Eine Modifikation des White-Tests bei großen Stichproben. Biometr. Zeitschr. 8, 
42-54. 

RöhHrs, M. (1961): Biologische Anschauungen über Begriff und Wesen der Domestikation. Z. 
Tierzüchtg. Züchtgsbiol. 76, 7-23. 


162 Christel Schmidt 


RoTH, G. (1974): Kritik der Verhaltensforschung. München: Beck. 

Sachs, L. (1978): Angewandte Statistik, Statistische Methoden und ihre Anwendung. Berlin, Heidel- 
berg, New York: Springer. 

SMITH, R. H. (1972): Wildness and Domestication in Mus musculus: A Behavioral Analysis. J. Comp. 
Physiol. Psych. 79, 22-29. 

— (1978): Open-Field Freezing as a Stable Parameter of Wildness in Mus musculus. Behav. Biol. 23, 
67-74. 

SMITH, R. H.; ConnoR, J. L. (1974): The Inheritance of Behavioral Wildness in House Mice (Mus 
musculus L.). Anım. Learn. Behav. 2, 249-256. 

— — (1978): Behavioral Effects of Laboratory Rearing in Wild Mus musculus. Behav. Biol. 24, 
387-392. 

STRESEMANN, E. (1970): Exkursionsfauna von Deutschland. Wirbeltiere. Berlin: VEB Volk und 
Wissen. 

WALSH, R. N.; Cummins, R. A. (1976): The Open-Field Test: A Critical Review. Psychol. Bull. 83, 
482-504. 

WHITE, C. (1952): The Use of Ranks in a Test of Siginificance for Comparing Two Treatments. 
Biometrics 8, 33—41. 

WIESER, W. (1976): Konrad Lorenz und seine Kritiker. München: Piper. 


Anschrift des Verfassers: Prof. Dr. M. WALTER SCHÄFER, Fachbereich Biologie (Zoologie) der 
Universität, Siesmayerstr. 70, D-6000 Frankfurt a. M. 


Behavioural and neurophysiological studies of the olfactory 
sensitivity in the albino mouse 


By CHRISTEL SCHMIDT 


Zoological Institute, University of Bonn 
Receipt of Ms. 8. 12. 1981 


Abstract 


Studied was’the olfactory sensitivity of dd mice by means of training experiments and electrophy- 
siological investigations. Three mice were investigated with both methods, and the behavioural and 
neural thresholds were found to be in the same range. Neural thresholds were determined in 24 
animals for geraniol (5 X 107 — 5 X 10° molecules/cm? air), butyric acid (7 X 107° — 7 X 10° mol./cm?) 
and butyric methyl ester (4 x 10° — 4 x 10!! mol./cm?) by studying evoked potential measurements 
and analysıs of bulbar oscillations. 


Introduction 


Considering the complex significance of the olfactory sense ın rodents, it is surprising that 
only few investigations deal with the olfactory sensitivity of thıs mammalian order. 
Behavioural experiments are performed with guinea pigs (MATTHES 1932) and rats (GRUCH 
1957; Mourrton and Eayrks 1960; MourLton 1961) to establish olfactory thresholds, 
whereas in albino mice neural response thresholds have been determined (SCHMIDT 1978; 
SCHMIDT and SCHMIDT 1980). 

The results of these few investigations are even difficult to compare, as the threshold 
values were obtained with different methods. Whereas in the behavioural studies the 
perceptive ability of the anımals is investigated, the electrophysiological experiments 
document a neural reaction of a certain brain area. To evaluate these results it is necessary 
to compare behavioural and neural threshold data. Investigations of the same individual 
with different methods especially allow a more accurate estimation of its olfactory ability. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4703-0162 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 162-168 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Studies of the olfactory sensitivity in the albino mouse 163 


Material and methods 


The experiments were carried out with dd albino mice, 6-8 months old (body-weight 30-38 g). Three 
mice were trained for the behavioural studies; the neural thresholds of these anımals were determined 
after completing the trainıng experiments. In addition, neural olfactory thresholds were established in 
24 mice. 


Training experiments 


The transparent Y-maze for the behavioural tests (fig. 1) consisted of a starting box and two 
destination boxes, connected by tunnels, and a central triangle. Shock grids were placed in front of 
both destination boxes (stimulation frequency 20 Hz; voltage 20-50 V). The clean (charcoal filter, 
water) and temperature constant (22-23 °C) air stream was divided before reaching the bottles with 
odour and with the control substance (solvent of the odourant). The flow velocity of the air stream 
was 10 cm?/s. The outlets ended Icm in front of the shock grids. Olfactory thresholds were 
established for butyric acıd (BA) and geraniol (GER). The choice apparatus was cleaned between each 
trial with Extran® and washed with fresh air; stale air was permanently sucked out by a funnel at the 
bifurcation. 

The anımals were trained to select the odourous air stream. Every day the mice performed 3 warm 
up runs and 20 test trıals. The location of odour- and control stream was randomly changed. For each 
concentration tested (log. unit steps) 60 trials were analysed (chance level: p< 0.001; z-test). 
Olfactory adaptation was diminished by an interval of 8-10 minutes between the trials. 

At high odour concentrations, the 3 mice already made their decision at the bifurcation, while at 
low concentrations after sniffing at the outlets. Correct choices were rewarded with food, whereas 
wrong ones were punished with a slight electrical shock. 


Electrophysiological experiments 


The electrical activity from the olfactory bulb was recorded with electrodes (tungsten wire) insulated 
up to the tip (® 50 um; resistance 20-30 kQ), permanently implanted in the brain and held in position 
by a mass of dental cement that also enclosed two small steel hooks anchored in the parıietal bone. A 


P F W FM 


Fig. 1. Two choice training 
apparatus. SB = starting box; 
S = shock generator; CT = 
central triangle; SG = shock 
grid; DB = destination box; 
C = control substance; O = 
odourant; WB = wash bottle; 
FM = flow meter; W = wa- 
ter bottle; F = charcoal filter; 
P = pump 


metal clamp attached to the ipsilateral ear served as the reference electrode (SCHMIDT 1978). Oltactory 
responses in the waking mice were recorded 24 hours after implantation. The olfactory stimulus 
(duration 1 s) was presented with a motor-driven syringe-olfactometer. Three substances were tested 
with the electrophysiological methods — butyric acıd (BA), butyric methyl ester (BME) and geraniol 
(GER) - at 3 minutes intervals, beginning with subthreshold concentrations. Control stimuli (pure air) 
were presented between the odour stimuli at irregular intervals. A continuous flow of air from the 


correct choices 


164 Christel Schmidt 


side, together with a funnel mounted above the animal to draw air away, ensured the odour neutrality 
of the surroundings. The activity of the olfactory bulb was recorded with the standard apparatus 
(oscilloscope, pen writer). The criterion for a positive neural response to an odour stimulus was the 
occurence of evoked potentials in the form of a negative deflection at the end of the inspiration. 


Results 
Behavioural experiments 


Pilot tests had shown that the animals behave differently when confronted with diverse 
odour concentrations. They rejected butyric acid and geraniol in low dilutions (10! 
vol.%) and avoided even high dilutions of butyric methyl ester, so that this odourant was 
unsuitable for these experiments. The training started with geraniol (10 ? vol.% = 5 x 
10'! molecules/cm? air). In the beginning the door of the negative destination box was 
closed, so that the anımals were forced to correct misjudgements. Nevertheless both boxes 
had to be opened after some time, since the mice localized the closed box, possibly by 


100 


80 


60 


molecules 


Fig. 2. Performance of 3 male albino mice trained to select geraniol (A) and butyric acid (B). Ordinate: 
percentage of correct choices; abscissa: concentration of the test substance (odour molecules/cm’ air); 
z = chance level (p<0.01; z-test). The 3 males are identified by different symbols 


acustical cues. The first obvious preference for the odourous stream could be recognized 
after ca. 130 trials. Threshold measurements started when the anımal showed more than 
80 % correct choices within 60 consecutive trials. This training level was reached after 
240-300 trıals. In the maın experiments, each concentration was tested by 60 trıals. 

The performance level is significantly positive (p<0.001) in all anımals up to a 
concentration of 5 X 10° molecules geraniol/cm? air. At the next lower concentration 
(5 x 107 mol./cm?) all anımals choose randomly (fig. 2a). Although the second substance 
(BA) was immediately accepted, the performance level remains lower than for geraniol 
(ca. 80 %) even after 10 days training. Individual differences between the anımals are 
obvious with lower concentrations. While the performance with a concentration of 7 X 10° 
molecules BA/cm? is significantly positive in all mice, only 2 of them are able to detect 
7 x 10° mol./cm?; with 7 X 107 mol./cm? choices are at random (fig. 2b). 


Neurophysiological threshold measurements 


With low resistance electrodes two characteristic potentials are recorded from the olfactory 
bulb of the mouse: 1.) a slow fluctuation consisting of a negative-going wave during 


Studies of the olfactory sensitivity in the albino mouse 165 


inspiration and a positive wave during expiration, and 2.) regular oscillations beginning at 
the end of the inspiration phase and continuing for most of the expiration phase (phase B; 
ScHMIDT 1978). 

Simultaneous recordings from the 2 bulbs reveal that both the background fluctuation 
and the oscillations result from stimulation of the nasal mucosa. When one side of the nose 
is reversibly blocked, both the respiration-linked and the superimposed oscillations in the 
ipsilateral bulb vanısh, whereas ın the contralateral bulb the normal resting activity persists 
(fig. 3). 

As pilot experiments have shown, the bulbar potentials are very sensitive to anesthetics 
and neuroleptics. For instance 160 mg/kg body weight Evipan® (sodium hexobarbital; 
15-20 min anesthesia) abolish all potentials; with Megaphen* (chlorpromazine) evoked 
potentials can only be elicited by intense odour stimuli. Therefore, all the experiments are 
carried out with anımals unaffected by any drugs. 

The evoked potential, a negative deflection occuring at the onset of the oscillations 
when stimulated with an odourant, can be used to determine the neural olfactory 
thresholds. The pattern of the bulbar potentials change, when recorded ın different areas of 
the olfactory bulb. Recordings from deeper layers are characterized by large oscillations 
and relatively inconspicious evoked potentials; when the tip of the electrode is placed in the 
mitral-cell region, the evoked potentials are extremely large, while the oscillations are small 
and sometimes not detectable at all. Nevertheless, the threshold concentration, at which 
the evoked potential ıs first discernible, ıs independant of the location of the electrode in 
the bulb. 

In the 3 mice investigated with both methods, for geraniol neural and behavioural 
thresholds are exactly the same (5 X 10° mol./cm? air). For butyric acid two animals 
showed the same thresholds (7 X 10° mol./cm?) independant of the method used, while the 
neural threshold of one animal is lower by a factor of 10 (7 X 10’ mol./cm?). The 24 males 


a 
Fig. 3. Normal activity of the 
olfactory bulbs, recorded 
simultaneously on both sides. 
A: both nostrils open; B: one 
nostril closed. In the bulb ip- 
silateral to the closed nostril 
the characteristic potential 
fluctuations have disappeared. 
C: when the nostril is 
reopened both bulbs again ex- 
hibit the normal activity pat- 


tern 


x r 
us 4 Harpı Ay 
| R 1% au LITER EUR BR yartıı ENT... 
3 54 wir W nal \ A “3 N Aw F ke A! f Mm Tr 
i { 3 i 5% 


N: 


number of oscillations 


TTS 


166 Christel Schmidt 


number of animals 


Fıg. 4. Neural thresholds measured by means 
of evoked potentials. Ordinate: number of 
anımals with the indicated thresholds; ab- 
scissa: threshold concentration (odour 
molecules/cm? air); BA: butyric acid; BME: 
butyric methyl ester; GER: geraniol 


5x 
9 
107 10 10 10° 10 10" 107 10° 10 


threshold concentration (molecules /cm?) 


7x 4 


x 
SERAg 92.107,71 


used only for the electrophysiological studies, show the first reactions to BA between 
7x 107’ and 7 X 10° mol./cm?; the thresholds measured for BME are between 4 X 10° and 
4 x 10!! mol./cm? and with geraniol 5 X 107t0 5 X 10° mol./cm? air are necessary to elicit a 
neural response in the olfactory bulb (fig. 4). 

The evoked potential is not the only change in bulbar activity to be observed during 
olfactory stimulation, the number and frequency of the oscillations change as well. In the 
presence of subthreshold odour concentrations, the number of oscillations per breath is 
nearly uniform before, during and after odour presentation, but with above-threshold 
stimuli, there is a distinct reduction in the number of oscillations during the stimulus. The 
maximal reduction wıth BME amounted to 71 % (fig. 5a) from the lowest concentration 
(10°” vol.% =4x 10% mol./cm?; 4.82 oscillations per breath) to the highest (10°! 
vol.% = 4x 10!* mol./cm?; 1,4 0/b). This change in oscillation frequency, in inverse 
proportion to the odour concentration, is just as striking for the other substances (BA: 
7 x 10° mol./cm?; 5.22 o/b and 7 x 10'? mol./cm?; 0.97 0/b — GER: 5 X 10° mol./cm?; 
4.28 o/b and 5 X 10? mol./cm?; 1.43 o/b). 


a j ' 
„90 b — 
A| 
—ı 7 Ar n m | 1 I , | 
A, 949 9 = 20 SE 7 = % = ° 2} 7 SE WM =} 74 : W N) E 
7 7 A| 3 ah AN > AH Dali Ali Mi A 2 
2 2 Y + 29 = 79 =] 7 SE v 4 H 5 W = D = 2% = 9 = D, 
nl A ee a || 
\ A : z } 3 ll _ | 2 = 
| |, SIE El 
N Mm AN ll dl. = = = 1 09 7 m 
7 = | = = | — = = z 2 = = 
7 Al il ı on 2 = = = W/ 4 AN 
7, = 2 = 7, ES 7 = DS = 1 7), 7 = 
G 7, EZ Z BE ? EEE 7, = | 3% = 7 N, N 2 
_ NN = 7? BR 7, E 7, == Be 3 23 7, , Be 7, Bi 
7 4 . E [| & 2 = , = 0 . BB BE / 1 Bl J :E 
7 -6 -5 2-1 -4 - -2 1 
log vol. conc. log vol. conc 


Fig. 5. a: number of oscillations/breath (x) at different odour concentrations (BME); b: Mean 

frequency of oscillations/breath (x) at different concentrations of BME. Abscissa: concentration of 

BME (log. vol.%); hatched: 1 s before stimulation; shaded: 1 s during stimulation; punctuated: 1s 
after odour presentation; vertical lines: standard deviation 


Studies of the olfactory sensitivity in the albino mouse 167 


The mean oscillation frequencies calculated for the same periods of time are also closely 
correlated with stimulus intensity. In all 5 mice tested, an increase in odour concentration 
is accompanied by a reduction in frequency during stimulus presentation. In the case of BA 
and BME this phenomenon first appears when the concentration is raised to 7 X 10° mol./ 
cm? BA resp. 4 X 10° mol./cm? BME (= 10° vol.%; fig. 5b); (probability of error in 
assuming that the frequency during the second preceding olfactory stimulation is greater 
than that during stimulation: BA, p<0.001; BME, p<0.01). With geraniol the first 
statistically significant frequency reduction occurs with 5 X 10° mol./cm? (p< 0.001). 

The absolute frequency reduction obtained with the highest concentration presented 
(107! vol.%), as compared with the unstimulated or subthreshold situation, is 11 Hz for 
BA, 13 Hz for BME, and 15 Hz for GER. In the second following the stimulus, the 
oscillation frequency rises above the unstimulated level by almost the amount of the 
preceding fall (BA, 10 Hz; BME, 14 Hz; GER, 10 Hz). This overshooting recovery, which 
leads to a gradual increase in unstimulated frequency as the series of stimuli at increasing 
concentrations proceeds, are interpreted as effects of adaptation. 


Discussion 


Investigations of the efficiency of sensory organs in mammals may lead to very diverse 
results, when different methods are applied. Especially the differences between neurophy- 
siological and behavioural methods may amount to great values, since the former give 
electrical responses of small neural areas or units and the latter consider the whole anımal. 
Therefore, the results of such experiments are not always comparable. For instance, a 
behavioural threshold gives an indication of the perceptual ability, whereas a neural 
threshold does not mean per se that the anımal can actually perceive a given threshold 
stimulus. Various methods - behavioural as well as electrophysiological ones - should be 
used for evaluating threshold values. 

Training experiments or conditioning methods have mainly been used to estabish 
threshold values for olfaction; data on neural thresholds are only available for mice 
(SCHMIDT 1978; SCHMIDT and SCHMIDT 1980). As the potentials in the olfactory bulb are 
strongly affected by anesthetics, only anımals which are fully awake are suitable for these 
studies. Therefore, permanently implanted electrodes prove to be most suitable for 
threshold measurements. They have the advantage of registering the bulbar activity in 
unanesthetized anımals and allow to work with the same individual for several weeks 
(SCHMIDT und SCHMIDT 1982). 

Two parameters were used in the present study to determine threshold values: The 
evoked potentials and the change of oscillation frequency. Both potentials seem to be 
generated in different parts of the olfactory bulb. While the negative potential, evoked by a 
sudden olfactory stimulus, can be recorded best in the outer layers of the Bulbus 
olfactorius, the amplitude of the oscillations increases in the deeper layers of the bulb. 
Nevertheless, the olfactory thresholds obtained with both potentials match. 

Comparing the results of the behavioural and the electrophysiological studies it 
becomes evident that all the thresholds are in the same order of magnitude. So it seems 
admissible to conclude that the threshold concentrations obtained are ın a reliable range. 

As olfactory threshold measurements in mammals are very scarce, comparable data are 
only available for butyric acıd. The results obtained in these studies indicate that the 
olfactory sensitivity of the laboratory mouse for BA is in the same range as in other small 
macrosmatic mammals, for instance the european hedgehog (BrETTInG 1972). These low 
threshold values emphasize the importance of olfaction in the life of the mouse. 


168 E. Pulliainen 


Zusammenfassung 


Verhaltensphysiologische und elektrophysiologische Untersuchungen zur Riechfähigkeit 


von Albinomäusen 


Bei dd Albinomäusen wurde die Riechleistung mit verhaltensphysiologischen und elektrophy- 
siologischen Methoden untersucht. Die olfaktorischen Schwellen (Geraniol, Buttersäure) der drei 
Tiere, bei denen anschließend an eine Geruchsdressur Ableitungen vom Bulbus olfactorius durch- 
geführt wurden, lagen bei beiden Methoden im gleichen Größenordnungsbereich. Bei 24 weiteren 
Mäusen wurden neurale Schwellen mit Hilfe evozierter Potentiale sowie durch eine Frequenzanalyse 
der für den Bulbus olfactorius charakteristischen Oszillationen bestimmt. Die Übereinstimmung der 
in verhaltens- und elektrophysiologischen Experimenten ermittelten Ergebnisse rechtfertigt es, auch 
bei den neuralen olfaktorischen Reaktionsschwellen von Riechschwellen zu sprechen. 


References 


BrETTInG, H. (1972): Die Bestimmung der Riechschwellen bei Igeln (Erinaceus europaeus L.) für 
einige Fettsäuren. Z. Säugetierkunde 37, 286-311. 

GrucH, W. (1957): Über die Riechfähigkeit bei Wanderratten. Zool. Jb., Abt. Physiol. 67, 65-80. 

MATTHESs, E. (1932): Geruchsdressuren an Meerschweinchen. Z. vergl. Physiol. 16, 766-788. 

MouLTton, D. G. (1961): Studies in olfactory acuity. Ill. Relative detectability of n-alıphatic acetates 
by the rat. Quart. J. Exp. Psychol. 13, 203-213. 

MouLTton, D. G.; Eayrs, J. T. (1966): Studies ın olfactory acuity. II. Relative detectability of n- 
aliphatic alcohols by the rat. Quart. J. Exp. Psychol. 12, 99-109. 

SCHMIDT, C.; SCHMIDT, U. (1980): Changes of olfactory sensitivity during the estrus cycle in female 
laboratory mice. Chem. Senses 5, 359-365. 

— — (1982): Olfaction ın pregnant and lactating mice. Z. Säugetierkunde 47, 50-55. 

SCHMIDT, U. (1978): Evoked-potential measurements of olfactory thresholds of laboratory mice (Mus 
musculus) to carboxylıc acıds. Chem. Senses Flav. 3, 177-182. 


Author’s address: Dr. CHRISTEL SCHMIDT, Zoologisches Institut der Universität, Poppelsdorfer 
Schloß, D-530 Bonn 


Habitat selection and fluctuations in numbers in a population 
of the arctic hare (Lepus timidus) on a subarctic fell in 
Finnish Forest Lapland 


By E. PuULLIAINEN 
Receipt of Ms. 20. 11. 1981 


Abstract 


Habitat selection and population fluctuations were studied in the arctic hare (Zepus timidus) ın the 
Värriötunturi fell area, East Finnish Forest Lapland, in the winters of 1968/69-1980/81. A transect 
survey indicated that the density of the population fluctuated in cycles of at least four years. Only the 
preferred habitats (i.e. lowland conifer-dominated mixed forests with birch and/or characterized by 
juniper) were inhabited during a population low, while in peak winters hares also occurred in less 
favoured habitats (e.g. mountain birch forests on windy slopes). The population decline is not 
triggered by predation. The movement activity of the hares when searching for food may increase 
considerably ın the late winter at times of high populations. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4703-0168 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 168-174 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Habitat selection and population fluctuations in the arctic hare 169 


Introduction 


Cyclic fluctuations and their causations in populations of the snowshoe hare (Lepus 
americanus) in the boreal forests of North America are relatively well known (see GREEN 
and Evans 1940; KeıtH 1963; KEITH and WINnDBERG 1978; FINERTY 1980), whereas 
fluctuations ın those of the arctic hare (Zepus timıdus) in the boreal and subarctic forests of 
Fennoscandia are poorly documented (for data, see SIIVonEN 1948; HÖRNFELDT 1978). 
Very little is known even about habitat selection among arctic hares under conditions in 
which different fell zones are available, although the composition of the winter food of the 
species under such conditions is better known (see PULLIAINEN 1971, 1972; LINDLÖF etal. 
1974a, b). 

The present author began a census of hare tracks on a subarctic fell in eastern Finnish 
Forest Lapland in the winter of 1968/69 and results from 13 successive winters are now 
available. The purpose of the present paper is to provide records on habiıtat selection and 
fluctuations in numbers during these winters. 


Study area 


The study area lies in eastern Itäkaira, Finnish Forest Lapland. The area includes two fell ridges 
(Pulkkatunturi — Sautunturi and Värriötunturi) running ın a north-south direction about 5 km apart. 
It is in a virgin state, with the following proportions of natural habitats in an area of approxımately 
200 km?: spruce forests (Picea abies) 26.5 %, Scots pine forests (Pinus sylvestris) 18.5 %, birch forests 
(Betula spp.) 3.0 %, mixed forests 22.0 %, open bogs 13.5 %, mountain birch forests (Betula 
pubescens tortuosa) 7.0 % and treeless fell summits 9.5 %. The forests are old and mature, including a 
large number of standing and fallen dead trunks. The density of the trees ıs generally very low, and ın 
some places there are young spruce, pine and birch saplıngs growing among the very old trees. Some 
of the lowland birch forests were killed by larvae of the geometrid Zpirrita autumnata ın 1965-1966 
(see PULLIAINEN 1976). The study area lies 266-475 m above sea level. 


Methods 


The numbers of arctic hare and mammal predator tracks crossing an observation line 5970 m long 
running westwards from the summit of Värriötunturi were recorded in the winters of 1968/69 to 
1980/81 by skiing along this line once a week. Only tracks which were two days old or fresher when 
recorded in November-April (identified on the basıs of the snowfall and wind data from the nearby 
meteorological screen and the states of freezing of the tracks) are taken into account in the present 
calculations. If the same individual had crossed the line in a zig-zag manner between two recording 
sticks 30 m apart, this was regarded as only one crossing. 

Some months contained four recordings and some five, and in some cases a heavy snowfall had 
covered the fresh tracks of the mammals in question. All values representing each habitat type were 
first calculated per 100 m of the observation line, rendering the recordings made in different habitats 
comparable one with another. All monthly means were then standardized to four recordings. If a 
heavy snowfall or strong wind had recently covered the tracks, these cases were omitted, and the 
monthly means were calculated on the basis of those recordings from the month in question ın which 
recent tracks were visible. 

The observation line ran through the following habitats (distances covered in parentheses): 

(A) Almost treeless summit of the Värriötunturi fell (510 m). 

(B) Mountain birch forest on the western slope of Värriötunturi, were mountain birches (B) 
comprised a half of the trees, the other half consisting of pines (P; 20-40 %), spruces (S; 0-10 %) 
and junipers (J; Janıperus commaunıs; 10-20 %) (570 m). 

(C) Spruce-dominated forest: S 40-85 %, P 0-40 %, B 5-30 % and J 0-10 % (1860 m). 

(D) Pine-dominated forest: P 70-80 %, S 0-10 %, B 5-10 % and ] 10-15 % (1200 m). 

(E) Mixed forest characterızed by juniper: ] 55-75 %, S5-15 %, P 0-20 % and B (living, since some 
had been destroyed by Zpirrita autumnata) 10-20 % (1200 m). 

(F) A bog: willows (Salıx spp.) 80 %, B 10 % and S 10 % (30 m). 

(G) A treeless marsh (600 m). 


170 


soo>d csoo>d 
6lE'O Z91'0 8980 erTTz LEET 6990 1891-9 
(oreto) er seco + 810 ITOF 8TOo + eco F 'q’S F ueaw 
+ "SF ueIW 


SS 
EIOr 900 = 02.0 °OEOZFZTOT SO FEN 200 FE SEO FEN 60078550 vet 


orage) ZIO LLO ala c60 18/0861 
= Z20’0 sro c8’0 60 08/6/61 
seo 090 c8’0 LS’O 801 6//8/61 
eeO Seo O1’O 0€E°O [a 8//LL61 
= O1’O lo sro trO LL/IL6l 
= 80°0 0,30) sıo 9L/SL6l 
81’0 cl’0 8970 7440) 0s’O S//v/6l 
= Ko) LS’O re) [7290] bL/El6l 
= 810 ss’o ve’o eL/CL6l 
ZlO O1’O 20’0 810 c//\Z6l 
syo seo er’O LIO LS’O 12/0261 
so’o er’ Or’0 syo 801 0//6961 
orage) ce 0 ce’d er’O LEO 69/8961 


c0’0 


15910} «38910} ..5910F .S910} 
3oq uado turuuns SS9[991 «30 MOIJIM,, yaııq uregunow "wop->aund,, “ıeyo-ıodıunl,, "wop-a9nuds,, 
judy-"AoN SIzAoN 9) V J a [ei £| 0) FU A 
sypen Jo ‘ou [elO], 
sıeitgeH 


18/0861-69/8961 JO s13JuIM ayJ ur (yJuour y9ea saw 3urpA10991 p ad payepnsje>) syY9e1J 3s9yJ Jo staquınu [eJo} pue (3%X3 398 
“suorJdr19sap 107) sJeJIgey Juaıajpp ur ur uoNeAAIIsgo 3yJ ZuIsso.19 P3PA10991 31ey I1e 3yJ Jo (Jurj uorJeAIssgo 3yJ Jo u O0I 19d) syae.ıy jo sıaquınu ueaW 


AAEL, 


Habitat selection and population fluctuations in the arctic hare 171 


Results 
Habitat selection 


The spruce-dominated and juniper-characterized mixed forests were clearly the most 
favoured habitats for the arctic hares (Table). The pine-dominated mixed forests formed an 
intermediate group. Arctic hares were recorded every winter in these three habitats, 
whereas they were absent from the rest of the habıtats available ın at least one winter, and 
the mean numbers of tracks were also smaller than in the first three habitats. In three 
winters (1969/70, 1977/78 and 1978/79) the arctic hares occurred in all the habitats studied, 
and in one winter (1975/76) they were found only in the three most favoured ones. 


Fluctuations in numbers 


Track counts during 13 winters indicated two lows, namely the winters of 1971/72 and 
1975/76, and three peaks, the fırst in the winter of 1969/70, the second in 1974/75, and the 
third in 1978/79. Thus the two lows followed each other at an interval of four years, and 
the three peaks at intervals of five and four years. The timing of the third peak may be 
somewhat later, however, as thıs has now lasted three winters. The pattern stands out more 
clearly if the conclusions are made on the basıs of the total numbers of tracks recorded in 
November-February in each winter (Table). Then the timing of the lows and the second 


UD 


240 


160 


Total numbers of tracks of arctic hares (solid line) 
and small carnivores (columns: A = Vulpes vulpes, 
B = Mustela erminea, € = Martes martes and 
D = Mustela nivalıs) recorded crossing the observa- 
tion line during the winters of 1968/69-1980/81 
(numbers calculated per 24 recording times). Triang- 
les indicate the years in which Aguila chrysaetos 
nested in the nearby aerie 


120 


80 


TOTAL NUMBER OF 
TRACKS / 24 RECORDINGS 


40 


68- 70- UZ 
69 71 


WINTERS 


peak are the same, but the first peak ıs located one year later (in 1970/71) and the third two 
or more years later. The intervals between the two lows and between the fırst two peaks are 
then four years, and that between the second and third peak at least six years. 

The reason why the two methods of calculation give partly different results ıs that the 
mean number of tracks per 100 m of the observation line in each of the most favoured 
habitats increased significantly in late winter ın the peak winters, but not during the other 
winters. 

Of the carnıvores ıinhabiting the area studied here, the following can be regarded as 
potential predators upon young and/or adult arctic hares: the red fox (Vulpes vulpes; see 
SEQUEIRA 1980), the pine marten (Martes martes; PuLLıaInEn 1981a), the stoat (Mustela 
erminea; ERLINGE 1979) and the pygmy weasel (Mustela nivalıs; PULLIAINEN and 
IivanAINEN 1979), together with the great predators, among which the wolf (Canis lupus; 
PULLIAINEN et al. 1980) and wolverine (Gulo gulo; PuLLIAINEN 1978) are only occasıonal 
visitors to the area and the Iynx (Felis lynx; PULLIAINEN 1978) a very rare visitor. 
Observations made at the nest of the local pair of Golden Eagle (Aguila chrysaetos) showed 


172 E. Pulliainen 


that arctic hares are an important source of food for them, at least during the nestling 
period. The years ın which the nearest pair hunting in the area also nested there are marked 
in the Fig. as also are the total numbers of tracks of the red fox, pine marten, stoat, pygmy 
weasel and arctic hare crossing the observation line during the winters concerned. The 
numbers of both the hare and the predator tracks were greatest in the winters of 1969/70, 
1974/75 and 1978/79. 


Discussion 


Arctic/subarctic hares are regarded as typical floaters ın their food intake in winter (see 
PruıTT 1970), although they may also occasıionally dig out food items from under the snow 
(PuLLIaınen 1972). According to BRyAnT and KuroPpar (1980), the forage preferences of 
snowshoe hares and arctic hares follow the pattern wıllow > aspen > larch > dwarf birch 
> tree birch > pine (including Scots pine) > fir > spruce > alder. Arctic hares eat the 
woody internodes of Betula pubescens and Populus tremula, but reject the foliar buds 
(BARIKMO 1976). Juvenile-growthform twigs in the browse material appear to be less 
palatable than the mature-growthform twigs of the same plant species (BRYANT 1980). 

Since there are no larches and very few willows, aspens or dwarf birches available in the 
present area (PuULLIAINEN 1971, 1972), ıt is understandable that about 65 % of the feeding 
observations made on arctıc hares during two successive winters should have concerned the 
twigs, bark, catkins and decayed wood of birch, and the shoots and bark of juniper 
(PuLLiaınen 1972). When juniper was plentiful, its importance seemed to increase at the 
expense of birch. During snowy winters, however, the majority of the juniper bushes (and 
willows and dwarf birches) are covered by the snow. 

Especıially in the mountain birch forest zone of Värriötunturi, but also to some extent in 
the lowlands, accumulations of hoarfrost and snow on the tops of birches and other trees 
and bushes bend the tıps of the branches and other parts to levels at which they can be 
reached by the hare (Purziaınen 1972). Taking this into account, together with the 
structure of the forest and the relatively high temperatures on the slopes of the fell as 
compared with the lowlands (see HELIMÄKI 1974), the mountain birch forests might have 
been expected to be the preferred habitats of the present arctic hares, but this was not the 
case (Table). Among the biotopes of southern Scandinavia studied by LinDLör et al. 
(1974a), clearıngs and pasture-land overgrown with birch, all of which had a rich 
undergrowth, were most utilized by hares, while mixed coniferous forests and open bogs 
were less favoured. 

The arctic hares of this area clearly preferred the conifer-dominated and juniper- 
characterized mixed forests as their habıtats (Table), although the conifers themselves are 
only exceptionally of nutritional importance for them, and then only in the form of fallen 
pine branches (see PuLLIAInEN 1971, 1972). It may be that, on the one hand, the hares 
avoided the warm but windy slopes and summit of the fell (HeLımAkı 1974), and on the 
other hand, they had found other acceptable food in addition to birches and junipers (e.g. 
grass and herbs, see PuLLıaınen 1972) and obtained shelter under the lower branches of 
the spruces and fallen trunks of the conifers. The prevailing low temperatures, often below 
—30 °C due to temperature inversion, seem to be of little or no importance in the habıtat 
selection of the local arctic hares. 

The snowshoe hare populations of the North American boreal forests exhibit a long- 
term periodicity, mainly of 8 to 11 years (KEITH 1963), but not those in the southern part 
of the range (Howeıı 1923; LEoroLD 1933; WoLFr 1977). In both areas increased hare 
densities lead to dispersal into less favourable habitats (KErTH 1974; DOLBEER and CLARK 
1975; WOoLFF 1977). Assuming that the changes in the total numbers of tracks crossing the 
present observation line per winter reflect corresponding changes in the density of the hare 
population, a similar trend can be said to be recorded here. In the winter of the deepest low 


Habitat selection and population fluctuations in the arctic hare 173 


(1975/76), hares were found only in the three most preferred habitats, whereas in the two 
peak winters (1969/70 and 1978/79; and also during the increasing phase in 1977/78) 
observations were made in all the habitats available (Table). Rapıd withdrawal from 
marginal habitats at times of declining populations has also been demonstrated in voles 
(TAMARIN 1978). WOLFF (1977) writes that “during the low phase of the hare cycle, hares 
are found only in the refuges, suboptimal habitats frequently having more food seasonally, 
but there being less cover and hares being more exposed to predation”. 

HÖRNFELDT (1978) found signs of a periodicity of 3-4 years in Swedish arctıc hare 
populations on the basıs of old catch records. The present data indicate one interval of four 
years between two lows, one interval of 4 or 5 years between two peaks and one interval of 
at least four years between two peaks (Table). Further studies are in progress to clarıfy the 
existence of fluctuations of 4 years or longer in this hare population. 

KEITH and his assocıiates (KEITH 1963, 1974; KEITH et al. 1977; KEITH and WINDBERG 
1978) studied the 10-year cycle in the snowshoe hare population of central Alberta, 
Canada. KEITH and WINDBERG (1978) conclude their results to date as follows: “The cycle 
is repeatedly generated intrinsically through a hare-vegetation interaction (dominant 
herbivore vs. winter food supply) that triggers the population decline. This elevates the 
predator-hare ratio, thereby intensifying a second interaction that extends the period of 
decline and drives the hare population still lower. Interregional synchrony ıs caused 
basıcally by mild winters that moderate mortality in peak hare populations, and permit 
others that are lagging to attaın peak densities. Such synchrony is reinforced by highly 
mobile predator populations”. Predators are thus not the driving force (see also FINERTY 
1980). This ıs also evidently true in the present study area, where the great predators played 
no role in the population dynamics of the hare population. Concerning the smaller 
carnivores and the predator-prey cycle, one would expect the predator cycle to follow the 
cycle in hare population. The figure shows that this was not the case. The local small 
predator populations were much more dependent on the small rodent populations for their 
nutrition (for further details, see PuULLIAINEN 1981b). 


Acknowledgements 


The author wishes to express his sincere gratitude to the staff of the Värriö Subarctic Research Station 
for their technical assistance. 

This paper constitutes Report No. 121 from the Värriö Subarctic Research Station of the 
University of Helsinki. 


Zusammenfassung 


Zur Selektion des Habitats und zur Fluktuation des Bestandes von Schneehasen (Lepus timidus) auf 
einem subarktischen Fjall im Waldlappland Finnlands 


Selektion des Habitats und Fluktuation des Bestandes von Schneehasen (Lepus timidus) wurden im 
Gebiet des Fjälles Värriö ım östlichen Teil des finnischen Waldlapplands in den Wintern 1968/69 bis 
1980/81 untersucht. Es ergab sich, daß der Hasenbestand zyklisch variierte. Die Dauer der Zyklen 
betrug mindestens vier Jahre. Bei geringer Bestandsdichte waren nur die bevorzugten Habitate (d.h. 
Mischwälder der niedrigen Gelände mit vorherrschendem Fichtenholz, in denen Birken und/oder 
Wacholder wuchsen) bewohnt, während in den Spitzenjahren Schneehasen auch in den weniger 
bevorzugten Habitaten, wie z.B. auf den windigen Abhängen der Fjällbirkenzone vorkamen. Be- 
standsverminderungen werden nicht durch Beutegreifer verursacht. In Jahren mit großer Bestands- 
dichte stieg die Bewegungsaktivität der Schneehasen während der Nahrungssuche im späten Winter 


erheblich an. 


References 


BARIKMO, ]J. (1976): Harens utnyhelse av björk som vinter föde. Inst. Nat. Cons. N. L. H., Norway. 
Trondheim: N. L. H. 

BRyANT, J. P. (1980): The regulation of snowshoe hare feeding behavior during winter by plant 
antiherbivore chemistry. In: Proc. Int. Lagomorph Conf., 1st., Guelph Univ., Canada, 1979 (in 
press). 


174 E. Pulhiainen 


BRYANT, J. P.; KuroPAT, P. J. (1980): Selection of winter forage by subarctic browsing vertebrates: 
The Role of Plant Chemistry. Ann. Rev. Ecol. Syst. 11, 261-285. 

DOLBEER, R. A.; CLark, W.R. (1975): Population ecology of snowshoe hares in the central Rocky 
Mountains. J. Wildl. Manage. 39, 535-549. 

ERLINGE, $. (1979): Adaptive significance of sexual dımorphism in weasels. Oikos 33, 233-245. 

FINERTY, J. P. (1980): The Population Ecology of Cycles in Small Mammals. New Haven and 
London: Yale University Press. 

GREEN, R. G.; Evans, C. A. (1940): Studies on a population cycle of snowshoe hares on the Lake 
Alexander area. I. Gross annual censuses, 1932-1939. J. Wildl. Manage. 4, 220-238. 

HELIMÄRT, U. 1. (1974): On the climate of Lapland, especially on the hills. Acta Lapponica Fenniae 8, 
23-27. 

HoweLı, A. B. (1923): Periodic fluctuations in the numbers of small mammals. J. Mammalogy 4, 
149-155. 

HÖRNFELDT, B. (1978): Synchronous Population Fluctuations in Voles, Small Game, Owls, and 
Tularemia in Northern Sweden. Oecologia (Berl.) 32, 141-152. 

KEITH, L. B. (1963): Wildlife’s Ten-Year Cycle. Madison: Univ. of Wisconsin Press. 

— (1974): Some features of population dynamics in mammals. In: Proc. Int. Congr. Game Biol., 
Stockholm. 

KEITH, L. B.; Topp, A. W.; BRAND, C. J.; ADamcık, R. $.; RuscH, D. H. (1977): An analysıs of 
predation during a cyclic fluctuation of snowshoe hares. In: Trans. XIIIth Congr. Game Biol., 
Atlanta. 

KEITH, L. B.; WINDBERG, L. A. (1978): A demographic Analysıs of the Snowshoe Hare Cycle. Wildl. 
Monographs 58, 1-70. 

LEoPoLD, A. (1933): Game Management. New York: Scribner’s. 

LinpLör, B.; Linpström, E.; PEHrRson, Ä, (1974a): On activity, habitat selection and diet of the 
mountain hare (Lepus timıdus L.) in winter. Viltrevy 9, 27-43. 

— (1974b): Nutrient content in relation to food preferred by mountain hare. J. Wildl. Manage. 38, 
875-879. 

PRUITT, W. ©. Jr. (1970): Some ecological aspects of snow. In: Proc. Helsinki Symposium (Unesco), 
Ecology of the subarctic regions. Paris: Unesco. 

PULLIAINEN, E. (1971): Winter nutrition of the snow hare (Lepus timidus) ın northeastern Lapland. 
Suomen Riııista 23, 72-79. 

— (1972): Nutrition of the arctic hare (Lepas timidus) in northeastern Lapland. Ann. Zool. Fennici 9, 
17-22. 

— (1974): Suomen suurpedot. Helsinki: Tammi 

— (1976): Birch forest damage caused by Oporinia autumnata Bkh. (Lep., Geometridae) in 
1965-1966 in eastern Itäkaira, northeastern Lapland. Ann. Et. Fennici 42, 166-170. 

— (1978): Wolverine - Our Most Endangered Carnivore. Suomen Luonto 37, 310-312, 344. 

— (1981a): Food and feeding habits of the pine marten in Finnish Forest Lapland in winter. In: 
Worldwide Furbearer Conference Proceedings. Ed. by J. A. CHapman; D. PursL£y. Frostburg: 
Worldwide Furbearer Conf., Inc. 

— (1981b): A 13-winter transect survey of small land carnivore populations on a subarctic fell ın 
Finnish Forest Lapland. Ann. Zool. Fennici 18, 270-278. 

PULLIAINEN, E.; IIVANAINEN, J. (1979): Young Arctic Hare killed by a Pygmy Weasel. Luonnon 
Tutkija 83, 151. 

PULLIAINEN, E.; IIVANAINEN, J.; LAITINEN M.; VAaRALA, K. (1980): A Wolf’s life amongst the 
reindeer. Suomen Luonto 39, 118-119, 143. 

SEQUEIRA, D. (1980): Comparison of the diet of the Red Fox (Vulpes vulpes L.) in Gelderland, 
Holland, Denmark and Finnish Lapland. Biogeographica 18, 35-51. 3 

SIIVONEN, L. (1948): Structure of short-cyclic fluctuations in numbers of mammals and birds in the 
northern hemisphere. Papers Game Res. 1, 1-166. 

Tamarın, R. H. (1978): Dispersal, population regulation and K-selection in field mice. Am. Nat. 112, 
545-555. 

WOoLFF, J. ©. (1977): Habitat utilization of snowshoe hares (Zepus americanus) in interior Alaska. 
Berkeley: University of California. 


Author’s address: Prof. Dr. ErKKı PuLLiaInen, Department of Zoology, University of Oulu, 
Kasarmintie 8, SF-90100 Oulu 10, Finland 


Quelques observations ethologiques sur ’Hydropote de Chine, 
Hydropotes inermis (Swinhoe, 1870) en captivite 


Par F. FEER 


Laboratoire d’Ecologie Generale du Museum, Brunoy, France 
Reception du Ms. 5. 10. 1981 


Abstract 


Some ethological observations on Hydropotes inermis (Swinhoe, 1870) in captıvity 


Studied the behaviour of Aydropotes inermis ın captıvity. Anımals use the vegetation as a hiding place. 
A particular ritual always preceeds resting. They live in relative isolation. Male relationships are 
characterized by mutual intolerance. Bonds are stronger among related ?2. Rivalty among dd is 
displayed through marking, threatening and fighting, mostly with their long canines. The coexistence 
of this archaic behaviour and of a more evoluted one such as the “Laufschlag“ in the sexual display 
characterizes the intermediate status of Aydropotes among Cervids. Though resembling other solitary 
Ruminants, A/ydropotes dısplays some of the signs of the group pulsion which is at the origin of 
gregarism. 


Introduction 


Nombre d’especes de Ruminants de petite taille habitent les zones forestieres ou, d’une 
maniere plus generale, les regions ou la vegetation est dense pres du sol. Ce type de milıieu 
impose certaines caracteristiques morphologiques et ethologiques que devaient posseder 
egalement les especes fossiles qui habitaient des biotopes comparables. Il semble que cela 
corresponde aux premiers stades de l’evolution de l’ordre des Artiodactyles vers les milieux 
ouverts. 

A ce titre, la connaissance de la biologie de ces petites especes de milieu dense est 
necessaire A une meilleure comprehension de l’evolution de l’ordre tout entier. Comme la 
plupart d’entre elles, AZydropotes inermis, orıginaire des zones mar&cageuses de Chine et de 
Coree, est mal connue. A cause de sa morpholigie originale et primitive (absence de bois, 
canines developpees, grande bulle osseuse, fosse lacrymale profonde et membres telemeta- 
carpiens), ’Hydropote forme ä& lui seul la sous-famille actuellement la plus archaique des 
Cervides, les Hydropotinae. Aucune &Etude de cette espece n’a Ete r£alisee sur le terrain et 
on. ne dispose que de quelques observations en captivit€ (ALLEN 1938-1940; KIRCHSHOFER 
1961; SCHERPE 1971). 


Materiel et methodes 


Un groupe de 2 mäles et 3 femelles a &t€ observ& pendant 2 moıs (Mai et Juin 1980) au parc zoologique 
de Wuppertal dans un enclos de 1300 m? comprenant une prairie centrale avec des massifs de 
vegetation herbacee dense entour&e d’arbres et d’arbustes. 

L’äge exact et les relations de parente des differents individus nous sont inconnus mais nous 
sommes assures que tous sont des adultes d’au moins 3 ans. 

Les 2 mäles ont Ete gardes dans le m&me enclos jusqu’ au 22 Mai, date ä laquelle ıl a fallu ısoler !’un 
d’entre eux, probablement le plus vieux, & la suite d’un combat avec l’autre mäle, au cours duquel il fut 
serieusement blesse. 

Les emplacements de repos et les lieux de marquages ont &t& cartographies pour pouvoıir analyser le 
mode d’utilisation de l’espace des differents groupes sociaux en fonction de la vegetation. Nous avons 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4703-0175 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 175-185 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


176 F. Feer 


arbitrairement choisi 5 metres comme distance interindividuelle en desa de laquelle nous avons 
considere qu’il existait une association entre 2 individus. 

Les frequences des differents marquages des 2 mäles quand ils Etaient ensemble dans l’enclos ont 
te calculees par des comptages successifs effectues sur des periodes pendant lesquelles ıl etait possible 
d’observer les anımaux en continu. La comparaison des pourcentages a &t& faite par le test de X?. 


R£sultats 
Comportement individuel 
Repos 


Nos observations sur le mode de deplacement, les positions de repos, les mouvements de 
coucher et de lever et les comportements de confort ne sont pas differentes de celles de 
SCHERPE (1971) sauf pour un comportement qui precede le coucher et qui n’a jamais te 
remarque& jusqu’icı. Avant de se coucher l’anımal flaire la place puis souleve en flechissant 
une patte anterieure. Il fait un mouvement de balancement du metacarpe vers l’avant et 
legerement sur le cöt€ comme s’il voulait donner un coup d’onglons dans le vide. Il procede 
ainsı alternativement des 2 pattes anterieures puis fait Ja m&me chose avec les 2 pattes 
posterieures qu’il souleve en flechissant puis repose. Apres avoir A nouveau contröle 
nasalement la place et fait un demi-tour sur lui-m&me, il refait parfois le m&me comporte- 
ment. Il est tres rare qu’il n’effectue pas ce comportement avant de se coucher. Il le fait 
aussi bien quand il se prepare A se coucher sur la pelouse rase que dans la vegetation haute, 
mais ı| semble que dans ce dernier milieu il leve les pattes plus haut. On constate alors que 
cela a pour effet d’Ecarter la vegetation ä l’endroit choisi par l’anımal. Nous pensons donc 
qu’il s’agit d’un comportement destine A preparer la place de repos dans la veg£tation tres 
dense, telle qu’il en existe dans les mar&cages qu’habite cette espece. 

Aucun comportement semblable n’a jamais Et€ observ& chez d’autres petites especes de 
milieux denses nı chez aucun autre Ruminant. Un certain rite avant le coucher existe chez la 
plupart des especes, mais l’anımal se contente alors souvent de sentir le sol, de plier 
plusieurs foıs un metacarpe en hesitant, parfoıs de se tourner legerement. 

La preparation de la place de repos par grattage du sol est connue chez quelques especes 
dont Moschus moschiferus (FRÄDRICH 1966). Ce comportement est probablement un 
marquage comparable ä celui effectu& dans d’autres situations. Il est possible que le 
pietinement de l’Hydropote, outre la preparation de la place de repos, serve egalement de 
marquage comme chez les especes precedentes, puisqu’il existe des glandes interdigitales 
aux pattes anterieures comme aux post£erieures. 

Le choix des emplacements de repos depend de la valeur protectrice de la vegetation. 
Les anımaux choisissent un couvert dans 90 % des cas (N = 344), avec une nette preference 
pour les zones oü la veg£etation herbacee est assez Elevee et dense pour les cacher 
completement. Ce besoin de sEcurite se traduit aussı par l’utilisation des troncs d’arbre, des 
murs de clöture ou des creux du sol. L’anımal se place toujours le dos ä l’obstacle, 
regardant dans la direction d’oü peut provenir un danger, ce qui est un comportement 
general chez les Ruminants. 

L’utilisation du couvert vegetal par ’Hydropote est tout & fait semblable ä celle qui 
caracterise d’autres Ruminants forestiers comme Hyemoschus aquaticus, Capreolus capreo- 
lus, Mazama, Muntiacus, Cervus axis, Cephalophus, Neotragus, ect.... 


Rythme d’activite 


Nos observations sur le rythme journalier confirment ce que d’autres auteurs ont deja 
observe. Il y a chez cette esp£ce une augmentation nette de l’activite le soir comme cela est 
frequent chez les Ruminants, mais pas de longue periode de repos en milieu de journee. 


Quelques observations ethologiques sur ’’Hydropote de Chine en captivite 1277. 


Cette activite en milieu de journ&e pourrait Etre düe a l’apport de nourriture A ce moment 
läl. Comme SCcHERPE (1971), nous pensons donc I’Hydropote doit &tre une espece forte- 
ment diurne dans son milieu naturel. Les conditions meteorologiques influencent le rythme 
d’activite. Pendant les journees chaudes, ıl y a un leger ralentissement de l’activite en milieu 
d’apres-midi. Avec des tempe£ratures fraiches, par contre, les anımaux £talent leur activite. 
On observe &galement des differences selon le sexe. Les mäles sont plus actifs que les 
femelles: 43 % de leur temps (N = 1283) contre 27 % pour ces dernieres (N = 247) de 6 
heures ä 22 heures. 


Vie sociale 


L’Hydropote est decrit comme “solitaire”’ en dehors de la periode de reproduction ou 
comme vivant en couple et en troupe familiale (HALTENORTH 1963). Cela signifie donc que 
nous ne connaissons rien de la vie sociale de cette espece dans la nature. Les seules 
observations disponibles concernent des individus obliges de vivre ensemble dans un 


Tableau 1 


Groupement des mäles et des femelles en pourcentage des observations: A/ aus repos; 
B/ en activite 


avec une ou plusieurs 9 9 Nombre d‘observations 


110 
91 


dud A 123 
378 


120 


en l’absence du d A 
RD 


espace limite. On constate que dans ces conditions les anımaux gardent une distance 
interindividuelle relativement grande par rapport aux especes plus gregaires. Cette ten- 
dance & l’isolement est particulierement nette chez les mäles au repos (tableau 1). Quand ils 
sont actifs, les anımaux ont au contraire une lEgere tendance A se regrouper, le plus souvent 
par deux (1 mäle avec 1 ou 2 femelles). 


Male = femelle 


Les rapprochements d’un mäle et d’une femelle sont certainement plus frequents pendant 
l’epoque du rut. En dehors de cette Epoque, les contacts observes sont rares et brefs. Le 
lechage des organes genitaux d’une femelle par un mäle n’a &t€ observ& que deux foıs. Le 
contröle urinaire avec “Flehmen” n’a jamais eEt€ observe. Le mäle dominant semblait avoır 
des contacts plus frequents que l’autre avec les femelles et plus specialement avec l’une 
d’entre elles: la seule femelle qui ait mis bas pendant la periode d’observation. Des liens 
privilegies d’un vieux mäle avec une femelle ont et& deja remarques par SCHERPE (1971) 
chez la m&me espece. Comme chez Muntiacus muntjak (Dusost 1970), ıls pourraient 
representer les indices d’une vie de couple rudimentaire. Les comportements agressifs d’un 
mäle vis-a-vis des femelles semblent indiquer la presence d’une hierarchie sexuelle. 


178 F. Feer 


Male = male 


Les deux mäles faisaient preuve entre eux d’une intolerance accusee. Elle se traduisait par 
des parades et des chasses permanentes par le mäle dominant jusqu’ä ce qu’un combat 
decisif ait necessite sa mise a l’Ecart. KIRCHSHOFER (1961) avait d&ja remarqu& ce pheno- 
mene et si SCHERPE (1971) ne l’a pas observ£, c’est que l’un des mäles &tait jeune. Comme 
nous l’avons vu, les mäles Evitent de se rapprocher au repos. Ils utilisent en commun une 
large zone de repos mais ne s’y trouvent ensemble que dans 10 % des cas (N = 69). La 
vegetation de cette zone £tait assez Epaisse pour que le contact visuel soit impossible. Cecı 
explique que nous y avons observ£ les plus petites distances interindividuelles. Le mäle 
dominant chasse l’autre mäle des meilleurs emplacements de repos, le plus abrites. Le mäle 


Fig. 1. Groupe de trois femelles au repos avec un mäle 


domine se couche donc ä decouvert dans 23 % des cas (N = 92) contre 10 % seulement 
(N = 110) pour le mäle dominant (P< 0,025). Quand ce dernier a Ete retire de l’enclos, le 
mäle precedemment domine ne se trouve couche ä decouvert que dans 15 % des cas 
(N = 123, P< 0,025). Le mäle dominant &Eloigne egalement son rival des femelles. Il occupe 
dans 57 % des cas (N = 110) 3 zones que les femelles frequentent 4 53 % (N = 233) (fig. 
1), tandıs que le mäle domine n’a acces qu’a une zone que les femelles frequentent rarement 
(9 % des observations). Une fois seul ıl frequentait bien plus souvent (42 % des observa- 
tions, N = 123) les zones que les femelles utilisaient beaucoup (56 % des observations, 
N =370). Une telle presence du mäle dans les zones communes des femelles existe 
egalement en semi-liberte chez une autre espece “solitaire”, Muntiacus muntjak (DuBost 
1970). 


Femelles = femelles 


Les femelles ont entre elles des contacts relativement plus frequents que les mäles. Elles ont 
tendance ä se regrouper temporairement surtout lors de la päture. Au cours de la recherche 
de la nourriture, il y a alternance plus ou moins rapide d’Eclatement et de resserrement du 
groupe des femelles parfoiıs accompagn&es d’un mäle. Pendant le regroupements il arrıve 


Quelques observations ethologiques sur !’Hydropote de Chine en captivite 179 


que 2 ou 3 individus se trouvent disposes en parallele et restent ainsi quelque temps 
(environ 30 secondes, maximum 5 minutes). Quand les femelles mangent des feuilles 
d’arbres, elles se trouvent parfois ensemble, l’une contre l’autre: la premiere femelle 
commengant ä manger avait alors attıre ä elle les autres femelles. Ce genre de comporte- 
ment a deja &t€ observ& par SCHERPE (1971) qui y voit une forme de comportement 
allomimetique analogue ä celui des Gazelles (WALTHER 1968). Quand un anımal du groupe 
se couche les autres restent un moment ä päturer autour de luı avant de se coucher eux- 
memes a cöte. 

Seule une femelle s’isolait plus souvent lors du repos, surtout avant la periode precedent 
sa mise bas. Elle se retrouvait alors 77 % du temps seule (N = 52), contre 51 % (N = 95) 
apres la naissance de ses jeunes (P < 0,005). Cette femelle avait des relations privilegiees 
avec une autre femelle. Ces deux anımaux pratiquaient des lechages (6 observations dont un 
lechage mutuel, contre O et 3 pour deux autres couples de femelles): sur le front , le tour de 
l’oeil, P’arrıere des oreilles et le cou; ils se frottaient egalement le front contre le cou de 
l’autre. Ils etaient plus souvent rapproches lors du repos (50 % des observations de couples 
de femelles au repos, N = 58). Par analogie avec ce qui a Ete observ& chez d’autres especes, 
comme Muntiacus, on peut supposer qu’il s’agissait de deux femelles parentes. La prolon- 
gation des liens mere-jeune chez les femelles est en effet un phenomene courant chez les 
Ruminants sociaux; elle est sans doute ä la base de la formation de groupes familıaux assez 
importants. D’apres SHERPE (1971) le lien mere-jeune chez l’Hydropote est plus solide que 
chez Muntiacus, ce qui aurait pour effet de renforcer les relations qui sous-tendent le 
groupe familial. Les naissances multiples (en captivite en general deux jeunes) ont sürement 
aussi une influence puisque le jeune vit, des le depart, dans des groupes familiaux plus 
importants que chez beaucoup d’autres especes. 

Bien que non constants, tous ces comportements prouvent l’existence d’une pulsion 
sociale qui est ä la base de la formation des groupes sociaux plus vastes que la cellule 
familiale stricte. En particulier, si nous avons remarque& un certain isolement chez la femelle 
prete a mettre bas, comme cela se remarque chez beaucoup de Ruminants gregaires, ıln’y a 
jamais eu d’indice de comportement territorial m&me ä cette &poque, contrairement donc ä 
ce qui a et& decrit chez Capreolus (KuRT 1968), Alces alces (ALTMANN 1959) et Cervus 
elaphus (BURCKHARDT 1958). Cette absence de territoire femelle en captivite va de pair avec 
une grande rarete des interactions agonistiques, contrairement donc aux mäles. 


Marquages 
Description des marquages 


Comme chez toutes les especes solitaires de milieu ferme&, les marquages olfactifs revetent 
une grande importance pour la communication entre les individus puisque ces derniers 
vivent disperses et ne sont que rarement en contact visuel. 

L’Hydropote possede une glande pre&-orbitaire peu marquee qu’il utilise tres rarement 
(1 seule observation de marquage pre&-orbitaire pour 1 mäle). Par contre, il marque 
beaucoup avec le front; on ne lui connait pourtant pas de glande frontale, mais il possede 
peut-Etre dans cette region un champ de glandes sudoripares non visibles ä l’oeil nu comme 
c’est le cas pour Odocosleus hemionus columbianus (Quay et MÜLLER-SCHWARZE 1970) et 
peut-Etre aussi pour Mazama (VOLKMAN et Raus 1973) et Pudu pudu (Obs. pers.). 

Apres un bref contröle nasal, le mäle frotte son front de 1 3 6 fois contre une branche ou 
un petit tronc vertical, a hauteur du garrot (fig. 2). Ce marquage bref peut &tre repete apres 
un 2eme contröle nasal. Il peut y avoir un räclage altern& du support par les incisives mais il 
est rare: 4 fois sur 72 observations. Les femelles se frottent aussi le front et parfois le bord 
des oreilles et l’arriere de la tete contre des branches pendantes. Le contröle nasal est rare et 
il a lieu apres. A cause de son manque de pre&cision, ce comportement ressemble davantage ä 


180 F. Feer 


un comportement de confort qu’ä un vrai marqua- 
ge. Il a et& rarement observ& chez les mäles. 
Les glandes interdigitales anterieures et poste- 


rieures pourraient avoır un röle de marquage com- | fe 

me nous l’avons vu lors du grattage du sol. Le mäle 3 g 

une somporteipent souvent seul (29 % des N U h Y 7 
cas, N = 1473447 %, N = 105), ou combine& avec la N a 
miction (35 % des cas, N=147337%, N=105, | N ’ 


la defecation (24% des ca, N=147 2 7%, | 9 % W 


N =105) ou les deux ä la foıs (12% des cas, 4 
N = 147 3 9,5 %, N = 105). Il flaire le sol, gratte 7 J 
plus ou moins longuement avec une patte ou les / 
deux alternativement puis avance de facon ä uriner 7 : 
et defequer sur la place grattee. Le grattage du sol et N in 
> ea 
son association avec la miction et la defecation n’a 
jamais et& observ& chez la femelle. Fig. 2. Marquage frontal du mäle 


Evolution du comportement de marquage entre les males 


La rivalite entre les deux mäles s’est exprim&e par une augmentation de la frequence des 
marquages et leur repetition au m&me endroit. Les marquages frontaux des deux mäles ont 
augmente mais sont restes egaux (tableau 2). Par contre la frequence des grattages du sol a 
surtout augmente chez le mäle dominant A (tableau 3). A la fın de la p£riode d’observation 


Tableau 2 


Evolution des frequences des marquages frontaux des mäles en nombre de marquages par heure 


(entre parenthöses: dur&e totale des comptages en heures) 


mäle A mäle B 


3.0 (3.25) 3.0 (4.25) 


6.0 (3.25) 6.6 (4.50) 
9.0 (5.50) 9.0 (5.00) 


Tablean 3 
Evolution des frequences des mictions, des defecations et des grattages du sol des mäles en nombre 


de comportements par heure 


miction defecation grattage dur&e des comptages 
en heures 


18.V. 


SB 
en l’absence du d A 
du 22 au 24.V. 
du 26.V au 2.VI. 


Quelques observations ethologiques sur ’’Hydropote de Chine en captivite 181 


les mictions du mäle A deviennent plus frequentes et elles sont proportionnellement plus 
souvent associees au grattage du sol (44%, N=77 462%, N=37, P<0,005). On 
observe le contraire pour le mäle B (79 %, N = 56 4 57 %, N = 44, P< 0,025). 

Apres le combat et la separation des mäles, le mäle restant diminue rapidement la 
frequence de ses mictions-Efecations et cesse completement de faire des grattages du sol et 
des marquages frontaux (tableau 3). 

Le mäle dominant se distingue €Egalement du domine puisqu’il superpose plus souvent 
ses propres marquages frontaux sur ceux de son rival (16 %, N = 78contre 1%, N = 89). 
Mais au plus fort de la concurrence entre les deux mäles, le nombre de marquages effectues 
par un mäle sur ceux de l’autre est presque identique chez les deux concurrents (83 %, 
N = 46 pour A et 97 %, N = 34 pour B). La proportion de ces sites utilises en commun 
dans la journee passe de 23% (N = 13) au debut de la periode d’observation 457 % 
(N = 35, difference significative P< 0,05) & la fin de la periode, oü apparaissent des series 
de marquages alternes plus ou moins rapproch&s sur ces sites communs. Il y a &galement 
des successions de grattages du sol et de mictions-defecations au m&me endroit, en 
particulier les places de repos par un seul mäle ou par les deux alternativement. 

Comme chez d’autres especes, les marquages revetent donc une grande importance dans 
les relations hierarchiques entre mäles. 


Comportement agonistique 


Les deux mäles que nous avons observ& se sont continuellement livres a une lutte dont la 
seule issue £tait l’elimination de l!’un d’entre eux par un combat decıisif. 

Le dominant chassait continuellement l’autre mäle en marchant simplement derriere luı 
sans attitude particuliere a une distance variant entre 3 et 10 metres. Cette “poursuite” au 
pas ne cessait que quand le mäle dominant s’arr£tait et se couchait ou quand le mäle domine 
prenait suffisamment de distance. Sı ce dernier se couchait le premier, le mäle dominant 
venait sur lui, le faisait lever a distance et reprenait la poursuite. 


Menaces 


Il existe cependant un certain type de menace utilise par l’anımal poursuivi quand il se 
retourne pour affronter son poursuivant, ou par ce dernier quand il charge. Le cou est &tire 
a l’horizontale, le t&te dans le prolongement et les oreilles tendues vers l’adversaire. 
L’animal se dirige d’un pas rapide ou au galop droit sur l’adversaire en Emettant une trille 
aigüe, la mächoire ouverte. Il peut y avoir entre les deux adversaires un Echange rapide de 
ces trilles, face a face & courte distance, ce qui debouche sur le combat ou la poursuite. Au 
cours de la poursuite quand le poursuivant est encore loın de l’adversaire et qu’il est tres 
excite, ıl gemit. 

Il existe une attitude de menace oreilles rejetees en arrıere semblable ä celle de beaucoup 
de Ruminants. Cette menace, moins serieuse que la precedente, n’a Et observ&e que chez 
un mäle vis-A-vis d’un groupe de femelles et chez une femelle qui essayait d’en faire lever 
une autre. 

Nous n’avons observe aucun comportement de menace qui puisse se rattacher & la 
menace tete basse, ce qui n’est pas Etonnant chez cette espece inerme. La menace t£te haute 
et la parade laterale (WALTHER 1960-1961) n’ont Egalement pas Et& observ£es, bien qu’elles 
soient repandues chez tous les Cervides. La menace tete haute a Et€ interpretee comme une 
intention de frapper avec la patte anterieure ou de mordre: ces deux comportements 
existent chez I’Hydropote mais ä un niveau agressif sup£rieur; et ne sont pas integres A la 
menace proprement dite. La parade laterale est une demonstration de toutes les capacites 
agressives de l’ındividu avec presentation des armes ä& l’adversaire. 

Chez I’Hydropote il ne semble pas y avoir de prelude ou combat par presentation des 


182 F. Feer 


canines comme cela est &galement observe chez Moschus (MÜLLER-UsInG et SCHLOETH 
1967). 


Soumissıon 


Dans le combat observ£, le vaincu s’est finalement couch& avec le cou et la tete complete- 
ment aplatis sur le sol dans l’attitude de complete soumission commune ä beaucoup de 
Ruminants. L’autre mäle ne le mordait que s’ıl avait r&ussı a luı soulever un peu la t£te. 
L’attitude completement couchee avait un effet inhibiteur de la morsure. 

L’Hydropote peut utiliser une patte anterieure seule lorsqu’il veut donner des coups sur 
la croupe ou le dos d’un individu couche pour le faire lever. Pour obtenir le m&me r&sultat il 
monte parfoıs avec les deux pattes anterieures sur le dos de l’anımal couche. Dans ce 
comportement le coup de patte est porte de haut en bas. Sı l’adversaire est debout les 
femelles effectuent un coup de patte anterieure de bas en haut, dans sa direction. Ce 
comportement ressemble ä un rudiment de “Laufschlag’ pratique par les mäles lors de la 
parade sexuelle (SCHERPE 1971). 


Combat 


Les deux pattes anterieures sont lancees en avant lorsque l’anımal veut frapper son 
adversaire avec les sabots lors de la poursuite. Cette forme primitive de combat se retrouve 
chez beaucoup de Cervides evolues inermes: femelles et meme mäles entre la chute des boıs 
et la perte du velours. 

Les femelles &changent des coups de front dans les flancs, sur le cou et la t&te et plus 
rarement entre les pattes posterieures. La morsure n’a jamais Et€ observ£&e chez les femelles 


u 


on. 


Fig. 3. Combat lateral avec morsure des mäles 


bien qu’il existe de courtes canines chez ce sexe. C’est par contre la base du combat chez les 
mäles. Les coups de dents sont donn&s de haut en bas par un rapide mouvement en crochet 
de la tete. Les combattants sont en position parallele inversee et se mordent au cou, & 
l’epaule ou & la croupe (fig. 3). Comme chacun essaye de derober son flanc et en m&me 
temps de mordre l’autre, les deux adversaires tournent en cercle. Ils essayent de se renverser 
avec de violents coups de front dans la region des Epaules, le mäle le plus fort en appuyant 
son cou sur le garrot de son adversaire pendant qu’il le mord sur le flanc. Le m&me “combat 
de cou’” avec morsure est connu chez d’autres Ruminants inermes ou ä cornes courtes: 
Moschus (KIRCHSHOFER 1961), Tylopodes (PıLTers 1956), Giraffa et quelques Bovides qui 
mordent (WALTHER 1960-1961). Des trilles de menace et quelques gemissements accom- 
pagnent le combat. 

Le combat de l’Hydropote est tres dangereux car les blessures ınfligees sont nombr- 
euses et profondes et le seul moyen d’echapper aux coups est la fuite. Il est probablement 
rare dans la nature ou les rencontres doivent &tre peu frequentes et oü le domine peut fuir. 

Avant le combat de&cisif, il n’a pas et€ observe de menaces avec la presentation 


Quelques observations ethologiques sur ’’Hydropote de Chine en captivite 183 


symbolique des armes. On ne peut cependant pas conclure ä l’absence de tels comporte- 
ments car ıls peuvent peut-£tre apparaitre chez des individus ayant une position hierar- 
chique relative differente de celle observee icıi. Pendant le combat ıl n’y a eu ni ceremonial 
ni stylisation des mouvements, visant ä le rendre plus symbolique, plus hierarchique et 
moins meurtrier. Le combat de I’Hydropote est analogue & celui de Hyermoschus aquaticns 
(Dusost 1975), Tragulus napn (Raııs et al. 1975), Muntiacus (DuBost 1970) et probable- 
ment ä ceux des autres Cervides mäles possesseurs de canınes (Moschus, Elaphodus). 

Le comportement agonistique de l’Hydropote correspond bien ä la morphologie 
rudimentaire et au type d’armes qu’il porte. Il est ainsı proche de celui des autres especes 
qui possedent €galement des canınes pro&minentes (Hyemoschus, Tragnlus, Moschus, Mun- 
tiacus, Elaphodus) et s’en servent dans un combat de type lateral, que l’on interprete 
comme le point de depart de l’Evolution du comportement de combat des Artiodactyles 
(WALTHER 1960-1961). Le comportement agonistique de ’Hydropote apparait egalement 
comme peu diversifi€ et peu stylise, ce qui peut Etre une caracteristique d’une espece 
primitive et d’un mode de vie solitaire. 


Discussion 


Differents niveaux Evolutifs apparaissent dans le comportement de l’Hydropote. Les plus 
archaiques comme la morsure et le combat de type lateral sont lies a son anatomie primitive 
(cf. WALTHER 1960-1961; FRÄDRICH 1966). Le combat sans menace nı stylisation est 
egalement archaique puisque les parades se compliquent chez les Artiodactyles au fur et ä 
mesure que les appendices frontaux se developpent. Mais l’absence de parade visant ä 
montrer les armes ä l’adversaire est peut-etre düe ä ce que les canines, tres longues, 
depassent la mandibule ce quı rendrait inutile un comportement demonstratif comme chez 
les especes dont les canınes sont nettement moins longues. Il subsiste un marquage par 
P’urine sans cer&monial visuel, qui peut Etre considere comme ancien puis quil est 
generalise dans tous les groupes de Mammiferes et qu’il ne fait pas intervenir de glandes 
bien specialisees comme le marquage preorbitaire. Le coup de patte lors du combat semble 
caracteriser un niveau moyen car ıl est absent chez les Suides et les Bovides; ıl est, par 
contre, courant chez la plupart des Cervides. Il en serait de m&me pour le combat de cou 
(WALTHER 1960-1961). Le coup de front des femelles est plus Evolue car on le trouve chez 
les Cervides et les Bovides. 

L’Hydropote semble £tre le seul Cervide ä effectuer le “Laufschlag” au cours de la 
parade sexuelle, un comportement largement repandu dans le groupe des Bovides. 

En raison de l’eventail des comportements presentes par I’Hydropote, on peut conclure 
qu’il se place A un niveau moyen ce qui est conforme ä la position de son groupe. 

L’Hydropote possede des caracteristiques Ethologiques qui le rapprochent des autres 
especes de Ruminants solitaires ou subsolitaires de milieu couvert. Cette &tude en amisen 
evidence quelques unes, ä savoir l’utilisation intense du couvert vege£tal, l’importance d’un 
systeme de communication ä distance (marquages olfactifs), la faiblesse des comportements 
demonstratifs et hierarchiques. Neanmoins, une legere pulsion de groupe existe chez les 
femelles; sı elle n’est pas düe aux conditions de captivite, elle pourrait £tre l’indice d’une 
tendance & la formation de troupes familiales de faible importance ou au moins d’une 
certaine tolerance chez les femelles. Les mäles adultes pourraient &tre, au contraire, 
disperses et territoriaux sur le domaine commun ä plusieurs femelles. 

L’Hydropote offrirait ainsi l’exemple d’un stade legerement superieur A ceux de 
Muntiacus et peut-etre de Neotragus (mäle territorial sur le domaine de plusieurs femelles) 
et se rapprocherait ainsi de celui de Ourebia ourebi, une autre petite espece de prairie: ce 
serait Ja premiere tape menant vers la structure sociale commune ä beaucoup de Rumi- 
nants sociaux: mäle territorial sur le domaine d’une troupe de femelles. Il conviendrait 
donc d’etudier l’organisation sociale de ’Hydropote dans la nature. 


184 F. Feer 


Remerciements 


Ce travail a Ete realıse gräce a une bourse du DAAD. Je remercie vivement le Dr. Haas et le Dr. 
SCHÜRER auxquels je dois aide et accueil au Zoo de Wuppertal. Je remercie enfin G. Dusosrt et A. 
HEYMER pour leurs critiques et leurs conseils. 


Resume 


Le comportement de deux mäles et de trois femelles d’A7ydropotes inermis a ete observ& en captivite au 
Zoo de Wuppertal (RFA). 

L’Hydropote est une espece de milieu couvert qui recherche la protection de la vegetation pour le 
repos. Il existe un rite precedant le coucher qui sert a preparer la place de repos et peut-£tre a la 
marquer. 

Les individus vivent relativement isoles. Des liens privilegies d’un mäle avec une femelle pourrait 
representer les indices d’une vie de couple rudimentaire. Les mäles font preuve entre eux d’une 
intolerance accusee. Le mäle dominant carte son rıval de la proximite des femelles et des zones de 
repos les meilleures. Les femeles ont tendance ä se regrouper quand elles sont actives et des liens 
sp&ciaux semblent exister entre femelles parentes. Ces comportements sont a la base de la formation de 
groupes familıaux. 

Les mäles effectuent des marquages avec le front et par grattage du sol combin€ ou non avec la 
miction-defecation. Leur rivalit@ se manifeste par une augmentation de la frequence des marquages 
mais surtout des menaces et des combats serieux dont les armes sont leurs tr&s longues canınes. 

L’ensemble des comportements de I’Hydropote situe donc cette espece dans une position moyenne 
au sein des Cervides: ä cöt& des comportements archaiques comme le combat de type lateral on trouve 
en effet des comportements plus €volues, comme en particulier le “Laufschlag” de la parade sexuelle 
du mäle. 

L’Hydropote possede des caracteristiques Ethologiques propres aux Ruminants solitaires de 
milieux couverts. Mais on discerne dans sa vie sociale des rudiments de pulsion de groupe qui sont A 
l’origine du gregarisme des especes plus Evoluees. 


Zusammenfassung 


Einige ethologische Beobachtungen am chinesischen Wasserreh, Hydropotes inermis (Swinhoe, 1870), 
in Gefangenschaft 


Das Verhalten von 2 dd und 3 2? des chinesischen Wasserrehes konnte während zweier Monate 
(Mai und Juni 1980) im Zoo Wuppertal beobachtet werden. Das Wasserreh sucht für den Ruheaufent- 
halt die Deckung der Vegetation, und es zeigt ein dem Niederlegen vorausgehendes Ritual, welches 
vielleicht auch der Liegeplatzmarkierung dient. Die Tiere leben relatıv isoliert, bevorzugte Beziehun- 
gen eines d mit einem bestimmten ? könnten Indizien für ein rudimentäres Paarleben sein. Die dd 
untereinander sind außerordentlich intolerant. Das dominante & hält seinen Rivalen von den 2% fern 
und vertreibt sie auch aus der Nähe der besten Ruheplätze. Die ?? zeigen eine Tendenz zur 
Gruppenbildung, außerdem scheinen bevorzugte Beziehungen unter verwandten 2? zu bestehen. 
Dies sınd Verhaltensweisen an der Basıs einer Familiengruppenbildung. 

Die &d markieren mit der Stirn und durch Bodenkratzen, kombiniert mit oder ohne Miktion- 
Defäkation. Ihre Rivalität untereinander manifestiert sich durch Zunahme der Markierungsfrequenz, 
besonders aber auch durch Drohen und Ernstkampf, wobei die sehr langen Eckzähne als Waffe 
eingesetzt werden. 

Die Summe aller Verhaltensweisen geben dem Wasserreh eine Mittelstellung unter den Cerviden. 
Neben ursprünglichen Verhaltensweisen wie Lateralkampf findet man auch solche, die als höher 
entwickelt gelten: z. B. der Laufschlag des d während der Werbung. Das Wasserreh ist nach den 
ethologischen Merkmalen ein solitärer Wiederkäuer geschlossener Biotope. Man findet aber ım 
Sozialleben Rudimente eines Gruppenstrebens, welche als Ansatz zum Herdenleben höher entwickel- 
ter Arten angesehen werden können. 


Bibliographie 


ALLEN, C.M. (1938-1940): The Mammals of China and Mongolia. American Mus. of Nat. Hist., New 
Mork, 11,1 et22. 

ALTMAnN, M. (1959): Group dynamics in Wyoming moose during the rutting season. J. Mammalogy 
40, 420-424. 

BURCKHARDT, D. (1958): Observations sur la vie sociale du Cerf (Cervus elaphus) du Parc National 
Suisse. Mammalıa 22, 226-244. 


A note on the sexnal behaviour in red deer hind 185 


DusosT, G. (1970): L’organisation spatiale et sociale de Muntiacus reevesi Ogilby, 1839, en semi- 
liberte. Mammalia 34, 331-335. 

Dusost, G. (1975): Le comportement du Chevrotain afrıcain, yemoschus aquaticus Ogilby (Artio- 
dactyla, Ruminantia). Sa signification Ecologique et phylogenetique. Z. Tierpsychol. 37, 403-501. 

FrRÄDRICH, M. (1966): Einige Verhaltensbeobachtungen am Moschustier (Moschus moschiferus L.). 
Zool. Garten 33, 65-78. | 

HALTENORTH, TH. (1963): Klassifikation der Säugetiere: Artiodactyla. In: Kükenthal, Zool. 8, 32, 1 
(18). Berlin: de Gruyter. 

KIRCHSHOFER, R. (1961): Hirschferkel und andere primitive Hirsche. Kosmos (Stuttgart) 57, 76-81. 

Kurr, F. (1968): Das Sozialverhalten des Rehes (Capreolus capreolus L.) Mammalia depicta. Hamburg 
und Berlin: Paul Parey, 102 p. 

MÜLLER-Usıng, D.; SCHLOETH, R. (1967): Das Verhalten der Hirsche (Cervidae). In Kükenthal, Hb. 
Zool. 10, (28) 1-60. Berlin: de Gruyter. 

Pırters, H. (1956): Das Verhalten der Tylopoden. In Kükenthal, Hb. Zool. 8, 10 (27), 1-24. Berlin: 
de Gruyter. 

Quay, W. B., MÜLLER-SCHWARZE, D. (1970): Functional histology of integumentary glandular 
regions in black-tailed deer (Odocozleus hemionus columbianus). J. Mammalogy 51, 675-694. 
Rauıs, K.; BArAsH, CH.; Mınkowski K. (1975): Behaviour of captive mouse deer, Tragulus napn. Z. 

Tierpsychol. 37, 356-378. 

SCHERPE, W. P. (1971): Beobachtungen am Verhalten des chinesischen Wasserrehs (Aydropotes 
inermis Swinhoe), des indischen Muntjaks (Muntiacus muntjak muntjak Zimmermann) und des 
chinesischen Muntjaks (Muntiacus muntjak reevesi Ogılby). Univ. Berlin, 762. 

VOoLkMman, N.; Raus, K. (1973): Scent Marking in the Red Brocket, Mazama americana. Zoologica 
58, 55-56. 

WALTHER, F. (1960-1961): Entwicklungszüge im Kampf- und Paarungsverhalten der Horntiere. Jb.d. 
G. v. Opel-Freigehege für Tierforschung 3, 90-115. 

WALTHER, F. (1968): Verhalten der Gazellen. Die neue Brehmbücherei 373, Wittenberg-Lutherstadt: 
A. Ziemsen. 


Adresse de l’auteur: FRANCOIS FEER, Attach& de Recherches au CNRS, Laboratoire d’Ecologie du 
Museum d’Histoire Naturelle, 4, Avenue du Petit Chateau, F — 91800 Brunoy 


NZISSEINSCTEAEFEICHE KURZMITTEILUNGEN 


A note on the sexual behaviour in red deer hind 


By L. BarTos 


Department of Genetics, Research Institute of Anımal Production, Prague, Czechoslovakia 


Receipt of Ms. 18. 1. 1982 


The following observation was made on a semi-tame socıally stabilized “white” red deer 
population in Zehusice game reserve, Czechoslovakia. Additional details are described 
elsewhere (BArTos 1981). 

At the beginning of the rutting season during the autumn 1979, the alpha stag was 
immobilized and removed. This resulted in a big social friction among the remaining deer. 
The beta stag and one hind were injured and died. The top hierachy position was then 
occupied by a 6-year-old stag, socially and probably sexually unexperienced. The observed 
behaviour occurred during the first part of the rutting season. At 0900 one hind was 
noticed as she bounded away at great speed, merely circling and returning to the stag if he 
had not followed her. However, the stag showed no tendency to keep within the proximity 
of the hind. The hind then solicited the stag by trotting past him with her head low and 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4703-0185 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 185-187 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


186 L. Bartos 


neck extended while making a characteristic chewing action with her mouth open (comp. 
Guinness et al. 1971). The same posture was performed towards the approaching stag 
when the hind was lying. In this situation preorbital glands of both animals were open. 
Within 2 h the hind was seen three times running towards the stag, suddenly stopping in 
front of him, and returning around and running away. Her steps were shorter than usual 
when running. At 1125 h the hind followed by her calf left the herd to the distance of 
approximately 200 m. The stag tried to “herd” her back, but when she ignored him, he 
gave up and followed her, although the herd was immediately joined by another stag, 
displaying full rutting behaviour except roarıng. After 30 min of standing on one place, the 
hind moved back followed by her calf and the stag. The stag displayed no aggressiveness 
towards the other male, who then left the herd. The whole herd then rested. At 1530 h the 
deer started to graze and the hind again became active. She repeatedly “attacked’”’ the stag, 
running fastly towards him in the same manner as described above. When she stopped in 
front of the stag, she almost squatted, before she darted away. When she came close to the 
stag he made a sudden dash towards her, flicking out his tongue with head streched 
forwards and nose directed towards her tail. The hind invarıably retreated quickly from 
these advances with her taıl sometimes raised in the air. The stag then chased the hind 
approximately 20 m, after giving a series of load roars. The stag placed his chin on her rump 
or licked her vulva during the chases. During one of the chases the hind came to a sudden 
halt and stead with her back slightly arched, ears back and tail raised. The stag then 
mounted with penile protrusion and erection. Mounts usually lasted approximately 1 to 5 
seconds, but the female avoided the male’s intromission by lowering her hindquarters and 
running out from beneath him. T'welve mounts occurred within a 1.5 h interval. After the 
13th mounting, the stag achıieved intromission and after a few mild pelvic thrusts, he 
lunged violently upwards from his hind quarters so that his legs left the ground and hıs 
body took in an almost vertical position. The force of his ejaculatory thrust pushed the 
hind forwards a few steps and the stag dismounted. The hind’s calf usually followed the 
chases of the pair and was barkıng, but was ignored by the hind (comp. GALLIA 1966). 
After copulation the hind stayed in a typical postcopulatory posture (GUINNESS et al. 1971) 
for 20 s, and then went to her calf and nursed it. During the last part of the described 
behaviour the stag completely lost control of the herd of hinds, chasıng off spikers only 
and ignoring adult subordinate stags. One of the other stags separated a small group of 
hinds, and the rest of hinds dispersed on a large space. The next morning, however, the 
herd was again integrated. 

The described active precopulatory behaviour of the red deer hind ıs in a good 
agreement with other reports ın red deer (Guinness et al. 1971; Tayror 1973), elk 
(Morrıson 1960), chital (SCHALLER 1967), carıbou (BERGERUD 1974), and brow-antlered 
deer (BLAKESLEE et al. 1979). YAHNER (1979) suggested that multiple mating bouts in 
muntjac are determined by the courting male. However, in red deer, during using an 
artificial vagina for semen collection (KrzıwınsKI and JaczEwskı 1978) the stag’s ejacula- 
tion was achieved usually just seconds after the beginning of the first mounting attempt 
(Krzıwinskı 1979, pers. comm.). Hence it seems in the red deer (Lucas 1973), sımilary as 
in other Cervids as in fallow deer (Espmark and BRUNNER 1974), that the female 
determines the precopulatory pattern, as well as the choice of sexual partner (MORRISON 
1960; Lucas 1973; BArTos 1981). Nevertheless, she usually chooses the dominant indi- 
vidual (LincoLn and Guinness 1977). 

The knowledge of the female deer sexual behaviour is usually based on incidental 
observation only. The observed behaviour reminds a well documented precopulatory 
display of some female laboratory species (e. g. MapLarousek and HLinAk 1977). It is 
suggested that similarly as in these species there might exist plastic changeable female 
ability in deer, which may gradually increase the soliciting impression of the precopulatory 
behaviour. Also BLAKESLEE et al. (1979) have stated that the females of most species have 


Mausschläfer (Myomimus Ognev, 1924) in Afghanistan 187 


the capacity for some form of proceptive behaviour (BEACH 1976) and they demonstrate 
the behaviour “when they must encourage more vigorous sexual behavior from a male who 
is inexperienced, disinterested, tired, old or lethargic”. Moreover, by means of this 
behaviour a female may stimulate and time her internal environment for an optimal stage 
for insemination. The chewing action of the soliciting hind’s open mouth also was 
observed out of rutting season. It was usually performed by a female as she approached a 
feeding place by a stag. This suggests a general active manifestation of submissiveness of a 
hind towards a stag. 
References 


BARTOoSs, L. (1981): Reproductive and social aspects of the behaviour of “white” red deer. Säugetierkd. 
Mitt. (in press). 

BEACH, F. A. (1976): Sexual attractivity, proceptivity and receptivity in female mammals. Horm. 
Behav. 7, 105-138. 

BERGERUD, A. (1974): Rutting behaviour of Newfoundland carıbou. In: The behaviour of ungulates. 
Ed. by V. Geist and F. WALTHER. ICUN Publs. n. s. Vol. I, 345-395. 

BLAKESLEE, C. K.; Rıce, C. G.; Ras, K. (1979): Behavior and reproduction of captıve brow- 
antlered deer Cervus eldı thamın (Thomas, 1918). Säugetierkde. Mitt. 27, 114-127. 

ESPMARK, Y.; BRUNNER, W. (1974): Observations on rutting behaviour in fallow deer, Dama dama 
(Linne, 1758). Säugetierkde. Mitt. 22, 135-142. 

GALLIA, G. (1966): Roe polygamy. Deer News 1, 18-19. 

Guinness, F.; LincoLn, G. A.; SHORT, R. V. (1971): The reproductive cycle of the female red deer, 
Cervus elaphus L. J. Reprod. Fert. 27, 427-438. 

KRZIwINsKI, A.; JACZEwsK1ı, Z. (1978): Observations on the artificıal breeding of red deer. Symp. 
Zool. Soc. Lond. 43, 271-237. 

Lıncorn, G. A.; Guinness, F. (1977): Sexual selection in a herd of red deer. In: Reproduction and 
Evolution. Ed. by. J. H. CAaragy; C. H. TynDALE-Biscoe. Canaberra City: Aust. Acad. Scı. 
33-38. 

Lucas, C. J. (1973): The world of park deer. Deer 2, 988-990. 

MADLAFOUSER, J.; HıınAx, Z. (1977): Sexual behaviour of female laboratory rat - invertory, 
patterning, and measurement. Behaviour 63, 129-174. 

MOoRRISoNn, J. A. (1960): Characteristics of estrus in captive elk. Behaviour 24, 84-92. 

SCHALLER, G. B. (1967): The deer and the tiger: a study of wildlife in India. Chicago: Univ. Chicago 
Press. 

TAyLoRr, K. (1973): The courtship of stag and hind. Deer 2, 1010. 

YAHNER, R. H. (1979): Temporal patterns in male mating behaviour of captive reeve’s muntjac 
(Muntiacus-reevesı). J. Mammalogy 60, 560-567. 


Author’s address: LuDEK BARTOS, Department of Genetic, Research Institute of Anımal Production, 
CS-251 61 Praha 10-Uhrineves 


Mausschläfer (Myomimus Ognev, 1924) in Afghanistan 


Von J. NIETHAMMER 


Eingang des Ms. 8. 3. 1982 


Schädelreste kleiner Säugetiere, die ich am 10. April 1965 etwa 30 km südlich von Herat in 
Afghanistan vom Boden einer Höhle auflas, enthielten neben anderen Arten (Tab. 1) auch 
die Reste von mindestens 4 Schläfern. Als ich die von mir Dryomys zugeordneten 
Fundstücke Herrn R. Daams, Utrecht, schickte, erkannte dieser, daß es sich dabei um 
Myomimus handelte (Daams 1981). Vorhanden sind 4 rechte und 4 linke zahnlose, zum 
Teil beschädigte Mandibeln und je ein rechter und linker Molar, vermutlich M,. Zwar 
ähneln die Unterkiefer denen von Dryomys, doch zeigen der kräftigere und höhere 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4703-0187 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 187-189 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


188 J. Niethammer 


Kronenfortsatz (Abb. 1a), das Tabelle 1 
einfachere Kauflächenmuster der 
Molaren (Abb. 1b) und beson- 
ders die Zahl von jeweils 3 Al- 
veolen der unteren Molaren 


Artzugehörigkeit der 1965 bei Herat in Afghanistan in 
einer Höhle gefundenen Säugetier-Reste 


(Abb. 1c) die Zugehörigkeit zu Art Nldeszeul 
Myomimus. Die Unterkiefermo- BEN D 25 
laren von Dryomys besitzen nur 2 Otonycteris hemprichi 2 
Wurzeln. Außerdem sınd Kiefer Lepus capensıs 1 
und Zahnreihen der afghanischen Ocporona mejzcens \ 
Myomimus kleiner als solche af- Mo # 
IH Alactaga elater 5 
ghanischer Dryomys (Tab. 2). Meriones (verschiedene Arten) 6 
Auch die beiden M3 sind verhält- Cricetulus migratorins 6 
nısmäßig klein. Maße (Länge X Calomyscus bailwardı : 
N: a Nesokia indica 1 
Breite ın mm): 1,12x0,84 un Microtus (Blanfordimys) afghanus 2 
N: Ellobius fuscocapillus 1 
Die Maße passen zu denen 


von Myomımus _personatus 

Ognev, 1924, dessen nächster Fundort bei Kutschan etwa 500 km nordwestlich von Herat 
liegt (RossoLımo 1976a, Abb. 2). Freilich schließen die spärlichen Reste eine Zugehörig- 
keit zu Myomimus setzeri Rossolimo, 1976 nicht aus, der aus den Provinzen Kurdistan und 
Luristan ım Iran beschrieben wurde (RossoLımo 19766). 


Tabelle 2 


Mandibellänge (Vorderrand ohne Schneidezahn bis Hinterfläche des Processus articularis) und 
Alveolarlänge der unteren Backenzahnreihe (P, - M,) der 4 linken Mandibeln der Myomimus 
von Herat 


Zum Vergleich die Maße von 2 adulten Dryomys nitedula aus Ost-Afghanistan (Paghman bei Kabul 
und Salang-Paß) 


Myomımus Dryomys 
Mandibel Zahnreihe Mandibel Zahnreihe 


mm mm 


355 4,2 
33 2) 
3,5 
3,6 


Tab. 1 enthält durchweg Arten, die von Myomimus abgesehen rezent aus Afghanistan 
nachgewiesen sind. Allerdings sind zwei von ihnen, der Pfeifhase Ochotona rufescens und 
die Wühlmaus Microtus afghanus bisher nur aus größeren Höhen bekannt. Der Herat 
nächstgelegene, bisher bekannte Fundort ist der Sauzak Kotal, ein 2400 m hoher Paß ım 
Nordosten von Herat (HassınGEr 1973). Dagegen liegt der Fundort der Myomimus-Reste 


Abb. 1. Fundstücke von Myo- 
mimus 30 km südlich von He- 
rat, Afghanıstan a= linke 
Mandibel von buccal; 
b = Aufsicht auf linken, unte- 
ren Molaren, vermutlich M,; 
c=linke, untere Alveolen; 
rostral ist oben im Bild 


Mausschläfer (Myomimus Ognev, 1924) in Afghanistan 189 


BES iR 60° 


Abb. 2. Fundorte von Myomi- 
mus personatus (schwarze 
Kreise), dem Myomimus- 
Fund aus Afghanistan (Kreuz) 
und M. setzeri (weiße Kreise). 
P=RamesKassır, 2= Kara 
Kala am Sumbar-Fluß, Turk- 
menien, UdSSR, 3 = 90 km 
westlich von Kutschan, 
Nordiran (RossoLıMmo 1976a), 
4 = 30 km südlich von Herat, 
Afghanıstan, 5=Baneh, 6 = 
Sanandal in Kurdistan, 7= 
Borudscherd, Luristan, Iran 
(RossoLımo 1976b) 


bei nur 1000 m. Zwar können Ochotona und Microtus afghanus in ihrer Häufigkeit 
erheblich fluktuieren. Trotzdem ist es unwahrscheinlich, daß sie in den vergangenen 
Jahrzehnten in unmittelbarer Nachbarschaft der Fundhöhle gelebt haben. Zwar handelt es 
sich vermutlich um Fraßreste des Uhus (Bubo bubo), die aus weiterer Entfernung 
herbeigetragen worden sein können. Unwahrscheinlich ist das aber bei einem relativ hohen 
Anteil der Beutetiere. Daher vermute ich, daß die Reste einige Jahrhunderte alt sind. Sie 
können aus dem Mittelalter stammen, als offenbar ein feuchteres Klima herrschte, wie es 
die Existenz zahlreicher verlassener Ortschaften in den Trockengebieten Afghanıstans 
vermuten läßt. 


Literatur 


Daams, R. (1981): The dental pattern of the dormice Dryomys, Myomimus, Microdryomys and 
Peridyromys. Utrecht Micropaleontological Bull., Special Publ. 3, 1-115. 

HAsSINGER, J. (1973): A survey of the mammals of Afghanistan resulting from the 1965 Street 
expedition (excluding bats). Fieldiana Zool. 60, 1-195. 

RossoLıMo, ©. L. (1976a): Taxonomic status of the mouse like dormouse Myomimus (Mammalıa, 
Myoxidae) from Bulgaria. Zool. Z. (Moskva) 55, 1515-1525. 

— (1976b): Myomimus setzerı (Mammalia, Myoxidae) a new species of mouse-like dormouse from 
Iran. Vestn. Zool. (Kiev) 10, 51-53. 


Anschrift des Verfassers: Prof. Dr. JoCHEN NIETHAMMER, Zoologisches Institut, Universität Bonn, 
Poppelsdorfer Schloß, D-5300 Bonn 


BUCEHBESPIRECERNUINIGEN 


ZÖRNER, H.: Der Feldhase. Die Neue Brehm-Bücherei 169. Wittenberg-Lutherstadt: 
A. Ziemsen Verlag 1981. 172 S., 48 SW-Fotos, 12 Diagr. DM 15,70. 


Nachdem man bis vor kurzem in einer Fachbibliothek vergeblich nach einer zusammenfassenden 
deutschsprachigen Darstellung über die Biologie des Feldhasen gesucht hätte, sind nun innerhalb 
kurzer Zeit gleich zwei Beiträge zu diesem Thema erschienen: 1978 das Buch von SCHNEIDER und 
1981 in der Neuen Brehm-Bücherei das vorliegende von ZÖRNER. Die Autoren beider Bücher haben 
eigene Forschungsarbeiten am Feldhasen mit unterschiedlichen Schwerpunkten durchgeführt: 
SCHNEIDER war mit ethologisch orientierten Freilandarbeiten befaßt, ZÖRNER vor allem mit Untersu- 
chungen zur Populationsstruktur und Fortpflanzungsleistung anhand großer Hasenstrecken. Grund- 
lage des neuen Brehm-Heftes - das an die Stelle des 1956 erschienenen Heftes von KoEnen über den 
Feldhasen tritt - ıst eine 1978 angefertigte Dissertation, deren Ziel es war, ‚‚dem interessierten Laien, 
Jäger und Wildbiologen neues Wissen über die Biologie und Lebensweise, das Verhalten und die 
Bejagung des Feldhasen zu vermitteln.“ Entstanden ist ın der Tat eine umfassende Darstellung über 
den Europäischen Feldhasen, in der neben den rein biologischen Aspekten wie Lebensweise, Verhal- 
ten, Altersstruktur, Lebensdauer, Fortpflanzung, Geschlechterverhältnis, Mortalität, Ernährung und 
Feinde einleitend auch die Stammesgeschichte, Systematik und Verbreitung, anschließend die Mor- 
phologie und Anatomie abgehandelt werden. Den Schluß der Ausführungen bilden Hinweise zur 
Bewirtschaftung, Jagd und Hege. Der Text wird durch fotografische Aufnahmen ergänzt, die den 
Feldhasen in verschiedenen Verhaltenssituationen zeigen. Zahlreiche Tabellen und Diagramme infor- 
mieren rasch über die quantitativen Untersuchungsergebnisse wie Körpergröße und Körpergewicht, 
Schädellänge und Gewicht innerer Organe, Geburtsgewicht und Wachstumsverlauf, Altersstruktur, 
Sexualproportionen, Testes- und Ovarıen-Gewicht, Zahl der Corpora lutea, Embryonenzahl, Wurf- 
größe u. a. Über 330 Literaturnachweise zeugen vom Bemühen des Autors, eine möglichst umfassende 
Abhandlung vorzulegen, die neben den Ergebnissen eigener Untersuchungen auch sonst alles Wis- 
senswerte über dieses zoologisch interessante und wirtschaftlich bedeutsame Säugetier enthält. Der 
Autor scheut nicht vor kritischen Auseinandersetzungen mit überholten Vorstellungen zurück, so 
etwa, wenn es darum geht, die Bedeutung der Marderartigen, der Hauskatzen und der Greifvögel und 
anderer Räuber für die Feldhasenverluste auf das wirklich gesicherte Maß zu reduzieren. Auch die auf 
STIEvE zurückgehenden Angaben zur Superfötation werden kritisch beleuchtet: „STIEvE (1952) will zu 
allen Zeiten der Trächtigkeit reifende oder fast sprungreife Follikel im Ovar gesehen haben. Die 
eigenen Untersuchungen ergaben aber, .. .““ und etwas später heißt es: „Das Auffinden von Embryo- 
nen verschiedener Größe kann ebenso wie das Überschneiden zweier Tragzeiten oder das Aufeinan- 
derfolgen zweier Trächtigkeiten ohne dazwischenliegende neuerliche Begattung auf einer verspäteten 
Nidation, einer verzögerten Entwicklung einzelner Keime oder aber dem Tod einzelner Früchte 
beruhen. So kann eine Superfoetation vorgetäuscht werden.‘ Es bedarf keiner besonderen Betonung, 
daß dieses neue Brehm-Heft in jede säugetierkundliche Fachbibliothek gehört, aber auch Wildbiolo- 
gen, Jäger, Biologielehrer und interessierte Laien werden aus der Lektüre Gewinn ziehen. Man muß 
dem Heft nicht erst eine weite Verbreitung wünschen, der niedrige Preis wird ein übriges dazu 
beitragen. H. REICHSTEINn, Kiel 


Pucekx, Z. (ed.): Keys to vertebrates of Poland: Mammals. Warszawa: Polish Scientific 
Publishers 1981. 367 pp.; figs. ISBN 83-01-02553-0 


In einer 2. Auflage ist dieses Bestimmungsbuch für Säugetiere in Polen erstmals in englischer Sprache 
erschienen. Es basiert auf einer polnischen 1. Auflage von Kowauskı vor 15 Jahren, ist aber 
modernisiert und überarbeitet. Mehrere Autoren teilen sich die Aufgaben. Auf den ersten 50 Seiten 
sind in einem Kapitel von Z. Pucek allgemeine Kennzeichen der Säugetiere abgehandelt; des weiteren 
finden sich Instruktionen zum Anlegen wissenschaftlicher Sammlungen, für Vermessungen, Präpara- 
tion und Fixierung kleinerer Arten, Häuten größerer und zur Aufbewahrung. Auch die benutzten 
Schädelmafe werden angegeben. Der Hauptteil beginnt mit einer systematischen Übersicht und dem 
Bestimmungsschlüssel zur Identifikation von Ordnungen, und zwar an äußeren Merkmalen des 
Tierkörpers und speziell am Schädel. Entsprechende Angaben findet man bei weiterführenden 
Schlüsseln zu den Familien, Gattungen und Arten. 111 Spezies aus den Ordnungen Insectivora, 
Chiroptera, Lagomorpha, Rodentia, Cetacea, Carnivora, Pinnipedia, Perissodactyla und Artiodactyla 
werden behandelt. Darunter sind solche Arten besonders gekennzeichnet, die nur sporadisch auftre- 
ten, gleichzeitig andere, deren Vorkommen in Polen bislang nicht nachgewiesen aber möglıch ist, und 
schließlich solche, die in historischer Zeit ausgestorben sind. Zu den Arten findet man Beschreibungen 
über anatomische Merkmale, biologische Besonderheiten und Vorkommen. Die Ausführungen sind 
durch viele Zeichnungen und einige Verbreitungskarten veranschaulicht. Das Buch ist sicherlich für 
Säugetierkundler von Interesse, insbesondere deshalb, weil es in englischer Sprache einem größeren 
Kreis in Europa verständlich ist. D. KruskA, Hannover 


Buchbesprechungen 191 


MiLıtTzer, K.: Haut und Hautanhangsorgane kleiner Laboratoriumstiere. Teil 1: 
Morphologie. Schriftenreihe Versuchstierkunde Bd. 9. Berlin - Hamburg: Paul Parey, 
1982. 174 S., 27 Abb., 28 Tab., DM 84,-. 


Zusammenfassende Darstellungen zur Morphologie der Haut und ihrer Anhangsorgane sind für die 
Säugetiere im wesentlichen auf den Menschen beschränkt. Eine Vielzahl von artspezifischen Unter- 
schieden in Bau und Funktion des Integuments erschweren allerdings für diese Tiergruppe eine 
umfassende vergleichende Betrachtung erheblich. Für kleine Laboratoriumstiere wird in der vorlie- 
genden Arbeit zum erstenmal eine Übersicht der zwar recht zahlreichen, aber auch verstreuten 
Literatur zur Morphologie der Haut gegeben. 

Der in drei Hauptabschnitte gegliederte Band bietet für allgemeine Fragen als auch für tierartspezi- 
fische Detailprobleme eine gute Informationshilfe. Der erste, allgemeine Teil behandelt die grundle- 
genden Aspekte der Haut- und Haarmorphologie, wobei versucht wird, die biologischen Grundlagen 
knapp, aber verständlich darzulegen. Hierzu werden auch Untersuchungsergebnisse von anderen 
Säugergruppen einbezogen, da nicht zu allen wichtigen Themenkreisen ausreichende Informationen 
von kleinen Laborsäugern vorliegen. Im zweiten, speziellen Teil sind in jeweils eigenen Abschnitten 
die Befunde an Haut und Haaren für die kleinen Versuchstierarten (Maus, Ratte, Goldhamster, 
Meerschweinchen, Kaninchen) wiedergegeben. Arttypische Besonderheiten, quantitative Angaben 
und die Beschreibung am entsprechenden Versuchstier gewonnener Forschungsergebnisse stehen im 
Vordergrund. 

Aufgrund der sorgfältigen Literaturarbeit des Autors hat die Publikation zweifellos hervorragende 
Bedeutung in der Behandlung einer für die dermatologische Forschung außerordentlich wichtigen 
Thematik. Besonders wertvoll ist dabei der ausführliche tabellarische Schlußteil der Arbeit sowie das 
umfangreiche Literaturverzeichnis. Andererseits macht der Verfasser auch deutlich, daß noch eine 
Reihe von Fragen ungeklärt bleiben. Dies bezieht sich z. B. auf eine differenzierte Beschreibung der 
Hautentwicklung, der regionalen Hautunterschiede und der Behaarung u.a. in Abhängigkeit von 
verschiedenen einflußnehmenden Parametern. W. MEYER, Hannover 


CHIVERS, D. ]J. (ed.): Malayen forest Primates. Ten years study in tropical rain forest. 
New York, London: Plenum Press 1980. 364 S. US $ 42.50. ISBN 0-306-40626-8 


Der vorliegende Sammelband berichtet über Erfahrungen und Ergebnisse zehnjähriger, kontinuierli- 
cher Studien an den Primaten des Regenwaldes der malayischen Halbinsel. Die Arbeit konzentrierte 
sich auf Studien an folgenden sechs Arten: Symphalangus syndactylus, Hylobates lar, Hylobates agılıs, 
Presbytis melalophos, P. obscura, Macaca fascicularis. Der besondere Wert des Buches ist darin zu 
sehen, daß die Untersuchungen zur Ökologie und Ethologie an den gleichen Primatengruppen am 
gleichen Ort langzeitlich verfolgt werden konnten. Dadurch war es möglich, Veränderungen in der 
Gruppenzusammensetzung und Struktur nicht nur zu konstatieren, sondern auch zu erklären. 
Hervorzuheben ist, daß die Vegetation nach Zusammensetzung und Struktur, qualitativ und quantita- 
tiv, sehr eingehend berücksichtigt wurde. Dabei fanden saisonale und langzeitliche Änderungen 
ebenso Beachtung wie adaptive Besonderheiten der Blätter, Früchte und Vegetationsorgane in 
Korrelation zur Nutzung durch die Tiere. Das Buch ist ein Musterbeispiel für Notwendigkeit und 
Wert aufeinander abgestimmter botanischer und zoologischer Arbeit ım gleichen Lebensraum. Die 
Resultate bringen zweifellos für die Vegetationskunde ebenso viele neue Erkenntnisse wie für die 
Primatologie. Dies erfreuliche Resultat ist allerdings auch der Tatsache zu danken, daß die beteiligten 
neun Autoren, Mitarbeiter eines wohlkoordinierten Teams verschiedener Fachdisziplinen waren, die 
es verstanden haben, ein in sich kohärentes, einheitliches Gesamtwerk zu schaffen. 

Durch die sorgfältige Analyse der Umwelt und des Verhaltens der Tiere in dieser gelang es, die 
ökologischen Nischen der sympatrischen Arten weitgehend zu klären und die Einsicht in viele 
Detailprobleme zu fördern (Ernährungsbiologie, Fortpflanzung, Familien- und Gruppenbildung). So 
läßt sich kein Zusammenhang zwischen Ernährungsweise und Lokomotionstyp nachweisen. Die 
große Anpassungsbreite der Makaken beruht nicht auf morphologischen oder physiologischen 
Spezialisationen, sondern auf Eigenschaften des Sozialverhaltens und kognitiven Fähigkeiten. Es wird 
überzeugend nachgewiesen, daß die grobe Finteilung in Frucht- und Blattfresser nicht ausreicht, um 
die Differenzierung der sechs Arten für die Nutzung verschiedener Nischen zu erklären. Hervorgeho- 
ben seien die ausführlichen Ausführungen zum Problem der Anpassung an Blätternahrung (Folivorie). 

Besonders zu danken ist dem Herausgeber dafür, daß er ein ausführliches Schlußkapitel beigefügt 
hat, ın dem er aufgrund seiner langjährigen Erfahrungen unter Beigabe von reichlich statistischem 
Material die Schlußfolgerungen zum Thema: Zerstörung des Primärwaldes, künftiges Schicksal des 
Biotops und Möglichkeiten zur Erhaltung zieht. Dem Buch sind eine umfangreiche Liste der 
identifizierten Pflanzen und ein erschöpfendes Literaturverzeichnis beigegeben. Es sollte wegen der 
Fülle neuen Materials und neuer, origineller Ideen von allen Primatologen, Ökologen, Ethologen und 
Naturschützern gelesen werden. D. STARcK, Frankfurt/M. 


192 Buchbesprechungen 


CHIARELLI, A. B.; CORRUCINI, R. S. (eds.): Primate Evolutionary Biology. Selected 
Papers (Part A) of the VIIIth Congress of the International Primatological Society. 
Florence, 7.-12. July 1980. Berlin, Heidelberg, New York: Springer 1981. 119 pp., 73 figs. 
DM 56,-. ISBN 3-540-11023-2. 


Der Sammelband enthält 13 (von 30 gehaltenen) Vorträge eines Symposions des Internationalen 
Primatologenkongresses 1980, die im weiteren Sinne unter dem Sammelbegriff der Evolutionsmor- 
phologie zusammengefaßt werden können. Die Vorträge stammen aus sehr verschiedenen Teildiszipli- 
nen und zeigen die außerordentliche Breite des heute von der Primatologie erfaßten Bereiches. 
Behandelt werden: Ontogenese der Art. carotis int. von Galago, Vergleich eozaener Prosimii (Nicht- 
Adapidae) mit Tarsius, adaptive Bedeutung von Größenunterschieden bei madagassischen Archaeole- 
muridae, Wachstumsänderungen bei Macaca, deskriptive Anatomie einiger Unterschenkelmuskeln, 
craniofaciale Morphologie, Schmelzstrukturen und Hirnrindenstruktur. Vielfach stehen methodische 
und funktionelle Überlegungen im Vordergrund. Demgegenüber treten evolutionsmorphologische 
Fragestellungen stärker in den Hintergrund, als der Titel vermuten läßt. Eine Synthese unter 
übergeordneten Gesichtspunkten enthält der Band nicht. Die einzelnen Beiträge dürften vor allem für 
Spezialisten von Wert sein. D. STARcK, Frankfurt/M. 


CHIARELLI, A. B.; CORRUCINT, R. S. (eds.): Primate Behavior und Sociobiology. Selected 
Papers (Part B) of the VIIIth Congress of the International Primatological Society. 
Florence, 7.-12. July 1980. Berlin, Heidelberg, New York: Springer 1981. 182 pp., 65 figs. 
DM 79,-. ISBN 3-540-11024-0. 


Der zweite Teilband (B) der Veröffentlichung des VIII. Internationalen Primatologenkongresses 
enthält 23 ausgewählte Vorträge über Verhalten und Soziobiologie der Primaten. Untersuchungen 
zum Sozialverhalten stehen im Vordergrund. Feldbeobachtungen an freilebenden Primaten wurden 
bei der Auswahl der Vorträge gegenüber Laboruntersuchungen bevorzugt berücksichtigt. Zwei 
Vorträge befassen sich mit der Sozialorganısation und dem Sozialverhalten von Tupaiidae. Drei 
Beiträge behandeln Territorialverhalten und Händigkeit bei Prosimiae. 12 Vorträge beziehen sich auf 
Cercopitheciden (Cercopithecus, Macaca, Papio; Sozialverhalten, Mutter-Kind-Beziehungen, Domin- 
anzverhalten, Lautäußerungen von Pavianen, Körperpflege). Langzeitbeobachtungen über Sexual- 
und Fortpflanzungsverhalten freilebender Orangs in Sumatra bringen eine Fülle neuer Beobachtun- 
gen. Das Verhalten von Schimpansen gegenüber Raubfeinden (Leopard, Löwe, Fleckenhyäne und 
Hyänenhund) trägt zur Erweiterung der Kenntnisse bei. Die Entwicklung und Organisation des 
Intelligenzverhaltens von Schimpansen erweist sich weitgehend als den Verhältnissen beim Menschen 
(PıAGET) vergleichbar. Der gut ausgestattete und inhaltsreiche Band dürfte nicht nur für Primatologen, 
sondern auch für Ethologen und Ökologen von großem Interesse sein. D. STARcK, Frankfurt/M. 


GUBERNICK, D. J.; KLoPprer, P. H. (eds.): Parental Care in Mammals. New York, 
London: Plenum Press 1981. 459 pp., US $ 39.50, ISBN 0-306-40533-4. 


Im Anschluß an die Bücher über „Maternal Behavior in Mammals“ (H. L. RHEINGOLD 1963) und 
„Maternal Care in Mammals“ (P. H. KroprEr, L. MCGEORGE, R. BARNETT 1973) liegt nun eine 
Fortführung dieses Themenkreises aus der Verhaltensontogenese vor. Darin finden sich elf neuere 
Übersichtsreferate von 14 nordamerikanischen Wissenschaftlern. Dargestellt wird nicht nur die 
zentrale Bedeutung der Mutter-Kind-Beziehung, sondern die gesamte elterliche Fürsorge. Die Bei- 
träge behandeln: Hormonale und nichthormonale Faktoren im Mutterverhalten bei Nichtprimaten; 
elterliche Beeinflussung der Jungen; Einwirkungen der Jungen auf elterliches Verhalten; mütterliche 
Aggressionen; Erlangung der Unabhängigkeit bei nesthockenden Säugern; Wechselbeziehungen 
zwischen Eltern und Jungen; geschwisterliche Interaktionen; väterlichen Einfluß; jugendliches Für- 
sorgeverhalten bei Primaten; sozialen Zusammenhang. Insgesamt wird ein interessanter Überblick 
über einen besonderen Teilbereich der Verhaltensforschung an Säugetieren geboten. Ein abschließen- 
der Index nach Stichworten erleichtert die Suche, ein zusätzlicher Artenindex wäre ebenfalls 
erwünscht. D. Kruska, Hannover 


Neuerscheinung 


Beiträge zur Verdauungsphysiologie des Pferdes 


Contributions to Digestive Physiology of the Horse. Von HELMUT MEYER, unter Mitwir- 
kung von BERNHARD HUSKAMP, VERA GÜLDENHAUPT, GERD LINDEMANN, HEINKE Muus, 
MOoNIKA PFERDEKAMP, MICHAEL SCHMIDT, KARIN SCHWABENBAUER. Heft 13 der „Fort- 
schritte in der Tierphysiologie und Tierernährung‘“; Beihefte zur „Zeitschrift für Tierphy- 
siologie, Tierernährung und Futtermittelkunde‘“. 1982. 69 Seiten mit 20 Abbildungen und 
34 Tabellen. Kartoniert 40,-— DM, für Bezieher der Zeitschrift 36,- DM 


Eine artgerechte Ernährung setzt entsprechende Kenntnisse der Verdauungsphysiologie 
voraus, wie sie zwar für das Rind und andere Wiederkäuer gegeben sınd, für das Pferd 
bislang aber kaum verfügbar waren. Erst durch Untersuchungen mit Caecumfisteln ist es 
gelungen, die Verdauungsvorgänge im Dünn- und Dickdarm des Pferdes unter besonderer 
Berücksichtigung der Wechselwirkungen zwischen beiden Abschnitten besser zu erfassen. 
In Beiheft 13 wird darüber eingehend berichtet. 

Die Untersuchungen erstrecken sich im einzelnen auf Futterpassage im Dünndarm, 
Entleerungsmechanismen des Hüftdarms (Ileum), Bildung organischer Säuren und anderer 
Abbauprodukte ım Blinddarm (Caecum) in Abhängigkeit von Futterart, Futtermenge und 
Fütterungstechnik; sie gelten besonders aber auch dem intestinalen Stickstoff-, Wasser- 
und Mineralstoffwechsel. 

Die Untersuchungsergebnisse erleichtern eine optimale Gestaltung von Futterrationen 
sowie eine sinnvolle Fütterungstechnik beim Pferd. Sie geben auch Hinweise auf die 
Entstehung der bei Pferden relativ häufigen Verdauungsstörungen und auf Möglichkeiten 
ihrer Vorbeuge. Darüber hinaus leisten sie einen wertvollen Beitrag zur vergleichenden 
Biologie der Verdauungssysteme bei Pflanzenfressern mit und ohne Vormagen. 


Neuauflage 
Tierproduktion 


Von ERNST GRANZ. 9., überarbeitete Auflage. 1982. 512 Seiten mit 207 Abbildungen, 
teilweise mehrfarbig und 119 Tabellen. Balacron broschiert 38,- DM 


Das bewährte Buch gibt eine gründliche Einführung in die landwirtschaftliche Tierproduk- 
tion. Ausgehend von den allgemeinen Grundlagen, der Anatomie und Physiologie des 
Tierkörpers, der Vererbung und Züchtung sowie der Tierernährung und Fütterung werden 
Erzeugung, Haltung und Vermarktung der einzelnen Nutztierarten ausführlich und unter 
verschiedenen Gesichtspunkten dargestellt, wobei nach Möglichkeit betriebs- und markt- 
wirtschaftliche Erkenntnisse im Vordergrund stehen. 

Neben neuen Produktionstechniken, verbesserten Methoden der Leistungsprüfung und 
Zuchtwertschätzung und rationellen Fütterungsverfahren berücksichtigt die nun bereits 
vorliegende 9., überarbeitete Auflage des Buches die Nettoenergie-Laktation, die neue 
Indexbewertung für Bullen, Eber und Schafböcke sowie die Güterverordnung für Milch 
und Änderungen im Futtermittelrecht. 


Bere PP NUL PAREY :-HAMBURG UND BERLIN 


Erscheinungsweise und Bezugspreis: Die Zeitschrift erscheint alle 2 Monate; 6 Hefte bilden einen Band 
mit insgesamt 384 Seiten. Der Abonnementspreis beträgt je Band 232,- DM zuzüglich Porto. Das 
Abonnement verpflichtet zur Abnahme eines ganzen Bandes. Es verlängert sich stillschweigend, 
wenn nicht unmittelbar nach Erhalt des letzten Heftes eines Bandes Abbestellung erfolgt. 
Einzelbezugspreis der Hefte: 43,- DM. Die Preise verstehen sich im Inland incl. Mehrwertsteuer. 
Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, 
Spitalerstraße 12, D-2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesell- 
schaft für Säugetierkunde“ erhalten die Zeitschrift unberechnet im Rahmen des Mitgliedsbeitrages. 


BARASH 


Sozıobiologie 
und Verhalten 


Von Prof. Dr. David P. Barash, 
Washington. Aus dem Amerikani- 
schen übersetzt vonDr.IngridHorn, 
Washington. 1980. 340 Seiten mit 
94 Abbildungen. Balacron bro- 
schiert DM 49,— 


Als Synthese aus den Erkenntnissen 
der Entwicklungsbiologie und der 
Sozialwissenschaft eröffnet die So- 
ziobiologie neue Möglichkeiten zur 
Erforschung von Verhaltensweisen 
mit dem Ziel, Gesetzmäßigkeiten 
herzuleiten, die sich wertungsfrei 
auf das Verhalten des Menschen 
übertragen lassen. 

Soziobiologie: Das ist die systema- 
tische Untersuchung der biologi- 
schen Grundlagen des Sozialverhal- 
tens in seinen unterschiedlichsten 
Erscheinungsformen, die sich auf 
alle Tierarten erstreckt. Sie ist da- 
her in erster Linie eine verglei- 
chende Wissenschaft. Während sich 
die Sozialwissenschaft ausschließ- 
lich mit einer einzigen Art, dem 
Menschen, beschäftigt, nutzt die 
Biologie die überreiche, organische 
Vielfalt unseres Planeten: Über eine 
Million Tierarten sind bisher be- 
schrieben worden; davon liefern 
Zehntausende, die eine gewisse so- 
ziale Lebensweise entwickelt haben, 
der Soziobiologie die Daten und 


Informationen, die sie sucht, wobei 
jede Spezies ein Evolutionsexperi- 
ment mit einer breiten Palette von 
Auswertungsmöglichkeiten verkör- 
pert. Die Vereinigung beider Diszi- 
plinen in der Soziobiologie hat be- 
reits zu einigen überraschenden 
Resultaten geführt, wobei nicht nur 
die Mechanismen des Sozialver- 
haltens in einem neuen Licht er- 
scheinen, sondern sich erstmals 
auch Ansatzpunkte für eine biolo- 
gische Analyse von Gesellschafts- 
strukturen bieten. 

David P. Barash, nach Edward O. 
Wilson einer der Wegbereiter der 
Soziobiologie, gibt mit diesem Buch 
eine leichtverständliche Einführung 
in dieses neue Wissensgebiet. Seine 
Darstellung der Biologie in ihrer 
Relation zum Sozialverhalten ist zu- 
gleich ein überzeugendes Plädoyer 
für die Bedeutung der Evolution 
als der eigentliche Schlüssel zum 
Verständnis der Eithologie aller 
Lebewesen, einschließlich des Men- 
schen. Er wendet sich nicht nur an 
Biologen, Ethologen, Zoologen, 
Psychologen, Soziologen und die 
Studierenden dieser Disziplinen, 
sondern darüber hinaus an den 
großen Kreis aller, die an natur- 
geschichtlichen Themen, Proble- 
men und Zusammenhängen inter- 
essiert sind. 


Berlin 


"Vol. 47 (4), 193-256, August 1982 ISSN 0044-3468 C 21274 F 
Mammals 


ZEITSCHRIFT FÜR 
SAUGE TTIERKUNDE 


INTERNATIONAL JOURNAL 
DF MAMMALIAN BIOLOGY 


Organ der Deutschen Gesellschaft für Säugetierkunde 


Kürten, L.; Schmidt, U.: Die Nasengruben der Vampirfledermaus Desmodus rotundus: Sinnesorgane zur Wahrneh- 
mung von Wärmestrahlung. — The nose pits of the vampire bat Desmodus rotundus: sense organs for infrared 


reception 193, 
Nadler, C. F.; Hoffmann, R. S.; Vorontsov, N. N.; Koeppl, J. W.; Deutsch, L.; Sukernik, R. I.: Evolution in ground 
squirrels. Part II. — Evolution der Erdhörnchen. Teil Il 198 


Bethge, E.: Zyklische Bestandswechsel (Gradationen) bei der Feldmaus (Microtus arvalis), festgestellt durch 
Analyse von Eulen-Gewöllen. — Cyclic population changes of the common vole (Microtus arvalis) studied by 
analysis of owl pellets 215 


Schäfer, M. W.: Analyse des Explorationsverhaltens verschiedener Stämme und Generationen der Hausmaus (Mus 
musculus domesticus Rutty). Teil Il. - Analysis of exploratory behaviour in different strains and generations of the 
house mouse (Mus musculus domesticus Rutty). Part Il 220 


Giagia, Eva; Savic, I.; Soldatovic, B.: Chromosomal forms of the mole rat Microspalax from Greece and Turkey. — 
Chromosomenformen von Blindmäusen (Microspalax) aus Griechenland und der Türkei 231 


Peters, G.: A note on the vocal behaviour of the Giant panda, Ailuropoda melanoleuca (David, 1869). - 
Beobachtungen zum Lautgebungsverhalten des Großen Panda, Ailuropoda melanoleuca (David, 1869) 236 


Born, E. W.; Clausen, B.; Henriksen, Sv. Aa.: Trichinella spiralis in walruses from the Thule district, North 
Greenland, and possible routes of transmission. — Trichinella spiralis in Walrossen aus dem Thule-Gebiet, 
Nordgrönland, und mögliche Wege der Übertragung 246 


Wissenschaftliche Kurzmitteilungen 


Schmidt, Christel; Schmidt, U.: Der Einfluß der Domestikation auf die Riechleistung der Hausmaus (Mus musculus). 
— The influence of domestication on the olfactory performance of the house mouse (Mus musculus) 251 


Schlosser-Sturm, Evelyn: Zur Funktion und Bedeutung des sekundären Schultergelenks der Microchiropteren. - On 
the function and significance of the secondary articulation in the shoulder joint of Microchiroptera 253 


Buchbesprechungen aa TETTE 255 
<ÄNUINDSUNGG 
Bu pr 


U LIBRAR! 


Verlag Paul Parey Hamburg und An 


HERAUSGEBER/EDITORS 


P. J. H. van BreEE, Amsterdam - W. FIEDLER, Wien - H. Frick, München - W. HERRE, 

Kiel - H.-G. Kıös, Berlin - H.-J. Kunn, Göttingen - B. Lanza, Florenz - T. C. S. 

MORRISON-ScOTT, London - J. NIETHAMMER, Bonn - H. REICHSTEIN, Kiel - M. Rönrs, 

Hannover - D. STARcK, Frankfurt a. M. — F. Strauss, Bern — E. THENIUS, Wien — 
W.VERHEYEN, Antwerpen 


SCHRIFTLEITUNG/EDITORIAL OFFICE 


H. SCHLIEMANnN, Hamburg — D. Kruska, Kiel 


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. Harald Schliemann. 


Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z.Hd. Prof. Dr. Dieter 
Kruska, Institut für Haustierkunde, Biologie-Zentrum, Neue Universität, Olshausenstr. 40-60, 
D-2300 Kiel. Für die Publikation vorgesehene Manuskripte sollen gemäß den „Redaktionellen 
Richtlinien“ abgefaßt werden. Diese Richtlinien sind in deutscher Sprache Bd. 43, H. 1 und in 
englischer Sprache Bd. 43, H. 2 beigefügt; in ihnen finden sich weitere Hinweise zur Annahme von 
Manuskripten, Bedingungen für die Veröffentlichung 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 
Vervielfältigung auf photomechanischem oder ähnlichem Wege oder im Magnettonverfahren 
sowie der Speicherung ın Datenverarbeitungsanlagen, bleiben, auch bei nur auszugsweiser Ver- 
wertung, vorbehalten. Werden von einzelnen Beiträgen oder Beitragsteilen einzelne Vervielfälti- 
gungsstücke in dem nach $54 Absatz 1 UrhG zulässigen Umfang für gewerbliche Zwecke 
hergestellt, ist dafür eine Vergütung gemäß den gleichlautenden Gesamtverträgen zwischen der 
Verwertungsgesellschaft Wort, vereinigt mit der Verwertungsgesellschaft Wissenschaft GmbH, 
rechtsfähiger Verein kraft Verleihung, Goethestr. 49, D-8000 München 2, und dem Bundesver- 
band der Deutschen Industrie e.V., dem Gesamtverband der Versicherungswirtschaft e.V., dem 
Bundesverband deutscher Banken e.V., dem Deutschen Sparkassen- und Giroverband und dem 
Verband der Privaten Bausparkassen e.V., an die Verwertungsgesellschaft zu entrichten. Erfolgt 
die Entrichtung der Gebühren durch Wertmarken der Verwertungsgesellschaft, so ist für jedes 
vervielfältigte Blatt eine Marke im Werte von 0,40 DM zu verwenden. Die Vervielfältigungen sind 
mit einem Vermerk über die Quelle und den Vervielfältiger zu versehen. 


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 


CcCcC 


Fortsetzung 3. Umschlagseite 


© 1982 Verlag Paul Parey, Hamburg und Berlin. - Printed in Germany by Westholsteinische Verlagsdruckerei 
Boyens & Co., Heide/Holst. 


Z. Säugetierkunde 47 (1982) 4, 193-256 
InterCode: ZSAEA 7 47 (4), 193-256 (1982) 


Mit zwei Beilagen des Verlages Paul Parey 


Die Nasengruben der Vampirfledermaus Desmodus rotundus: 
Sinnesorgane zur Wahrnehmung von Wärmestrahlung 


Von L. KÜRTENn und U. SCHMIDT 


Zoologisches Institut der Universität Bonn 


Eingang des Ms. 29. 4. 1982 
Abstract 


The nose pits of the vampire bat Desmodus rotundus: sense organs for infrared reception 


Studied was the distribution of thermoreceptive areas in the nose leaf with two vampire bats 
(Desmodus rotundus) that were trained to locate a source of infrared radiation. Applicating the local 
anaesthetic Xylocaine (4%) in the three nose pits the percentage of correct responses fell to chance 
level (Xylocaine 1% resp. water did not influence thermoperception). These results indicate that the 
nose pits of Desmodus are specific sense organs for the reception of infrared radiation. 


Einleitung 


Die sanguivore Vampirfledermaus (Desmodus rotundus) ıst ın der Lage, thermische Reize 
wahrzunehmen. SCHMIDT und MAnskKE (1982) wiesen nach, daß Desmodus bei der 
Auswahl einer Bißstelle durch Wärmereize beeinflußt werden kann. KÜRTEN und SCHMIDT 
(1982) untersuchten die Reaktionsschwellen und stellten fest, daß die Vampirfledermäuse 
Infrarotstrahlung bis zu einer Intensität von 0,5 x 10* Wcm” perzipieren können. 
Histologische und thermographische Ergebnisse weisen darauf hin, daß die thermorezepti- 
ven Strukturen (freie Nervenendigungen; KÜRTEN in Vorber.) in drei Gruben konzentriert 
sind, die im Nasenaufsatz von Desmodus das zentrale Nasen-,,‚blatt“ von den umliegenden 
Polstern trennen (Abb. 1). Die hier vorgestellte Arbeit soll diese Hinweise untermauern. 


Material und Methode 


Die Untersuchungen wurden mit zwei dd Desmodus rotundus durchgeführt, die aus der Zuchtkolo- 
nie des Zoologischen Institutes, Universität Bonn, stammten. Sie waren zur Zeit der Versuche 
zweieinhalb Jahre alt. 

Die Dressuren fanden in einer Zweifach-Wahlanlage (150 X 65 X 70cm) statt. An deren Rückseite 
befanden sich die Individualkäfige, davor ein Aufenthaltsraum, sowie der eigentliche Wahlraum (Abb. 
2). Alle Teile der Anlage waren durch Schiebetüren getrennt, die vom Experimentator über Seilzüge 
betätigt werden konnten. An der Vorderseite des Wahlraumes waren zwei Anweiser (AW) ange- 
bracht, die je aus einem Heizelement (Fläche 224 cm?), einer dahinterliegenden Kupferplatte (1 mm 
dick) sowie einem 2cm starken Holzrahmen bestanden. Beide AW waren leicht austauschbar und in 
Form, Farbe und Abmessung völlig identisch. Zwischen ihnen befand sich, vom Boden bis zur Decke 
der Anlage reichend, eine Trennwand (4cm dick), durch die die AW gegeneinander wärmeisoliert 
waren. Die Trennwand reichte bis zur Entscheidungslinie in die Anlage hinein (8 bzw. 12cm). Unter 
jedem AW war eine 1 X 2cm große Öffnung ausgespart, durch die ein Fütterungsröhrchen eingescho- 
ben werden konnte. Die Unterkante des Heizelementes lag 5cm über dem Boden des Wahlraumes. 
Dieser Boden war so in Stufen angeordnet, daß an der Entscheidungslinie sich der Kopf der Tiere in 
Höhe der Heizelemente befand. Der Versuchsraum hatte konstante Temperatur (24 °C) und Luft- 
feuchte (80 %). Die Beleuchtungsintensitäten innerhalb der Anlage lagen unter 0,3 Lx. 

Bei den Versuchsläufen war stets ein AW erwärmt (positive Seite), der andere besaß Raumtempera- 
tur. Nach dem Öffnen der Schiebetür lief das Versuchstier zur Entscheidungslinie und wählte einen 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4704-0193 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 193-197 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


194 L. Kürten und U. Schmidt 


Abb. 1. Kopf einer Vampirfledermaus (Desmodus rotundus). Die Pfeile weisen auf die Nasengruben 
(1 = apikale Grube; 2 = linke laterale Grube) 


Abb. 2. Versuchsanlage (schema- 
tisch). AW = Anweiser (Wärmeplat- 
ten); W = Wahlraum; A = Aufent- 
haltsraum; E = Einzelkäfige zur indi- 
viduellen Haltung der Versuchstiere 


der AW. Bei richtiger Wahl wurde ein Trinkröhrchen eingeschoben, und die Fledermaus konnte ca. 
10s fressen; Negativwahlen durften korrigiert werden. Der Seitenwechsel der AW war zufallsgemäß 
vorgegeben. Täglich konnten etwa 20 Läufe pro Tier durchgeführt werden. Während der Andressur 
betrug der Wahlabstand 8cm; die Temperatur der Heizelemente war auf ca. 50°C eingestellt. Nach 
1200 bzw. 1400 Läufen lag die Wahlsicherheit bei 85 % (Tier A) bzw. 75% (Tier B). 

In einer Eingewöhnungsphase wurden dann den Tieren die Nasengruben mit einem dünnen Pinsel 
bestrichen. Dieser war in täglichem Wechsel angefeuchtet bzw. trocken. Anschließend erfolgten die 
Versuchsläufe, bei denen der Wahlabstand 12 cm (Tier A) und 8 cm (Tier B) betrug. Die AW waren auf 
eine Strahlung von 0,8-1,0 x 10*W cm” eingestellt; diese Reizintensität hatten die Tiere in der 
Andressur sicher beantwortet. Um eine akustische Beeinflussung auszuschließen, wurde vor Beginn 
jedes Laufes der Strom des erwärmten AW ausgeschaltet; der Abfall der Temperatur in der Zeit, die 
das Tier zum Wählen benötigte, war so gering, daß er vernachlässigt werden konnte. 

In der eigentlichen Versuchsphase wurden die Nasengruben mit einem Lokalanästhetikum (Xylo- 
cain®) eingepinselt (1%ige bzw. 4%ige Lösung). Die Flüssigkeit wirkte jeweils zwei Minuten ein, 


Wärmesinnesorgane bei Desmodus 195 


Richtigwahlen 


Br 7578 7,8 910 1112 


Versuchstag 


Abb. 3. Veränderung der Wahlsicherheit bei den Aus- 
schaltversuchen. a: Tier A; b: Tier B. Die Haut der 
Nasengruben wurde folgendermaßen behandelt: w = 
1% Xylocain; @ = Wasser; m = trocken; O = 4% 


Xylocain 


Reste wurden anschließend abgetupft. In 
einem 12 Tage dauernden Versuch fand 
abwechselnd Wasser, 1%ıge oder 4%ige 
Lösung Verwendung; an zwei Tagen war 
der Pinsel beim Bestreichen trocken. Bei 
Tier B zeigte das Anästhetikum allein 


“ keine Wirkung. Deshalb fand in einer 


weiteren Versuchsserie vor Applikation 
des 4%igen Xylocains eine Vorbehand- 
lung der Haut in den Nasengruben mit 
einer 1%igen Lösung Salicylsäure statt, 
die ein Aufweichen der Haut bewirkt. 


Ergebnisse 


In den Läufen der Eingewöhnungs- 
phase, bei denen die Nasengruben 
mit Wasser eingepinselt waren, zeig- 
ten die Tiere anfangs eine leichte 
Verunsicherung. Bereits nach weni- 
gen Tagen verhielten sie sich jedoch 
wieder völlig normal und wählten 
signifikant den warmen AW. Auch 
im weiteren Verlauf des Versuches 
war ıhr Verhalten in keiner Weise 
verändert. Dies zeigt, daß die Sub- 
stanzen keine schädlichen Auswir- 
kungen auf den Allgemeinzustand 
der Tiere hatten. Beim Auftragen der 
Flüssigkeiten wurde darauf geachtet, 
daß ausschließlich das Innere der 
Nasengruben benetzt war. Abbil- 


dung 3 gibt den Verlauf des Versuches mit dem täglichen Wechsel des jeweils aufgetrage- 
nen Mittels an. Sie zeigt, daß bei Applikation von Wasser, Xylocain 1%, sowie nach 
Bestreichen mit trockenem Pinsel keine Beeinträchtigung der Wahlsicherheit eintritt, daß 
jedoch bei den Läufen, bei denen das 4% ige Anästhetikum appliziert worden war (bzw. bei 
Tier B: 1% Salicylsäure + 4% Xylocain), die Prozentzahl der Richtigwahlen gegenüber 
allen anderen Tagen stark abfällt. Die Unterschiede sind statistisch hochsignifikant (Vier- 
feldertest, Abb. 4). Bei Applikation des 4%igen Anästhetikums ın die Nasengruben ist die 
Perzeption der Wärmestrahlung offensichtlich drastisch reduziert. 


A 
60% 
n = 124 
rev 0 | 
FEUCHT n=99 
P«<0,001 


76% 

XYLOCAIN 1% n = 94 
P«0,01 

81°, 

TROCKEN n- 91 
P < 0,001 


XYLOCAIN 4% 


B 
62°% 
n = 91 
84°% 
n= 114 
P «0,001 


Abb. 4. Trennung der Applika- 
tionsarten im _ Vierfeldertest 
(SacHs 1978). A: Tier A; B: Tier 
B; % = Prozent Richtigwahlen; 
n = Stichprobenanzahl 


196 L. Kürten und U. Schmidt 


Um zu kontrollieren, ob das Xylocain tatsächlich rezeptive Strukturen in der Haut der 
Nasengruben ausschaltet, oder aber durch Diffusion in tiefergelegenes Gewebe einen 
ableitenden Nerv blockiert, wurden bei Tier A an zwei Tagen die Nasenpolster, die 
halbkreisförmig um die Nasengruben ziehen, mit 4%igem Xylocain behandelt. In diesen 
Versuchen war die Wärmewahrnehmung nicht eingeschränkt (81% Richtigwahlen; 
p<0,01). 


Diskussion 


Die Fähigkeit zur Wahrnehmung von Wärme ist im Tierreich weit verbreitet (PRECHT et al. 
1973); vor allem kutane Thermorezeptoren sind gut untersucht. HEnseL (1975) betont, daß 
in allen diesen Fällen ein Temperatursinn vorliegt, daß also - unabhängig von der Art der 
Wärmequelle - die Erwärmung des die Rezeptoren umgebenden Gewebes der adäquate 
Reiz ist. Die physikalische Ursache dieser Erwärmung kann sehr unterschiedlich sein: in 
den meisten Fällen dient die Thermoperzeption zur Feststellung der Temperatur des 
umgebenden Mediums oder zur Perzeption von Kontaktreizen (Erwärmung der Haut 
durch Konvektion und Leitung); bei einigen Tieren existiert jedoch die Fähigkeit zur 
Wahrnehmung von Wärmestrahlung. Um die kleinen Energiemengen, die z. B. von einem 
warmblütigen Beutetier abgestrahlt werden, noch wahrnehmen zu können, ist bei diesen 
Tieren die rezipierende Region anatomisch und histologisch so gestaltet, daß der Reiz 
optimal genutzt werden kann. Ein bekannter Fall sind die Grubenorgane der Crotaliden, 
deren nur 15 um dicke Membran schon von sehr geringen Strahlungsintensitäten erwärmt 
wird (BuLLock und DiEckE 1956; DE Cock Bunıng et al. 1981). Die Anordnung dieser 
Membran in zwei Gruben ermöglicht außerdem eine Richtungswahrnehmung. Dies ist 
auch bei einigen Boiden durch Anordnung der Rezeptoren in Lippengruben der Fall 
(WARREN und PROSsKE 1968; DE Cock Buninc et al. 1978). 

In den hier vorgestellten Versuchen gelang es — durch begrenzte Applikation eines 

Anästhetikums ın die Nasengruben -, die Fähigkeit der Vampirfledermäuse, Infrarotstrah- 
lung wahrzunehmen, reversibel auszuschalten; die Auftragung des Anästhetikums auf 
benachbarte Gebiete (Kontrollversuch mit Xylocain-Behandlung der Polster) führte dage- 
gen nicht zu einer Beeinträchtigung. Dies weist darauf hin, daß 
1. die thermosensitiven Strukturen auf die Nasengruben beschränkt sind, und daß 
2. das Xylocain nicht in benachbarte oder tiefergelegene Gebiete diffundiert. 
Die von uns verwendete Lösung des Anästhetikums (4%, als Hydrochlorid) ist vom 
Hersteller für die Anwendung auf Schleimhäuten vorgesehen; bei Verwendung auf kerati- 
nisierter Haut werden dagegen eine Öl/Wasser-Emulsion und 60 min Einwirkzeit empfoh- 
len (pers. Mitteilung der Fa. ASTRA, Schweden). Wir mußten bei Tier B zusätzlich 
Salicylsäure auftragen, um einen Fffekt zu erzielen. Man kann daher davon ausgehen, daß 
das Xylocain nur in oberflächliche Gewebe eindringt und tieferliegende afferente Nerven 
nicht beeinflußt. 

Die Beschaffenheit der Haut in den Gruben erfüllt alle Voraussetzungen für eine 
optimale Ausnutzung eines Infrarotreizes (KÜRTEN und SCHMIDT 1982). So fehlen dort 
z. B. jegliche Drüsen, deren Sekrete die Aufnahme der Strahlung behindern könnten. Das 
dichte Bindegewebe, das der Nasenregion unterlegt ist, hat in diesem Zusammenhang zwei 
Funktionen: Durch seine isolierenden Eigenschaften verhindert es zum einen die Leitung 
der von der Haut aufgenommenen Wärme in Gewebeschichten unterhalb der Rezeptoren, 
zum anderen schirmt es die Gruben von der Körperwärme ab und erleichtert so die 
Rezeption niedriger Strahlungsenergien. 

Die Ergebnisse lassen den Schluß zu, daß die Nasengruben von Desmodus ein Sinnesor- 
gan zur Wahrnehmung von Wärmestrahlung darstellen. Die Anordnung der rezeptiven 
Strukturen in einem Organ, das eine optimale Ausnutzung des Reizes ermöglicht, gleicht 
den Nachteil der geringen Größe der Tiere und der deshalb geringen rezeptiven Fläche aus. 


Wärmesinnesorgane bei Desmodus 197. 


Die Vampire vermögen die Wärmestrahlung, die von einem Warmblüter ausgeht, noch in 
13-16cm Entfernung wahrzunehmen (KÜRTEN und SCHMIDT 1982). Diese Fähigkeit 
könnte bei der Lokalisation und Selektion einer geeigneten Bißstelle eingesetzt werden. 
Freilanduntersuchungen haben z.B. gezeigt, daß die Vampirfledermäuse bei ihren Beute- 
tieren (Rinder, Pferde) oft direkt oberhalb des Hufes beißen oder alte Bißwunden neu 
öffnen (SCHMIDT 1978). Die Temperaturdifferenz zwischen dem Huf und dem durchblute- 
ten Gewebe kann ohne Schwierigkeiten von den Vampiren perzipiert werden. Desmodus 
ist es damit gelungen, sich beim Nahrungserwerb die Wärmestrahlung der Beutetiere als 
zusätzliche Orientierungshilfe nutzbar zu machen. 


Zusammenfassung 


Bei zwei Vampirfledermäusen (Desmodus rotundus), die in einer Zweifach-Wahlanlage darauf dres- 
siert waren, eine Wärmequelle zu lokalisieren, wurde untersucht, welche Bereiche des Nasenaufsatzes 
für die Perzeption von Infrarotstrahlung verantwortlich sind. Bei Applikation des Lokalanästheti- 
kums Xylocain® (4%ig) in die drei Nasengruben fiel die Wahlsicherheit in den Zufallsbereich 
(1 %iges Xylocain bzw. Wasser beeinflußten die Thermoperzeption nicht). Aus diesen Ergebnissen 
wird geschlossen, daß die Nasengruben von Desmodus ein spezifisches Sinnesorgan zur Wahrneh- 
mung von Wärmestrahlung darstellen. 


Literatur 


BuLrock, T. H.; DiEcke, F. P. J. (1956): Properties of an infrared receptor. J. comp. Physiol. 134, 
47-87. 

DE Cock Buning, T.; POELMANN, R. E.; DULLEMEIJER, P. (1978): Feeding behavior and the 
morphology of the thermoreceptors in Python reticulatus. Netherlands J. Zool. 28, 62-93. 

DE Cock Buning, T.; TERASHIMA, $.; GoRIS, R. C. (1981): Crotaline pit organs analysed as warm 
receptors. Cell. Mol. Neurobiol. 1, 69-85. 

Henseı, H. (1975): Static and dynamic activity of warm receptors in Boa constrictor. Pflügers Arch. 
253, 191-199. 

KÜRTEn, L.; SCHMIDT, U. (1982): Thermoperception in the common vampire bat (Desmodus 
rotundus). J. comp. Physiol. 146, 223-228. 

PRECHT, H.; CHRISTOPHERSEN, J.; HENSEL, H.; LARCHER, W. (1973): Temperature and life. Berlin, 
Heidelberg, New York: Springer. 

Sachs, L. (1978): Angewandte Statistik. 5. Aufl. Berlin, Heidelberg, New York: Springer. 

SCHMIDT, U. (1978): Vampirfledermäuse. Die neue Brehm Bücherei 515. Wittenberg: Ziemsen. 

SCHMIDT, U.; MaAnskE, U. (1982): Thermopräferenz bei der Gemeinen Vampirfledermaus (Desmodus 
rotundus). Z. Säugetierk. 47, 118-120. 

WARREN, J. W.; PRoskE, U. (1968): Infrared receptors in the facial pits of the australian python 
Morela spilotes. Science 159, 439441. 


Anschrift der Verfasser: LupwıG KÜRTEn, Prof. Dr. UwE SCHMIDT, Zoologisches Institut der 
Universität Bonn, Poppelsdorfer Schloß, D-5300 Bonn 


Evolution in ground squirrels 
II. Biochemical comparisons in Holarctic populations of Spermophilus 


By C. F. NADLER, R. S. HoFFMANnN, N. N.VORONTSOV, J. W. KoEppL, L. DEUTSCH, 
and R. I. SUKERNIK 


Department of Medicine, Northwestern University Medical School, Chicago, Museum of Natural 

History and Department of Systematics and Ecology, University of Kansas, Lawrence, Institute of 

Developmental Biology, Academy of Sciences of the USSR, Moscow, and Institute of Cytology and 
Genetics, Siberian Branch, Academy of Sciences of the USSR, Novosibirsk 


Receipt of Ms. 16. 12. 1981 
Abstract 


A comparison of biochemical similarities among 21 of 23 recognized species in the subgenus 
Spermophilus was made. Thirteen proteins, representing at least 18 locı were studied, and electromor- 
phic variations described. A phenogram and a Wagner network of genetic resemblance are presented, 
and, together with morphological, chromosomal, and distributional data, used to interpret the 
systematics of Spermophilus. Two lineages diverged in North America, the Nearctic ““small-eared’” 
and “big-eared’” ground squirrels. Most of the Recent species in these lineages evolved during the late 
Pleistocene. 


Introduction 


Ground squirrels (Spermophilus = Citellus) are widely distributed in and restricted to 
North America and Eurasia, and probably comprise the most recently evolved of the 
terrestrial squirrels (BLACK 1963, 1972). Up to twenty-three species are currently recog- 
nized in the subgenus Spermophilus (sensu lato): 12 ın the Palearctic, and 10 in the 
Nearctic, with one additional species shared (GROMoVv et al. 1965; Harı 1981; LyAPunovA 
and VORONTSoVv 1970; RoBInsoN and HOFFMANN 1975). GROMOV et al. (1965) subdivided 
Spermophilus ınto three subgenera, Spermophilus (= Citellus) (sensu stricto), Urocıitellus, 
and Colobotis, but others have not recognized this subdivision. The subgenus Spermophilus 
(s. s.) may be further subdivided into informal “groups”. Davıs (1939) proposed that the 
Nearctic species could be allocated to a ““big-eared” group and a “small-eared” group. 
Evidence concerning these groups, and a subgeneric classification will be presented below. 
Taxa included within the genus Spermophilus illustrate the effects of varying periods of 
geographical isolation by the Bering Strait on evolutionary patterns of cytological, 
morphological and biochemical characters. Among the long-tailed ground squirrels 
(Urocitellus) is found the only species with a Holarctic distribution, the arctic ground 
squirrel (S. parryu). This species occupied Beringia during the last glacial period (Würm- 
Wisconsin) (REPENNING et al. 1964). Eastern Siberian and western North American 
populations were isolated by the Bering Strait about 13,000 years ago as ocean levels rose 
with glacial retreat (Horkıns 1972). Since then, morphological and biochemical differences 
have accumulated (Rosınson and HorrMmAann 1975; NADLER et al. 1976), but gross 
chromosomal morphology has remained unchanged (NADLER 1966; LyapunovA 1969). 
The other two species of long-tailed ground squirtels, S. undulatus and S. columbianus, 
comprise sister-species (sensu HENNIG 1966) in the Palearctic and Nearctic respectively. 
They also retain apparent chromosomal identity while diverging in morphology and 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4704-0198 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 198-215 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Evolution in ground squirrels II 199 


biochemistry to a greater degree than parryıi (NADLER et al. 1975; RoBınson and 
HoFrFMANN 1975). 

The Palearctic subgenus Colobotis has no representative in the Nearctic so other 
Holarctic relationships must be sought within the subgenus Spermophilus (s. s.). In an 
earlier paper we presented detailed comparison of transferrins (NADLER et al. 1974). In this 
paper, we extend this comparison to other biochemical systems and relate these data to 
other systematic characters of Holarctic ground squirrels. A revised classification of the 
subgenus Spermophilus in presented, in which we employ the concept of the superspecies, 
and the notation proposed by Amanon (1966). This is necessary because of the many allo- 
and parapatric populations in the subgenus, and the occurrence of sporadic hybridization 
between allospecies (= semispecies) (LORKOVIc 1960; Mayr 1969; PETROV and ZıvkovIc 
1977; Reıc et al. 1980). 


Materials and methods 


Specimens analyzed 


A total of 1087 specimens were examined. Voucher specimens of most North American specimens 
reported are deposited in the Museum of Natural History, University of Kansas, and Furasian 
specimens are in the collection of the Zoological Institute, Academy of Sciences of the USSR, 
Leningrad. 

Spermophilus townsendil townsendii (Bachman). USA, Washington: Benton County, S. of Yakıma 
River, vic. of Prosser, 27 specimens; Yakima County, Yakıma airport, 15 specimens. 

S. mollis mollis (Kennicott). Idaho: Cassıa County, Burley, 47 specimens. 

S. m. idahoensis (Merriam). Idaho: Elmore County, vic. of Mountain Home, 9 specimens. 

$. m. nancyae Nadler. Washington: Benton County, N. of Yakima River, vic. of Prosser, 26 
specimens; Yakıma County, Yakima, 10 specimens. 

S$. vigilis vigilis (Merriam). Oregon: Malheur County, vic. of Ontario, 16 specimens; vic. of Vale, 
39 specimens. 

$. v. canus (Merriam). Oregon: Harney County, Narrows, 3 specimens; Hines, 1 specimen. 

S. washingtoni (A. H. Howell). Washington: Franklin County, Scootenay Reservoir, 31 speci- 
mens. 

$. brunneus (A. H. Howell). Idaho: Adams County, 3 mi. $, 0.5 mi. E of Bear Post Office, 8 
specimens. 

S. beldingi beldingi Merriam. California: Inyo County, Rock Lake, 10 mi. SW of Tom’s Place, 13 
specimens; Mono County, Rock Creek Lodge, 9-10 mi. SW of Tom’s Place, 3 specimens. 

S. b. creber (Hall). Idaho: Cassia County, Burley, 2 specimens. 

S. b. oregonus (Merriam). Oregon: Grant County, 3.5 mi. $ of Mount Vernon, 5 specimens; 
Harney County, Burns, 3 specimens; Multnomah County, vic. of Portland, 1 specimen. 

$. armatus (Kennicott). Montana: Gallatin County, 3.5 mi. S, 1.5 mi. W of Gallatin Gateway, 2 
specimens; Madison County, 6.5 mı. S, 1.5 mi. W of Harrison, 19 specimens; 3 mi. W of McAllister, 7 
specimens; 4 mi. N, 3 mi. E of McAllister, 9 specimens. Utah: Cache County, vic. of Logan, 5 
specimens; Tony Grove Lake, 15 mi. NE of Logan, 10 specimens. Wyoming: Lincoln County, 12 mı. 
NE of Cokeville, 8 specimens. 

S. columbianus columbianus (Ord). Montana: in Granite County, 15 mi. $ of Clinton, 1 specimen; 
Jefferson County, 0.5 mi. E of Whitehall, 2 specimens; in Madison County, 12.5 mı. S, 4 mi. E of 
Whitehall, 1 specimen; Madison County, 1.5 mi. N, 2 mi. E of Harrison, 83 specimens; 1.5 mi. N of 
Harrison, 4 specimens; 1.25 mi. $S, 1 mi. W of Harrison, 19 specimens; 6.5 mı. $, 1.5 mi. W of 
Harrison, 1 specimen; Missoula County, 12 mi. E of Potomac, 1 specimen; 5 mi. N of Missoula, 56 
specimens. Idaho: Adams County, 1 mi. N of Bear Guard Station, 3 specimens; 5 mi. E of Cuprum, 1 
specimen; Idaho County, 4 mi. $, 4 mi. E of Lowell, 1 specimen. Canada, Alberta: vic. of Edmonton, 
6 specimens. 

S. richardsonii (Sabine). Montana: in Broadwater County, 2 mi. N 0.5 mi. W of Three Forks, 3 
specimens; 0.5 mi. N of Jefferson R., 5 mi. SW of Three Forks, 1 specimen; 1.5 mi. N., 2 mi. E of 
Sappington, 23 specimens; Gallatin County, 5 mi. NW of Bozeman, 32 specimens; W bank, Gallatın 
R., 7 mı. W of Bozeman, 7 specimens; 3 mi. N of Gallatin Gateway, 3 specimens; 0.5 mı. SW of Three 
Forks, 3 specimens; 0.5-1 mi. E of Madison R., 15 mi. S. of Three Forks, 5 specimens; S bank, 
Jefferson R., 0.6 mi. N of Three Forks, 1 specimen; 4.25 mi. S, 2.25 mi. W of Willow Creek, 23 
specimens; Jefferson County, 15 mi. SE of Boulder, 1 specimen. Canada, Alberta: vic. of Edmonton, 3 
specimens. 

S. elegans elegans (Kennicott). Colorado: Gilpin County, 1.75 min. N of Rollingsville, 10 
specimens; Eagle County, 6.5 mi. SE of Eagle, 3 specimens; Moffat County, 14 mı. N, 3 mi. W of 


200 C.F.Nadler, R.S. Hoffmann, N. N. Vorontsov, J. W. Koeppl, L. Deutsch and R. I. Sukernik 


Greystone, 1 specimen. Wyoming: Lincoln County, 5 mi. S of Cokeville, 2 specimens; 5 mi. N of 
Fontenelle, 5 specimens; Sweetwater County, 1.25 mi. NE of Superior, 1 specimen. 

S. e. aureus (Davis). Montana: Madison County, 5 mi. E of Cameron, 3 specimens; Wall Creek, 17 
mi. S, 1 mi. W of Cameron, 3 specimens; 0.5 mi. S, 2 mi. E of Harrison, 32 specimens; 1.5 mı. $, 3-3.5 
mi. E of Harrison, 3 specimens; 3.5 mı. $, 1.5 mi. E of Harrison, 1 specimen; 4.5 mı. $, 6 mi. E of 
Harrison, 22 specimens; 6.5 mi. S, 1.5 mi. W of Harrison, 8 specimens; 3 mi. N, 2 mi. W of McAllister, 
1 specimen; 1 mi. N, 2 mi. E of Norris, 1 specimen; 1.25 mi. N, 0.25 mi. E of Pony, 19 specimens. 

S. e. nevadensis (Howell). Nevada: Elko County, 6.5 mi. W of Halleck, 5 specimens. 

S. citellus citellus (Linnaeus). USSR, Moldavian ASSR, no exact locality, 2 specimens. 

S. suslicus suslicus Guldenstaedt. USSR, Ukraine SSR, Kiev Obl., 75 mı. S of Kiev, 10 specimens; 
vic. of Odessa, no exact locality, 2 specimens. 

S. major major Pallas. USSR, Tatarsk ASSR, Ul’yanovsk. Obl., vic. of Cherdakly, 19 specimens. 

S. major ungae Martino. USSR, RSFSR, Sverdlovsk. Obl., no exact locality, 4 specimens. 

S. dauricus dauricus Brandt. USSR, RSFSR, Chitinisk. Obl., 15 mi. SW of Chindet, 9 specimens. 

S. musicus Menetrie. USSR, RSFSR, Karachaevo-Cherkessk. Aut. Obl., no exact locality, 56 
specimens. 

S. pygmaeus brauneri Martino. USSR, RSFSR, Rostovsk. Obl., vic, of Zernograd, 21 specimens; 
Rostovsk. Obl., no exact localıty, 4 specimens. 

S. erythrogenys erythrogenys Brandt. USSR, RSFSR, Novosibirks. Obl., Toguchin, 24 specimens. 

S. fulvus nigrimontanus Antipin. USSR, Kazakh SSR, Dzhambilsk. Obl., vic. of Otar, 2 speci- 
mens; in Kazakh SSR, 28 mi. NE of Frunze, 6 specimens. 

S. relictus relictus Kashkarov. USSR, Uzbek SSR, Tashkentsk. Obl., Kuraminsk. Mts., Angren 
Plateau, 23 specimens. 

S. relictus ralli Kuznetsov. USSR, Kirgiz SSR, Issyk-Kul’sk. Obl., vic. of Przheval’sk, 2 specimens. 

S. undulatus undulatus Pallas. USSR, RSFSR, Irkutsk. Obl., vic. of Angarsk, 4 specimens. 

$. u. eversmanni Brandt. USSR, RSFSR, Altaisk. Krai, Gorno-Altaisk. Aut. Obl., 1.5 mi. N of 
Shebalino, 1 specimen. 

$. u. jacutensis Brandt. USSR, RSFSR, Yakutsk. ASSR, 52 mi. N of Yakutsk, W bank, Lena R., 5 
specimens. 

S. u. menzbieri Ognev. USSR, RSFSR, Amursk. Obl., 15 mi. S of Belogorsk, 1 specimen. 

S. parryıi parryü Richardson. Canada, Manitoba: 5 mi. N of Seal River, W shore, Hudson Bay, 12 
specimens; Northwest Territories: Thelon R., 22 mi. NW of Baker Lake, 17 specimens; 32 mi. NW of 
Baker Lake, 29 specimens; Thelon R. at E end of Schultz Lake, 1 specimen. USA, Alaska: East Fork, 
Chandalar R., Arctic Village, 2 specimens; 25 mi. N of Arctic Village, 22 specimens 30 mi. N of Arctic 
Village, 5 specimens. 

S. p. ablusus Osgood. Alaska: Alaska Peninsula, Cold Bay, 8 specimens; 13 mi. N of Egegik, 3 
specimens; Alaska Range, Otto Lake, 2 specimens; Tikchik Lake, 6 specimens. 

S. p. plesius Osgood. Alaska: Wrangel Mts., no exact locality, 6 specimens. 

S. p. leucostictus Brandt. USSR, RSFSR, Khabarovsk. Krai, Chukotsk. Nats. Okr., Pevek, 14 
specimens; Magadansk. Obl., vic. Atka, 19 specimens; 125 mi. NNW of Magadan, 1 specimen; 
Yakutsk. ASSR, Cherskii, 10 specimens; E bank, Kolyma R., 15 mi. below Sredne Kolymsk, 10 
specimens. 


Laboratory methods 


Blood was drawn from live anımals into heparinized syringes, then centrifuged at ambient tempera- 
tures in the field or in a refrigerated centrifuge in the laboratory to separate plasma and red cells. Red 
cells were washed twice in normal saline and following centrifugation were hemolyzed with an equal 
volume of distilled water and 0.1 M of CCl,. The hemolysate was shaken vigorously, centrifuged and 
the supernatant stored at —20° to —60°C. Multiple small samples of both plasma and red cell 
hemolysates were stored in separate vials for multiple electrophoretic analyses, thus avoiding thawıng 
and refreezing. Livers and kidneys removed in the laboratory were minced in distilled water and 
ground in a glass homogenizer on ice; after centrifugation (10,000 rpm at 04°C) for 30 min, the 
supernatant was lyophillized and stored at —20°C until use. Tissues collected in the field were placed 
directly into plastic vials and stored in a liquid nitrogen freezer until return to the laboratory where 
they were thawed, homogenized, and lyophillized as described above. At the time of electrophoresis 
whole plasma and red cell hemolysates were thawed and analyzed in undiluted form; 10 mg of 
lyophillized tissue was reconstituted with 0.1 ml of distilled water. Samples were absorbed on 
5 x 9 mm pieces of Whatman #3 filter paper prior to insertion in the gel. 

Starch-gel electrophoresis on horizontal or vertical apparatus in a cold room was used for the 
analysis of all proteins; size of the horizontal tray was 6 X14 x 0.9 cm and that of the vertical 
18 x 13 x 0.9 cm. Each gel could be cut into three slices which permitted the examination of several 
proteins separated simultaneously during a sample run. Reference samples of known electrophoretic 
mobilities were included in each run and isozymes of each polymorphic protein were arbitrarily 
designated by numbers or lettes according to their time of discovery, rather than their relative 
mobilities. 


Evolution in ground squirrels II 201 


Serum transferrin (Tf-one locus) was analysed by horizontal starch-gel electrophoresis on 14 % 
Connaught starch according to the methods of PouLik (1957) and PouLık and SMITHIES (1958); these 
data were reported previously (NADLER et al. 1974). Serum albumin (ALB-2 loci?) were examined on 
the half of the gel used to separate transferrins after staining with Buffalo Black. Hemoglobin (Hgb- 
two loci), serum leucine aminopeptidase (LAP-one locus), red cell aldolase (ALD-one locus), red cell 
alpha glycerophosphate dehydrogenase (aGPD-one locus), and red cell lactic dehydrogenase (LDH- 
two locı) were run on a 14 % electrostarch gel made with pH 8.1 gel buffer (0.05 M Tris, 0.0018 M 
EDTA disodium salt, 0.0075 M boric acid) and pH 8.1 electrode buffer (0.3 M boric acid, 0.0625 M 
NAH) and stained according to the techniques of SHaw and PrasAaD (1970). Glucose-6-phosphate 
dyhydrogenase (G6PD-one locus), 6-phosphogluconate dehydrogenase (6PGD-one locus), red cell 
and liver malate dehydrogenase (MDH-two locı), liver and kidney isocitric dehydrogenase (ICD-two 
locı), and liver glutamate oxalate transaminase (GOT-two loci) were analyzed on 12,5 % electrostarch 
utilizing the buffers of EnGEL et al. (1970) and the stains of Suaw and PrasanD (1970). Red cell 
phosphoglucomutase (PGM-two locı) was analyzed in the vertical apparatus utilizing 14 % electro- 
starch, the buffer system of KoeEn (1971) and the stain of SHaw and Prasan (1970). 


Computation methods 


Heterozygosity (h), average heterozygosity (H), and its variance (V};) were all calculated following 
Nei (1975). 

For systematic analysis, electromorphs at each locus for each taxon were coded for presence (1) or 
absence (0) and arranged in two-dimensional arrays for analysis on a Honeywell 66/60 Computer 
using FORTRAN IV programs. Two methods for determining relationships were employed, the 
association coefficient (S) of Jaccard (SNEATH and SokAL 1973), and a rootless Wagner network 
(FARRIS 1970; SNEATH and SokAL 1973). The association coefficient is given by S= a/(a + u), where a 
and u are the number of respective matched (1:1) and mismatched (0:1 or 1:0) electromorphs, 
respectively. S values were arranged in a similarity matrix and using the NT-SYS program package at 
The University of Kansas Computation Center, we performed an unweighted pair group cluster 
analysis (UPGMA, SokAL and SNEATH 1963; SNEATH and SoKAL 1973) from which a phenogram was 
derived. We chose Jaccard’s measure of relationship rather than other measures that have been 
proposed (SELANDER 1970; HEDRICcK 1971; NEı 1972; ROGERs 1972) because (1) preliminary analysis 
showed it was highly correlated with the four measures we tested (see also HEDRıcK 1975); (2) sample 
sizes for many of the populations considered were small and variable, and Jaccard’s $ is less sensitive 
to sample size differences; and (3) it was more appropriate for systematics in that presence of 
electromorphs is more important than relative frequency (Avıse 1974). 


Results 
Protein mobility 


Ten of the 13 variable proteins (representing at least 11 locı) in Holarctic Spermophilus are 
listed in the table together with electromorphs observed at the various loci. Figure 1 
illustrates their relative mobilities. Interspecific variation was observed in ALD, LAP, Hgb 
and ICD,, with different electromorphs fixed in different species. Polymorphism within 
populations was observed in Tf, Alb, PGM,, PGM,, G6PD, 6GPD, and MDH,.. Seven 
additional loci were monomorphic in all species examined, and are not listed (see also 
SEROV et al. 1974). 


Monomorphic loci 


Two lactate dehydrogenase lociı, LDH, and LDH,, were monomorphic in all North 
American and Eurasian species, as was the malate dehydrogenase found in red cell 
hemolysates, (MDH ,.) (cf. SEROV et al. 1974). A second locus, MDH, of liver, was 
polymorphic (see below) (cf. SUKERNIK 1975). Isocitrate dehydrogenase (ICD,) from 
kidney was invariant, whereas the liver enzyme (ICD,) displayed three electromorphs (see 
below). GOT exhibited a fast fraction migrating toward the anode and a slow fraction 
migrating cathodally from the origin, each considered to be under the control of a separate 
locus. These two fraction appeared with identical mobilities in all Holarctic taxa, as did the 
single a GPD fraction observed. 


202 


BErTr 
< 


mmuDmm 


aAd9D 


pn am 
d<d< 


an 


a 


a 


mamma 
d<<<d< 


a 


an 


mmm 
<<< 


WII 


<<< <<< 


<<< 


I-V 0-2 

IZV OD) 

IV. lzV I 

IZV. (0) Ol 
EVa0zV OL Ga Zr 


V 
V 
V 
V 


[eo] 


IV: 0-4 
12V0=V2.°292.03 
ISVa0zV; (28) 

IZV 1a 
IZVSOZV I-qa 


IV. I 
I-V ED 
I-V ID 
a ee) 
I-V (40) 


IwISd qdId9 2 "HAW 


93H 


dVv1I 


aIV 


snpydourssds 913731e]07 Jo syd.ıou10.1393[3 JıydıouAjog 


SIARL. 


snaynf 'S 
solo 'S 
shuadosgyakıa 'S 


‚„dda3s-119s>p,, nS1eafeg 


< 


snavudld 'S 

snasnwu ‘Ss 

INES 

SnIsmsS 'S 
„‚P>4e9-ews,, snoresfeg 


snauunig'S 
snMunvp 'S 
SAP 'S 
„DIETPaULISIU],, 


auy N dvd 'S 
erragis nduvd 'S 
sısusgnaol “usıgzusu 'n 'S 
(dss snujvnpun 'S 
snuviqwn]03 'S 


p>preı-3uo] 


1uogduıgsvm 'S 
suv33J2 (4) 'S 
Nuospavg9ur (4) 'S 
Bupaq (v) 'S 
snpwAv (‘v) 'S 
..P2re3-Z1q,, 99183 N 


aphduru zw (I) 'S 
sısuaogupi wu (3) 'S 
2) S 

Ba (2) 'S 
ıpuasumol (‘I) 'S 


..P2Ies-[fews,, In91e3N 


'q9I uoxe [/dno1ıy 


Evolution in ground squirrels II 203 


+ + + + 
4 a r b N c : d 
1} 
I 
ICD, 1 2 3 ALD 12 2-3 LAP 1 2 Hgb 0 AB BO A cDd EO EF 
{6} 0 o o 
+ + f + 
e ’ en ! g 
Tf 14 1 21239815 734% 105 6 9 7 8 9 1 ALB A-1 B-0 B-1 c-0 c-2 c-3 MDH,; AO A-1 
+ + + 
h i J 


PGM, AA AB BB AC BC cc =— 


6PGD 1-1 1-3 3-3 2-2 24 44 5-5 PGM, A Aor B —— GePDp D A B c 
° ° 


Fig. 1. Relative mobilities of isozymes ın the subgenus Spermophilus. a Isocitrate dehydrogenase; b 

Aldolase; c Leucine aminopeptidase; d Hemoglobin; e Transferrin; f Albumin; g Malate dehydrogen- 

ase; h 6-phospogluconate dehydrogenase; i Phosphoglucomutase; j Glucose-6-phosphate dehyd- 
rogenase 


Interspecifically variable loci 


Isocitrate dehydrogenase. ICH, in liver displayed three electromorphs, each control- 
led by a single fraction (Fig. 1a). 

Adolase. Two different, double-banded patterns, of unknown genetic control, were 
observed (Fig. 1b). 

Leucine aminopeptidase. Two LAP electromorphs, 1 and 2, characterized the 
species studied and they occurred monomorphicaly as single, presumably homozygous 
fractions ın all taxa (Fig. Ic). 

Hemoglobin. Electrophoretic analysıs of hemoglobin revealed species patterns con- 
taining either one or two fractions. Where two fractions were found (S. townsendü group 
except S. mollis; S. suslicus, S. dauricus; S. relictus), intrapopulational polymorphism was 
absent (Fig. 1d). The lack of polymorphism indicates that the double fractions are not 
heterozygotes resulting from segregation of two alleles at a single locus and instead 
suggested that two locı were operative in the production of double-banded hemoglobin 
pattern. Subsequent studies, to be published in detail elsewhere, indicate presence of a ß- 
chain gene duplication. For the purpose of the present taxonomic analysis we have scored 
hemoglobin according to phenotypic pattern with fractions designated by letters A 
through F depending on electrophoretic mobility; single fractions were designated A-O, 
B-O and E-O and double fractions A-B, A-C, C-D and E-F (Fig. 1d). 

Both Palearctic and Nearctic species exhibited two hemoglobins (Table) which suggest 
that the gene duplication responsible for the phenomenon occurred at an early time in the 
evolutionary history of these ground squirrels or that the duplication occurred indepen- 
dently in the two regions. In the case of S. suslicus (Hgb C-D) and S$. relictus (Hgb E-F) 


204 C.F.Nadler, R.S. Hoffmann, N. N. Vorontsov, J. W. Koeppl, L. Deutsch and R. I. Sukernik 


both locı have undergone mutational divergence from the Hgb A-B pattern seen in 
Nearctic species. Furthermore, the Hbg B-O pattern of one species of the townsendii 
group, S. mollis, indicates that the locus responsible for Hgb A production was deleted 
during evolution of that group. 


Polymorphic loci 


Transferrin. Nineteen alleles plus two additional variants of undetermined relative 
mobility (TfA and TfB of S. erythrogenys) have now been identified (Fig. 1e). $. parryii 
populations exhibited sıx alleles (Tf 5, 6, 7, 8, 9, 19), the greatest number of alleles observed 
at any locus or in any species (NADLER and HoFFMANN 1977). Transferrin polymorphism 
was generally associated with those species examined most extensively both in numbers 
and distribution (Nearctic $. parryu, S. columbianus). 

Albumin. Albumin patterns appeared electrophoretically as single or double fractions 
(Fig. 1f). In species exhibiting double fractions, specimens were never observed with single 
fast or slow fractions which would be anticipated if the double fractions represented the 
heterozygous expression of two alleles. We therefore scored albumins on the basis of their 
patterns of phenotypes, the faster fractions designated by letters and the slower fractions 
by a numerical designation or, if absent, a zero. Two locı probably account for the two 
observed fractions in some species. Only $. columbianus, characterized predominantly by 
Alb C-0, displayed intraspecific variation with Alb C-3 appearing in two populations at 
Harrison and Missoula, Montana. 

Malate dehydrogenase. The locus for this enzyme in liver exhibited two patterns of 
undetermined genetic control: pattern A-O consisted of a single fraction whereas pattern 
A-1 was comprised of fraction A and a second, more slowly migrating fraction (Fig. 1g). 

6-Phosphogluconate dehydrogenase. Both red cell hemolysates and liver 
homogenates yielded single fractions with identical mobilities in the case of homozygotes 
and triple fractions in the case of heterozygotes (Fig. 1h); five electromorphs were 
observed. 

Polymorphism occurred in one large population of S. columbianus from Missoula, 
Montana (6PGD-1 and -3), but not in the other populations from Montana. A similar 
polymorphism involving 6PGD-2 and -4 occurred in S. t. townsendiil from the Yakıma 
Airport colony, Yakima County, Washington. All populations of other species were 
monomorphic for a singel electromorph. 

Phosphoglucomutase. The multiband PGM pattern was considered to be control- 
led by at least two loci, the faster constellation of fractions assignable to PGM, and the 
slower group to PGM, (Fig. 11). In PGM,, ground squirrels followed the pattern of genetic 
regulation described in humans where each allele controlled two electrophoretic fractions 
(Horkınson and Harrıs 1969); thus, homozygotes displayed two fractions whereas 
heterozygotes exhibited either four fractions, or three, when one fraction controlled by 
each allele had similar mobilities. Three electromorph designated A, B and C were 
observed at the PGM, locus (Fig. 11). 

The slower fractions were all assıgned to PGM, in the present study; two patterns of 
unknown genetic control designated PGM, A and B were observed (Fig. 11). Attempts to 
explain intrapopulation of variations by postulating control of two fractions by each allele, 
as was possible with PGM,, were not successful, and a third homozygous locus with its 
resultant two fractions added to those of PGM, may exist. Nevertheless, PGM, was scored 
as either PGM, phenotype A or B for the taxonomic purposes of this study. 

At the PGM, locus two pattern, A and B, were recorded in $. columbianus and S. vigılıs 
whereas the other Nearctic taxa and Palearctic $. parryıi and S. undulatus were monomor- 
phic for the A pattern. SUKERNIK (1975) reported both A and B from S. major. PGM gene 
frequency data for certain extensively analyzed species including S$. parryii, S. columbianus, 
S. richardsonii, S. elegans, and S. armatus will be published elsewhere. 


Evolution in ground squirrels II 205 


Glucose 6-phosphate dehydrogenase. Four electromorphs, G6PD A, B, C, and 
D, appeared in Spermophilus, each reflecting a sex-linked allele controlling a single fraction 
(VoRoNTSov et al. 1980), and with identical electrophoretic mobility in both red cell 
hemolysates and liver (Fig. 1j). Although both G6PD-A and -B occurred in the same 
population of $S. armatus, heterozygotes with double-banded patterns were not identified, 
even ın females. 


Discussion 
Biochemical comparisons and systematic relationships 


Biochemical similarity. A phenogram based on the association coefficient of Jaccard 
(proportion of electromorphs in common) between pairs of species and subspecies is 
displayed in Figure 2a. The major division separates most Palearctic from most Nearctic 
species. Among Palearctic Spermophilus, S. relictus is quite isolated; Nearctic $. brunneus 
and $. dauricus are also ısolated, but form a small cluster of their own. The remainder from 
two clusters, S. erythrogenys, $S. major and S. fulvns in one (“subgenus Colobotis”), and the 
remaining species in the other (subgenus Spermophilus [s. s.]). 

Among Nearctic Spermophilus, the townsendii complex is quite isolated from the other 
species. Almost as distant is $S. washingtoni, usually considered on morphological grounds 
as close to $. townsendü (HoweELL 1938). The remaining Nearctic “big-eared” and 
Holarctic “long-tailed’ ground squirrels do not separate along conventional lines, as 
proposed in the subgeneric classification of GROMoVv etal. (1965). The closest relationships 
are shown by Siberian and North American populations of $. parryı, only recently 
separated by the Bering Strait (NADLER et al. 1973b; NADLER and HOFFMANN 1977), and 
the sibling species $. richardsonii and S. elegans, whose divergence is probably also 
postglacial in age (NADLER et al. 1971; NEUNER 1975). 

The above description also fits the results of the unrooted Wagner network, the only 
difference being the somewhat greater distance between the two S$. parryıı groups (Fig. 2b). 

Systematic relationships. Other available evidence-morphological and 
chromosomal as well as biochemical-should be considered in evaluating systematic rela- 
tionships within the Holarctic ground squirrels. We make summary comparisons here, 
since a detailed phylogeny of the subgenus of Spermophilus ıs not yet feasible. 

Spermophilus in the Old’World exhibit somewhat less diversity in chromosome number 
and form, and in external and cranıal morphology, than do New World ground squirrels of 
this subgenus (s. 1.). The majority belong to a group morphologically similar, allopatrically 
distributed taxa sharing the same gross chromosomal morphology (2n = 3642) (Fig. 3). S. 
xanthoprymnus is sometimes considered conspecific with S$. citellus on the basis of external 
morphology (CoRBET 1978; FLiNT et al. 1965) but the two differ in chromosome number 
(2n =42 and 40, respectively). S. suslicus ıs morphologically distinctive with its white 
dorsal spotting, but is known to hybridize with S. pygmaeus (DEnIsov 1963; DENISOV and 
SMIRNOVA 1976) with which it shares a chromosome number of 2n = 36, ın the zone of 
potential contact (Fig. 3a). $S. musicus consists of several isolated, montane populations 
(Fig. 3a) that have often been considered conspecific with $. pygmaeus (CORBET 1978; 
FLinT et al. 1965). Biochemically S. musicus ıs very close to $. citellus but shares 2n = 36 
with S. suslicus and $. pygmaeus, and resembles the latter in external and cranial morphol- 
0gy (GROMoVv et al. 1965). 

$. pygmaeus ıs slightly more divergent biochemically, and morphologically appears to 
link the Palearctic small-eared ground squirrels with the larger, more specialized “desert- 
steppe” ground squirrels some-times referred to the subgenus Colobotis (GROMOV et al. 
1965). This latter group includes three taxa with largely allopatric ranges. $. fulvus to the 
south shares a zone of potential contact with S. major to the northwest and S. erythrogenys 
to the northeast (Fig. 3b); the ranges of the former pair of species overlap in the steppe 


206 


.595 .645 .695 


S. (richardsoniil richardsonii 


S. relictus 


S. columbianus 


S. (citellus) suslicus 


S. [C.) pygmaeus 


\ S. Ic.] musicus 
S. Ic.| citellus 


S. [tlm.?nancyae 


S. (m) major 


S. [t.Jmollis 


S. (major) erythrogenys 


S. [townsendii] vigilis 


S. dauricus 


S. (m) fulvus 


S. brunneus 


S. [tJt. townsendii 


S. [townsendii] vigilis 

S. [t.)t. townsendii 

S. [t.Im.?idahoensis 

S. [tJm.?nancyae 

S. [t.] mollis 

S.[armatus] armatus 

S. [a.Jbeldingi 

S. undulatus ssp. 

S. u. jacutensis, menzbieri 
S. parryii SIBERIA 

S. parryii NORTH AMERICA 
S. columbianus 

S. (richardsoniil richardsonii 
S. {r]) elegans 

S. washingtoni 

S. relictus 

?S. alashanicus 

S. dauricus 

S. brunneus 

S. (citellus) suslicus 
?S.Ic.| xanthoprymnus 

S. Ic.) citellus 

S. [c.] musicus 

S. [C.) pygmaeus 

S. (major) erythrogenys 

S. (m) major 


S. (m) fulvus 


.895 .945 


S. {r) elegans 


S. parryii NORTH AMERICA 


S. parryii SIBERIA 


S. undulatus ssp. 


— S. u. jacutensis, menzbieri 
S. [a.lbeldingi 


S.[armatus] armatus 


S. washingtoni 


S. [t.m. idahoensis 


Fig. 2. Dendrograms based on biochemical similarity. a: Jaccard’s Coefficient; b: rootless Wagner 


network 


80 100 120 140 80N 180 


170E 


N6O 10E 


650MI 
1000KM 


F) IN) ie 3 Y De 
N EN Ele % ; 180 
nellusı 10 40 D p) . R 


| NEE: “ > I 


== 


= 150 
‚ xanthoprymnus 40 
a er 
= en 
Sr 36 
3 SE “ 
alashanicus (?) ee h 140E 
N25 I o _— > e 
25N 
70 80 90 E10O 110 120 EIZO 
Fig. 3a 
N60 10E 70 30 N80 60 80 100 120 140 80N 170E 180 N7O 170W 
N60 2% FL 170 
E1O RS \ 60N 
an 
N) 
\_\p 180 
ENT e 
20 
| 170€ 
Lröpeh [Sn 
LIT 
Ki SS), 
LITE so 
40 IN 160 
III 9 
=> 2n=36 erythrogenys 2n=36 (c : 
major re ; 
de : 
als > Er 
EN D Ä Te 


erythrogenys (?) intermedius Fbrevicauda) 
NL — 150 


e) 
Arne 


B 
30 
ee = 
5 140E 
N25 2 
“ 25N 
E60 70 80 EI 


Fig. 3b 


Fig. 3. Range of the superspecies S. [citellus), including S. /c.] citellus, S. [c.] xanthoprymnus, S. [c.] 

suslicus, S. [c.] pygmaeus, and S. [c.] musicus, and the dauricus species group including $. dauricus, S. 

alashanicus, and $. relictus (a); the superspecies $. [major], including $. /m.] erythrogenys, S. [m.] 
fulvus, and $. [m.] major (b) 


208 C.F. Nadler, R.S. Hoffmann, N. N. Vorontsov, J. W. Koeppl, L. Deutsch and R. I. Sukernik 


zone of central Kazakhstan, while the latter two taxa are reported to overlap in distribution 
between the Toböl and Ishim rivers southeast of Tyumen’, also in Kazakhstan (Fig. 3b). 
Within these zones of sympatry, hybridization is known (DEnIsov 1963; OGNEVv 1947). 
The range of S. pygmaeus overlaps those of two semidesert-steppe taxa (fulvus, major; 
Fig. 3a, b) but though they too share 2n = 36 chromosomes, hybridization between them 
and $. pygmaeus in the zone of sympatry ıs rare (BAzHAnoVv 1944, 1945; DEnIsov 1964). 

Thus, both of these groups of allopatric Palearctic ground squirrels may best be 
regarded as superspecies, listed, in AMADon’s (1966) notation, as $. /[citellus] and S. [major]. 
The low level of morphological differentiation, close biochemical resemblance, and identi- 
cal chromosome morphology argue against conferring subgeneric recognition to the 
desert-steppe ground squirrels; accordingly we do not recognize the subgenus Colobotis. 

Of the remaining three taxa of Palearctic Spermophilus, S. dauricus (Fig. 3a) is mor- 
phologically sımilar to the $. /cıtellus] superspecies, and was considered conspecific with S$. 
citellus by ELLERMAN and MORRISON-SCOTT (1951); ıt also shares the common chromo- 
some number, 2n = 36. Detailed morphological study of $. dauricus by GROMoVv et al. 
(1965) subsequently supported its specific distinction while confirming its close morph- 
ological to S. cıtellus. However, $. dauricus shares a sufficient number of electromorphs in 
common with Nearctic $. brunneus that it clusters with it, rather than its Palearctic 
congeners, although at a low level. S. alashanicus ıs apoorly known species with a montane 
distribution (Fig. 3a), sometimes considered conspecific with dauricns (ALLEn 1940; 
ELLERMAN and MORRISON-ScOTT 1951), and sharing certain morphological characters 
with that species and with S$. relictus (GROMoVv et al. 1965). Its biochemical characteristics 
are unknown, but it possesses 2n = 38 (ORLOV and Davaa 1975), a karyotpe otherwise not 
seen in Palearctic ground squirrels, but fairly common in Nearctic taxa, including 
S. brunneus (NADLER et al. 1973a). S. relictus (2n = 36), also with a montane distribution 
(Fig. 3a), although clustering biochemically with the Palearctic species, does so at a very 
low level; morphologically ıt bears some resemblance to the Nearctic species S. richardsonu 
(2n = 36) (GROMoV et al. 1965). 

The Nearctic townsendii group proper forms a cluster of its own on both biochemical 
and morphological evidence (Fig. 2). However, chromosomal differentiation is consider- 
able. No chromosomally intermediate individuals have been described among townsenduü 
(2n = 36), mollis (2n = 38), and vigilis (2n = 46), and judging from other species within the 
genus, a difference of this magnitude ıs likely to indicate a specific difference (LYAPUNOVA 
and VoronTsov 1970). We consider these allopatric taxa as species or semispecies 
belonging to a single superspecies, $. [townsendi] (Fig. 4a). Although Howet (1938) in 
the most recent revision of North American ground squirrels affıliated $. brunneus with S. 
washingtoni and the townsendi group of “small-eared” squirrels, Davıs (1939) noted that 
brunneus shares many morphological characters with the “big-eared”’ squirrels. As noted 
above, $. brunnens ıs also distinct biochemically, and its restricted range (Fig. 4a) indicates 
a relict distribution. $. washingtoni may also be a relict species (Fig. 4a). Its morphology 
places it with the small-eared group (Howe 1938), but biochemically it clusters with the 
big-eared group, albeit at a faırly low level. 

The remaining Nearctic big-eared ground squirrels include species placed by some 
authors in two subgenera Spermophilus (s. s.) and the long-tailed ground squirrels, 
Urocitellus. The first includes $. beldingi (2n = 30), $. armatus (2n = 34), $. elegans 
(2n = 34), and S. richardsoni (2n = 36). Within this chromosomally diverse group, sym- 
patry and ecological niche segregation is considerable (DURRANT and Hansen 1954). 
However, the pairs exhibiting parapatric distributions (armatus-beldingi; elegans-richard- 
soni) (Fig. 4b, c) are also most similar to one another biochemically (Fig. 2) and mor- 
phologically (Rosınson and HorrMmAnn 1975). While hybridization has been confirmed 
between elegans and richardsonii, it is very limited (NADLER et al. 1971), and between 
armatus and beldingi, it has only been inferred (Davıs 1939; but see Hansen 1956). No 


w130 N60 120W 110 100W 


rn 


N 0 
richardsonii 2n=3 
0 ü 
W ENEERE, 
elegans 2n=34 
AO 
N 
== AO 
N 


beldingi 2n=30 


209 


210 C.F.Nadler, R.S. Hoffmann, N. N. Vorontsov, J. W. Koeppl, L. Deutsch and R. I. Sukernik 


U, 
SS 
mollis ? nancyae 2n=38 


- ; / 
townsendiı 2n=36 ; Rt 
< 


| “ "Washington! 2n=36] 
ı Is En 

; 5 brunneus 2n=38 

u; en | 

<4T 


Fig. 4c 


Fig. 4a-c. Range of the townsendi superspecies including S$. /t.] townsendii, S. [t.] mollıs, and S. /t.] 

vigılis, with the range of the relict species $. brunneus and S. washingtoni also shown (a); the 

superspecies $. [armatus), including $. [a.] armatus and S. [a.] beldingi (b); and S. frichardsonii], 
including S$. /r.] elegans and S. [r.] richardsonii (c) 


hybrids between elegans and beldingi or armatus are known, although they are sympatric 
in some places. Thus, each pair may best be regarded as comprising a distinct superspecies, 
S. [armatus] and S. [richardsoni). 

The last Nearctic species placed by Davıs in the big-eared group was S. columbianus. 
However, this taxon, together with $. parryı and $. undulatus, ıs best regarded as forming 
the group of long-tailed ground squirrels sometimes placed in a distinct subgenus, 
Urocitellus. Together they have allopatric ranges from central Sıberia to northwestern 
North America, centering on the Beringian region (Fig. 5). 

Of the three extant species, S. undulatus ıs the most variable, and has the widest 
distribution (Fig. 5). All subspecies of S. undulatus have a 2n = 32 karyotype which is 
indistinguishable in gross morphology from that of North American $. columbianus, and 
GROMOV et al. (1965) regarded the two species as derived from the same lineage, suggesting 
that the sister species may have diverged as early as the Pliocene (VORONTSOV and 
LyarpunovA 1976). NADLER et al. (1975) suggested, however, that the ısolation between 
Siberian and North American populations probably dates from the end of the Mindel- 
Kansas glacial period. In any event, the level of biochemical similarity between S. 
undulatus and S. columbianus ıs quite low, and parallels the pattern of cranial dıvergence 
(RoBınson and HOFFMANN 1975). 

The only of taxon ground squirrels with a Holarctic distribution at the species level is $. 
parryüi (Fig. 5). Its morphological resemblance to $. undulatus is attested to by the fact that 


Evolution in ground squirrels II 211 


2n=32 (u. menzbieri 
u. jacutensis 
parryli 2n=34 
columbianus 2n=32 


B undulatus 


NORTH POLAR REGIONS 
EQUAL AREA PROJECTION 
1:20,000,000 


Fig. 5. Range of the long-tailed species group, including S. undulatus, S. parryü, and S. columbianus 


they were considered conspecific (HEPTNER 1941; OGnEv 1947; RauscH 1953; HALL and 
Keıson 1959) until GROMOoVv et al. (1963) separated them on morphological grounds. Their 
specific distinctness was later supported by chromosomal evidence (VORONTSOV and 
Lyapunova 1969). Biochemically, the Siberian and North American populations of S$. 
parryü have a very high coefficient of similarity, not surprising in view of their very recent 
(13,000 yr. B. P.) post-glacial ısolation (NADLER et al. 1975). However, as in the case of S$. 
columbianus and S. undulatus, S. parryü shows rather low biochemical resemblance to the 
other two long-tailed ground squirrels, and this is also true of cranial morphology 
(Rogınson and Horrman 1975). In fact, while S. parryıi and S. undulatus cluster with S. 
(armatus), S. columbianus clusters with S. (richardsonı). 


A revised classification of the subgenus Spermophilus 


Genus Spermophilus 
Subgenus Spermophilus 


Nearctic “small-eared’” group 


Spermophilus [towsendii] townsendi 


[t.] vigalis 
[t.] mollıs (incl. idahoensis and nancyae?) 
Spermophilus washingtoni 


Nearctic “big-eared’” group 


Spermophilus [armatus] armatus 


[a.] beldingi 


212 C.F.Nadler, R.S. Hoffmann, N. N. Vorontsov, J. W. Koeppl, L. Deutsch and R. I. Sukernik 


Spermophilus [richardsonit] richardsonii 
[r.] elegans 
Spermophilus brunneus ıncertae sedis 


Intermediate; dauricus group 


Spermophilus dauricus 
Spermophilus alashanicus 
Spermophilus relictus 


Palearctic “small-eared’”’ group 


Spermophilus [citellus] cıtellus 
[c.] xanthoprymnus 
[c.] suslicus 
[e.] musıcus 


[e.] pygmaenus 
Spermophilus [major] major 

[m.] fulvus 

[m.] erythrogenys 


Long-tailed group (= subgenus Urocitellus?) 


Spermophilus undulatus 
Spermophilus columbianus 


Spermophilus parryü 


Acknowledgements 


This paper ıs part of a joint research project co-sponsored by the Academies of Sciences of the US and 
USSR, and the bilateral Environmental Protection Agreement (Project 0.2.05-7104). Funding was 
provided by National Science Foundation grants DEB 72-01851 and 73-01449, the University of 
Kansas General Research Fund, the Institutes of Biology and Pedology, Vladivostok; Cytology and 
Genetics, Novosibirsk; and Biological Problems of the North, Magadan, all of the Academy of 
Sciences of the USSR. We thank the directors of the latter two institutes, Acad. D. K. BELYAEv, and 
Dr. V. L. KonTrimavicHus, for their support. E. A. Lyapunova, L. V. Frısman, Yu. W. Droz- 
Dova, N. I. KALABUKoOV, N. Y. NEKIPELOV, A. F. Popkov, and E. G. PoTAPova assisted us in the 
Altai and eastern Siberia; R. R. PATTERSON, K. GREER, R. L. Rausch, P. B. ROBERTSON, L. TURNER, 
R. BAarRnESs, E. H. CARLsoN, J. BAUMGARNER, in North America, and N. W., C., and C. F. NADLER, 
Jr., S. A. J. F., D. R., and B. E. Horrmann in both Siberia and North America, either in the field or 
by supplying specimens. L. P. MorLokova, D. Pozın, O. L. SErov and I. V. KARTAVTSEvA provided 
laboratory advice or assistance. We also acknowledge the cooperation of the various state and 
provincial game departments in granting us permission to collect. 


Zusammenfassung 


Evolution der Erdhörnchen II. Biochemische Vergleiche holarktischer Populationen von Spermophilus 


Biochemische Ähnlichkeiten innerhalb 21 von 23 anerkannten Arten der Untergattung Spermophilus 
wurden verglichen. Dreizehn Proteine, wenigstens 18 Loci vertretend, wurden untersucht und die 
elektromorphen Variationen beschrieben. Phänogramm und Wagner Netzwerk der genetischen 
Ähnlichkeiten sind dargestellt und zusammen mit morphologischen Daten, mit Daten ihrer 
Chromosomen und Verbreitung benutzt, um die Systematik des holarktischen Spermophilus zu 
interpretieren. Zwei Entwicklungslinien spalten in Nordamerika voneinander ab, die nearktischen 
„Klein-ohrigen‘“ und „groß-ohrigen‘ Erdhörnchen. Die meisten der rezenten Arten in diesen 
Entwicklungslinien haben sich im späten Pleistozän entwickelt. 


References 


ALLEN, G. M. (1940): The mammals of China and Mongolia. Natural history of Central Asia. Vol. 11, 
2. New York: Amer. Mus. Nat. Hist. 
Amapon, D. (1966): The superspecies concept. Syst. Zool. 15, 245-249. 


Evolution in ground squirrels II 213 


Avıse, J. C. (1974): Systematic value of electrophoretic data. Syst. Zool. 23, 465-481. 

AyaLa, F. J.; Tracey, M. L.; HEDGEcocK, D.; RiICHMOND, R. C. (1975): Genetic differentiation 
during the speciation process in Drosophila. Evolution 28 (for 1974), 576-592. 

BAzHanov, V. S. (1944): Gibridy suslikov. Dokl. Akad. Nauk SSSR 42, 321-322. 

BAZHanov, B. S. (1945): K sistematike dvukh vidov suslikov ız Kazakhskoi SSR. Vest. Kazakh. Fil., 
Akad. Nauk SSSR 5, 37-39. 

Back, C. C. (1963): A review of the North American Tertiary Sciuridae. Bull. Mus. Comp. Zool. 
(Harvard Univ.) 130, 109-248. 

Back, C. C. (1972): Holarctic evolution and dispersal of squirrels (Rodentia: Sciuridae). In: 
Evolutionary biology Vol. 6, 10. Ed. by T. DoszHansky, M. K. HEcHT, and W. C. STEERE. 
Appleton-Century-Crofts, New York. Pp. 305-322. 

CoRBET, G. B. (1978): The mammals of the Palaearctic region: a taxonomic review. London: Brit. 
Mus. Nat. Hist. and London and Ithaca, N. Y.: Cornell Univ. Press. 

Davıs, W. B. (1939): The recent mammals of Idaho. Caldwell, Idaho: Caxton Printers. 

Denısov, V. P. (1963): O gibridizatsii vidov roda Citellus Oken. Zool. Zhur. 42, 1887-1889. 

Denısov, V.P. (1964): Rasprostranenie malova (Citellus pygmaeus Pallas) ı rizhatovo (Citellus major 
Pallas) suslikov v Zavol’zhe. Nauch. Dokl. Vysshe Shkol., Biol. Nauki 2, 49-54. 

Denısov, V. P.; SMIRNOVA, N. 1. (1976): Immunological relationships of sousliks genus Citellas in the 
Povolgje region. Acta Theriol. 21, 267-278. 

DURRANT, $. D.; Hansen, R. M. (1954): Distribution patterns and phylogeny of some western 
ground squirrels. Syst. Zool. 3, 82-85. 

ELLERMAN, J. R.; MORRISON-SCOTT, T. C. $. (1951): Checklist of Palaearctic and Indian mammals 
1758-1946. London: Brit. Mus. Nat. Hist. 

EnGEL, W.; BENDER, K.; Kapır, $.; Op’r Hos, J.; WoLk, U. (1970): Zur Genetik der 6-Phosphog- 
luconate-dehydrogenase (EC:1.1.1.44) bei Säugern. Humangenetik 10, 151-157. 

FARRıS, J. S. (1970): Methods for computing Wagner trees. Syst. Zool. 19, 83-92. 

FLiNT, V. E.; CHUGUNov, Yu. D.; SMmIRIn V.M. (1965): Mlekoptitayushchie SSSR. “Misl“””, Moscow. 

GROMOvV, I. M.; GUREEV, A. A.; Novikov, G. A.; SOKOLOV, 1. I.; STRELKOV, P. P.; CHapskı, K.K. 
(1963): Mlekopitayushchie fauni SSSR. Vol. 2, Moscow-Leningrad: Acad. Sci. 

GRoMov, I. M.; BıBıKov, D. I.; KALABUKHov, N. 1.; MEIER, M.N. (1965): Fauna SSSR. Mlekopitay- 
ushchie. Vol. 3, 2. Nazemnye Belich’ı (Marmotinae). Moscow-Leningrad: Nauka. 

Haut, E. R. (1981): Mammals of North America. Vol. 1, New York: J. Wiley. 

Hau, E. R.; Keıson, K. R. (1959): The mammals of North America. Vol. 1, New York: Ronald 
Press. 

Hansen, R. M. (1956): Remarks on reported hybrid ground squirrels, Citellus. J. Mammalogy 37, 
550-552. 

HEDRrIıcK, P. W. (1971): A new approach to measuring genetic similarity. Evolution 25, 276-280. 

— (1975): Genetic similarıty and distance: comments and comparisons. Evolution 29, 362-366. 

HEnnIG, W. (1966): Phylogenetic systematics. Urbana: Univ. Illinois Press. 

HEPTNER, V. G. (1941): Taxonomic and zoogeographical interrelations of some Asiatic and North- 
American ground-squirrels. Zoogeographica 4, 21-27. 

Horkınson, D. A.; Harrıs, H. (1969): Red cell acid phosphatase, phosphoglucomutase and 
adenylate kinase. In: Biochemical methods in red cell genetics. Ed. by G. Yunıs. New York: 
Academic Press. 

Hopkins, D. M. (1972): The paleogeography and climatic history of Beringia during late Cenozoic 
time. Inter-Nord, Paris, 12, 121-150. 

Howeıı, A. H. (1938): Revision of the North American ground squirrels, with a classification of the 
North American Sciuridae. North American Fauna 56, 256 pp. 

Koen, A.L. (1971): Use of blood clots as enzyme source in gel electrophoresis. Biochemical Genetics 
5, 449-450. 

Lorkovı6, Z. (1962): Wesen, Anwendungsbereich und Nomenclatur des Taxon Semispecies. XI 
Intern. Kongr. Entom., Wien, 1960, 3, 325-328. 

Lyapunova, E. A. (1969): Opisanie khromosomnovo nabora i podtverzhdenie vidovoi samostoyatel’- 
nosti Citellus parryi (Marmotinae, Sciuridae, Rodentia). In: Mlekopitayushchie. Ed. by N. N. 
VoRronTsov. Novosibirsk: Acad. Scı. USSR, Pp. 53-54. 

Lyapunova, E. A.; VORONTsov, N. N. (1970): Chromosomes and some issues of the evolution of the 
ground squirrel genus Citellus (Rodentia: Sciuridae). Experientia 26, 1033-1038. 

Mayr, E. (1969): Principles of systematic zoology. New York: McGraw-Hill. 

NADLER, C. F. (1966): Chromosomes and systematics of American ground squirrels of the subgenus 
Spermophilus. J. Mammalogy 47, 579-596. 

NADLER, C. F.; HOFFMANN, R. S. (1977): Patterns of evolution and migration in the arctic ground 
squirrel, Spermophilus parryii (Richardson). Can. J. Zool. 55, 748-758. 

NADLER, C. F.; HOFFMANN, R. $.; GREER, K. (1971): Chromosomal divergence during evolution of 
ground squirrel populations (Rodentia: Spermophilus). Syst. Zool. 20, 298-305. 

NADLER, C. F.; HOFFMANN, R. S.; VORONTSovV, N. N.; SUKERNIK, R. I. (1976): Evolyutsionnye 


214 C.F.Nadler, R.S. Hoffmann, N. N. Vorontsov, J. W. Koeppl, L. Deutsch and RR. I. Sukernik 


svyazi nekotorykh Beringiiskikh mlekopitayushchikh. Pp. 325-336. In: Beringia in Cenozoic. Ed. 
by V. L. KontrımavicHus. Vladivostock: Acad. Scı. 

NADLER, C. F.; LYAPUNOVA, E. A.; HOFFMANN, R. $.; VORONTSov, N. N.; Marycina, N. A. (1975): 
Chromosomal evolution ın Holarctic ground squirrels (Spermophilus). 1. Giemsa-band 
homologies in Spermophilus columbianus and S. undulatus. Z. Säugetierkunde 40, 1-7. 

NADLER, C. F.; SUKERNIK, R. I.; HoFFMann, R. S.; VORONTsov, N. N.; NADLER, C. F. Jr.; 
FOMICHova, 1. I. (1974): Evolution in ground squirrels. I. Transferrins in Holarctic populations of 
Spermophilus. Comp. Biochem. Physiol. 47a, 663-681. 

NADLER, C. F.; TURNER, L. W.; HOFFMANN, R. $.; DEUTSCH, L. (1973a): Chromosomes and Giemsa- 
bands of the Idaho spotted ground squirrel, Spermophilus brunneus (Howell). Experientia 29, 
893-894. 

NADLER, C. F.; VORONTSoOVv, N. N.; HOFFMANN, R. $.; FOMICHOVA, 1. I.; NADLER, C. F. Jr. (1973b): 
Zoogeography of transferrins in arctic and long-tailed ground squirrel populations. Comp. 
Biochem. Physiol. 44b, 3340. 

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

— (1975): Molecular population genetics and evolution. Frontiers of biology. Vol. 40, Amsterdam: 
North-Holland. 

NEUNER, M. (1975): Evolution and distribution of the Spermophilus richardsonüi complex of ground 
squirrels in the middle and late Pleistocene: a multivariate analysıs. Unpubl. M.A. thesis, Univ. of 
Kansas, Lawrence. 

Ocnev, S. I. (1947): Zveri SSSR ı prilezhashchikh stran. Vol. 5. Gryzuny. Moscow-Leningrad: Akad. 
Nauk. 

Orıov, V. N.; Davaa, N. (1975): O sistematicheskom polozhenii Alashanskovo suslika Citellus 
alashanicus Buch. (Sciuridae, Rodentia). In: Sistematika ı tsıtogenetika mlekopitayushchikh. Ed. 
by V. N. Orıov. Moscow: Nauka. Pp. 8-9. 

PETRov, B.; Zıvkovic, $. (1977): Some questions of the actual systematics in the light of use of 
karyological methods. Biosistematika (Beograd) 3, 13-22. 

Pourik, M. D. (1957): Starch-gel electrophoresis in a discontinuous system of buffers. Nature 180, 
1477-1479. 

Pouuik, M. D.; SMITHIES, ©. (1958): Comparison and combination of the starch-gel and filter-paper 
electrophoretic methods applied to human sera: two-dimensional electrophoresis. Biochem. ]. 68, 
636-643. 

Rausch, R. (L.) (1953): On the status of some arctic mammals. Arctic 6, 91-148. 

REıc, ©. A.; AQUILERA, M.; BARROSs, M. A.; UsEcHE, M. (1980): Chromosomal speciation in a 
Rassenkreis of Venezuelan spiny rats (genus Proechimys, Rodentia, Echimyidae). In: Animal 
genetics and evolutions. Ed. by N. N. VoRONTSov and J. M. Van Brınk. The Hague: Junk. Pp. 
291-312. 

REPENNING, C. A.; Horkıns, D. M.; Rusın, M. (1964): Tundra rodents in a late Pleistocene fauna 
from the Tofty placer district, central Alaska. Arctic 17, 177-197. 

Rosınson, J. W.; HOFFMANN, R. S. (1975): Geographical and interspecific cranial variation in big- 
eared ground squirrels (Spermophilus): a multivariate study. Syst. Zool. 24, 79-88. 

RoGeRrs, J. S. (1972): Measures of genetic sımilarity and genetic distance. Studies in Genetics VII 
(Univ. Texas Publ. 7213), 145-153. 

SELANDER, R. K. (1970): Bichemical polymorphism in populations of house mouse and old field 
mouse. In: Variation in mammalıan populations. Ed by R. J. BERRY and H. N. SOUTHERN. New 
York: Acad. Press. 

SEROV, O. L.; VORONTsov, N. N.; FRISMAnN, L. V. (1974): A comparison of enzyme and other protein 
spectors in some species of Citellus (Rodentia). Trans. Ist Instern. Theriol. Congr. Moscow, 1974, 
2, 145. 

SHaw, C. R.; Prasap, R. (1970): Starch-gel electrophoresis of enzymes — a compilation of recipes. 
Biochem. Genetics 4, 297-320. 

SNEATH, P. H. A.; Sokar, R. R. (1973): Numerical taxonomy. San Francisco: W. H. FREEMAN. 

SoKAL, R. R.; SNEATH, P. H. A. (1963): Principles of numerical taxonomy. San Francisco: W. H. 
FREEMAN. 

SUKERNIK, R. I. (1975): Geneticheskie vzaimootnosheniya mezhdu vosem’yu vidami suslikov (Citel- 
lus) po resul’tatamıi elektroforeza belkov i izofermentov krovi. Izv. Akad. Nauk SSSR, Ser. Biol. 5, 
713-718. 

VoroNTsov, N. N.; LyapunovaA, E. A. (1969): Stroenie khromosom Citellus undulatus ı istoriya 
stanovleniya arealov C. undulatus ı C. parryi. Dokl. Akad. Nauk SSSR 187, 207-210. 

— — (1976): Genetica i problemy transberingiiskikh svyazei Golarkticheskikh mlekopitayushchikh. 
In: Beringia in Cenozoic. Ed. by V. L. KontrimavicHus. Vladivastok: Acad. Sci. Pp. 337-353. 

VORONTSov, N. N.; FRISMAN, L.V.; LyapunovaA, E. A.; MEZHOvA, OÖ. N.; SERDYUK, V. A.; 
FoMIcHova, 1. I. (1980): The effect of isolation on the morphological and genetical divergence of 
populations. In: Animal genetics and evolution. Ed. by N. N. VoRONTsov and J. M. Van BRINK. 
The Hague: Junk. Pp. 339-359. 


Evolution in ground squirrels II 215 


VoRoNTSov, N. N.; LyaApunova, E. A.; Borıssov, Yu. M.; Dovcaı, V.E. (1980): Variability of sex 
chromosomes in mammals. In: Animal genetics and evolution. Ed. by N.N VoronTsov and J.M. 
Van Brink. The Hague: Junk. Pp. 361-372. 


Authors’ addresses: CHARLES F. NADLER, LJERKA DEUTSCH, Dept. of Medicine, Northwestern Univ. 
Medical School, 303 E. Chicago Ave., Chicago, IL 60611, USA; Roßerr S. 
HOFFMANN, JAMES W. KoepPpL, Mus. Nat. Hist., Univ. Kansas, Lawrence, KS 
66045, USA; NıkoLaı N. VORONTSoOv, Inst. Develop. Biol. Acad. Sciences, 26 
Vavilov St., Moscow 117334, USSR; Rem I. SUKERNIK, Inst. Cytol. and Genetics, 
Acad. Sciences, Novosibirsk 630090, USSR. 


Zyklische Bestandswechsel (Gradationen) bei der Feldmaus 
(Microtus arvalis), festgestellt durch Analyse 
von Eulen-Gewöllen 


Von E. BETHGE 


Eingang des Ms. 29. 12. 1981 
Abstract 


Cyclic population changes of the common vole (Microtus arvalıs) studied by analysis of owl pellets 


Studied over 13 years (1968/69 until 1980/81) the pellets of longeared owls, which regularly winter ın a 
park at the edge of Würzburg. The pellets contained the remains of 51 682 small vertebrates. The 
numbers of the main prey anımals common vole (Microtus arvalıs) and longtailed fieldmouse 
(Apodemus sylvaticus) and of the captured birds fluctuated regularly with peaks of 3 year intervals, 
except for one 2 year interval. The yearly portion of the common vole and the longtailed fieldmouse 
together was rather constantly 82-95 %. Therefore there was a reciprocal relationship of the 
occurrence of these two species. It was concluded that these fluctuations reflect the real fluctuations of 
the population density of the common vole. The 3 year cycle of the longtailed fieldmouse seemed to be 
dependent upon the course of the cycle of the common vole. 


Einleitung 


Jedes Jahr erscheinen in Deutschland Waldohreulen, um hier den Winter zu verbringen. 
Die Ankunftszeiten dieser wohl aus dem Norden und Osten einfliegenden Eulen liegen in 
Norddeutschland 2-3 Wochen früher als in Unterfranken. Frühester Zeitpunkt war nach 
meinen Beobachtungen in Hamburg der 20. August, in Würzburg der 10. September. Die 
Verweildauer beträgt 5-6 Monate. 


Untersuchungsgebiet und Methodik 


Seit dem Winter 1968/69 wurden an verschiedenen Stellen um Würzburg, seit 1974/75 vor allem ın 
einem Park am Stadtrand von Würzburg, Ansammlungen von Waldohreulen regelmäßig beobachtet 
(BETHGE 1975). Dieser auf der „Sieboldshöhe“ gelegene Park ist etwa 8ha groß und erhält jeden 
Herbst Besuch von 14 bis 40 Eulen. Unter den Ruhebäumen, 20 m hohen Schwarzkiefern (es sind fast 
immer die gleichen Bäume besetzt) wurden regelmäßig Gewölle aufgesammelt, um ihren Inhalt 
auszuwerten. So entstand in 13 Wintern zwischen 1968/69 und 1980/81 eine Gesamt-Beuteliste von 
51 682 Wirbeltieren in etwa 30000 Gewöllen. Käfer und andere Wirbellose spielen bei der Winternah- 
rung keine Rolle. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4704-0215 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 215-219 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


216 E. Bethge 


Ergebnisse und Diskussion 


Die Tabellen 1 und 2 zeigen die Zahlen der von den Waldohreulen der Sieboldshöhe 
erbeuteten Kleinsäuger. Die Feldmaus stand, wie vorauszusehen war, in der Nahrung der 
Waldohreulen an der Spitze. Sie war mit einem Anteil von 69,5 % (im Durchschnitt der 13 
Jahre) in der Beute vertreten. Von den Apodemus-Arten kommen Gelbhalsmaus (A. 
flavicollis) und Waldmaus (A. sylvaticus) in Frage. Sie sind im Knochenbau, der für die 
Bestimmung aus den Gewöllen maßgebend ist, schwer zu unterscheiden. Jedoch kann man 
nach MoHRr (1954) die beiden Arten durch Messung der Unterkieferzahnreihen trennen. 
Es wurden 1600 Messungen an den Apodemus-Exemplaren aus der Waldohreulen-Beute 
der Sieboldshöhe vorgenommen und mit 800 Apodemus-Schädelteilen aus Waldkauz- 
Gewöllen eines 10km entfernten Waldes verglichen (Tab.3). Aus den Zahlenwerten läßt 
sich folgern, daß die Apodemus-Beute der Waldohreulen von der Sieboldshöhe vorwiegend 
aus Waldmäusen bestand. Bedenkt man weiter, daß die Waldohreule im offenen Gelände 


Tabelle 1 
Waldohreulen-Winternahrung von 1968/69 bis 1980/81 = 13 Winter 


Zahl der in den Gewöllen gefundenen wichtigsten Beutetiere 
(nur Wirbeltiere), aufgeteilt nach Jahren 


Feldmaus Waldmaus sonstige Säuger 


1968/69 
1969/70 
1970/71 
1971/72 
1972/73 
1973/74 
1974/75 
1975/76 
1976/77 
1977/78 
1978/79 
1979/80 
1980/81 


Summe 
= Anteil 


! davon 2 Anuren. 


Tabelle 2 


Gesamtzahlen erbeuteter Kleinsäuger, die nicht in Tabelle 1 aufgeführt sind 


Endsummen von 13 Wintern 


Erdmaus Ratte spec. 
Rötelmaus Mauswiesel 
Schermaus Maulwurf 
Haselmaus Hamster 


Zwergmaus Kaninchen 
Hausmaus Siebenschläfer 
Weißzahnspitzmaus 

Waldspitzmaus 

Zwergspitzmaus 


Zyklische Bestandswechsel bei der Feldmaus 2177 


Tabelle 3 


Länge der Unterkieferzahnreihe (UKZ) bei Apodemus-Schädeln aus Eulen-Gewöllen 
Würzburger Herkunft und Vergleichswerte bei Mohr (1954) 


Eulenart Zahl der Messungen UKZ-Längen in mm (Mittelwerte) 


Waldohreule 1600 
Waldkauz 800 
Vergleichswerte von MoHR: 


Waldmaus 
Gelbhalsmaus 


jagt, wo die Waldmaus lebt, so kommt man zu dem Schluß, daß die den Wald bewohnende 
Gelbhalsmaus bei der Ernährung der hiesigen Waldohreulen keine Rolle spielt und in der 
Aufstellung weggelassen werden kann. 

4,3% der erbeuteten Wirbeltiere waren Vögel, in mindestens 30 verschiedenen Arten. 
Am meisten kamen Haussperlinge (690 Exemplare), Kohlmeisen (230) und Feldsperlinge 
(208) vor. Gewichtsmäßig würden allerdings die Amseln (152 Stück) an die zweite Stelle 
rücken. 


Beute-Anteil (in %) von Feldmaus (F), Waldmaus (W) und Vögeln (V) im Verlauf der Jahre (Winter 
1968/69 bis Winter 1980/81) 


Die Zusammensetzung der Beute im Verlauf der Jahre ist in der Abbildung dargestellt. 
Der Beuteanteil sowohl der Feldmaus als auch der Waldmaus und der Vögel schwankt in 
sehr ausgeprägten Jahreszyklen, wobei die Kurven für die Waldmaus und die Vögel genau 
spiegelbildlich zur Feldmaus-Kurve verlaufen. Die Summe der Anteile von Feldmaus und 
Waldmaus bleibt während der ganzen sich über 13 Winter erstreckenden Beobachtungszeit 
annähernd konstant bei 82-95 %. (Vom Winter 1970/71 wurde dabeı wegen der geringen 
Beutemenge abgesehen.) Die Feldmaus allein deckt schon mit durchschnittlich 69,5 % über 
2/3 des Nahrungsbedarfs. Ihr Kurvenbild, also der jeweilige Beuteanteil, zeigt 1969/70 mit 
83 %, 1972/73 mit 88%, 1974/75 mit 84%, 1977/78 mit 91% und 1980/81 mit 64 % 
ausgesprochene Höhepunkte, die ın dreijährigem Abstand folgen. Dieser Rhythmus wird 
einmal unterbrochen: Zwischen 1972/73 und 1974/75 liegen nur 2 Jahre. Wenn auch das 
hier benutzte Material nur den Zeitraum von jeweils 5 Wintermonaten umfaßt, kann man 
doch schließen, daß in unserem Fall bei der Feldmaus die Höhe des Anteils an der Beute 


218 E. Bethge 


vom jeweiligen Bestand bestimmt wird (vgl. WENDLAND 1981). Damit weist die Popula- 
tionsdichte also ebenfalls einen im allgemeinen dreijährigen Rhythmus auf. 

Die Höhepunktsjahre und die Zeitabstände solcher Gradationen brauchen nicht in 
verschiedenen Gebieten die gleichen zu sein. Doch müssen solche Unterschiede umso 
kleiner sein, je geringer der räumliche und klimatische Abstand ist. Nun hat Kaus (1977) 
ebenfalls in Franken das Jahr 1974 als Gradationsjahr der Feldmaus besonders erwähnt. 
Für die Schwäbische Alb gibt ROcKENBAUCH (1978) als Höhepunktsjahre der Feldmaus 
1966, 1969/70, 1974, 1977 an, wobei auch er 1974 hervorhebt. Daß der Feldmausbestand 
im Würzburger Raum zwischen 1972 und 1974 nur eine zweijährige Erholungszeit bis zum 
nächsten Höhepunkt brauchte, hat vielleicht seinen Grund in besonders günstigen klimati- 
schen Bedingungen, die zu dieser Zeit hier herrschten. WENDLAND (1981) hat im Berliner 
Raum für die Feldmaus einen vierjährigen Zyklus und für die, der Waldmaus nahe 
verwandte Gelbhalsmaus einen dreijährigen Zyklus im Wechsel der Populationsdichte 
gefunden: In den Beutelisten nicht nur der Waldohreulen, sondern auch der Waldkäuze 
wies die Gelbhalsmaus alle 3 Jahre einen Höhepunkt auf, der in 3 gemeinsamen Beobach- 
tungsjahren fast die gleichen Prozentwerte erreichte. Einen dreijährigen Rhythmus beim 
Auftreten von Feldmausplagen hat BoHnsAck (1966) aus Kirchenbüchern Dithmarschens 
schon für die Zeit zu Ende des 18. Jahrhunderts (1772, 1775, 1778, 1781) ermittelt. Er 
selbst hat in Westholstein Wintergewölle von Schleiereulen untersucht. Hauptbeutetier 
war auch hier die Feldmaus mit den Höhepunktsjahren 1952, 1955, 1957, 1960, 1963. Wie 
im Würzburger Gebiet wird der Dreijahreszyklus auch hier einmal durch ein zweijähriges 
Intervall abgelöst. Auch dort besteht die Nahrung zu über 80 % aus 2 Wirbeltier-Gruppen. 
In diesem Fall sind neben der Feldmaus nicht die Waldmaus, sondern verschiedene 
Spitzmausarten die wichtigsten Beutetiere. So entstehen auch bei BoHnsAck (1966) 2 
spiegelbildlich verlaufende Kurven, und auch er hält die Kurve der Feldmaus für die 
ursprüngliche und entscheidende. Auch er schließt von den Höhepunktsjahren in der 
Beute auf echte Gradationsjahre der Feldmaus. Ähnlich schreibt BEcker (1955, vgl. auch 
BECKER 1958): „Von der Waldspitzmaus (Sorex araneus) sind ebenfalls seit langem Jahre 
mit großer Populationsdichte bekannt, die in auffälligem Gegensatz zu solchen mit dünner 
Siedlungsdichte stehen. ... Ob diese Bestandsschwankungen mit einem echten Massen- 
wechsel gleichzusetzen sind, ... dürfte allerdings noch unbekannt sein.“ 

In den Würzburger Untersuchungen war die Waldmaus das zweithäufigste Beutetier. 
Ihr Anteil an der Gesamtbeute zeigt zwar Höhepunkte, diese sind aber immer vom 
Feldmaus-Anteil abhängig und ihm jeweils entgegengesetzt gerichtet. Ein direkter Zusam- 
menhang mit eigenen rhythmisch verlaufenden Bestandsänderungen läßt sich demnach bei 
der Waldmaus aus dem Kurvenbild nicht ableiten. Vielleicht zeigt der Kurvenverlauf nur 
an, daß die Waldmaus für die Sieboldshöhen-Eulen einfach eine Ersatznahrung in Tief- 
punktjahren der Feldmaus darstellt, eine Nahrungsquelle, die immer in ausreichender 
Menge zur Verfügung steht und keine regelmäßig verlaufenden Schwankungen erkennen 
läßt. Das gleiche gilt für die Vögel, deren Gesamtkurve der der Waldmaus parallel verläuft. 

Die Frage nach den Ursachen solcher Gradationen bei Wühlmäusen wie den Feldmäu- 
sen und Lemmingen ist noch nicht geklärt. Äußere Einwirkungen, etwa durch Änderung 
des Kleinklimas, kommen wohl höchstens als Auslöser in Frage. FRAnK (1953, 1956) gibt 
innere Gründe, nämlich einen Wechsel im Hormonhaushalt, an. Auch Kress und Myers 
(1974) halten endogene Faktoren für ausschlaggebend. 


Danksagungen 


Meinem Freund VICTOR WENDLAND möchte ich für seine Hilfe bei Entwurf und Durchsicht dieser 
Arbeit herzlich danken, ebenso Herrn WALTER FÜCHTBAUER für Hilfe bei der Abfassung des 
Manuskripts und Herrn Davıp UNGeR für Mitarbeit beim Aufsammeln und Zerlegen der Gewölle. 


Zyklische Bestandswechsel bei der Feldmaus 219 


Zusammenfassung 


Von Waldohreulen, die regelmäßig ın einem Park am Stadtrand von Würzburg überwintern, wurden 
13 Jahre hindurch Gewölle gesammelt und analysiert. Die Zahlen der in den Gewöllen gefundenen 
Feldmäuse, Waldmäuse und Vögel schwankten in regelmäßigen Jahreszyklen. Die beiden wichtigsten 
Beutetiere, Feldmaus und Waldmaus, machten zusammen einen weitgehend gleich bleibenden Anteil 
von 82-95 % der Gesamtnahrung aus; ihre Kurven, die diese Anteile chronologisch darstellen, 
verlaufen demnach spiegelbildlich: In Wintern großer Feldmaus-Beute wurden nur wenige Wald- 
mäuse und Vögel erbeutet, und umgekehrt. Für die Feldmaus, deren Beuteanteil im Durchschnitt 
69,5% beträgt, kann man aus dem Beuteanteil der einzelnen Jahre auf den jeweiligen Bestand der 
Population schließen. Es ergaben sich die Höhepunktsjahre 1969, 1972, 1974, 1977, 1980; die 
Gradationen folgten also in einem Abstand von 3 Jahren, einmal von 2 Jahren. Die Schwankungen im 
Beuteanteil der Waldmaus sind eine Folge der Gradationen der Feldmaus und von diesen abhängig. 
Ein eigenständiger Rhythmus ist demnach bei der Waldmaus nicht erkennbar. 


Literatur 


BECKER, K. (1955): Über Art- und Geschlechtsunterschiede am Becken einheimischer Spitzmäuse 
(Soricidae). Z. Säugetierkunde 20, 78-88. 

— (1958): Die Populationsentwicklung von Feldmäusen (Microtus arvalıs) im Spiegel der Nahrung 
von Schleiereulen (T’yto alba). Z. Angew. Zoologie 45, 403-431. 

BETHGE, E. (1975): Eulen im Würzburger Raum und ihre Ernährung, vor allem im Hinblick auf das 
Vorkommen von Kleinsäugern. Abh. Naturwiss. Ver. Würzburg 16, 19-34. 

BoHnsack, P. (1966): Über die Ernährung der Schleiereule, 7yto alba, insbesondere außerhalb der 
Brutzeit, in einem westholsteinischen Massenwechselgebiet der Feldmaus, Microtus arvalıs. Corax 
1-4, 162-172. 

FRANK, F. (1953): Zur Entstehung übernormaler Populationsdichten im Massenwechsel der Feldmaus 
(Microtus arvalıs). Zool. J. 81, 610-624. 

— (1956): 5 Jahre Zyklenforschung an einheimischen Kleinsäugern. Beiträge z. Naturk. Niedersach- 
sens, Weigold-Festschrift, 88-95. 

Kaus, D. (1977): Zur Populationsdynamik, Ökologie und Brutbiologie der Schleiereule Tyto alba in 
Franken. Anz. Orn. Ges. Bayern 16, 18—44. 

KreBs, C. J.; Myers, J. H. (1974): Population Cycles in Small Mammals. Adv. Ecol. Res. 8, 267-399. 

MoHR, ERrNA (1954): Die frei lebenden Nagetiere Deutschlands und der Nachbarländer. Jena: VEB 
Gustav Fischer. 

ROCKENBAUCH, D. (1978): Brutbiologie und den Bestand steuernde Faktoren bei Waldkauz (Strix 
aluco) und Waldohreule (Aszo otus) in der Schwäbischen Alb. J. Orn. 119, 429-440. 

WENDLAND, V. (1981): Cyclic Population Changes in Three Mouse Species in the same Woodland. 
Oecologia 48, 7-12. 


Anschrift des Verfassers: ERICH BETHGE, Unterer Katzenbergweg 3/b, D-8700 Würzburg-H. 


Analyse des Explorationsverhaltens verschiedener Stämme und 
Generationen der Hausmaus (Mus musculus domesticus Rutty) 


II. Vergleichende Betrachtungen der Parental- und Filialgenerationen 


Von M. W. SCHÄFER 


Fachbereich Biologie (Zoologie) der Universität Frankfurt a. M. 


Eingang des Ms. 5. 11. 1981 
Abstract 


Analysis of exploratory behaviour in different strains and generations of the house mouse 
(Mus musculus domesticus Rutty). II. Comparative investigations of different generations 


Tested were white and grey strains of house mice kept under similar laboratory conditions demon- 
strating significant differences ın their exploratory and escaping behaviour under “open-field’ — 
conditions. 

The behavioral strategies of the grey individuals still included distinctive changes from sitting and 
hiding to fast movements and jumping, expressing adaptations to natural conditions. The white strain 
lost the ability of effective escapıng tendencies demonstrating phases of movement in continuity. 

The hybrids of the F,- and F,-generations showed behavioral patterns of both parents in opposite 
combinations; although, however, some results pointed out a certain “dominant” influence of “wild” 
behaviour, ı.e. the F,-hybrids especially appeared to be more equal to the grey parents, a fixed 
combination between coat colour and behaviour could not be found. 

The differences between white and grey parents are interpreted as results of different times - and 
therefore different genetic patterns! — of inbred conditions which are resulting ın decreasing avoidance 
strategies in the albinos. 

Hybrid-generations are containing individuals of “mixed” types combining parameters of the 
parental generations with much more varıability than fixed relations to their origin (combination of 
parental sexes) and coat colours. Therefore, parameters of exploration and neophobia are accepted as 
expressions of genetic effects but, based on a differenciated analysis of hybrid-behaviour, the 
interpretation of “wildness’” as an outcome of some dominant genes is rejected. 


Einleitung 


Verschiedene Stämme der Hausmaus (NMRI-Albinos, graue ‚„Wildformen‘“) aus identi- 
schen Haltungsbedingungen zeigen auf einer weitgehend deckungslosen Lauffläche 
(„open-field“) deutliche quantitative Unterschiede in ihrem Explorationsverhalten (ScHÄ- 
FER 1982). In der vorliegenden Arbeit wird das Verhalten von F,-Hybriden aus reziproken 
Kreuzungsansätzen beider Stämme sowie von F,-Hybriden aus Kreuzungen der ersten 
Filialgenerationen unter dem Aspekt genetischer Grundlagen vergleichend geprüft. 

Zur Methodik der Verhaltensaufnahme und zur Klassifizierung der abgekürzt aufge- 
führten Verhaltensparameter siehe Teil I, Kap. ‚Material und Methode“ 


Ergebnisse 
„Erstarrtes Sitzen‘ und ‚„klebendes Laufen“ 


Zwei gut abgrenzbare Verhaltensweisen sind das ‚‚erstarrte Sitzen‘, ein bewegungsloses 
Verharren in relativ beliebiger Körperstellung und das „klebende Laufen“, eine zögernde, 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4704-0220 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 220-230 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Analyse des Explorationsverhaltens verschiedener Stämme der Hausmaus II PEN 


mit Schnuppern und starker Körperstreckung einhergehende Fortbewegungsart. Bereits 
qualitativ sind die beiden Parentalstämme ‚‚weiß‘ und „‚grau“ nach Auftreten dieser beiden 
Verhaltensformen trennbar: Weiße Tiere (w, w’) zeigen nie erstarrtes Sitzen, graue (g, g’) 
nie klebendes Laufen. 

Prüft man nun jeweils den prozentualen Anteil der Individuen, die in den Filialgenera- 
tionen die genannten Verhaltensweisen zeigen, so lassen sich Hinweise auf Beziehungen 
zur Fellfärbung, zum Geschlecht der Elterntiere oder zum Heterozygotiegrad in der 
Parentalgeneration entnehmen. So liegt der Anteil erstarrt sitzender Individuen der 7, bei 
Tieren mit grauer Mutter (7, /g/) höher als bei solchen mit weißer (F, [w)), graue wie weiße 
Tiere der F, (gF,, wF,) mit dieser Verhaltensweise sind auffindbar, der Anteil grauer 
Individuen ist jedoch in beiden Fällen größer (Abb. 1B). 

„Klebendes Laufen“ findet sich unabhängig von der Kombination der Elterntiere 
jeweils etwa nur bei 10 % der F,, in der F, liegt der Anteil bei weißen wie grauen Tieren mit 
weißer Großmutter (im folgenden als wF,/w] und gF,/w] bezeichnet) höher als bei den 
Gruppen wF, und gF, mit grauer Großmutter. In jedem Fall scheint auch die Fellfärbung 
additiv Wirkung zu zeigen: So finden sich mehr Tiere der F,(w) mit „klebendem Laufen“ 
als solche der F,(g), jedoch liegt der Anteil der wF,(w) wie auch der wF,(g) jeweils höher als 
der der zugehörigen grauen Tiere gF,(w) und gF,(g). Ein Summationseffekt kann dahinge- 
hend vermutet werden, daß von den Tieren der wF,(w) 2/3, von jenen der gF,(g) jedoch 
nur etwa 1/20 „klebendes Laufen“ zeigen (Abb. 1B). 


> 


| 
) 


46,9 71,4 9,4 Fı| 10,7 
6 | 80 96 |9C 
y } } t 
wP,| 33, 3 40,0 WFal 66,7 wF>| 20,0 
gR| 73,3 gF2| 57,9 [gF2 | 33,3 Sal 38 


Abb. 1. Prozentuale Anteile an Individuen verschiedener Gruppen und Generationen, die ‚‚erstarrtes 

Sitzen“ (A) und ‚„klebendes Laufen“ (B) zeigen. Die jeweils linken Spalten geben die Verteilungen aus 

der Kreuzung weißer Weibchen mit grauen Männchen wieder, die rechten Spalten den reziproken 

Ansatz. w’ und g’ repräsentieren transplantierte Individuen, wF, und gF, weiße bzw. graue Tiere der 
zweiten Filialgeneration. (Stichprobenumfänge entspr. Tab. 1, SCHÄFER 1982) 


Vergleich der Parentalgenerationen w und g mit den Filialgenerationen F, und F, 


In den Abb. 2 und 3 sind die Ergebnisse eines relativ pauschalen Vergleichs aufgeführt, der 
die Werte der Gruppen w und g sowie der gesamten F, (F,/w] + F,/g]) und der gesamten F, 
(wF/[w] + gF/w] + wF,[g] + gFılg)) umfaßt. 

Pro Vierergruppe wurde auf signifikante Unterschiede mittels des H-Tests (KRUSKAL 
und Waruıs, modifiziert nach RAATZ 1966) geprüft, im Falle eines signifikanten A/-Wertes 
wurde paarweise der Vergleich nach NEMmEnYI (in SAacHs 1978) angewandt; Signifikanzni- 
veau 0,01. 


222 M. W. Schäfer 


Die zusammenfassende Wertung der statistischen Vergleichsverfahren innerhalb der 
Vierergruppen sowie paarweiser Vergleiche ergibt im einzelnen folgendes Bild: 

Für die Parameter a.B., t/a.B. und t/i.B. sowie „Urin“ sind keine Unterschiede 
sicherbar. 

Für die Gruppen w und g sind in sieben Fällen paarweisen Vergleichs keine Unter- 
schiede sicherbar. 

Signifikant größere Anteile zeigen die weißen Tiere bei „Schnuppern” und „Putzen“ (t/ 
Sch, t/P; Sch/ Akt, P/Akt), auch erreicht die aktıv verbrachte Zeit höhere Werte (Akt), die 
mittlere Dauer t/Akt ıst allerdings nicht von jener der grauen Tiere signifikant verschieden, 
da offenkundig die weißen Individuen eine größere Anzahl aktiver Verhaltensweisen 
zeigen. Umgekehrt sitzen die grauen Mäuse mehr und länger (Sit und t/Sıt, Abb. 2 und 3), 
wodurch die mittlere Dauer der gesamten Verhaltensweisen größer wird (t/Ges, Abb. 3). 
Auch nimmt dadurch der Anteil der Bewegungen an der verbleibenden aktiven Zeit 
genauso zu wie die Zahl überlaufener Felder der Versuchsfläche während der Aktivität 
(Bew/Akt und F/akt.min). Schließlich nehmen Aktivitäten am Becher (a.B./Akt) ebenfalls 
einen größeren Anteil ein. Besonders zu betonen ist die hohe Sprungaktivität der grauen 
Tiere (Abb. 3), während die weißen fast nie springen (ein Tier einmal). 


100 100 


60 60 


Y 


N 


= 
=9 
=7 
=9 
=7 
=9 
=7 
=7 
=7 
=g 
=9 
= 
= 
EV 
EH 


LLLITELLTIITITITTD 
III 
INBRURERDRDRN 
SU 


III 


.B. i.B. Si Sch A Sch a w /Akt 
EN 2 EN Deren 


Abb. 2. Vergleich der Gruppen w (weiße Säulen), g (schwarze Säulen), F, (waagerechte Schraffur) und 

F, (Schrägschraffur) bezüglich der Dauer verschiedener Verhaltensweisen, aufgetragen als mittlerer 

Prozentwert mit Standardabweichung der Gesamtbeobachtungszeit (linkes Diagramm) und bezüglich 

der Dauer als mittlerer prozentualer Anteil der Aktivitätszeit (rechtes Diagramm). Die Existenz 

signifikanter Unterschiede innerhalb der jeweiligen Vierergruppen wird pro Verhaltensparameter 

durch einen Pfeil gekennzeichnet. Weitere Erläuterungen im Text sowie ım Kapitel Material und 
Methode, SCHÄFER 1982 


S 


IIIIIIIIIIISSIIISINSS 


IIIIIIINI 
IIABNUIHIUHURHNNUTIINUNGNI 
IIIIIIIIIIIIINI 
IMERENNRINMENNNNUNWERDLNNN 


ST Z Se FEN) 5 
= HAND 3 = 
> 5 S - 73 >) — 
3, 
p>) 


Abb. 3. Vergleich der Gruppen w, g, F, und F, (Säulen entspr. Abb. 2 zugeordnet) bezüglich der 

mittleren Dauer verschiedener Verhaltensweisen in Sekunden (linkes Diagramm) sowie der mittleren 

Anzahl der Ereignisse „Kot“, „Urin“, „überlaufene Felder‘ pro verschiedener Zeitintervalle und 

„Springen“ (rechtes Diagramm). Signifikante Unterschiede durch Pfeile gekennzeichnet. Weitere 
Erläuterungen ım Text 


Analyse des Explorationsverhaltens verschiedener Stämme der Hausmaus II 223 


Setzt man die mittlere Dauer der Aktivität ins Verhältnis zur Dauer der gesamten 
Verhaltensweisen, so ergibt sich für w: /Akt:t/Ges= 0,98, für g: 0,49. Die aktiven 
Verhaltensweisen dauern also bei den weißen Tieren etwa genauso lang wie die inaktiven, 
bei den grauen sind letztere deutlich länger. 

Weiße Mäuse putzen sich mehr, schnuppern mehr und sind insgesamt weniger mobil als 
graue Tiere, diese verbringen einen größeren Teil der Beobachtungszeit bewegungslos, 
zeigen jedoch größere Mobilität in den verbleibenden aktiven Phasen. Die beiden Parental- 
generationen sind damit deutlich im Explorationsverhalten unterscheidbar. Hinweise auf 
eine genetische Basis dieser Befunde sollten den Daten der Filialgenerationen entnommen 
werden können. 

Aus den in Abb. 2 und 3 dargestellten Daten ergeben sich unter Anwendung des A- 
Testes lediglich Hinweise auf mögliche signifikante Unterschiede innerhalb der Vierer- 
gruppe w-g-F,-F, der sich anschließende paarweise Vergleich läßt eine Gruppierung 
der Befunde aller jeweils signifikanten Unterschiede wie folgt zu: 

1. wund g sind unterschiedlich, F, und F, sind voneinander und von einer der Parentalge- 
nerationen nicht unterscheidbar. Dies gilt zum einen für P/Akt, t/Sch und t/P, wo die 
Filialgenerationen nicht voneinander und nicht von g zu unterscheiden sind und ihre 
Werte einheitlich unter denen der weißen Tiere liegen; es gilt also: 


Ge ze) u 


Umgekehrt gilt für Springen: (, = RB =g)>w. 
Für Sch+ Agılt: (F,= BR = w)>g, für t/Ges und a.B./Akt: (F,= F, = w) <g, und mit 
Einschränkung für t/Akt: FR = BR = g<w bzw. für F/bew.min: FR =R,=g>w. 

2. w und g sind unterschiedlich, F, gleicht einer Parentalgeneration, F, liegt wertemäßig 
dazwischen; dies gilt für Sch: w>(g= F,<F,), für 1/Sıt: w< (g= F,> F,), allerdings 
sind hier F, und F, nicht signifikant trennbar, und schließlich mit einer gewissen 
Abweichung für Bew/Akt: gS>p(w=R,<F,=FR,), d.h. die F,-Werte liegen zwar 
zwischen.jenen der grauen und weißen Tiere, allerdings ist die A, zwischen w und g 
liegend, von F, nicht signifikant trennbar und auch nicht von w! 

3. w und g sind unterschiedlich, die Filialgenerationen liegen mit + mittleren Werten 
zwischen den Eltern; entsprechend gilt für Pund Akt: w>F,= F,>g und für Sit und 
Flakt.min.: w<F,= F<g. 

4. w und g sind relativ gleich, ın den Filialgenerationen treten stärkere Merkmalsausprä- 
gungen auf. So gilt für A:w=g<F,=F, für 4/Akt und YA: w=g<F,>F,. 

Schließlich gilt für die Parameter Sch + A/Akt und „Kot“, daß w, g und F, nicht 

voneinander sicherbar sind, auch F, und F, nicht, wohl aber F, signifikant größere Werte 

zeigt als g, und für Sch/Akt, daß die weißen Tiere mehr schnuppern als g und 7, allerdings 

nicht signifikant trennbar von F,sind, die wiederum einen größeren Wert zeigt als gund 7’. 


Datenverteilungen der Filialgenerationen 


Sicherbare Unterschiede zwischen F,(w) und F,(g) sind geeignet, mögliche Beziehungen 
zur Kombination der Elterntiere zu liefern, wobei mit dem U-Test (in MÜHLENBERG 1976 
und Sachs 1978) für p<0,01 geprüft wurde. In der vorliegenden Stichprobe sind die F/- 
Hybriden weißer Mütter durch eine höhere „Schnupperrate‘ ausgezeichnet. D.h. gegen- 
über F,(g) haben die Parameter Sch, Sch+ A, Sch/Akt und Sch + A/Akt den höheren 
Anteil, entsprechend der längeren Dauer einer „Schnupperfolge“ (t/Sch). Gleichsinnig 
verhalten sich die mobilitätsbezogenen Parameter Bew, Bew/Akt, F/min, F/akt.min und 
F/bew.min, die alle jene der F,(g) übertreffen. 

Auch in der F, finden sich für „Bewegung“ und überlaufene Felder (Bew, Bew/Akt und 
F/min) die größeren Werte bei den von weißen Weibchen abstammenden Tieren (F/w/)). 
Die F;(g) liefert hingegen größere Werte für Putzen als Aktivitätsanteil (P/Akt) und für 


224 M. W. Schäfer 


Sitzen (t/Sit). Unter Berücksichtigung zunehmender Heterozygotie der Hybriden und 
möglicher Heterozygotie der Eltern müssen mögliche Hinweise auf eine ‚„‚Geschlechts- 
kopplung“ (besser heterosomaler Einfluß) spärlich und widersprüchlich erscheinen. 

Sicherbare Unterschiede zwischen wF, und gF, (U-Test) prüfen abstammungsunabhän- 
gig die Beziehungen zwischen Fellfarbe und Verhalten, also eine mögliche Albinismus- 
kopplung: Es ergibt sich eine höhere Schnupperrate (Sch) der weißen Tiere, sowie eine 
unterschiedliche, allerdings so nicht sicherbare Anzahl der Sprünge ($p). Eine Sichtung der 
Individuen ergab, daß 37% der Tiere der gF, nicht sprangen, bei wF, sprangen 57 % der 
Tiere nicht. Der unterschiedliche Anteil dieser „‚Null-Ereignisse“ in den beiden Gruppen 
ist nach LACHENBRUCH (1976) mit p<0,01 signifikant (Z = 9,47 >9,21 = y?), die grauen 
Tiere der F, sind also „‚sprungfreudiger“. 

Damit sind lediglich zwei Parameter zwischen wF, und gP, sicherbar, Verhaltensweisen, 
die in stärkerer Differenz zwischen den Parentalgenerationen ganz entsprechend ausge- 
prägt sind. 

Eine quası kombinierte Aufschlüsselung der F, kann schließlich nach Abstammung und 
Fellfärbung vorgenommen werden: wF,(w), gF,(w), wF,(g) und gF,(g). Die Prüfung auf 
Signifikanz auftretender Unterschiede innerhalb der genannten Vierergruppe liefert zwar 
in vier Fällen (lediglich!) statistisch begründbare Aussagen (F-Test), und zwar bezüglich 
Schnuppern (Sch), mittlere Sitzdauer (t/Sit), Bewegungsdauer (Bew) und Aktivitätsanteil 
(Bew/Akt). Insgesamt folgen jedoch die Datenverteilungen der Verhaltensparameter nicht 
jener groben Einteilung der zweiten Filialgeneration nach Abstammung und Fellfarbe, 
vielmehr nehmen die auftretenden „widersprüchlichen Kopplungen“ gegenüber den 
Unterschieden zwischen den Parentalgenerationen den größten Werteanteil ein. 

Tabelle 1 faßt die sicherbaren Unterschiede in den Datenverteilungen der Filialgenera- 
tionen zusammen. 


Diskussion 


Ausgangspunkte und Begründung der vorgenommenen Verhaltensprüfungen an Filial- 
generationen reziproker Kreuzungsansätze weißer und grauer Hausmäuse ergaben sich 
zunächst aus dem Nachweis deutlicher Unterschiede in den Explorationsstrategien der 
Parentalstäimme und weiterhin aus der nachgewiesenen Bedeutungslosigkeit von Auf- 
zuchtbedingungen und sozialem Umfeld (SCHÄFER 1982). 

Neben postnatalen Umwelteinflüssen sind weiterhin Faktoren der Ontogenie als mög- 
liche Einwirkungen der „Umwelt“ des Uterus auf die Ausdifferenzierung körperlicher wie 
verhaltensphysiologischer Eigenschaften zu bedenken. Zweifellos wurden diese Einflüsse 
oft unterschätzt bzw. nur im Falle aufgetretener Mißbildungen, die die Ausprägung 
genetisch ‚intakter‘“ Programme verhinderten oder modifizierten, beachtet. Tatsächlich 
sind jedoch bis hinaus zum Menschen morphologische wie psychologische Folgen pränata- 
ler Entwicklungsbedingungen z.B. bei der Ausbildung der Geschlechtsidentität zu berück- 
sichtigen (MonEY und EHRHARDT 1975). Neben anderen, später darzustellenden Befun- 
den, sind bezüglich der Verhaltensmuster ‚„erstarrtes Sitzen‘ und „klebendes Laufen“ 
Unterschiede auffindbar, die neben pauschaler Korrelation zur Pigmentierung auch solche 
zur Herkunft, vereinfacht zur Fellfarbe der Eltern bzw. Großeltern in reziproker 
Geschlechtskombination erkennen lassen könnten. Vergleiche der Filialgenerationen F,(w) 
und F,(g) sınd zwar nicht geeignet, Aussagen über postnatale Einflüsse zu machen, da bis 
zur Entwöhnung beide Elternteile anwesend waren. Der Vergleich solcher Generationen 
aus reziproken Kreuzungsansätzen kann jedoch zur Prüfung pränataler Einflüsse im Sinne 
„geschlechtsgekoppelter‘ (besser heterosomal fixierter) Vererbung herangezogen werden 
(siehe auch BROADHURST 1961). Voraussetzungen sind der Nachweis der Bedeutungslosig- 
keit postnataler Einwirkungen (s. Transplantationen) und das Fehlen geschlechtsspezifi- 
scher Verhaltensunterschiede. Tatsächlich unterscheiden sich Weibchen und Männchen 


225 


Analyse des Explorationsverhaltens verschiedener Stämme der Hausmaus II 


Tabelle 1 


Beispiele zur Datenverteilung in den Filialgenerationen 


Aufgeführt sind jene Parameterwerte, die im paarweisen Vergleich signifikante Unterschiede auf- 
weisen. Nähere Erläuterungen s. Text bzw. Kap. Material und Methode, SCHÄFER 1982 


125 85=E74,7. 
9,3:222,9 


Bew/Akt 


Sch+A 


37, 08-2102 
92,32=21.052 


t/Sch 


A ae 
2,05:22055 


P/Akt 


Parameter 
Bew. 


20,6 
173 


F/min 


6,95-202,2 
551 159 


Bew/Akt 


Sch/Akt 
16,6 
14,7 
F/akt.min 


‚8 
8 


8 252 
6, 1,8 


b) 


Sch+ A/Akt 


46,5 + 7,6 
4322=20755 


F/bew.min 


| 
DI IE 338 


F/min 


14,8 52 
2255 1257. 


ZZ, 
18,1 4, 


7 
9 


253 4,3 
955 15,9 


+ 5,4 
1227:274,5 


Bew/Akt 


23,51=274,3 
17, 20=E7452 


(gleicher Gruppen, w oder g) im open-field-Verhalten nicht. Die Tatsache, daß bei den 
F,(w)-Hybriden eine weiße Mutter bis zur Entwöhnung anwesend war, ım Falle der F,(g) 
jedoch ein weißer Vater ist deshalb nicht von Belang. Liegen also verhaltensrelevante 
Genomanteile ganz oder zu einem Teil auf dem weiblichen Geschlechtschromosom, so 
sollten die Nachkommen mehr ihren Müttern ähneln. Die heterosomale Fixierung kann 
jedoch nur als wahrscheinlich gelten, wenn uterusspezifische Einflüsse auf den Keim 
ausgeschlossen wurden. Hierbei kommt erschwerend der notwendig höhere Heterozygo- 
tiegrad der Hybriden gegenüber ihren Eltern hinzu. Dieses Problem versuchten DEFRIES et 
al. (1967) durch die Transplantation von Ovarien ingezüchteter Spender in hybride 
Empfänger zu lösen. Als Ergebnis entstanden so ingezüchtete Nachkommen, die von 
Hybrid-Müttern ausgetragen wurden und deren open-field-Verhalten mit dem ingezüchte- 
ter Nachkommen, die von ihren genetischen Müttern geboren worden waren, verglichen 
werden konnte. Es ergaben sich in keinem Fall Hinweise darauf, daß prä- oder postnatale, 
durch die Mutter erzeugte Effekte das Verhalten beeinflußten. In diesem Sinne könnte man 
bezüglich in dieser Arbeit aufgefundener Unterschiede zwischen Nachkommen reziproker 
Kreuzungsansätze der Deutung als heterosomal fixierter Vererbung den Vorzug geben. 
Dabei ist selbstverständlich nicht an die „klassischen“ Beispiele geschlechtsgebundener 
Vererbung gedacht, die ja auf der Basis zweier Allele (z.B. „gesund“ und „Bluter‘“) die 
Gültigkeit Menneuscher Regeln lehrbuchhaft demonstrieren. Es sollte lediglich herausge- 
stellt werden, daß die reziproke Gruppierung weißer und grauer Elternteile zuordbare 
unterschiedliche Merkmalsverteilungen in der F, und/oder F, bewirken kann, wobei 


226 M. W. Schäfer 


natürlich der Heterozygotiegrad von Filialgeneration zu Filialgeneration ansteigt und 

mangelnde Homozygotie schon für die Parentalgenerationen der weißen und ganz beson- 

ders der grauen Tiere zu berücksichtigen ist. So steht zwar fest, daß Tiere der Gruppe w 

niemals „erstarrt sitzen‘ und Tiere der Gruppe g niemals „klebend laufen“, das heißt 

jedoch umgekehrt nicht, daß diese Eigenschaften bei sämtlichen Individuen der jeweils 
anderen Gruppe (also 100%) vorhanden wären. Auch ist keine Aussage darüber möglich, 
welche Allel-Anzahlen und -Kombinationen letztlich die phänotypische Ausprägung 
bedingen und wie sich diese bei F,- und F,-Heterozygotie durchzusetzen vermögen. So ist 
der Nachweis abstammungsspezifischer Einflüsse bezüglich der Parentalgenerationen 

(neben solchen der Pigmentierung) zwar anhand der Befunde der Abb. 1A und B in 

Erwägung zu ziehen, nicht jedoch die Isolation einfach ‚„mendelnder“ Erbgänge. Für 

Rattus norvegicus wurde übrigens eine an die Geschlechtschromosomen gekoppelte Ver- 

erbung bezüglich des open-field-Verhaltens von wilden und Labor-Ratten in reziproker 

Kreuzung weitgehend ausgeschlossen (PrıcE und Loomıs 1973). 

Offenbar erweist sich, daß der Homozygotie-Grad beider Parentalgenerationen für die 
gemessenen Verhaltensparameter nicht eindeutig festlegbar ist; das Ziel der vorliegenden 
Untersuchung wird primär durch die Frage des wahrscheinlichen Nachweises erbgebunde- 
ner Verhaltensweisen bestimmt und nicht durch die Aufdeckung von Erbgängen, die erst 
im zweiten Schritt, aufbauend auf wahrscheinlicher Genomfixierung zu erfolgen hat. Der 
Homozygotiegrad bezüglich der Pıgmentierung (grau versus Albino) liegt jedoch eindeutig, 
fest. Die Frage, ob Verhaltensweisen in Kopplung entsprechend vorliegen, also analog der 
dominant-rezessiven „‚grau-weiß-Vererbung“, muß jedoch weitgehend verneint werden, 
da mit zunehmender Hybridisierung widersprüchliche Kombinationen (z.B. weiße Tiere 
mit erstarrtem Sitzen und höherer Sprungrate) auftreten, also zunehmend auch Austausch- 
prozesse und Aufhebungen möglicher Kopplung erreicht werden. 

Auch für die in Abb. 1 dargestellten Abhängigkeiten zwischen reziproken Elternkombi- 
nationen und ungleichen Merkmalskombinationen der Hybriden können alternativ unglei- 
che (und unbekannte) Homozygotiegrade der Parentalindividuen verantwortlich gemacht 
werden: Da z.B. weiße Tiere der Parentalgeneration niemals ‚erstarrt sitzen“, graue 
jedoch überwiegend, können u.a. folgende Gen-Beziehungen hypothetisch formuliert 
werden: 

1. „Erstarrtes Sitzen‘ wird durch ein dominantes Gen FE bedingt, Tiere ohne diese 
Eigenschaften sind homozygot rezessiv (ee). Da die Tiere der F, nicht alle „erstarrt 
sitzen“, müssen heterozygote graue Tiere (Ze) in der Parentalgeneration vorhanden 
gewesen sein, und zwar neben homozygoten Individuen (ZE). 

2a. Aus P=eexXEE entsteht die F, einheitlich als Ze, 100% der Individuen ‚sitzen 
erstarrt‘, die F, (aus F, X F, = Ee X Ee) liefert zu 75 % „erstarrt sitzende“ Tiere (Ze 
und ZE) und zu 25% Tiere ohne diese Eigenschaft (ee). 

2b. Aus P=ee X Ee entsteht die F, uneinheitlich als Ze (50 %) und ee (50 %), die F, kann 
zu 100% nicht „erstarrt sitzen“ (ee aus F, X F,=eeX ee) oder zu 50% ‚erstarrt 
sitzen“ (Ee neben ee aus F,X F,=Ee X ee) oder schließlich zu 75 % „erstarrt sitzen“ 
(Ee und EE neben ee aus F,X FR,=Eex Ee). 

Da nun aus laufenden Zuchten jeweils Stichproben der einzelnen Gruppen und Generatio- 

nen entnommen wurden, muß bezüglich der jeweils vorhandenen Parentalgenerationen 

mit unterschiedlichen Genotypen in verschiedenen Kombinationen gerechnet werden. Es 
läßt sich dann leicht ableiten, daß die in Abb. 1 dargestellten Unterschiede lediglich Folgen 
dieser unterschiedlichen Kombinationen sind, wobei die Annahme eines dominanten Gens 
für „‚erstarrtes Sitzen‘ natürlich ohne weiteres als Simplifizierung abgetan werden könnte. 

Für die Werteverteilungen des „klebenden Laufens“ sind analoge Betrachtungen anzustel- 

len. Gewichtig erscheint dennoch, daß die Pigmentierung in der F, jeweils die Wichtung 

der Merkmalsausprägungen parentalkonfom verschiebt (weiße Tiere in jeder 7, enthalten 
weniger „erstarrt sitzende“ und mehr ‚klebend laufende‘ Individuen als ıhre grauen 


Analyse des Explorationsverhaltens verschiedener Stämme der Hausmaus II ZI] 


Geschwister), eine strenge Kopplung mit der Pigmentierung ist jedoch eindeutig abzu- 
lehnen. 

Zur Problematik möglicher Kopplungen zwischen Verhaltensmustern, maternalen 
Effekten und/oder Albinismus sei ergänzend auf weitere Befunde hingewiesen. In der F, 
ermittelbare Unterschiede könnten bei fünf Parametern auf maternale Einflüsse hindeuten, 
wobei diese Parameter sämtlich das Schnuppern betreffen. Fünf andere Mefßßwerte bezüg- 
lich der Bewegungsaktivität stehen dem jedoch entgegen. In der F, schließlich ergibt sich 
eine positive Korrelation zu deren Großmüttern ın nur einem Fall. In vier anderen 
Vergleichsmöglichkeiten ist das Resultat negativ. Auf die möglichen Beziehungen von 
mangelnder Homozygotie und entsprechenden (scheinbaren) Widersprüchlichkeiten der 
Merkmalsverteilungen wurde bereits oben hingewiesen. 

Die weißen Nachkommen in der F, springen weniger und schnuppern mehr als ihre 
grauen Geschwister. Auch zeigen die Parameter der Fortbewegungsart und des Sitzens 
(siehe oben) sowie die der Latenzzeiten (SCHÄFER 1982) Entsprechungen zwischen den 
Individuen der zweiten Filialgenerationen und den ihnen jeweils gleich gefärbten Parental- 
stämmen. Hingegen sind die Gruppen wF, und gF, bezüglich ihrer Bewegungsaktivität 
gleich. Allerdings gibt FuLLErR (1967) hier einen Unterschied an, berichtet jedoch von 
gleichen Aktivitäten beider Gruppen im Laufrad. Nun ist zu bedenken, daß die Pigment- 
losigkeit der Albinos auch Schutzlosigkeit gegenüber Strahlung bedeutet, hohe Lichtinten- 
sitäten entsprechend für Albinos aversive, mit Verharren beantwortete Stimuli darstellen 
und Unterschiede in der Bewegungsaktivität zwischen grauen und weißen Formen durch 
die Anwendung von Rotlicht egalisiert werden können (FuULLER und THOMPsoN 1978). 
Bezüglich erhöhter Streßanfälligkeit der Albinos schließt z.B. FuLLER (1967) durchaus auf 
eine Kopplung des Locus für Albinismus mit verhaltensbestimmenden Loci. Aus (bislang 
unveröffentlichten) vergleichenden Untersuchungen des Labyrinthlernens grauer und 
weißer Stämme ist auch dem Verfasser eine solche erhöhte Streßsensibilität der Albinos 
bekannt, die durchaus zu Todesfällen führen kann. Ob die erhöhten Kot- und Urinwerte 
der Albinos streß- oder lediglich stoffwechselbezogen sind, wäre differenziert zu prüfen. 

Der primäre Geneffekt der Pigmentlosigkeit liegt jedenfalls im Mangel an Phenoloxy- 
dase (Iyrosinase), der das Ausbleiben der Melanınbildung zur Folge hat. Auch wird die 
Aktivität der Dopa-Oxydase gesenkt und damit die Noradrenalin-Bildung beeinflußt. 
Beziehungen zu Streß- und anderen Verhaltensreaktionen sind zu vermuten. Allerdings 
sind die zentralnervösen Auswirkungen der genannten Veränderungen bislang nicht 
aufgedeckt worden (Karıson 1967; FULLER 1967; FULLER und THOMPSoN 1978). 

Domestizierte Formen zeigen allgemein in höherer Frequenz Abwandlungen der 
Pigmentierung und Albinismus, da die fitnessmindernde Auffälligkeit bei Übernahme in 
den Hausstand bedeutungslos wird. Geringere Erregbarkeit (auch in den wF,-Gruppen im 
open-field erkennbar) und reduziertes Fluchtverhalten sind geradezu züchterisch 
erwünscht, die künstliche Selektion wird also Genfrequenzen dieser Wirkungen erhöhen. 
So deutet etwa HEMMER (1978) auch die gering ausgeprägte Tagesperiodik weißer Mäuse 
gegenüber pigmentierten und Wildformen als Haustiermerkmal ım Sinne erwünschter 
permanenter Aktivierbarkeit. DEFRIES et al. (1978) berichten schließlich über Selektions- 
wirkungen auf hohe bzw. niedrige Aktivität über 30 Generationen: die auf niedrige 
Aktivität gezüchteten Linien bestanden ab der 8. bzw. 23. Generation nur noch aus 
Albinos. Ob hierin direkte Genwirkungen des Albinismus-Locus, Gen-Kopplungen im 
weiteren Sinne oder Parallelerscheinungen der inzuchtbedingten Homozygotie-Steigerung 
zu sehen sind, ist einstweilen nicht entscheidbar. 

Da Gefangenschaft, Haltungs- und Zuchtbedingungen im Labor durchaus Verhalten zu 
ändern vermögen, muß natürlich allgemein nach der Zulässigkeit des Vergleichs weißer 
und grauer Formen im Sinne des Gegensatzes „domestiziert-wild“ gefragt werden. Nach 
SMITH und Connor (1978) ergaben sich beim Verhaltensvergleich ‚‚wilder‘ Mäuse der 3. 
und der 13. gezüchteten Generation keine Änderungen, die nicht auch in mindestens einem 


228 M. W. Schäfer 


Stamm parallel gehaltener Inzuchtstämme auftraten. Laborhaltung vermag also heterozy- 
gote Wildpopulationen über zahlreiche Generationen genetisch stabil und gegenüber 
freilebender Vorfahren identisch zu halten. Untersuchungen an frischen „Wildfängen“ 
sind wegen der Stressoren der Gefangenschaft und des Eingewöhnungsprozesses auf 
keinen Fall sinnvoll. 

Die Resultate der vorliegenden Arbeit weisen die grauen Mäuse als Individuen aus, die 
einen größeren Teil der Versuchsdauer bewegungslos sitzend verbringen, in der übrigen 
Zeit jedoch höhere Mobilität, Springen und intensivere Beschäftigung mit dem Becher 
zeigen als weiße Mäuse. Diese schnuppern mehr und sind fast ständig aktiv, erreichen 
jedoch durch ihre kontinuierlich-verzögernde Fortbewegungsart ein geringeres Mittel 
zurückgelegter Strecken sowie erhöhte Putzfrequenzen. Die hier untersuchten grauen 
Tiere ordnen sich dabei durchaus dem Typus einer Wildart zu, auch wenn einschränkend 
die kommensale Hausmaus eher als Zwischenstadium zwischen Wild- und Haustier mit zu 
berücksichtigendem Verhaltenspolytypismus anzusehen ist (Frick und NorD 1963; PrLo- 
MIN und MAnosEVvITZz 1974). Wildtiere sind allgemein durch hohe Fluchttendenz (Hepı1- 
GER 1938) zu charakterisieren, für die unter starkem Feinddruck stehenden Muriden gilt 
dies besonders. So ergibt sich für die freilebende Maus die grundsätzliche Ambivalenz 
zwischen kryptischem Verhalten und der Notwendigkeit, neue Territorien und Nahrungs- 
quellen durch Exploration unter Aufgabe der Deckung zu erschließen. Das aktive Erkun- 
den wird dabei immer wieder von Deckungssuche und Verharren (bewegungsloses Sitzen 
in gespannter Aufmerksamkeit) unterbrochen (BRock 1934; SCHLEIDT 1951). Die Ver- 
trautheit mit der zunächst unbekannten Umgebung wird dabei durch Markierungsverhal- 
ten gesteigert (EIBL-EIBESFELDT 1958). Entsprechend betont die „Zwei-Faktoren-Theorie“ 
nach MONTGOMERY (zit. ın RusseLL 1973) die Ambivalenz zwischen Erkundungs-,, Trieb“ 
und „Furcht“, als deren Resultierende das Ausmaß der Explorationsaktivität erscheint. In 
dieser Situation unterbleiben andere Verhaltensweisen, wie z. B. Komfortverhalten. Indivi- 
duen ohne genetische Fixierung der geschilderten angepaßten Strategien erfahren „automa- 
tisch“ Fitnessminderungen, wenn etwa Feindvermeidung erst gelernt werden müßte. 

Demgegenüber zeigt die weiße Maus deutliche Verluste kryptischer Reaktionen (,,Sit- 
zen“, „erstarrtes Sitzen‘), Zunahme des Komfortverhaltens (‚Putzen‘) bei schneller 
Habituation („Gewöhnung“) und hohe Schnupperraten zur Kompensation der größeren 
Unbeweglichkeit als Folge der Größen- und Gewichtszunahme. Die Ursache der letzteren 
Eigenschaft ist dabei durch Verhaltensprotokolle am normalen Haltungskäfig mit unbe- 
grenztem Futtervorrat leicht zu ermitteln. Die weißen Tiere der Parentalgeneration fressen 
ganz einfach mehr und länger als graue Formen. Bemerkenswert ist, daß andererseits weiße 
Tiere der F, sich gewichtsmäßig nicht mehr von ihren grauen Geschwistern unterscheiden 
und alle Filialgenerationen etwa intermediär (nach ihren Gewichtsdaten, s. SCHÄFER 1982) 
zwischen den Eltern liegen. Als Kuriosum sei angemerkt, daß F,-Hybriden nicht nur etwa 
halb so viel wie graue Tiere springen, sondern — bedingt durch ihr höheres Gewicht - auch 
nur halb so hoch! 

Prüft man die Verteilung von Merkmalen, die signifikante Unterschiede zwischen Filial- 
und Parentalgenerationen erkennen lassen, so entfallen 56,8% auf Unterschiede zu den 
weißen Eltern. Pauschal erscheinen demnach Hybriden den grauen Stammformen ähnlı- 
cher. Die Zuordnung wird allerdings mit zunehmender Hybridisierung (73!) auch zuneh- 
mend uneinheitlicher. Verhaltensmuster beider Elterngruppen treten ın Mischung und 
„widersprüchlichen‘ Kombinationen (,Springen‘ + „klebendes Laufen“) auf, wobei die 
Streuung der Gesamt-F, im Bereich der w-, g- und F,-Gruppen bleibt. Auf angedeutete 
Zweigipfligkeiten kann bislang nicht eingegangen werden. Bei der einleitend angesproche- 
nen Alternative, ob „Wildheit‘ auf wenige dominante Gene rückführbar ist oder vielmehr 
ein multifaktorielles Charakteristikum darstellt, muß zweifellos letzterer Möglichkeit der 
Vorzug gegeben werden. Der Nachweis der genetischen Fixierung von Anteilen des 
Erkundungsverhaltens gelingt gerade mittels des Nachweises gesteigerter Heterozygotie in 


Analyse des Explorationsverhaltens verschiedener Stämme der Hausmaus II 22 


den Filialgenerationen. Die relativ beliebige Kombinierbarkeit der Verhaltensmerkmale 
weist dabei den experimentellen Ansatz nicht als Nachweismöglichkeit starrer Kopplungen 
an den Albinismus aus, sondern vielmehr als Vergleichsmöglichkeit eines stärker domesti- 
zierten Stammes (der „‚zufällig‘‘ weiß ist) mit einem Stamm, der Wildformeigenschaften 
unter kurzzeitiger Laborhaltung bewahrte. 

Im Vergleich erweisen sich zahlreiche Komponenten des Explorationsverhaltens ent- 
sprechend ihrem Stellenwert unter natürlichen Bedingungen als genetisch fixiert. Eine 
höhere Differenziertheit der Verhaltensanalyse bewirkt jedoch, daß ‚„Wildtierverhalten‘“ 
als komplexes Mosaik erscheinen muß, für dessen einzelne Meßwerte und Parameter sich 
durchaus „Erbgänge“ vorschlagen lassen, die jedoch höchst unterschiedlich sind. Dies gilt 
sogar für den Vergleich von Meßwerten eines Verhaltensmusters, wenn dieses nach „Dauer 
als Anteil der Gesamtzeit“, „Dauer pro Aktivitätszeit‘“ oder „mittlerer Dauer“ definiert 
und quantifiziert wird. Folglich wird eine Verhaltensweise nicht durch ein Gen bestimmt, 
vielmehr entscheiden komplexe Steuerungs- und Regelmechanismen in Interaktionen von 
Genen und Umwelt (passende Reizkonfiguration) darüber, ob, wie häufig, wie lange und 
mit welchen mittleren Anteilen an einer übergeordneten Sequenz ein Verhaltensmuster zu 
Tage tritt. 

Unter Verwendung der im Ergebnisteil aufgelisteten unterschiedlichen Merkmalsvertei- 
lungen in den Parental- und Filialgenerationen (Vergleich von w, g, F, und F,) gibt Tab. 2 


Tabelle 2 


Zusammenfassender Versuch, aus signifikanten quantitativen Beziehungen und Unterschieden 
zwischen den Gruppen und ihren Verhaltensparametern „Erbgänge‘“ von Komponenten des 
Explorationsverhaltens zu postulieren 
Heterosis kennzeichnet dabei eine relativ zu beiden Elternteilen stärkere Merkmalsausprägung der 


Hybridengenerationen; s. BRESCH, HAUSMANN 1970 u. a. Zur näheren Erläuterung der Parameter 
s. Kap. Material und Methode, SCHÄFER 1982 


Parameter Beziehungen zwischen den Gruppen Erbgang 
w:F, g:Fı g:R 


S 
ke 


A 

A/Akt 

WA 
Sch+A/Akt 
Kot 


Heterosis 


IAAN 
IAAN 
IA AA 
EVA 


IN NINA IN 
I 


P 

Akt 

Sit 
F/akt.min 


intermediär 


ANVNV 
IN IN MM 
VAL 
VaAVEINEN 


Sch 
v/Sıt 
Bew/Akt 


partiell dominant 


t/Sch 

v/P 
P/Akt 
Sch+A 
t/Ges 
a.B./Akt 


Sp 
tv/Akt 


F/bew.min 


I VVV 
I VVV 


dominant- 
rezessiv 


NN NV MVM 
INN | 
IBIEVZAEN EI 


N 


A\AV 
I AV 
vVarı 
VEN 
| 
m 222 EEE Tr EEE S—— og 


AV 
IN 


230 M. W. Schäfer 


abschließend Denkmöglichkeiten unterschiedlicher ‚Erbgänge“ einzelner Verhaltenspara- 
meter wieder. 
Danksagung 


Mein besonderer Dank gilt Frau Craupıa RÖTER für die im Rahmen ihrer Diplom-Schrift im 
Fachbereich Biologie, Frankfurt, geleistete Mitarbeit sowie Frau MarGoT NAUMANN für die tier- 
pflegerische Betreuung. 


Zusammenfassung 


Die Explorationsstrategien von weißen und grauen Hausmaus-Stämmen sowie zwei Hybrid-Filial- 
generationen wurden im „open-field-Verfahren‘“ nach zahlreichen Verhaltensparametern quantifiziert 
und verglichen. 

Graue und weiße Stammformen erwiesen sich als trennbar, wobei die Erkundungsstrategien der 
grauen Tiere durch höhere Anteile erstarrten Sitzens im Wechsel mit hoher Mobilität und Sprungver- 
suchen als unverändert angepaßt an selektive Anforderungen des natürlichen Biotops charakterisiert 
werden konnten. 

Hingegen waren die weißen Tiere durch Verlust kryptischer Verhaltensanteile sowie verzögernde 
Bewegungsabläufe unter hoher Schnupperaktivität gekennzeichnet. 

Die Filialgenerationen erwiesen sich zunehmend durch erhöhte Heterozygotie (gegensätzliche 
Merkmalskombinationen aus dem Verhaltensrepertoire beider Elternteile) charakterisiert. Kopplun- 
gen zur reziproken Geschlechtskombination der Elterntiere oder zur Pigmentierung wurden disku- 
tiert und weitgehend ausgeschlossen. Vielmehr ergab die Auflösung der Verhaltensabläufe unter- 
schiedliche Möglichkeiten der Vererbbarkeit, wobei Unterschiede auch innerhalb einer Verhaltens- 
weise deren unterschiedliche Meßparameter betreffen konnten. 

So erscheinen Anteile des Explorationsverhaltens von Mus musculus genetisch fixiert, die eindeutig 
multifaktorielle Bestimmtheit erlaubt jedoch keine Zuordnung von ‚„Wildheit“ versus ‚„Domestika- 
tion‘ zu wenigen dominant-rezessiven Mechanismen, wie dies in der Vergangenheit mehrfach für 
Stämme der Hausmaus postuliert wurde. 


Literatur 


BROADHURST, P. L. (1961): Analysis of Maternal Effects in the Inheritance of Behaviour. Anım. 
Behav. 9, 129-141. 

Brock, F. (1934): Jahrmarktsdressuren wilder Mäuse als Grundlage einer wissenschaftlichen Verhal- 
tensanalyse. Verh. d. Dtsch. Zool. Ges., Suppl. Bd. 7, 235-246. 

DErFRIES, J. C.; THomas, E. A.; HEGMAnNn, ]J. P.; WEıRr, M. W. (1967): Open-Field Behaviour in 
Mice: Analysis of Maternal Effects by Means of Ovarian Transplantation. Psychol. Sci. 8, 207-208. 

DErrıss, J. C.; GERvaIs, M. C.; THomas, E. A. (1978): Response to 30 Generations of Selection for 
Open-Field Activity in Laboratory Mice. Behav. Genetics 8, 3-14. 

EısL-Ei1BEsrELpT, I. (1958): Das Verhalten der Nagetiere. In: KüKENTHAL, Hb. Zool. 8 (10), 13, 1-88. 

Frick, H.; Norp, H. ]J. (1963): Domestikation und Hirngewicht. Anat. Anz. 113, 307-316. 

FULLER, J. L. (1967): Effects of the Albino Gene upon Behaviour of Mice. Anım. Behav. 15, 467-470. 

FULLER, J. L.; Thompson, W. R. (1978): Foundations of Behavior Genetics. Saint Louis: Mosby 
Comp. 

HEDIGER, H. (1938): Tierpsychologie und Haustierforschung. Z. Tierpsychol. 2, 29-46. 

HEMMER, H. (1978): Zusammenhänge zwischen Fellfarbe und Ausprägung des Aktivitätsrhythmus 
bei Labormäusen und wilden Hausmäusen. Säugetierkdl. Mitt. 26, 256-259. 

Karısson, P. (1967): Kurzes Lehrbuch der Biochemie. Stuttgart: Thieme. 

LACHENBRUCH, P. A. (1976): Analysıs of Data with Clumping at Zero. Biometr. Zeitschr. 18, 
351-356. 

Money, J.; EHRHARDT, A. (1976): Männlich-weiblich. Die Entstehung der Geschlechtsunterschiede. 
rororo-Sexologie. Reinbek bei Hamburg: Rowohlt Taschenbuch Verlag. 

Promin, R.; MAnosEVvITZz, M. (1974): Behavioral Polytypism in Wild Mus musculus. Behav. Genetics 
4, 145-157. 

Price, E.; Loomis, $. (1973): Maternal Influence on the Responses of Wild and Domestic Norway 
Rats to a Novel Environment. Develop. Psychobiol. 6, 203-208. 

Russeıı, P. A. (1973): Relationships between Exploratory Behaviour and Fear: A. Review. Br. ]. 
Psychol. 64, 417-433. 

Sachs, L. (1978): Angewandte Statistik, Statistische Methoden und ihre Anwendung. Berlin, Heidel- 
berg, New York: Springer. 

SCHÄFER, M. W. (1982): Analyse des Explorationsverhaltens verschiedener Stämme und Generationen 
der Hausmaus (Mus musculus domesticus Rutty). I. Methodik, Datenzuordnung und Einflüsse des 
sozialen Umfeldes. Z. Säugetierkunde 47, 149-162. 

SCHLEIDT, W. M. (1951): Nest und „Zuflucht“ bei Mäusen. Z. Tierpsychol. 8, 137-140. 


Analyse des Explorationsverhaltens verschiedener Stämme der Hausmaus II 231 


SMITH, R. H.; Connor, J. L. (1978): Behavioral Effects of Laboratory Rearing in Wild Mus musculns. 
Behav. Biol. 24, 387-392. 


Anschrift des Verfassers: Prof. Dr. M. WALTER SCHÄFER, Fachbereich Biologie (Zoologie) der 
Universität, Siesmayerstr. 70, D-6000 Frankfurt a.M. 


Chromosomal forms of the mole rat Microspalax 
from Greece and Turkey 


By Eva Giacıa, I. Savıc and B. SOLDATOVIE 


Zoological Laboratory, University of Patras, and Institute for Biological Research “Sinisa Stankovic”, 
University of Belgrade 


Receipt of Ms. 29. 1. 1982 
Abstract 


The karyotype of several populations from different regions of Greece and Turkey was analysed. On 
the basıs of the analysis, the existence of five different karyotypes is recorded: Microspalax leucodon 
makedonicus with 2n =52 and NF=86, M. I. epiroticus 2n =56 and NF=84, M |. hellenicus 
2n =58 and NF = 88, M. |. turcicus 2n = 56 and NF = 78 and the newly described karyotype from 
the western part of Asia Minor and from the island of Lesbos (probably M. nehringi anatolicus) with 
2n = 38 and NF = 74. 

It is interesting to note that the karyotype of the population Polychnitos from the Aegean island of 
Lesbos is identical to the karyotype of the populations Havran and Selguk from the western part of 
Asıa Minor. 

Although there are great differences in respect of the number of chromosomes between the Asian 
and Aegean populations M. nehringi anatolicus on the one hand, and the population M. leucodon 
turcicus on the other, it can be assumed on the basis of the NF value and the number of subacrocentric 
chromosomes that these two karyotypes are phylogenetically closely linked. 


Introduction 


Karyotype studies of the populations of the complex species Microspalax leucodon 
(Nordmann, 1840) from south-eastern Europe reveal obvious chromosome polymorph- 
ism. Several different karyotypes from this region have been described until now. Their 
diploıd chromosome number vary from 46 to 58, while their NF ranges from 76 to 98 
(SavıC and SoLDATovIE 1979a; PEsHEV 1980). 

The first karyological analyses of mole rat populations in Greece were made by Savı© 
and SoL.DATovIıE (1978, 1979a) and SavıC (1981). They recorded two new karyotypes: M. |. 
epiroticus (the population of Lefkothea) and M. !. hellenicus (the population of Hag. 
Spyridon). 

Karyotype studies of the mole rat in Turkey were initiated by SOLDATOVIC and SavIc 
(1978) and Savı€ and SoLDATovıE (1979a, b). They described the karyotype M. I. turcicus 
in two populations in Balkan Turkey (the Corlu and Karaevlı populations). Another 
karyotype was discovered in Turkey in two populations in western Asıa Minor (the 
Havran ans Selcuk populations), most probably belonging to the species M. nehringi (M.n. 
anatolicus Mehely, 1909). Since the karyotype of these latter two populations was only 
analysed in the female however, the sex chromosomes could not be determined. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4704-0231 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 231-236 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


292 Eva Giagia, I. Savic and B. Soldatovic 


The analysis of the karyotype of mole rat populations on the island of Lesbos (the 
Polychnitos population), whose results are shown here, represents a contribution to 
knowledge of the karyotype of the mole rat populations on this Aegean island for the 
purpose of explaining problems of taxonomy and speciation of representatives of the genus 
Microspalax on the Balkan peninsula and in neighbouring regions. The study also includes 
a comparative survey of all the karyotypes found in Greece and Turkey. 


Results and discussion 


Savı@ and SoLDATovIE (1978, 1979a) and SavıC (1981) described two karyotypes from 
Greek territory: M. l. epiroticus (the Lefkothea population) with 2n = 56 and NF = 84 and 
M. I. hellenicus (the Hag. Spyridon population) with 2n = 58 and NF = 88 (Fig. 1). Both 
karyotypes have a great number of acrocentric autosomes — 14 pairs, six pairs each of 
subacrocentric and one pair each of metacentric autosomes. The only difference between 
these two karyotypes lıes in the number of submetacentric autosomes — M. I. epiroticus has 
sıx and M. |. hellenicus seven. Since the sex chromosomes of both these forms are identical, 
it may be concluded that their karyotype is extremely similar. 

SavıC and SoLDATOVIE (1979b) analysed two populations in the Balkan part of Turkey 
(Corlu and Karaevli) which belong to the subspecies M. I. turcicus and have an identical 
karyotype - 2n = 56 and NF = 78. In the western part of Asıa Minor, there was an analysis 
made of the karyotype of two populations (Havran and Selguk) which probably belong to 
a quite different species - M. nehringi (according to MURSALOGLU 1978 — M. lencodon, to 
MEHELYy 1909 and OnDkrIAs 1966 - M. I. anatolicus and to Lyapunova et al. 1974 - M. 
nehringi) and they were found to have an identical karyotype = 2n-38 and NF = 74. 
Compared with all the karyotypes of the mole rat described, the form M. I. turcicus has by 


Fig. 1. Survey of the localities 
of the populations analysed 
karyologically from Greece 
and Turkey. 1= Corlu, 
2 = Karaevli, 3 = Arnissa, 
4=Havran, 5 = Lefkothea 
(and Eleousai), 6 = Polychni- 
tos (Island of Lesbos), 7 = 
Hag. Spyridon, 8 = Selguk 


o 50 1ookm 


[+] % @ © & 


2n=38 2n=52 2n=56 2n=56 2n=58 
NF=74 NF=86 NF=78 NF=84 NF=88 


Chromosomal forms of the mole rat Microspalax from Greece and Turkey 233 


far the greatest number of acrocentric autosomes — 17 pairs. It has two pairs of metacentric 
chromosomes, five pairs of submetacentric and only three pairs of subacrocentric auto- 
somes, which is at the same time one of the lowest values for subacrocentric autosomes 
among all the other karyotypes of the mole rat in Europe. 

Of the studies of the karyotypes of mole rat populations in the Balkan peninsula and 
Asia Minor, studies of populations from the Aegean islands are of particular significance as 
they can offer appropriate information on mutual phylogenetic relations. 

On this occasion, the results of the analysis of the karyotype of one population of the 
Aegean islands are given for the first time. The analysis was made on the Polychnitos 
population from the island of Lesbos (Fig. 1), using two males and one female. Chromo- 
some preparations were obtained by the direct method from the bone marrow (Hsu and 
Kerroc 1960). 

This karyotype is characterized by a diploid number of chromosomes 2n = 38 and 
NF = 74 (Fig. 2). According to the morphology and position of the centromere, the 
chromosomes can be divided into four groups. The first contains six pairs of metacentric 
autosomes of greater and medium length, as well as two pairs of the smallest autosomes in 
the karyotype. The group of submetacentric chromosomes has eight pairs of mainly larger 
autosomes, only the last pair of which is considerably smaller. In the third - the group of 
subacrocentric chromosomes, there are three pairs of autosomes, of which the first two 
pairs are the longest chromosomes in the karyotype. It es interesting to point out that in 
the fourth group, the group of acrocentric chromosomes, there is only one pair of 
autosomes — which are very small. The X chromosome is larger and subacrocentric, while 
the Y chromosome is acrocentric, very small and corresponding in size to a pair of 
acrocentric autosomes. 


Unmun. 
“MR Wi 


a ds MR 
xY 


oM 


Fig. 2. Karyogram of the population Polychnitos from the island of Lesbos-male (probably Micro- 
spalax nehringi anatolicus) 


According to the morphology of the chromosomes of the population Polychnitos from 
the island of Lesbos, it can easily be perceived that this karyotype could be identical to the 
karyotype of the populations Havran ans Selguk from the western part of Asıa Minor 
(Savı@ and SoLDATovIE 1979b). It could therefore be assumed that the sex chromosomes 


234 Eva Giagia, I. Savic and B. Soldatovic 


fr DR.» 


6 


a AN Mm 


8 
ee 


Fig. 3. Karyogram of the population Havran from the western part of Asia Minor-female (probably 
Microspalax nehringi anatolicus) 


10 M 


of the two poulations from Asıa Minor are also the same as in the population Polychnitos 
(Fig. 3). 

The karyotype of the population Arnissa (two males and two females) was analysed in 
northern Greece. The diploid number of chromosomes in this population is 52 and 
NF = 86 (Fig. 4). According to the position of the centromere, we distinguish four groups 
of chromosomes. There are two pairs of autosomes in the group of metacentric chromo- 
somes. There are seven pairs each of submetacentric and subacrocentric autosomes. The 
subacrocentric autosomes are the longest chromosomes in the karyotype. There are nine 


m 3 R% 10M 
ee, 

AN hi \ Ax IR i€ An au 
) 7 


Mu LK. Kı 


| ————————— — [11 1777727 XY 


aln na ana no nA AM on aA an 


Fig. 4. Karyogram of the population Arnissa from northern Greece-male (Microspalax leucodon 
makedonicus) 


Chromosomal forms of the mole rat Microspalax from Greece and Turkey 235 


pairs of autosomes in the group of acrocentric chromosomes. These chromosomes are the 
smallest in the karyotype. The X chromosome is of medium size and submetacentric while 
the Y chromosome is small and acrocentric. The mole rat populations of western 
Macedonia in Yugoslavia have a similar karyotype. 

The table gives a comparative survey of all populations analysed to date of the genus 
Microspalax from Greece and Turkey. On the basis of comparative analysıs of karyotypes, 
it can be concluded that there is great chromosome polymorphism in this region which is 
manifested in the differences in respect of the number and morphology of the chromo- 
somes. The diploid number of chromosomes of the majority of the populations is over 50 


Comparative survey of the morphology of the karyotypes from Greece and Turkey 
Karyotypic form Autosomes Sex chromosomes 


M. nehringi 

M. Il. makedonicus 
M. 1. turcicus 

M. 1. epiroticus 

M. |. hellenicus 


metacentric, sm = submetacentric, sa = subacrocentric, a = acrocentric 


(52-58), which is within the borders of the corresponding values with other populations of 
the Balkan peninsula and other regions of their range (SavıC and SoLDATovıC 1979a; 
PEsHEv 1980). However, the extremely small number of chromosomes in the population of 
the island of Lesbos and two populations (Havran and Selguk) from Asıa Minor is 
particularly striking. This number also deviates greatly in comparison with the lowest 
number of chromosomes registered until now in representatives of the genus Microspalax 
(population Kozarevets 2n = 46, PEsHEv 1980; M. I. hungaricus 2n = 48, Savıd and 
SoLpAarovı& 1977). Apart from this, it is also significant to stress the fact that the 
karyotype of the populations from the island of Lesbos as well as from two localities in 
Asia Minor contains only one pair of acrocentric autosomes. In all the karyotypes of mole 
rats on the Balkan peninsula described, the number of acrocentric chromosomes was 
considerably greater and varied from 6 in the subspecies M. I. serbicus (SavıC and 
SOLDATOVIE 1974) to 17 in M. |. turcicus (SOLDATOVIE and SavıC 1978). Such a karyotype, 
which differs extremely from all karyotypes of Microspalax representatives known to-day, 
has introduced a degree of confusion into our knowledge of the cytogenetics of the mole 
rat. Nevertheless, if the NF is analysed, its relative closeness to the karyotype of the 
subspecies M. I. turcicus is to be seen. This fact leads us to believe that there is a close 
mutual phylogenetic link between these two karyotypes, whereby one could have sprung 
from the other mainly by means of Robertson’s fusion or fission, by the transformation of 
chromosomes in groups of meta- and submetacentric or acrocentric autosomes. Analysis of 
the phylogenetic links between the other karyotypes is mainly in line with the earlier 
conclusions of Savı@ and SoLDATovIC (1979a, b). 


Zusammenfassung 


Chromosomenformen von Blindmäusen (Microspalax) aus Griechenland und der Türkei 


Neu beschrieben werden die Karyogramme je einer Population von Microspalax aus Nordgriechen- 
land und von der Insel Lesbos. Aus Griechenland und der Türkei sind damit insgesamt fünf 
Karyotyp-Formen bekannt: 2n=52, NF=86 (Microspalax leucodon makedonicus); 2n = 56, 
NF = 84 (M. |. epiroticus); 2n = 58, NF = 88 (M. I. hellenicus); 2n = 56, NF=78 (M. |. turcicus); 
2n = 38, NF = 78 (wahrscheinlich M. nehringi anatolicus). Der Karyotyp von Lesbos stimmt mit dem 
aus West-Kleinasien (Havran und Selguk) überein. 


236 Eva Giagia, I. Savic and B. Soldatovic 


Obwohl sich die Chromosomenzahlen zwischen europäischen und kleinasiatischen Microspalax 
stark unterscheiden, dürften die Karyotyp-Formen namentlich von anatolicus und turcicus verwandt 
sein. NF-Werte und Zahlen subakrozentrischer Chromosomen lassen nämlich vermuten, daß die 
Unterschiede durch Robertsonsche Fusionen bzw. Fissionen zustande gekommen sind. 


Literature 


Hsu, T; KerLoc, D. (1960): Primary cultivation and continuous propagation in vitro of tissue from 
small biopsy specimens. J. Nat. Canc. Inst. 25, 221-235. 

LyarunovA, E. A.; VORONTSOV, V.; MARTYNOVA, L. (1974): Cytogenetical differentiation of 
burrowing mammals in the Palaearctic. Symp. theriol. II, Proc., Academia, Praha 203-215. 

MEHELY, L. (1909): Species generis Spalax. A földı kutyäk fajai szärmazases rendzertani tekintetben. 
A Magyar Tudomänyos Akademia Kıiadasa, Budapest. 

MursALoctıu, B. (1978): The taxonomic status and distribution of Spalax (Rodentia) in Turkey. II. 
Congressus theriologicus internationalis, Abstracts of Papers, 24, Brno. 

OnDrıas, J. C. (1966): The taxonomy and geographical distribution of the rodents of Greece. 
Säugetierkundl. Mitt. 14 (Sonderheft), 1-136. 

PEsHeEv, D. (1981): On the caryotypes in some populations of the mole rat (Spalax leucodon 
Nordmann) in Bulgaria. Zool. Anz. 1/2, 129-133. 
Savı6, I. R. (1982): Familie Spalacidae Gray, 1821 — Blindmäuse. In: NIETHAMMER, J.; Krapp, F. 
(eds.): Handbuch der Säugetiere Europas. Band 2/I. Wiesbaden: Akad. Verlagsges. (in press). 
Savı6, I.; SOLDATOVIG, B. (1974): Die Verbreitung der Karyotypen der Blindmaus Spalax (Mesospa- 
lax) ın Jugoslawien. Arh. biol. nauka 26, 115-122. 

— (1977): Prilog poznavanju ekogeografskog rasprostranjenja i evolucijje hromozomskih formi 
Spalacidaea Balkanskog poluostrva. Arh. biol. nauka 29, 145-156. 

— (1978): Studies on the karyotype and distribution range of the mole rat (Spalax leucodon 
Nordmann) in Greece. Caryologia 31, 63-73. 

— (1979a): Distribution range and evolution of chromosomal forms in the Spalacidae of the Balkan 
Peninsula and bordering regions. J. Biogeography 6, 363-374. 

— (1979b): Contribution to the knowledge of the genus Spalax (Microspalax) karyotype from Asıa 
Minor. Arh. biol. nauka 31, 1P-2P. 

SOLDATOVIG, B.; Savı, I. (1978): Karyotypes in some populations of the genus Spalax (Mesospalax) 
in Bulgarıa and Turkey. Säugetierkundl. Mitt. 26, 252-256. 


Authors’ addresses: Dr. Eva GıaGIa, Zoological Laboratory, University of Patras, Patras, Greece; 
Dr. Ivo Savı€ and Dr. BoGosAav SOLDATOVIG, Institute for Biological Research 
“Sinisa Stankovic”, University of Belgrade, 142, 29th November St., 11060 
Belgrade, Yugoslavia 


A note on the vocal behaviour of the Giant panda, 
Ailuropoda melanoleuca (David, 1869) 


By G. PETERS 


Department of Zoological Research, National Zoological Park, Smithsonian Institution, 
Washington, D.C. 


Receipt of Ms. 22. 12. 1981 
Abstract 


The vocal repertoire of adult giant pandas as observed in the pair living at the National Zoo, 
Washington, D.C., was studied with the help of sound spectrographic analysis. Vocal activity inthe 
and the 2 is nearly exclusively restricted to pro-oestrus and oestrus period of the ?. The vocal 
repertoire of the species consists of several vocal forms, the 2 exhibiting more vocal types than the d. 
Most of the vocalizations form a graded system with at least one other form in the repertoire, one 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4704-0236 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 236-246 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


A note on the vocal behaviour of the giant panda, Ailuropoda melanoleuca 237 


vocalization seems to be discrete. With the exception of the vocal forms typical of agonistic behaviour, 
the functional significance of the other forms is hard to establish. The vocalization accounting for most 
of the vocal activity in the d and the 2 during the oestrus period of the 2 is unlike any vocal form 
known in the Ursidae. It is structurally quite sımilar to a vocalization in the red panda (Ailurus 
fulgens) which is also its functional equivalent. Corresponding vocal forms are known in the 
Procyonidae. 


Introduction 


Since the discovery and first scientific description of the giant panda disagreement on the 
systematic position of this species has continued (for reviews see CHORN and HOFFMANN 
1978; TuEnıus 1979). For various reasons thıs little known species has remained appealıng 
to zoologists and laymen alıke. 

With the recent policy of the government of the People’s Republic of China to donate a 
pair of giant pandas to several major zoos in Europe, America and Japan, it has become 
possible to fill part of the gap in our knowledge of this species’ behavioural biology. 
Detailed field studies of giant panda behaviour are lacking (a large-scale field project has 
begun [SCHALLER 1981]) and the data from captive anımals have been rather fragmentary to 
date. Moreover, a detailed study of the vocal behaviour of this species in ıts natural habitat 
might be difficult, thus a study of this aspect of behaviour in captive anımals is advan- 
tageous. 

A few anecdotal reports on field observations of giant panda vocalızations were 
published by SHELDoNn (1937, 1975), SCHÄFER (1938) and SCHALLER (1981) and some notes 
on captive anımals by SCHNEIDER (1939) and Haas (1963). However, no structural details 
were presented. Good general accounts of the vocal behaviour of giant pandas in captivity 
were published by MoRrRIS and MoRrrıs (1966) and especially by KıEıman et al. (1979), the 
latter paper based on observations of the same two individuals that are dealt with here. 

The object of this paper is to contribute to the knowledge of the giant panda’s behaviour 
and to provide some additional characteristics to be considered in discussions of the 
systematic position of this species. 


Material and methods 


The vocal behaviour of the adult pair of giant pandas living at the National Zoological Park, 
Smithsonian Institution, Washington, D.C. was studied during spring and summer 1980. At the time 
of observation both animals were about 9 years old. Details on housing, maintenance, behaviour, and 
development of the two individuals have been published elsewhere (e.g. CorLıns and PAGE 1973; 
KLeEıman 1974; MELuN 1975; Kreıman et al. 1979). From April 21st 1980 until June 2nd 1980 the two 
giant pandas were observed daily - mainly during their early morning activity period after 0700 hours. 
A total of 131 hours was spent observing the animals during which 20 reels (13 cm reels = 5 hours) of 
vocal activity were recorded on tape. After June 2nd sporadic observations of at least one hour each 
were made once or twice a week until July 15 and few tape recordings were made during these. 

The recordings were made with a portable tape recorder UHER Report IC 4000 Automatic at 
19 cm/s and a microphone SENNHEISER MKH 816. The tape recordings were analyzed on a sound 
spectrograph KAY Sonagraph model 7029 A in wide band setting (some also in narrow band) and 
shaping flat with different frequency ranges. In the range most often used (20-2000 Hz) the effective 
temporal resolution is 0.013 s and the effective frequency resolution 75 Hz in the wide band mode. A 
total of 262 sonagrams was evaluated, most of them comprising several vocalizations (cf. Fig. 5). Tape 
recordings of the vocal behaviour of these individuals from previous years were available for 
comparison. Data on their vocal behaviour during the 1981 oestrus period of the ? as well as that of 
the & giant panda of the London Zoo which was at the National Zoo as a loan during this time were 
kindly provided by Dr. D. G. KrEıman and included in this paper. 

The following statements on the vocal repertoire of the giant panda, structure of its vocalizations 
and functional context(s) are limited by the fact that the study was based on only two individuals of 
this species, one of either sex, (there are only relatively few data for the London 4), living ın captivity. 
Preliminary results of field observations of giant panda vocal behaviour are in full agreement with the 
data presented here (G. SCHALLER, in litt.). 

All spectrograms figured are ın wide band. 


238 G. Peters 


Results 
The vocal repertoire of the giant panda 


As stated by KrEıman et al. (1979) vocal activity in the d and the 2 is mainly restricted to 
the pro-oestrus and oestrus period of the 2. Both individuals showed a significant increase 
in rate of vocalizing some days before the ?’s peak heat on May 16, 1980, and a steady and 
relatively rapid decrease after that date. Both the increase and the decrease in rate of 
vocalizing stretched over alonger time inthe d,the 2 being highly vocal only during a few 
days around her peak heat. The vocal activity of both individuals very probably was 
affected by the procedures of electro-ejaculation and artıficial insemination carried out on 
May 17 and 18 such that the decrease in rate of vocalizing observed was probably more 
rapıd than normal. During the remainder of the observation period the animals vocalized 
only infrequently; generally there is little vocal activity in both individuals except during 
the oestrus period of the 2. 

In the ? two types of vocalization occurred nearly exclusively during the oestrus 
period; they were rarely heard at other times of the year. Thus in addition to other 
behavioural changes, the 2’s oestrus was marked by a significant increase in rate of 
vocalızing and the occurrence of specific vocalizations, the bleat and the chirp (cf. 
Kıeıman et al. 1979). In the & one vocalization, the bleat, accounted for nearly all vocal 
activity, other vocalızations being uttered only occasionally. In 1981 the London d did not 
perform any bleating and was less vocal than the National Zoo d. 

The d’s repertoire comprised four vocal forms: honking, bleating, growling and 
squealing. The latter two vocalizations were not heard in the 1980 observation period but 
were ıdentified on tape recordings from previous years. Honking (Fig. 1) (duration: 
0.37-0.45 s; frequency range: 0.2-2 kHz, tonal, up to 4 harmonics, the fundamental has the 
highest amplitude) is a highly stereotyped short call of low to medium intensity with a 
rapid downfall in pitch towards its end. Usually it is produced in bouts with relatively 
regular intervals between the calls. Bleating (Figs. 2, 5) (duration: 0.06-2.8 s; frequency 
range: 0.18-5.4 kHz, up to 5 and more harmonics, the fundamental and the first harmonic 
usually have the highest amplitude) is a tonal call of medium intensity with a rapid 


mw kKHZ 


Fıg. 1. Series of honks by the d giant panda 


A note on the vocal behaviour of the giant panda, Ailuropoda melanoleuca 239 


m kHz 


Fıg. 2. Bleating by the d. Note the high rate of vocalizing and the variable duration of the bleat, due to 
the number of phases of frequency modulation 


frequency and amplitude modulation, its basic structural pattern being that of a twitter. 
The duration of the bleat varies according to the number of phases of regular frequency and 
amplitude modulation in the call. Compared with the size of the giant panda the bleat ıs 
rather high in pitch. 

The vocal repertoire of the 2 comprised more forms than that of the d, including all 
types that were observed in the d. In addition to bleating, honking, squealing and growling 
the ® articulates moans, barks and chirps; none of the latter three vocalizations was ever 
heard in the National Zoo & during this study and in previous years (B. BINGHAMm, ]. 
BLock, F. DREHER, D. KLEIMAN, pers. comms.). KLEIMAN (in lıtt.), however, recorded 
moans from the London d giant panda ın April 1981. 

Honking (Fig. 3) and bleating (Fig. 4) in the ? have the same basic structure as the 
equivalent vocalızations of the d. Growling is a drawn-out, relatively low-pitched, noisy 
and pulsed vocalization (duration: 0.6-2.2 s; frequency range: 0.1 - above 8 kHz, with 
some tonal structure below 4 kHz and purely noisy in the upper range). These structural 
characteristics apply to this vocaltype inthe ® as well as ın the London Zoo d. Squealing 
(duration: 0.19-1.1 s; frequency range: 0.16-8 kHz, noisy beginning and end, tonal in the 
middle, the fundamental has the highest amplitude) is a short to medium-length, high- 
pitched piercing call of high intensity. The noisy beginning and end of the call are lower in 
pitch than its more tonal middle section which is rapidly frequency modulated. Its 
structure is similar in the d and the 9; the data given are of vocalizations recorded in 
previous years because it could not be recorded during the 1980 observation period. 
Chirping (Fig. 5) (duration: 0.08-0.99 s; frequency range: 0.14-8 kHz, up to 5 and more 
harmonics, the fundamental and sometimes the first harmonic have the highest energy) is a 
short to medium-length, high-pitched and tonal call of variable intensity with a rapid 
frequency modulation; generally it descends in pitch towards its end. The frequency 
modulation in the beginning of chirping is highly variable. In relation to the size of the 
giant panda chirping ıs remarkably high in pitch, even more so than bleating. Moans 
(Figs. 6, 7, 9) (duration: 0.13-2.46 s; frequency range: 0.08-4 kHz, rarely above this range, 
up to 10 harmonics, but usually less, the fundamental and the first two harmonıcs have the 
highest amplitude) usually are low- to medium-intensity calls of variable duration with a 


240 G. Peters 


relatively low pitch and little frequency modulation. Moans quite often are truncated in the 
beginning, thus forming one or several short starting elements in front of the main portion 
of the call (cf. Fig. 6) or a vocalization may just consist of one or several such fragmentary 
elements (cf. Fig. 7). The same structural characteristics were found in the moans of the 
London Zoo & recorded by KıEıman in 1981. Barks (Figs. 6, 8) (duration: 0.13-0.29 s; 
frequency range: 0.06-6 kHz) are short, fairly tonal to totally noisy medium- to high- 
intensity vocalizations somewhat similar to the barking of a medium-sized dog. 

All vocalizations mentioned so far in the d and the ® can be clearly defined in their 
structural characteristics and thereby distinguished from one another. In the 2 all call 
types with the exception of honking formed a graded system with at least one other type of 
vocalization. Intermediate vocal forms were especially frequent during peak oestrus. No 


m— kHz 


Fıg. 3. Series of honks by the 2 giant panda 


m kHz 


Fig. 4. Bleating by the 2. This vocalization in the 2 is structurally as variable as ın the S 


A note on the vocal behaviour of the giant panda, Ailuropoda melanoleuca 241 


nn klNz 
ie) 
(@) 
®) 
19) 
© 
@) 
[®) 


. 
4 
; 


Fig. 5. Chirps (C) by the ® and bleats (B) by the d. The high rate of vocalizing in both sexes is typical 
of the peak oestrus of the ? 


grading was found in the vocal repertoire of the National Zoo d, the London Zoo d 
produced intermediate vocal forms between moaning and growling. Bleating, present in 
the repertoire of the d and the ?, formed a graded system with chirping and moaning 
which were not heard in the National Zoo d. In 1981 the London Zoo d did not perform 
bleating during the observation period, therefore no statement as to the occurrence of 
intermediates between bleating and moaning in dd can be made at present. 


Functional aspects of vocalization 


With the exception of vocalizations emitted during agonistic behaviour (barkıng, growling, 
squealing) the functional significance of the giant panda’s other vocal forms was difficult to 
establish. There was hardly any evidence as to consistent behavioural patterns associated 
with any other type of vocalızation, either in the sender of the signal or its addressee. 
Growling and barking probably function as aggressive threats, and squealing may denote 
defensive threatening and/or submission. As squealing occurs when an anımal is bitten it 
may also be an expression of pain. The nearly exclusive occurrence of chirping and bleating 
during the ?’s oestrus period strongly suggests a functional correlation with physiological 
status. The S’s increased rate of bleating during this period also is probably related to the 
mating season, and the male rut, as well as the S’s attempts to approach the 2. Bleating - 
among other functions - may also indicate appeasement and reassurance since the giant 
panda is a solitary anımal in which the sexes rarely meet except for mating. Neither bleating 
nor chirping (in the ?) probably function as long distance advertisement calls to attract a 
partner for mating nor help in mutual localization when anımals alternately vocalıze since 
the intensity of these calls seems insufficient to carry far ın the dense vegetation of the giant 
panda’s natural habitat. However, this needs experimental verification as to the relative 
absorption of the sound signal with increasing distance from the source under these 
conditions. Olfactory communication is probably the main method for bringing the 
partners together for mating, as indicated by the significant increase ın scent marking 
activity of the $ during pro-oestrus and oestrus (KLEIMAN et al. 1979). The quantitative 
change in vocal activity observed during the oestrus period of the ? in these two 


242 G. Peters 


m— kHZ 


Fig. 6. A series of moans and a bark (BA) by the 2. The beginning of the 2nd and 3rd moan are 
truncated, forming 2 resp. 1 short starting element(s) 


individuals very probably was influenced by the fact that they are living in adjacent 
enclosures all year round with visual and olfactory contact between each other, a situation 
different from the wild. 

Moaning may denote a slightly aggressive threat. Its structural affinity to barking and 
the fact that both vocalizations quite often occur in the very same behavioural context 
support this view. Moreover, the increasing aggressiveness of the 2, signified by lunges 
towards the d, was accompanied by the occurrence of intermediate forms of moans and 
barks and - with more aggressive behaviour - a progressive replacement of the former by 


w— kHz 
m KHNZ 


Fig. 7 (left). Two fully truncated moans by the 9, split into 5 resp. 6 short elements. - Fig. 8 (right). A 
series of noisy barks by the 2. In highly agonistic contexts barks are uttered in bouts in rapid 
succession 


A note on the vocal behaviour of the giant panda, Ailuropoda melanoleuca 243 


the latter. The structural relationship between growling and moaning and their occurrence 
in the same behavioural situations supports the functional interpretation of moaning given 
above. The situations in which honking was observed in both individuals gave no 
unequivocal hint as to the function of this vocalization. The 3 occasionally honked when 
he was locked in his den and appeared to want to be released to his main cage. Several times 
the d and the ? uttered bouts of honks when they were resting or even seemed to be 
sleeping. On the other hand, the d performed a long series of honks on seeing the zoo’s 
veterinarian after the first electro-ejaculation had been performed on him. Among other 
functions honking may probably denote frustration. Of the different vocal forms of 
Auluropoda observed and analyzed in this study honking is the most highly stereotyped call 
structurally, not showing any grading with other vocal types and usually i is produced ın 
relatively regular sequences even though no intra-specific communicatory function is 
obvious as it was never heard during interactions between the anımals. 

In only a few cases the occurrence of intermediate type vocalızations, i.e. the phenome- 
non of grading, in the ? (and in the London d in 1981) seemed to be correlated with 
motivational changes inferred from changes in the 2’s behaviour in her interaction with the 
&. The increase in aggressive threat behaviour accompanied by the occurrence of moan/ 
bark intermediates has already been mentioned. However, the highest incidence of vocal 
grading inthe ? was observed during long bouts of vocalizing which were accompanied by 
the 2 masturbating (rubbing the genital region with a forepaw). During masturbation she 
mainly performed bleat/chirp/(moan) intermediates without any changes in behaviour nor 
any interaction with the d. In 1981 the London Zoo & performed moan/growl intermedi- 
ates during agonistic interactions with the $ as did the 2 in these situations. 


Discussion 


It is not known whether the vocal repertoire of the giant panda presented here ıs complete. 
D. G. KLEIMAN (pers. comm.) and G. SCHALLER (pers. comm. through KrEıman) heard 
explosive breaths and jaw clappıng which may constitute vocal signals and perhaps there 
are further vocal types in this species not observed until now. However, if the differences 


mw kHZ 


Fıg. 9. Two intermediate moans by the ? with significant frequency modulation and relatively high in 
pitch. The beginning of the 1st call structurally represents a transition to bleating 


244 G. Peters 


in the vocal repertoire found in these two individuals (and the London Zoo 8) are species- 
typical, ıt would suggest a significant sex-specific difference in the vocal repertoire of 
Ailuropoda, the ? having more forms than the d. In most mammals the species-specific 
vocal repertoire is genetically fixed (PETERS 1978a; EHRET 1980) and thus not influenced by 
captivity. 

To my knowledge comparisons of vocalizations between mammalian families have not 
been evaluated as taxonomic evidence and therefore the phylogenetic significance of 
differences or conformities in vocalizations at this taxonomic level is not yet established. In 
the Pinnipedia STIRLING and WARNEKE (1971) and TRILLMICH and MAJLUF (1981) argued 
that the intermediate position of Arctocephalus pusillus between the two sub-families 
Arctocephalinae and Otariinae of the family Otariidae is also evident in the vocal repertoire 
of this species. In birds the significance of vocalization above the generic level has been 
questioned (THORPE 1961; THIELCKE 1964; Lanvon 1969). Nevertheless, the available data 
of vocaliızation of the giant panda (with vocalızations assumed to be genetically fixed 
behavior characteristics) should be incorporated in the discussions of the systematic 
position of this species. 

In no bear species (Ursidae) studied so far (PETERS 1978b; Jorpan 1979) has a 
vocalization like bleating, a predominant form in Ailuropoda, been described, whereas the 
twitter call of the red panda (Allurus fulgens) represents the same structural pattern as the 
giant panda’s bleat and appears to be functionally equivalent to it (ROBERTS and KEssSLER 
1979; ROBERTS 1981). Similar vocalizations in corresponding functional contexts are 
known in procyonid species like the ringtail (Bassariscus astutus) (WILLEY and RiCHARDS 
1981) and juvenile raccoons (Procyon lotor) (personal observations). Vocal forms with a 
basıc structural pattern of a twitter call in the relevant functional contexts are quite 
uncommon in the Carnivora (personal observations), therefore this sımilarity ıs striking. 
Although no conclusive arguments can be made without more details on ursid, procyonid 
and Ardlurus vocalizations, some comments can be made. 

It seems doubtful that the conformity in Azluropoda, Ailurus and procyonid twitter/ 
bleat vocalızations is a result of convergent evolution. Nor are these forms likely to 
represent symplesiomorphic characteristics. A convergence would imply the adaptive 
significance of this structural pattern in the vocalizations of Ailuropoda, Ailurus and 
procyonids as opposed to the different phylogenesis of the functionally equivalent 
vocalizations in other carnivores and especially in the Ursidae. On the other hand, to argue 
that the twitter/bleat pattern is a symplesiomorphic character requires an explanation why 
this supposedly original structural pattern was retained in some species (especially in those 
mentioned here) but not in the majority of carnıvore species known so far in this respect 
and specifically not in the Ursidae. Moreover, the bleating and chirping of Arluropoda (the 
latter ıs also not known in any ursid species) are extremely high in pitch compared with the 
species’ size, for which there is no plausible explanation, either functional or anatomical. 
According to Davis (1964) the larynx of the giant panda resembles that of the Ursidae. 

Although the data on vocalizations presented here cannot offer any conclusive evidence 
as to the phylogenetic position of the giant panda, they are, however, inconsistent with the 
view published by THEnIus (1979) that Azluropoda ıs an early offshoot of the Ursidae 
(deserving family rank because of its many peculiar characteristics) and is not related to the 
red panda (considered the sole species of the sub-family Ailurinae in the Procyonidae), 
more so with the view of CHorn and HorrMmann (1978) that the giant panda is a species 
belonging to the family Ursidae. 

Note added in proof: One d& was heard to chirp in a recent study in China (KLEIMman ın litt.). Thus, 


Sc and 2? seem to have the same basic vocal repertoire but differ in the frequency of use of the 
different vocalizations. 


A note on the vocal behaviour of the giant panda, Ailuropoda melanoleuca 245 


Acknowledgements 


This study was done under a contract within the Smithsonian Scholarly Studies Program “ Analysis of 
Animal Communication”. I am especially grateful to Dr. D. G. KrEıman for inviting me to do this 
study and to the Smithsonian Institution for providing the grant. My thanks go to the whole staff of 
the Department of Zoological Research at the National Zoo for the encouragement and help I received 
and for including me in the cooperative and friendly atmosphere of the ınstitute. Thanks are also due 
to the administration, curators and keepers of the zoo for their assistance, especially the keepers of the 
giant pandas for their understanding patience and help. Drs. J. F. EISENBERG, ] .G. FRAZIER, S.L. 
GisH, D. G. KLEIman, and E. MoRToN made valuable comments on earlier drafts of the manuscript, 
their help ıs highly appreciated. 


Zusammenfassung 


Beobachtungen zum Lautgebungsverhalten des Großen Panda, Ailuropoda melanoleuca 
(David, 1869) 


Das Lautgebungsverhalten des Paares Großer Pandas im National Zoological Park, Washington, 
D.C., wurde zwischen 15. April und 15. Juli 1980 131 h lang beobachtet. Die Lautäußerungen wurden 
auf Tonband aufgezeichnet und lautspektrographisch ausgewertet. Tonbandaufnahmen vom Lautge- 
bungsverhalten dieser beiden Individuen aus den vorhergehenden Jahren lagen zum Vergleich vor. 

Bei d und 2 ist die lautliche Aktivität im wesentlichen auf die Hitzeperiode des ? beschränkt, wo 
beide Tiere sehr lautfreudig sind. Während der übrigen Zeit des Jahres äußern sie nur sporadisch 
Laute. ? und d haben mehrere strukturell eindeutig definierte Lautformen, die jedoch bis auf einen 
Lauttyp jeweils mit zumindest einer anderen Lautform ein Kontinuum vielfältiger Übergangslaute 
bilden. Das 2 besitzt ein reichhaltigeres Lautrepertoire als das d ; anhand der bisherigen Beobachtun- 
gen ist allerdings keine Aussage möglich, ob schon alle Lautformen dieser Art erfaßt sind. 

Die Funktion der einzelnen Lauttypen ist bis auf diejenigen, die für agonistische Verhaltenszusam- 
menhänge charakteristisch sind, nur ungefähr zu bestimmen. Da beim ? zwei Lautformen nahezu 
ausschließlich während ihrer Hitzeperiode auftreten, dürfte ihr physiologischer Zustand dafür mit 
auslösend sein. 

Einige Lauttypen des Großen Panda sind angesichts der Körpergröße der Art von ganz unge- 
wöhnlich hellem Klang. Strukturell zeigt eine dieser Lautformen deutliche Übereinstimmungen mit 
einem auch funktional entsprechenden Laut von Procyoniden und dem Kleinen Panda (Aslurus 
fulgens), bei Ursiden sind keine derartigen Lautäußerungen bekannt. Im Hinblick auf das Problem der 
stammesgeschichtlichen Ableitung von Ailuropoda erscheint dies als ein Hinweis, Beziehungen dieser 
Art zu den Ursiden neu zu interpretieren. 


Literature 


CHORN, J.; HOFFMANN, R. S. (1978): Ailuropoda melanoleuca. Mammalıan Spec. No. 110. 6 pp. Am. 
Soc. Mammalogısts. 

Coins, L. R.; PAGE, J. K. (1973): Ling-Ling and Hsing-Hsing: Year of the Panda. New York: 
Anchor Press, Doubleday. 

Davıs, D. D. (1964): The giant panda: a morphological study of evolutionary mechanisms. Fieldiana: 
Zool. Mem. 3, 1-339. 

EHRET, G. (1980): Development of sound communication in mammals. In: Advances in the Study of 
Behavior. Vol. 11, pp. 179-225. Ed. by ROSENBLATT, J. S.; HınDE, R. A.; BEER, C.; BUSNEL, 
M.-C. New York, London: Academic Press. 

Haas, G. (1963): Beitrag zum Verhalten des Bambusbären (Ailuropus melanoleucus). Zool. Garten 
INSER279225233: 

JoRDAN, R. H. (1979): An observational study of the American black bear (Ursus americanus). Ph. D. 
Diss., Univ. of Tennessee, Knoxville (unpubl.). 

KLEIMAn, D. G. (1974): Activity rhythms in the Giant panda A:luropoda melanoleuca: an example of 
the use of checksheets for recording behaviour data in zoos. Int. Zoo Yb. 14, 165-169. 

KLEIMAnN, D. G.; KARESH, W. B.; CHu, P.R. (1979): Behavioural changes associated with oestrus ın 
the Giant panda A:luropoda melanoleuca with comments on female proceptive behaviour. Int. Zoo 
Yb. 19, 217-223. 

Lanyon, W.E. (1969): Vocal characters and avian systematics. In: Bird Vocalizations: Their Relations 
to Current Problems ın Biology and Psychology. pp. 291-310. Ed. by Hınpe, R. A. London: 
Cambridge Univ. Press. 

MErun, N. C. (1975): Giant panda exhibit at the National Zoological Park, Washington. Int. Zoo Yb. 
15, 263-265. 

MoRRıs, R.; Morrıs, D. (1966): Man and Pandas. London: Hutchinson. 


246 G. Peters 


PETERS, G. (1978a): Vergleichende Untersuchung zur Lautgebung einiger Feliden (Mammalia, Feli- 
dae). Spixiana, Suppl. 1, 1-206. 

— (1978b): Einige Beobachtungen zur Lautgebung der Bären: Bioakustische Untersuchungen im 
Zoologischen Garten. Z. Kölner Zoo 21, 45-51. 

ROBERTS, M. S. (1981): Reproductive biology of the red panda, Azlurus fulgens, in captivity. M. S. 
Thesis, Univ. Maryland, College Park (unpubl.). 

ROßBERTS, M. S.; KessLEr, D. $. (1979): Reproduction in Red pandas, Azlurus fulgens (Carnivora: 
Ailuropodidae). J. Zool. 188, 235-249. 

SCHÄFER, E. (1938): Der Bambusbär (A:rluropus melanoleucus). Zool. Garten N. F. 10, 21-31. 

SCHALLER, G. B. (1981): Pandas in the wild. Nat. geogr. Mag. 160, 735-749. 

SCHNEIDER, K. M. (1939): Einiges vom Großen und Kleinen Panda. I. Vom Großen Panda. Zool. 
Garten N. F. 11, 203-232. 

SHELDON, W. G. (1937): Notes on the giant panda. J. Mammalogy 18, 13-19. 

— (1975): The Wilderness Home of the Giant Panda. Amherst: Univ. Massachusetts Press. 

STIRLING, I.; WARNEKE, R. M. (1971): Implications of a comparison of the airborne vocalızations and 
some aspects of behaviour of the two Australian fur seals, Arctocephalus spp., on the evolution and 
present taxonomy of the genus. Aust. J. Zool. 19, 227-241. 

THENIUS, E. (1979): Zur systematischen und phylogenetischen Stellung des Bambusbären: Azluropoda 
melanoleuca Davıd (Carnıvora, Mammalia). Z. Säugetierkunde 44, 286-305. 

THIELCKE, G. (1964): Lautäußerungen der Vögel in ihrer Bedeutung für die Taxonomie. J. Orn. 105, 
78-84. 
THORPE, W. H. (1961): Bird Song: The Biology of Vocal Communication and Expression in Birds. 
Cambridge Monographs in Experimental Biology. No. 12. London: Cambridge Univ. Press. 
TRILLMICH, F.; MajLur, P. (1981): First observations on colony structure, behavior, and vocal 
repertoire of the South American Fur Seal (Arctocephalus australis Zimmermann, 1783) in Peru. Z. 
Säugetierkunde 46, 310-322. 

WILLEY, R. B.; RıicHARDs, R. E. (1981): Vocalizations of the ringtail (Bassariscus astutus). SWest. Nat. 
26, 23-30. 


Author’s address: Dr. GusTAav PETERS, Zoologisches Forschungsinstitut und Museum Alexander 
Koenig, Adenauerallee 150-164, D-5300 Bonn 1 


Trichinella spiralis in walruses from the Thule district, 
North Greenland, and possible routes of transmission 


E. W. Born, B. CLAusen and Sv. AA. HENRIKSEN 


Receipt of Ms. 5. 2. 1982 
Abstract 


Examined samples of muscle tissue from 126 Atlantic walruses, Odobenus rosmarus rosmarus (L.), 
collected in the Thule district, North Greenland (1975, 77 and 78), for Trichinella spıralıs. 

Two adult males were found to be infected with Trichinella spiralıs (prevalence of 1.6 %). 

It is suggested that the walrus in the Thule district contracts trichinosis from scavenging on 
carcasses of polar bears or/and sled dogs. 


Introduction 


The parasite Trichinella spiralis in the walrus (Odobenus rosmarus Linnaeus) is a potential 
threat to the Inuit health. An epidemic disease in 1947 among the Inuit of central West 
Greenland was caused by the eating of trichinous walrus meat (THORBORG et al. 1948; 
ROTH 1949). Since then, trichinous walruses have been reported from different arctic areas 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4704-0246 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 246-251 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Trichinella spiralis in the Atlantic walrus 247 


(e. g. ROTH and Mansen 1953; Rausch et al. 1956; BRown et al. 1949; Kurtunen 1954; 
THorsHAug and RosTeD 1956; Fay 1960; MAansen 1961; KozLov 1971; THıng et al. 1976; 
MaRGoLIS et al. 1979). 

In the Thule district, North Greenland, the walrus is a cornerstone in the subsistence 
catch and trichinous walruses do occasionally cause trichinosis in humans. In the fall 1977 a 
woman and two children were severely sick of trichinosis after having eaten raw walrus 
meat which was insufficiently frozen. 

The Inuit of Grise Fjord at Ellesmere Island, Canada, take a limited number of walruses 
from the same population (BowLER 1976, fide REEvEs 1978). 


Materials and methods 


Samples of diaphragma, the masseter and the intercostal muscles from 126 walruses (64 males, 59 
females and 3 with no sex recorded) in the Thule district were collected in 1975, 1977 and 1978 (Fig. 1). 
The total sample represents about 60 % of the Inuit annual subsistence catch of walrus on the stretch, 
Saunders Island (Agpat, 76°30’N) to Cape Inglefield (Anoritoog, 78°30’N) in the period 24 April until 
28 July. The samples were kept frozen at minus 12°C until examination in the laboratory where 
15-20 g muscle from each individual were examined in a trichinoscope and by digestion (HENRIKSEN 
1978). 

The ages of 106 walruses caught in 1977 and 1978 were determined from growth layers in the 
cement of the lower cheek teeth as described in MansrIELD (1958) and KryLov (1965). No teeth for 
age determination were obtained from 20 walruses ın 1975 (reported in ThHıng et al. 1976). However, 
an approximate age has been determined for these specimens by comparison of: 1. nosetail length and 
2. tusk length and circumference, with age-dependent growth curves established on measurement 
results of walruses of known age collected in 1977 and 1978. 


Results 


Trichinella spiralis was detected ın two adult males. One was killed 28 May 1975 off Nege 
in Murchison Sound. According to tusk and body dimensions this male was about 12 years 
old and in good condition. No larvae were detected in the trichinoscope, but 1-2 larvae per 
gram muscle tissue were found after digestion, this indicating a low infection level. The 
other infected walrus was a 20-year-old, apparently healthy, male killed 6 June 1978 off the 
northeastern point of Northumberland Island (Kiataq). This anımal was infected with 
about 50 Trichinella larvae per gram muscle tissue, indicating a moderate infection level. 

The results of the present study are shown in the Table together with the results of other 
studies for comparative purposes. In this study the overall infection prevalence in the 
walrus from the Thule district is 1.6 %. Including the 74 walruses recorded in MADSEN 
(1961) the infection prevalence in the Thule area is 1.0 %. 

Although there is no criterion for determining whether a walrus is trichinous or not, 
from physical appearance alone, ıt should be mentioned that the infected male taken in 
1978 had an abnormal hump on its shoulders. By a superficial inspection this hump 
consisted of normal-looking somatic muscle tissue and blubber. 

In the fall 1977 three persons in the Thule district became severely sick of trichinosis 
after having eaten infected meat from an old bull which had been caught at Northumber- 
land Island. Judged from the skull of this walrus it was estimated to be at least 20 years old. 


Discussion 
The finding of Trichinella spiralıs in the walrus and other marine mammals is puzzling. The 


walrus is an inshore bottom-feeder whose main diet consists primarily of bivalve molluscs. 
Occasionally an insignificant number of fish and decapod crustaceans are also found in the 


248 E. W. Born, B. Clausen und Sv. Aa. Henriksen 


Age composition of the com- 
bined sample of walruses ana- 
lyzed for Trichinella spiralıs 
(1975, 1977 and 1977), Thule 
district North Greenland. 
Black columns = males; Whi- 
te columns = females; Cross 
hatched columns = anımals 
with age determined from 
tusk and body measures 


DISS 


O [ ISIS SISST 
N 
\ 


age and sex unknown 


15 20 25 
Age in years 


Table 


Prevalence of Trichinella spiralis among walruses (Odobenus rosmarus) in different arctic areas 


No. No. % Source 
examined infected infected 


Alaska Rausch et al. (1956); Fay (1960) 
Eastern Canada 17 ; Brown et al. (1949); KuITuUNnEN (1954) 
The Svalbard Area 7 i THORSHAug and RosTED (1956) 


Greenland 


East Greenland 

Upernavik (NW. Greenl.) 
Egedesminde (W. Greenl.) 
Frederikshäb (W. Greenl.) 


Thule district (N. Green.) 


(1961) ! MaDsen (1961) 
THın et al. (1976); this study 
This study 
This study 


MaDsen (1961) 


! Trichinoscopical detection. — ? Trichinoscopical detection and digestion. 


stomach (Dunsar 1949; VıBE 1950; MAnsFIELD 1958; Fay and Lowry 1981). Because 
Trichinella is absent from these food items different alternative routes of infection in the 
walrus have been proposed. VıBE (1950), Rausch et al. (1956) and Fay (1967) suggested 
that Trichinella is transferred from carcasses of polar bears and sled dogs to marine 
mammals, with amphipods serving as „‚intermediate hosts“. The transfer of Trichinella via 
amphipods (Fav 1967) and via fish (KozLov 1971) has been experimentally demonstrated. 
Although the ringed seal (Phoca hispida Schreber) feeds extensively on amphipods and 
other crustaceans (e.g. MCLAREN 1958), only 0.06 % (one of 1775 examined) Trichinella 
infected ringed seals have been found in Greenland (MApsen 1961) and 0,7 % (two of 300 
examined) in Alaska (Fav 1960). These findings indicate that transfer of Trichinella via 
amphipods to marine mammals is very unusual and may be even more unlikely ın the case 
of the walrus, which does not feed on crustaceans to any extent. FAy (1960), underlining 
the occasional carnivorous habits of the walrus, suggested that walruses might be infected 


Trichinella spiralis in the Atlantic walrus 249 


from bearded seals (Erignathus barbatus Erxleben) and ringed seals. The walrus kills seals 
during periods of famine, and there are several reports of walruses with remnants of seals ın 
the stomach (e.g. JOHANSEN 1910; DEGERBOL and FREUCHEN 1935; VıBE 1950; LOUGHREY 
1959; Fay 1960). MAnnıInG (1960) drew attention to the low frequency of seals infected 
with Trichinella and suggested that a polar bear-walrus-polar bear cycle could not be ruled 
out. It is most likely that the walrus is infected when scavenging on polar bear carcasses, as 
suggested by ManniIngG (1960) or on polar bear and sled dog carcasses (VıBE 1950). In the 
Thule district 20 % of the polar bears (3 of 15 examined) was found to be infected with 
Trichinella and 91 % (38 of 42 examined) of the sled dogs was also infected (MADsen 1961). 

Kozrov (1971) was also of the opinion that the main source of Trichinella infection ın 
pinnipeds and whales is carcasses of terrestrial and marine mammals. 

In the Thule district the Inuit kill diseased and old sled dogs and leave them on the ice ın 
the tidal zone before ice break-up ın June-July (VıBE 1950 and own information). This 
activity is designed to eliminate the necessity of feeding surplus dogs during the open water 
period. About 200-300 dogs of a total of about 1500-1600 sled dogs in the Thule district 
are disposed in this manner every year. If the walrus contracts trıchinosis when scavenging 
on polar bear and dog carcasses this route of infection could also explain why trichinous 
bearded seals have been found, as reported by Mapsen (1961). The bearded seal is also a 
bottomfeeder (VısE 1950) which has never been reported to prey on other seals, to our 
knowledge. 

The average annual catch of walruses in the Thule district is about 200 animals of which 
about 53 % are males (own information). Thus, probably three to four trichinous walruses 
are taken in the district annually. The yield of the annual walrus catch is about 80 tons of 
meat, blubber and hide. While most of the yield is used for dog food an unknown 
proportion is eaten by humans. In the case of the walrus the epidemiological risk to 
humans is theoretically large because walrus meat is distributed to many households in the 
communities in contrast to the meat of other pinnipeds and of the polar bear. About 77 % 
of the annual catch of walrus ıs taken in the period from May until October (own 
information) where temperatures in the Thule district are often about or above freezing 
point which increases the rısk of walrus meat being infectious. 

Temperatures in household freezers are not low enough to reduce infectivity of arctic 
Trichinella ssp. (Anon. 1981). For example, portions of infected meat of the Alaskan black 
bear (Ursus americanus Pallas) held at minus 15 °C for up to 35 days showed no loss of 
infectivity for laboratory anımals (Anon. 1979, fide Anon. 1981). 

Thus, walrus meat must be cooked thoroughly, which according to RansoM and 
SCHWARTS (1919) means that all parts of the meat must be raised to 58 °C (137 F) in order 
to kill the encysted Trichinella larvae. In the Thule district the walrus meat is boiled on 
kerozene burners, and at least when on hunting trips the Inuit often eat the meat while 
parts of it are only half done. Furthermore, raw walrus meat which previously has been 
frozen for a variable period is also eaten. 


Acknowledgements 


We wish to thank the Inuit of the Thule district for their great help during our work in the area. 
Furthermore, we wish to express our gratitude to Dr. THERESE KRISTENSEN, Zoological Museum of 
Copenhagen, for collecting some of the samples, and Dr. HoLGEr Mapsen, Zoology Laboratory 
University of Copenhagen, for offering useful criticism during the preparation of this paper. 

This study was supported by grants from the Danish Natural Science Research Council and from 
the Commission for Scientific Research in Greenland. 


250 E. W. Born, B. Clausen und Sv. Aa. Henriksen 


Zusammenfassung 


Trichinella spiralıs in Walrossen aus dem Thule-Gebiet, Nordgrönland, 
und mögliche Wege der Übertragung 


Fleischproben von 126 Atlantischen Walrossen, Odobenus rosmarus rosmarus (L.) aus dem Thule- 
Gebiet, Nordgrönland (1975, 1977 und 1978), sind auf Befall durch den Parasiten Trichinella spiralis 
analysiert geworden. 

Zwei erwachsene Bullen waren mit Trichinella spiralıs infiziert (Häufigkeit 1,6 %). 

Es wird angenommen, daß das Walroß aus dem Thule-Gebiet durch Fressen von Eisbären- und/ 
oder Hundekadavern infiziert wird. 


References 


Anonymous (1979): Trichinosis surveillance annual summary 1978. Center for Disease Control, 
Sept.,.1979: 

ANoNnYMous (1981): Trichinois Associated with Meat from a Grizzly Bear. Morbidity and Mortality 
Weekly Report, Center for Disease Control, March 20, Vol. 30, 15-21 

Brown, M.; CRONK, B.; de SINNER, F.; GREEN, J. E.; GIBBONS, J. E.; KUITUNEN-EKBAUM (1949): A 
Note on Trichinosis in Anımals of the Northwest Territories. Can. Public Health 40, 20-21. 

DEGERBoL, M; FREUCHEN, P. (1935): Mammals. Report of the Fifth Thule Expedition 1921-1924, 
(4-5): 278 pp. Copenhagen: Gyldendalske Boghandel. Nordisk Forlag. 

DowLer, D. H. (1976): Average marine mammal harvest 1973 to 1975. Northwest Territories. 
(Unpubl. tables received from H. R. TRUDEAU. Fish. and Marine Service, Yellowknife, N. W.T.). 

DunBaR, M. J. (1949): The pinnipedia of the arctic and subarctic. Fisheries Research Board of Canada. 
Bulletin 85, 1-22. 

Fay, F. H. (1960): Carnivorous walrus and some arctic zoonoses. Arctic 13, 111-122. 

Fay, F. H. (1967): Experimental transmission of Trichinalla spiralis via marine amphipods. Can. ]. 
Zool. 46, 597-599. 

Fay, F. H.; Lowry, L. F. (1981): Seasonal use and feeding habits of walruses in the proposed Bristol 
Blay clam fishery area. Final Report to North Pacific Management Council P.O. Box 3136 DT, 
Anchorage, Alaska 99510, U.$.A. Contract No. 80-3. 

HENRIKSEN, Sv. Aa. (1978): A new tegnique for demonstration of Trichinella spiralis larvae in 
suspensions of digested muscle tissue. Acta vet. scand. 19, 466-68. 

JoHANSEnN, F. (1910): Observations on Seals (Pinnipedia) and Whales (Cetacea) on the Denmark 
Expedition, 1906-1908. Meddelelser om Grenland 45, 203-224. 

Kozıov, D. P. (1971): K voprosu o putyakh zarazheniya lastonogikh trikhinellezom. Trudy Gel’ 
mintologicheskoi Laboratorii Akademii Nauk SSSR 21, 36-40. (Sources of Trichinella spiralıs 
infection ın pinnipeds. Fisheries Research Board of Canada 1974, Translation Series No. 3010). 

Kryrov, V. I. (1965): Opredelenie vozrasta, temp rosta ı analız vozrastnoi struktury poboek 
tikhookeanskogo morzha. Morskie mlekopitayushchie. Akademiya Nauk SSSR. Izdatel’stvo 
Nauko, Moskva: 201-210. Determination of age, rate of growth and analysis of the age structure of 
the Pacific walrus catch. Fisheries Research Board of Canada 1968. Translation Series No. 1004). 

KuITunen, E. (1954): Walrus meat as a source of trichinosis in Eskimos. Can. J. Public Health 45, 30. 

LouGHREY, A. G. (1959): Preliminary Investigation of the Atlantic Walrus Odobenus rosmarus 
rosmarus (Linnaeus). Can. Wildl. Serv., Wildlife Management Series 1, No. 14. 

MAaDpsen, H. (1961): The distribution of Trichinella spiralis in sledge dogs and wild mammals ın 
Greenland under a global aspect. Meddelelser om Grenland 159. 

MAnNING, T. H. (1960): Comments on “Carnivorous Walrus and Some Arctic Zoonoses”. Arctic 14, 
76-77. 

MAnsFIELD, A. W. (1958): The Biology of the Atlantic Walrus (Odobenus rosmarus rosmarus L.) in 
Eastern Canadien Arctic. Journal of Fisheries Research Board of Canada. Manuscript Report Ser. 
Biol. 653. 

Marcousıs, H. $.; MıpDpAaucH, J. P; Burcess, R. D. (1979): Arctic Trichinosis: Two Alaskan 
Outbreaks from Walrus Meat. ]J. Infect. Diseases 139, 102-105. 

McLAREn, I. A. (1958): The Biology of the Ringed Seal (Phoca hispida Schreber) in the Eastern 
Canadıan Arctic. Journal of Fisheries Research Board of Canada. Bulletin 118. 

Ransom, B. H.; SCHwARTZ, B. (1919): Effect on heat on trichinae. J. Agricult. Res. 17, 201-221. 

RauscH, R.; BABERO, B. B.; Rausch, R. V.; SCHILLER, E. L.; (1956): Studies on the helminth fauna of 
Alaska. XXVII. The occurrence of larvae of Trichinella spiralıs in Alskan mammals. Parasitology 
42, 259-271. 

Reevzs, R. R. (1978): Atlantic walrus (Odobenus rosmarus rosmarus): A literature survey and status 
report. Unites Staates Department of the Interior. Fish and Wildlife Service. Wildlife Res. Report 
10. 

RoTH, H. (1949): Trichinosis in arctic anımals. Nature 163, 805-806. 


Trichinella spiralis in the Atlantic walrus 251 


RortnH, H.; Mapsen, H. (1953): Die Trichinose in Grönland, abschließender Bericht der Jahre 
1948-1953. Proc. 14th Int. Congr. Zoology. Section X, Copenhagen, 340-341. 

Thing, H.; CLausen, B.; HENRIKSEN, Sv. Aa. (1976): Finding of Trichinella spiralis in a walrus 
(Odobenus rosmarus L.) in the Thule district, Northwest Greenland. Nordisk Vet. Med. 28, 59. 

THORBORG, N. B.; Turınıvs, $.; ROTH, H. (1948): Trichinosis in Greenland. Acta Pathologica 25, 
778-794. 

THORSHAUG, K.; RoSTED, A. F. (1956): Researches on the prevalence of trichinosis in anımals in 
Arctic and Antarctic waters. Nordisk Vet. Med. 8, 115-129. 

VIiBE, C. (1950): The marine mammals and the marine fauna in the Thule district (Northwest 
Greenland) with observations on ice conditions in 1939 ın 1939-1941. Meddelelser om Gronland 
150. 


Authors’ addresses: E. W. Born, Greenland Fisheries Investigations, Tagensvej 135, DK-2200 
Copenhagen N; B. CLAausen and Sv. AA. HENRIKSEN, State Veterinary Serum 
Laboratory, 27 Bülowsvej, DK-1870 Copenhagen V 


NZESSEINSCHAFTLICHE KURZMITTEILUNGEN 


Der Einfluß der Domestikation auf die Riechleistung 
der Hausmaus (Mus musculus) 


Von CHRISTEL SCHMIDT und U. SCHMIDT 


Zoologisches Institut der Universität Bonn 


Eingang des Ms. 6. 1. 1982 


Domestizierte Tiere unterscheiden sich von ihren freilebenden Wildformen nicht nur 
durch Veränderungen im äußeren Erscheinungsbild (Größe, Fellfärbung, Schädelform 
etc.), sondern auch durch die Kapazität ihrer Hirnschädel. Darwın (1859) wies erstmals 
auf dieses Phänomen hin; inzwischen wird die Abnahme der Hirngröße als ein Charakteri- 
stikum der Domestikation angesehen (HERRE und RönHrs 1973). Nieder evoluierte 
Gehirne, wie die der Nager, weisen dabei mit bis zu 9 % (Laborratte: 8,3 %; KRUSKA 
1975b) jedoch weit geringere Gewichtsverluste auf als in der Evolution höher stehende 
(Hausschwein: 33 %; KruskA 1970). Eine Ausnahme macht die Labormaus, bei der sich 
im Vergleich mit der westlichen Hausmaus keinerlei Hirngrößenunterschiede ergaben 
(NoRrD 1963). 

Veränderungen des Hirngewichtes sagen jedoch nicht unbedingt etwas über die Lei- 
stungsfähigkeit einzelner Funktionsbereiche aus, da Teilstrukturen wesentlich stärkere 
Umwandlungen aufweisen können als das Gesamtsystem. So besitzt z. B. die Laborratte, 
trotz einer Gesamthirngewichtsabnahme von 8,3 %, einen um 4,5 % größeren Bulbus 
olfactorius und ein um 11,2 % vergrößertes Tuberculum olfactorıum als die Wanderratte 
(KruskA 1975). 

Elektrophysiologische Untersuchungen am olfaktorischen System der wilden Haus- 
maus und der Labormaus sollten klären, ob bei dieser Art, die als einzige keinerlei 
Abweichungen im Hirngewicht aufweist, Divergenzen in der Leistungsfähigkeit des 
olfaktorischen Systems auftreten. 

Als Versuchstiere dienten adulte Männchen der wilden Hausmaus (Wildfänge und F 1). 
Mit Hilfe fest implantierter Elektroden (Methode: ScHMIDT 1978) wurden neurale olfakto- 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4704-0251 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 251-252 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


252. Christel Schmidt und U. Schmidt 


BS BME GER rische Reaktionsschwellen 
10 für Geraniol, Buttersäure 
und DButtersäuremethyl- 
ester bestimmt. Die Duft- 
10 reize wurden in log Abstu- 
fung (Faktor 10) präsen- 
tiert. Die Versuche began- 
nen immer mit unter- 
10 schwelligen Konzentratio- 
nen. Als positive Reaktion 
auf einen Duftreiz wurde 
8 das Auftreten des olfakto- 
risch evozierten Potentials 
gewertet. Die Schwellen- 
werte der wilden Haus- 
10 mäuse wurden mit den bei 
dd Albinomäusen ge- 
wonnenen Daten vergli- 
chen (ScHMIDT 1982). 
Die neuralen Antwort- 


Schwellenkonzentration (Moleküle/cm) 


Abb. 1. Neurale olfaktorische Reaktionsschwellen bei wilden Haus- muster? sind bee 
mäusen (@; n = 7) und Labormäusen (A; n = 24), (geometr. Mittel Be 
xs+sc). Ordinate: Schwellenkonzentrationen (Moleküle/cm?); Ver suchsgruppen völlig 
BS: Buttersäure, BME: Buttersäuremethylester, GER: Geraniol identisch. Wie aus der Ab- 
bildung hervorgeht, besit- 
zen dıe wilden Hausmäuse jedoch für alle 3 Testsubstanzen eine ca. um den Faktor 10 
niedrigere Schwelle als die Labortiere. 

In vergleichenden elektrophysiologischen und verhaltensphysiologischen Untersuchun- 
gen konnte gezeigt werden, daß die neuralen Reaktionsschwellen bei der hier verwendeten 
Methode mit Duftkonzentrationen korreliert sind, die vom Versuchstier bei einer Zwei- 
fachwahldressur gerade noch positiv beantwortet werden (SCHMIDT 1982). Aus den hier 
vorliegenden Ergebnissen kann man somit schließen, daß die olfaktorische Sensitivität der 
wilden Hausmaus ca. um den Faktor 10 größer ist als die der domestizierten Form. 
Berücksichtigt man die Bedeutung des Geruchssinnes, so scheint eine bessere olfaktorische 
Sensitivität der bei Nahrungssuche, Orientierung und sozialen Interaktionen auf den 
Geruchssinn sehr viel stärker angewiesenen freilebenden Tiere biologisch überaus sinnvoll. 


Literatur 


Darwın, CH. (1859): The origin of species by means of natural selection. London: Murray. 

HERRE W.; Rönrs, M. (1973): Haustiere — zoologisch gesehen. Stuttgart: G. Fischer Verlag. 

Kruska, D. (1970): Vergleichend cytoarchitektonische Untersuchungen an Gehirnen von Wild- und 
Hausschweinen. Z. Anat. Entwickl.-Gesch. 131, 291-324. 

Kruska, D. (1975a): Vergleichend-quantitative Untersuchungen an den Gehirnen von Wander- und 
Laborratten. I. Volumenvergleich des Gesamthirns und der klassischen Hirnteile. J. Hirnforsch. 
16, 469-483. 

Kruska, D. (1975b): Vergleichend-quantitative Untersuchungen an den Gehirnen von Wander- und 
Laborratten. II. Volumenvergleich allokortikaler Hirnzentren. J. Hirnforsch. 16, 485496. 

Norp, H.-J. (1963): Quantitative Untersuchungen an Mus musculus dom. (Rutty 1772). Zool. Anz. 
170, 311-335. 

SCHMIDT, U. (1978): Evoked-potential measurements of olfactory thresholds of laboratory mice (Mus 
musculus) to carboxylic acıds. Chem. Senses Flav. 3, 177-182. 

SCHMIDT, C. (1982): Behavioural and neurophysiological studies of the olfactory sensitivity in the 
albino mouse. Z. Säugetierkunde 47, 162-168. 


Anschrift der Verfasser: Dr. CHRISTEL SCHMIDT und Prof. Dr. Uwe SCHMIDT, Zoologisches Institut, 
Universität Bonn, Poppelsdorfer Schloß, D-5300 Bonn 


Zur Funktion und Bedeutung des sekundären Schultergelenks 
der Microchiropteren 


Von EVELYN SCHLOSSER-STURM! 


Zoologisches Institut und Zoologisches Museum der Universität Hamburg 


Eingang des Ms. 24. 6. 1982 


Dreizehn der achtzehn rezenten Microchiropteren-Familien sind durch ein Körperbau- 
merkmal gekennzeichnet, das MıLLER (1907) erstmalig beschrieben und „sekundäres 
Schultergelenk““ genannt hat. Dieses Gelenk - als Parallelentwicklung unabhängig vonein- 
ander in allen vier Überfamilien entstanden - ist bisher zu wenig beachtet worden; denn 
schon auf Grund der mehrmaligen Entstehung kann man erwarten, daß ihm eine wichtige 
Bedeutung zukommt. Auffallend ist, daß alle Kleinfledermausarten, die ausgedehnt rüt- 
teln, große jahreszeitliche Wanderungen oder stundenlange Jagdflüge unternehmen, bzw. 
schwalbenartig schnell und geschickt fliegen, mit dem sekundären Schultergelenk ausgerü- 
stet sind. In den fünf Familien mit herkömmlichem Gelenk, die auch in anderen Merkma- 
len ursprünglich sind (WEBER 1928), finden sich dagegen solche Flugspezialisten nicht. 

Das sekundäre Schultergelenk besteht hauptsächlich aus einer zweiten Gelenkfläche auf 
der Dorsalseite der Scapula, die mit dem stark vergrößerten Tuberculum majus der 
proximalen Humerusepiphyse artikuliert. Es handelt sich dabei um den seltenen Fall, daß 
ein Anlagerungsgelenk sekundär an einem Abgliederungsgelenk entstanden ist (STARCK 
1979). Zu seiner Funktion sagt MILLER — ohne näher auf die biologische Bedeutung 
einzugehen -: “..... thus giving the shoulder joint a very unusual strength, while limiting its 
motion strictly to a single plane.” 

Bei Untersuchungen zur funktionellen Morphologie dreier Microchiropteren-Arten 
stieß auch VaucHan (1959) auf diese Struktur, interpretierte sie jedoch ganz anders. Seiner 
Meinung nach schließen sich eine Bewegungseinschränkung im Schultergelenk und ein 
funktionierender Flügelschlag aus. Er sieht in dem Gelenk eine Sperrvorrichtung, die den 
Flügelaufschlag im oberen Kulminationspunkt abbremst und über die passive Vordehnung 
des mächtigen M. serratus ventralis den Abschlag des Flügels einleitet. Die Funktion der 
einzelnen Schultermuskeln und auch des Serratus deutete VaAUGHAaNn jedoch ausschließlich 
aus deren Lage und Faserverlauf. Seine Interpretation wurde in den folgenden Jahren von 
mehreren Autoren übernommen. 

Eigene umfangreiche, vergleichend-morphologische Untersuchungen am Schultergürtel 
der Mirochiropteren lassen die Richtigkeit der VaucHanschen Vorstellungen bezweifeln. 
Neueste elektromyographische Befunde über die Aktivitätsmuster einiger Schultermuskeln 
während des Fluges widersprechen den von VAUGHAan postulierten Verhältnissen ebenfalls 
(HERMANSoN und ALTENBACH 1981), so daß es an der Zeit scheint, eine neue Deutung der 
Funktion des sekundären Schultergelenks der Microchiropteren anzubieten: 

Manipulationen an frischtotem Chiropterenmaterial und an herauspräparierten Schul- 
terknochen bestätigten im wesentlichen die von MILLER beschriebene Bewegungsein- 
schränkung durch das sekundäre Gelenk. Diese tritt jedoch nicht bei jeder Armbewegung 


! Diese Veröffentlichung ist Teil einer Dissertation, die am Fachbereich Biologie der Universität 
Hamburg angefertigt wird. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4704-0253 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 253-255 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


254 Evelyn Schlosser-Sturm 


auf. Bei Nyctalus noctula z. B. befindet sich nur in Abduktionsstellung des Humerus - in 
einem Winkel von etwas weniger als 90° zur Wirbelsäule - das Tuberculum majus in vollem 
Kontakt mit der dorsalen Gelenkfläche der Scapula. Gleichzeitig greift ein Zapfen oberhalb 
der Cavitas glenoidalis, der Processus supraglenoidalis, in eine Grube am proximalen 
Humerusende. In dieser Stellung ist jede Rotation um die Humeruslängsachse ausgeschlos- 
sen. Diese „‚starre‘“‘ Gelenkverbindung wird erst bei Adduktion, nämlich dann, wenn der 
Humerus in einen Winkel von etwa 50° zur Körperlängsachse gebracht wird, gänzlich 
gelöst und läßt dann wieder die volle Beweglichkeit des Humeruskopfes um drei Achsen 
zu. Die Wirkung des Gelenkes kommt also nur bei bestimmten Bewegungen der Tiere, bei 
weitgehend abduziertem Oberarm zum Tragen und bedingt eine Einschränkung der 
Rotationsmöglichkeit um die Humeruslängsachse. 

Die funktionelle Deutung dieser Rotationssperre ergab sich aus der sorgfältigen mor- 
phologischen Analyse des Flugapparates der Kleinfledermäuse sowie aus der Beschäftigung 
mit der Flugmechanik dieser Tiere. 


Abb. Humerus und Scapula von Molossus ater. a: Medialansicht der proximalen Epiphyse des rechten 
Humerus zur Darstellung von Caput humeri und der Gelenkfläche des Tuberculum majus; b: 
Laterodorsalansicht der Scapula mit Cavitas glenoidalis, dorsaler Gelenkfläche und Processus supra- 
glenoidalis des sekundären Schultergelenks. Bei a und b knorpelbedeckte Gelenkflächen schraffiert. 
Rand der Schraffur = Rand der Gelenkkapsel. c: Dorsalansicht des Schultergelenks in Abduktions- 
stellung des Humerus. Gelenkflächen des sekundären Schultergelenks in vollem Kontakt; Einwärts- 
rotation nicht möglich; d: Dorsalansicht des Schultergelenks mit adduziertem Humerus. Elemente des 
sekundären Schultergelenks ohne Kontakt zueinander; Humerus um drei Achsen beweglich. Ac = 
Acromion; Cg = Cavitas glenoidalis; Ch = Caput humeri; Dg = Dorsale Gelenkfläche des sekun- 
dären Schultergelenks; Pc = Processus coracoideus; Ps = Processus supraglenoidalis des sekundären 
Schultergelenks; Ta = Tuberculum majus; Ti = Tuberculum minus 


Zur Funktion und Bedeutung des sekundären Schultergelenks der Microchiropteren 255 


Der Flügelschlag der Kleinfledermäuse wurde schon 1936 von EISENTRAUT als kom- 
plexe Rotationsbewegung beschrieben: Die Schwingen werden dabei während des 
Abschlages ausgestreckt von oben hinten ‚am Körper“ vorbei, kräftig nach vorn unten 
geschlagen (siehe auch VAUGHAN 1970; ALTENBACH 1979). Zu Beginn des Aufschlages 
werden Arm und Hand leicht an den Körper des Tieres herangezogen (Adduktion des 
Humerus, Beugung in Ellenbogen- und distalen Gelenken), dann nach oben hinten geführt 
und dort wieder ausgestreckt. 

Nur während des Abschlages ist der Arm so stark abduziert, daß das sekundäre Gelenk 
zum Einsatz kommt. Ohne zusätzlichen oder zumindest mit stark reduziertem Muskelauf- 
wand kompensiert es dabei eine Reihe von Kräften, die auf den Flügel sämtlich einwärtsro- 
tierend wirken. Ein Drehmoment ım Sinne einer solchen Einwärtsrotation wird, um nur 
zwei Faktoren zu nennen, durch die Kontraktion des M. pectoralis major, des wichtigsten 
Abschlagsmuskels, zwangsläufig ausgeübt. Gleichsinnig wirkt der Druck der Luft unter 
den abwärts bewegten Schwingen, die nur an ihrem vorderen Rand durch das stützende 
Armskelett mit dem Rumpf verbunden sind. 

Eine derartige, unbeabsichtigte Einwärtsrotation hätte aber nachteilige Folgen für den 
Flug. Zum einen würde die volle Kontraktionskraft des M. pectoralis major nur zum Teil 
in Hubkraft umgewandelt, zum anderen würde die Einwärtsrotation den Anstellwinkel 
des Flügels verändern. Jedes Flügelprofil arbeitet aber nur bei Anströmung unter bestimm- 
ten, relativ eng begrenzten Winkeln optimal, d. h. es erzeugt möglichst viel Auftrieb bei 
möglichst geringem Luftwiderstand. Jede unkontrollierte Verschwenkung des Profils ın 
der Luftströmung ist also zu vermeiden, da sonst die aufgewendete Muskelkraft nicht 
effektiv in Flugleistung umgesetzt werden kann. 

Das sekundäre Schultergelenk der Microchiropteren stellt somit eine Zwangsführung 
dar, die ın der kraftkostenden, auftrieberzeugenden Abschlagsphase des Fluges (NORBERG 
1976) eine ungünstige Bewegung des Oberarmes unterdrückt, die optimale Übertragung 
der Muskelkraft auf den Arm gewährleistet und gleichzeitig eine vorübergehend mechani- 
sche Festigung der Nahtstelle Arm/Rumpf ım Sinne einer Steifigkeit des Flügels bewirkt. 
Durch seine besondere Konstruktion bleibt jedoch die volle Drehbarkeit des Humerus um 
drei Achsen grundsätzlich erhalten, so daß die Kleinfledermäuse beim Laufen, Klettern etc. 
durch dieses Gelenk nicht behindert werden. 


Literatur 


ALTENBACH, ]J. S. (1979): Locomotor Morphology of the Vampire Bat, Desmodus rotundus. Spec. 
Publ. Amer. Soc. Mamm. 6, 1-137. 

EISENTRAUT, M. (1936): Beitrag zur Mechanik des Fledermausfluges. Z. wiss. Zool. 148, 159-188. 

HERMANSOoN, J W.; ALTENBACH, J. S. (1981): Functional Anatomy of the Primary Downstroke 
Muscles in the Pallid Bat, Antrozous pallidus. J. Mammalogy 62, 795-800. 

MILLER, G. S. (1907): The families and genera of bats. US. Nat. Mus. Bull. 57. 

NoRBERG, U. M. (1976): Aerodynamics, kinematics and energetics of horizontal flapping flight ın the 
long-eared bat Plecotus aurıtus. J. exp. Biol. 65, 179-212. 

STARCK, D. (1979): Vergleichende Anatomie der Wirbeltiere auf evolutionsbiologischer Grundlage. 
Bd. 2: Das Skelettsystem. Berlin, Heidelberg: Springer. 

VaucnHan, T. A. (1959): Functional morphology of three bats: Eumops, Myotis, Macrotus. Publ. Mus. 
Nat. Hist. Univ. Kan. 12, 1-153. 

- (1970): Flight patterns and aerodynamics. In: Biology of Bats Vol. I. Ed.-by W. A. WımsATT. 
London, New York: Academic Press. 

WEBER, M. (1928): Die Säugetiere. Bd. II. Jena: Gustav Fischer. 


Anschrift der Verfasserin: EvELyn SCHLOSSER-STURM, Zoologisches Institut und Zoologisches 
Museum der Universität Hamburg, Martin-Luther-King-Platz 3, D-2000 
Hamburg 13 


BIUIGIERBIESIIRSEIGIERUFN ESSEN 


NIEVERGELT, B.: Ibexes in an African Environment. Ecology and Social System of the 
Walıia Ibex in the Simen Mountains, Ethiopia. Berlin-Heidelberg-New York: Springer 
1981. 189 S., 40 Abb. DM 89,-. 


Der Autor versucht eine Zusammenschau alles dessen, was er durch eigene Forschungstätigkeit (1968/ 
69 und zwei weitere Kurzaufenthalte) und aus den Beobachtungen der leitenden ‚Park Wardens“ 
(1971-77) zur Ökologie und Ethologie des Walia-Steinbockes erarbeiten konnte. Der Anlaß zur 
intensiven Beschäftigung mit diesem paläarktischen Ungulaten in einer ungewöhnlichen, afroalpinen 
Umgebung war vordringlich die Beschaffung des Grundlagenwissens zum Schutz und zur Erhaltung 
dieser inzwischen hochgefährdeten Tierart. Dabei mußten, wie aus der Darstellung vielfach deutlich 
erkennbar wird, erhebliche methodische Schwierigkeiten überwunden werden. Die faunistische 
Sonderstellung der Walia-Steinböcke regte aber auch zu mehr theoretischen Überlegungen zur 
Herkunft wie Geschichte, Einnischung und ethologischen Sonderentwicklung an. 

Die beiden Blickwinkel, das praktische Naturschutzinteresse und die methodisch theoretische 
Aufarbeitung wie Interpretation der Beobachtungen, lassen sich sicher nicht ganz nahtlos miteinander 
verbinden. Dennoch ist es schon etwas ungewöhnlich, daß es dem Autor notwendig erscheint, dem 
Leser einleitend eine „Gebrauchsanweisung‘“ geben zu müssen. Eine weniger auf die Behandlung von 
Einzelfragen konzentrierte, kontinuierlichere Darstellung hätte das wohl überflüssig gemacht. So muß 
man sich einige Zusammenhänge in mehreren Kapiteln zusammensuchen. 

Der angesprochene formale Mangel schmälert den sachlichen Gewinn nur sehr wenig. Zunächst 
erschließt sich mit der Schilderung der Simien-Berge ein wohl allgemein wenig bekannter, afrıkani- 
scher Lebensraum. Ein weiterer Schwerpunkt liegt bei der technischen Durchführung und der 
Auswertung der Steinbockbeobachtungen. Alle Daten wurden durch planvolle Beobachtung gewon- 
nen. Fine Markierung oder gar der Einsatz der Radiotelemetrie waren anscheinend nicht möglich. 
Erstaunlich ist, was bei gut durchdachtem Vorgehen auch mit einfachsten Mitteln möglich wird. 

Es wird soviel über die Habitatwahl, die Populationsstruktur und das Sozialverhalten der Walıa- 
Steinböcke und ihrer möglichen Konkurrenten, Klipspringer und Gelada, bekannt, daß ein recht 
umfassendes Lebensbild entsteht. Aus dem hier nicht lang ausgeführten Vergleich zwischen Walıa- 
und Alpensteinbock werden doch die aus den stark verschiedenen Klimabedingungen ableitbaren, 
divergierenden Entwicklungen sichtbar. 

Der abschließende Ausblick auf die Zukunft des Semien-Nationalparks und damit auch der 
außerhalb dieses Gebietes nur noch in geringsten Beständen vorkommenden Steinböcke zeigt Gefähr- 
dungen und Gefährdungsursachen auf, die auf dieser menschenüberfluteten Erde fast überall die 
gleichen sind. Die hier, wie bei manchem anderen ähnlichen Projekt, aus dem starken Engagement und 
dem tätigen Einsatz erstehende Hoffnung nötigt zur Anerkennung. H. HOERSCHELMANN, Hamburg 


CHAPMAN, J. A.; PuRSLEY, D. P. (eds.): Worldwide Furbearer Conference Proceedings. 
Vol. 1-3. Worldwide Furbearer Conference, Inc.: Baltimore 1981. 2056 pp. Total set 
US-$ 60,-. Zu beziehen durch: Worldwide Furbearer Conference, Inc., Book Distribution 
Center, 1111 E. Cold Spring Lane, Baltimore, Maryland 21239, USA. 


Vom 3. bis zum 11. August 1980 fand in Frostburg, Maryland, der 1. internationale Kongreß über 
Pelztiere statt. Die Referate sind in den vorliegenden Bänden für Interessenten veröffentlicht. 
Insgesamt 110 Einzelreferate sind 14 weitgefaßten Themenkreisen zugeordnet. Der Leser erhält 
vielfältige Informationen zur Biologie verschiedener Arten (Systematik, Zoogeographie, Evolution, 
Habitat, Physiologie, Morphologie, Reproduktion, Ernährung, Parasiten, Krankheit, Verhalten, 
Territorialität, Fluktuationen, Populationsdynamik) und darüber hinaus auch zu Management, Beja- 
gungstechnologie, toxischer Belastung, Schäden durch Pelztiere, Ökonomie, Geschichte des Pelzhan- 
dels in USA und sozialer Situation der Trapper. Die behandelten Arten sind in der überwiegenden 
Mehrzahl fissipede und pinnipede Carnivora sowie Rodentia, Lagomorpha und Marsupialia. Den drei 
Bänden sind viele interessante Ergebnisse an Arten aus mehreren Kontinenten zu entnehmen. Diese 
umfangreiche Dokumentation stellt eine willkommene Bereicherung dar. Sie kann allen Säugetier- 
kundlern empfohlen werden, zumal der Preis nicht zu hoch erscheint. D. Kruska, Hannover 


Neuerscheinung 


Beiträge zur Verdauungsphysiologie des Pferdes 


Contributions to Digestive Physiology of the Horse. Von HELMUT MEYER, unter Mitwir- 
kung von BERNHARD HUSKAMP, VERA GÜLDENHAUPT, GERD LINDEMANN, HEINKE Muus, 
MONIKA PFERDEKAMP, MICHAEL SCHMIDT, KARIN SCHWABENBAUER. Heft 13 der ‚‚Fort- 
schritte in der Tierphysiologie und Tierernährung“; Beihefte zur „Zeitschrift für Tierphy- 
siologie, Tierernährung und Futtermittelkunde“. 1982. 69 Seiten mit 20 Abbildungen und 
34 Tabellen. Kartoniert 40,- DM, für Bezieher der Zeitschrift 36,- DM 


Eine artgerechte Ernährung setzt entsprechende Kenntnisse der Verdauungsphysiologie 
voraus, wie sie zwar für das Rind und andere Wiederkäuer gegeben sind, für das Pferd 
bislang aber kaum verfügbar waren. Erst durch Untersuchungen mit Caecumfisteln ist es 
gelungen, die Verdauungsvorgänge im Dünn- und Dickdarm des Pferdes unter besonderer 
Berücksichtigung der Wechselwirkungen zwischen beiden Abschnitten besser zu erfassen. 
In Beiheft 13 wird darüber eingehend berichtet. 

Die Untersuchungen erstrecken sich im einzelnen auf Futterpassage im Dünndarm, 
Entleerungsmechanismen des Hüftdarms (Ileum), Bildung organischer Säuren und anderer 
Abbauprodukte im Blinddarm (Caecum) in Abhängigkeit von Futterart, Futtermenge und 
Fütterungstechnik; sie gelten besonders aber auch dem intestinalen Stickstoff-, Wasser- 
und Mineralstoffwechsel. 

Die Untersuchungsergebnisse erleichtern eine optimale Gestaltung von Futterrationen 
sowie eine sinnvolle Fütterungstechnik beim Pferd. Sie geben auch Hinweise auf die 
Entstehung der bei Pferden relativ häufigen Verdauungsstörungen und auf Möglichkeiten 
ihrer Vorbeuge. Darüber hinaus leisten sie einen wertvollen Beitrag zur vergleichenden 
Biologie der Verdauungssysteme bei Pflanzenfressern mit und ohne Vormagen. 


Neuauflage 
Tierproduktion 


Von ERNST GRANZ. 9., überarbeitete Auflage. 1982. 512 Seiten mit 207 Abbildungen, 
teilweise mehrfarbig und 119 Tabellen. Balacron broschiert 38,- DM 


Das bewährte Buch gibt eine gründliche Einführung in die landwirtschaftliche Tierproduk- 
tion. Ausgehend von den allgemeinen Grundlagen, der Anatomie und Physiologie des 
Tierkörpers, der Vererbung und Züchtung sowie der Tierernährung und Fütterung werden 
Erzeugung, Haltung und Vermarktung der einzelnen Nutztierarten ausführlich und unter 
verschiedenen Gesichtspunkten dargestellt, wobei nach Möglichkeit betriebs- und markt- 
wirtschaftliche Erkenntnisse im Vordergrund stehen. 

Neben neuen Produktionstechniken, verbesserten Methoden der Leistungsprüfung und 
Zuchtwertschätzung und rationellen Fütterungsverfahren berücksichtigt die nun bereits 
vorliegende 9., überarbeitete Auflage des Buches die Nettoenergie-Laktation, die neue 
Indexbewertung für Bullen, Eber und Schafböcke sowie die Güterverordnung für Milch 
und Änderungen im Futtermittelrecht. 


BEE NE PAUL PAREY : HAMBURG UND BERLIN 


Erscheinungsweise und Bezugspreis: Die Zeitschrift erscheint alle 2 Monate; 6 Hefte bilden einen Band 
mit insgesamt 384 Seiten. Der Abonnementspreis beträgt je Band 232,- DM zuzüglich Porto. Das 
Abonnement verpflichtet zur Abnahme eines ganzen Bandes. Es verlängert sich stillschweigend, 
wenn nicht unmittelbar nach Erhalt des letzten Heftes eines Bandes Abbestellung erfolgt. 
Einzelbezugspreis der Hefte: 43,- DM. Die Preise verstehen sich im Inland incl. Mehrwertsteuer. 
Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, 
Spitalerstraße 12, D-2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesell- 
schaft für Säugetierkunde“ erhalten die Zeitschrift unberechnet im Rahmen des Mitgliedsbeitrages. 


Die umfassende 
onographie 
as Schwarzwi 


Über 


ieses Werk schließt 
eine Lücke in der Jagd- 
literatur, denn es umfaßt 
alle Aspekte, die für den 
Jäger heute von Bedeutung 
sind: Wildschweinzoologie, das 
heimische Schwarzwild, Sau- 
jagden in vergangenen Zeiten, 
Sauparks, heutige Jagdarten, 
Schwarzwildhege und Be- 
standsregulierung, Altersschät- 
zung, Trophäen, Waffen- und 
Kaliberfrage, Brauchtum und 
jagdwirtschaftliche Bedeutung 
des Schwarzwildes. 
Die Monographie ist mit vielen 
Photos hervorragend ausge- 
stattet. Geschrieben wurde sie 
von zwei renommierten Ken- 
nern: Lutz Heck, international 
engagierter Jäger, Verfasser 
zahlreicher Jagdbücher, Zoo- 
loge, Wildbiologe und ehemali- 
ger Direktor des Zoologischen 
Gartens in Berlin; Günther 
Raschke, Forstamtsleiter in 
einem der besten Schwarz- 
wildgebiete Deutschlands und 
vorzüglicher Spezialist in der 
praktischen Wildstandsbewvirt- 
schaftung. 


Lutz Heck/Günther Raschke 
Die Wildsauen _ 
Naturgeschichte, Okologie, 
Hege und Jagd. 1980. 216 Sei- 
ten mit 134 Einzeldarstellungen 
in 90 Abbildungen, davon 

14 farbig, und 16 Tabellen. 
Kunstdruckpapier. Leinen 
58,- DM 


Zu beziehen 
durch Buchhandlungen 
und Fachgeschäfte 


Verlagsbuchhandlung 
Pau! Parey 

Postfach 10 63 04 
2000 Hamburg 1 


PAUL 
P/REY 


Vol. 47 (5), ne Oktober 1982 ISSN 0044-3468 C21274F 
- Mammals 


ZEITSCHRIFT FÜR 


Irgan der Deutschen Gesellschaft für Säugetierkunde 


| Je Jong, W. W.; Goodman, M.: Mammalian phylogeny studied by sequence analysis of the eye lens protein a- 
crystallin. — Untersuchung der Stammesgeschichte der Säugetiere durch Sequenzanalysen des Augenlinsen- 
proteins a-Kristallin 


ohannesson-Gross, Kristina; Gross, H.: Lernversuche mit Maulwürfen (Talpa europaeaL.) unter Anwendung einer 
speziellen Labyrintnmethode. — Learning experiments with moles (Talpa europaea L.) applying a special maze 
method 

Vendland, V.: Häufigkeitsschwankungen der Rötelmaus (Clethrionomys glareolus Schreber, 1780). — Fluctuations 

, of Bank vole (Clethrionomys glareolus Schreber, 1780) 


| leal, B. R.: Reproductive biology of three species of gerbils (Genus Tatera) in East Africa. — Fortpflanzungsbiologie 
von drei Gerbilliden-Arten (Gattung Tatera) in Ostafrika 


lelson, B. B.; Chapman, J. A.: Age determination and population characteristics of red foxes from Maryland. — 
Altersbestimmung und Populationsmerkmale beim Rotfuchs in Maryland 


eldhamer, G. A.; Dixon, K. R.; Chapman, J. A.: Home range and movement of sika deer (Cervus nippon) in 
Maryland. — Reviergröße und Ortswechsel beim Sika-Hirsch (Cervus nippon) in Maryland 
Vissenschaftliche Kurzmitteilungen 


‘ock, D.; Kinzelbach, R.: Der Dachs, Meles meles (Linnaeus, 1758), in NW-Syrien. — The badger, Meles meles 
(Linnaeus, 1758), in NW-Syria 


amperl, Roswitha: Tandem-Fusion als chromosomaler Evolutionsmechanismus bei Microtus agrestis (Rodentia, 
| Microtinae). - Tandem fusion in chromosomal evolution of Microtus agrestis (Rodentia, Microtinae) 


257 


DTM 


283 


287 


296 


311 


316 


317 


HERAUSGEBER/EDITORS 


P. J. H. van BrEE, Amsterdam - W. FIEDLER, Wien — H. Frick, München - W. HERRE, 

Kiel - H.-G. Kıös, Berlin - H.-J. Kunn, Göttingen - B. Lanza, Florenz - T. C. S. 

MORRISON-SCOTT, London - J. NIETHAMMER, Bonn - H. REICHSTEIN, Kiel - M. RöHRrs, 

Hannover - D. STArRcK, Frankfurt a. M. -— F. Strauss, Bern — E. THENIUS, Wien — 
W.VERHEYEN, Antwerpen 


SCHRIETLEITUNG/EDITORTATL-OFEIGE 


H. SCHLIEMANnN, Hamburg - D. Kruska, Kiel 


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. Harald Schliemann. 


Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z.Hd. Prof. Dr. Dieter 
Kruska, Institut für Haustierkunde, Biologie-Zentrum, Neue Universität, Olshausenstr. 40-60, 
D-2300 Kiel. Für die Publikation vorgesehene Manuskripte sollen gemäß den „Redaktionellen 
Richtlinien“ abgefaßt werden. Diese Richtlinien sind in deutscher Sprache Bd. 43, H. 1 und in 
englischer Sprache Bd. 43, H. 2 beigefügt; in ihnen finden sich weitere Hinweise zur Annahme von 
Manuskripten, Bedingungen für die Veröffentlichung 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 verötfenlichen 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 
Vervielfältigung auf photomechanischem oder ähnlichem Wege oder im Magnettonverfahren 
sowie der Speicherung ın Datenverarbeitungsanlagen, bleiben, auch bei nur auszugsweiser Ver- 
wertung, vorbehalten. Werden von einzelnen Beiträgen oder Beitragsteilen einzelne Vervielfälti- 
gungsstücke in dem nach $54 Absatz 1 UrhG zulässigen Umfang für gewerbliche Zwecke 
hergestellt, ist dafür eine Vergütung gemäß den gleichlautenden Gesamtverträgen zwischen der 
Verwertungsgesellschaft Wort, vereinigt mit der Verwertungsgesellschaft Wissenschaft GmbH, 
rechtsfähiger Verein kraft Verleihung, Goethestr. 49, D-8000 München 2, und dem Bundesver- 
band der Deutschen Industrie e.V., dem Gesamtverband der Versicherungswirtschaft e.V., dem 
Bundesverband deutscher Banken e.V., dem Deutschen Sparkassen- und Giroverband und dem 
Verband der Privaten Bausparkassen e.V., an die Verwertungsgesellschaft zu entrichten. Erfolgt 
die Entrichtung der Gebühren durch Wertmarken der Verwertungsgesellschaft, so ist für jedes 
vervielfältigte Blatt eine Marke im Werte von 0,40 DM zu verwenden. Die Vervielfältigungen sind 
mit einem Vermerk über die Quelle und den Vervielfältiger zu versehen. 


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 internal use 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 


© 1982 Verlag Paul Parey, Hamburg und Berlin. — Printed in Germany by Westholsteinische Verlagsdruckerei 
Boyens & Co., Heide/Holst. 


Z. Säugetierkunde 47 (1982) 5, 257-320 


InterCode: ZSAEA 7 47 (5), 257-320 (1982) 


Mit 3 Beilagen des Verlages Paul Parey. 


Mammalian phylogeny studied by sequence analysis 
of the eye lens protein a-crystallin 


W.W. DE Jong and M. GOODMAN 


Receipt of Ms. 3. 2. 1982 
Abstract 


Analyzed the amıno acid sequences of the eye lens protein a-crystallin in 41 mammalian species with 
the aım to resolve phylogenetic relationships among mammalian orders. The species represented 17 
orders of mammals. Chicken and frog (Rana esculenta) were included as outgroups. The observed 
amino acıd differences were used to construct cladograms, either solely on the basis of the lowest 
numbers of required nucleotide replacements in the DNA, or also taking into account certain well- 
established phylogenetic relationships. The a-crystallin A sequences indicate that: the paenungulate 
orders Proboscidea, Hyracoıdea and Sirenia are a monophyletic group to which the Tubulidentata 
(aardvarks) also belong; the paenungulates are not related to the ungulates, but together with the 
edentates represent the oldest offshoots of the eutherian stem; the pangolins show no relationship with 
edentates and are most parsimoniously placed close to the carnivores; the ungulates, whales, and 
carnivores form a monophyletic grouping; among the carnivores the seals and sea lions are monophy- 
letic; the investigated bat (Microchiroptera) appeared not to be related to insectivores or primates. The 
a-crystallin A sequences left the rodents, lagomorphs, insectivores, primates and Tupaia as an 
unresolved cluster of orders, but within the primates the prosimians are clearly set apart from the 
Anthropoidea. The results are compared with current opınions about mammalian phylogeny and 
related to other comparative protein sequence data. 


Introduction 


Evolutionary trees can be constructed from genealogical analyses carried out by the 
parsımony method on the amıno acıd sequences of homologous proteins of different 
organisms. Such trees are capable of offering important insights on phylogeny despite the 
limitations of current algorıthms (DavyHorr 1972; PEAcock and BouULTER 1975; FITCH 
1977, HEnDy et al. 1978; GooDMman et al. 1979; FITCH 1979; GOODMAN 1981). Previously, 
an analysis of the amino acid sequences of the A chain of the eye lens protein a-crystallin 
from 17 mammalian species showed its usefulness in the study of mammalian phylogeny, 
especially at the ordinal level (DE Jonc et al. 1977). We now have extended the number of 
mammalian species from which a-crystallin A sequences have been determined to 41. The 
aim was to obtain information about the phylogenetic affınities among mammalian orders 
and about certain phylogenetic relationships within orders. Representatives from all 
eutherian orders, except Dermoptera, and from many of the major subordinal groups have 
therefore been studied. A considerable number of species whose phylogenetic affınities are 
not at stake, has been included to provide a framework in which to place the phylogenetic 
problem cases. 

Evolutionary trees have been constructed from the aA sequence data using the chicken 
and frog Rana esculenta as outgroups. Our major conclusions are based on the most 
parsimonious trees found in a computer search which was not constrained in any way by 
evidence from other sources on the phylogenetic relationships of the 41 species represented 
by the sequences. This search was guided solely by those changes in branching arrange- 
ment of the trees which most lowered the number of nucleotide replacements (1.e. point 
mutations) required for the descent of the contemporary sequences from a common 
ancestor. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4705-0257 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 257-276 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


258 W. W. de Jong and M. Goodman 


Certain peculiar branching arrangements ın the lowest nucleotide replacement (NR) 
length trees for the aA sequences, such as the grouping of marsupials with chicken rather 
than other mammals, do not agree with the results obtained in lowest NR length trees 
constructed for other proteins or with classıcal zoological evidence. To resolve this 
dilemma, we have used a newly developed parsımony procedure (GOODMAan et al. 1979), 
which incorporates phylogenetic evidence from other proteins or from classical taxonomic 
sources to decide whether the branching arrangement of a set of homologous sequences (in 
this case, aA sequences) should be made concordant with other phylogenetic evidence on 
the species represented by the sequences or should show gene duplications in some regions 
of the phylogeny. Obviously we only incorporated in the decision making process 
concepts on phylogenetic relationships which we considered firmly established such as the 
monophyly of therian mammals (in the species phylogeny marsupials should group with 
other mammals not with a chicken). Since almost nothing was assumed about mammalıan 
relationships at such higher taxonomic levels as the interordinal, the parsimonious trees 
found by constraining the computer search with a limited number of a priori assumptions, 
lead to phylogenetic conclusions which were relatively independent of any prior bias. Our 
findings indicate that such a synthesis of classical taxonomy and amino acıd sequence data 
may contribute to the eventual elucidation of the most probable course of events in 
mammalıan evolution. 


Material and method 
o-erystallin A sequences 


a-crystallin is a water-soluble structural protein which occurs exclusively in the epithelial and fiber 
cells of the vertebrate eye lens (for a review see BLOEMENDAL 1981). It makes up a variable proportion 
of the total lens protein; the amount depends on the species and age of the animal. In many species 
a-erystallin constitutes 25 to 50 % of the total lens protein. a-crystallin forms large aggregates, of 
average molecular weight from 400,000 to 800,000, and is composed of two types of chains, aA and 
aB. The ratio of aA to aB chains varies between species, ranging from 9:1 in the kangaroo to 1:4 ın 
the spiny dogfish. The aA and aB chains of the ox show 58 % homology between their amıno acıd 
sequences, thus reflecting an ancient common ancestry of their genes. 

a-crystallin can easily be obtained in considerable quantities from most vertebrate species, and the 
amino acid sequence of the aA chain, 173 residues long in most species, is relatively simple to 
establish. The procedures involved in the isolation and sequence determination of the aA chaıns are 
described in DE Jong and TERWINDT (1976). In most cases aA chains were isolated from pooled lenses 
from different specimens of the same species. Apart from hyrax aA chain (where both alanıne and 
threonine were found at position 55), no polymorphisms were ever detected in the amıno acıd 
sequences of mammalıan aA chains. The amino acid sequences of bovine, kangaroo, chicken and frog 
aA chains have been completely established by identifying all residues in these chains by the Edman- 
degradation method. Because only 10 % of the residues in the sequences of bovine and kangaroo aA 
chains are different, a simplified procedure has been used to study other mammalıan aA chaıns. Their 
sequences have largely been deduced from the amino acid composition of small peptides obtained by 
enzymatic and chemical cleavage of the chains. When the amıno acid composition of such peptides 
were found to be the same as those from the corresponding peptides of bovine or kangaroo aA chain, 
it was assumed that their sequences were also identical. When a difference in composition was found, 
such a peptide was usually subjected to Edman-degradation in order to confirm the position and type 
of the underlaying substitution. 

Using this approach there is a risk of overlooking double substitutions which change the sequence 
but not the composition of a peptide. It has been established, however, that the risk of overlooking 
such reciprocal substitutions ıs extremely small if the analyzed peptides are small (van DRUTEN etal. 
1978). 

The choice of species to investigate was directed by their phylogenetic relevance; species were 
selected either because they pose particular taxonomic problems or to increase the denseness of the 
phylogenetic tree in appropriate places. An important limiting factor was obviously the availability of 
the desired lens material. For this reason several interesting taxa have not yet been studied. The names 
and sources of all mammalian species of which the aA chain sequences have now been determined are 
given in Table 1. As already mentioned, 17 of these sequences had been employed in a previous study 
(DE Jong et al. 1977). Details of the chemical determination of the 24 new mammalıan aA sequences 
added to the present study are being described elsewhere (DE Jong et al. in preparation). Certain 


Mammalian phylogeny and protein sequence analysıs 259 


findings concerning trends ın amıno acıd substitutions in the reconstructed phylogeny of mammalian 
aA sequences have already been reported (DE Jon et al. 1980). 

The observed differences between aA chain sequences are summarized in Table 2. From inspection 
of the sequence differences it is easy to identify at certain positions residues which apparently are 
ancestral, primitive ones, and others which are derived ones. For example, at position 3 Ile occurs both 
in the outgroups (frog and chicken) and ın the marsupials, whereas Val only occurs in several eutherian 
orders. It thus seems likely that at this position Ile ıs the primitive and Val the derived character. 
Similarly, at position 4 Thr seems to be primitive and Ala derived. Also some shared derived 
(synapomorphous) amino acıd substitutions in certain species can easily be recognized, as for instance 
13 Pro and 61 Val in the prosimians lemur, potto and galago. Table 2 also shows, however, that certain 
substitutions, such as 55 Ser and 61 Val, must have occurred more than once in entirely unrelated taxa, 
and that back substitutions, for instance 101 Asn— Ser, may complicate the interpretation. Because 
convergent substitutions and back substitutions occur frequently, it is obvious that rigourous 
computer handling of the data is required to assess objectively the numerous possible branching 
Patterns. 


Construction of cladograms 


A maximum parsimony approach (Moore et al. 1973; Goopman et al. 1979) was employed to 
construct cladograms for the 43 a-crystallin A chain sequences. In this approach the contemporary 
amino acid sequences, ı.e. the operational taxonomic units (OTUs), are mapped, through the genetic 
code, into corresponding messenger ribonucleic acid (mRNA) sequences. The object then is to find an 
ancestral order of branching and ancestral mRNA sequences which account for the descent of the 
OTUs by the fewest possible nucleotide replacements (NRs). Such a parsimony tree maximizes the 
number of nucleotide identities among descendant sequences ascribable to shared common ancestry 
rather than to convergence or parallelisms and back mutations. 

As indicated in the Introduction, it has been found (e.g. GooDman et al. 1979; MAEDA and FITcH 
1981) that the trees with the fewest NRs constructed from different proteins can yield non-concordant 
branching arrangements for the same anımal species and thus violate in each tree some of the features 
of the anımal phylogeny strongly supported by the evidence from other proteins. Such violations 
could be indicative of incorrect groupings of sequences that happen to have an excess of convergent 
resıdues. Alternatively, such violations could be due to real differences in the branching arrangement 
between the gene phylogeny and the species phylogeny. This latter possibility indeed opens the way 
for the construction of more accurate genealogical trees by an extension of the parsimony criterion. 
Not only are base replacements counted, but also the additional cost in gene duplication (GD) and 
gene expression (GE) events that must be assumed to fit the putative gene phylogeny into well 
established features of the species phylogeny. The object is to minimize the total NR+ GD+ GE 
length. (Computer algorıthms for counting the number of GDs and GES are described in Appendix A- 
2 and A-3 of GooDman et al. 1979.) 


Lowest NR length trees 


For a set of contemporary amıno acid sequences the only sure way to find the tree or trees of lowest 
NR length is by examining all unrooted trees that the OTUs can possibly form and then choosing the 
tree or trees with the lowest score. Unfortunately this method is prohibitive in computer time when 
there are more than 8 or 9 OTUs in the data set. For larger sets of data, such as that employed in the 
present study, heuristic approaches can be used which limit the search procedure to practical 
dimensions. 

We started the search by calculating a matrix of minimum mutation distances for the 43 a- 
crystallin A sequences by the method of FrrcH and MARGOLIASH (1967). Then two initial dendrog- 
rams were constructed from this matrix, the distance Wagner tree by the method of Farrıs (1972) and 
the unweighted pair group tree by the clustering algorıthm of SoKkAL and MICHENER (1958). With the 
initial dendrograms and with the computer file of a-crystallin A sequences, we employed the 
maximum parsimony branch swapping algorithm described in Appendix A-1 of GooDman et al. 
(1979) to search for the lowest NR length trees. In order to test a wide range of phylogenetic 
possibilities the search was continued using 33 further starting dendrograms, from which lıterally 
thousands of alternative dendrograms were examined in the progression of branch swaps. The trees 
found with the lowest NR score each required 152 NRs. Representative examples of the major 
different branching arrangements exhibited by these trees are shown in Figs 1-3. In a few local regions 
of each of these three trees minor changes in the branching arrangement also yield trees of the lowest 
NR length, 152 NRs. 

The lowest NR length tree shown in Fig. 1 differs from the distance Wagner tree only in the 
position of the bear (with the whale-porpoise branch in the Wagner tree and with the pangolın in 
Fig. 1) and in containing one less NR. The distance Wagner tree itself costs 153 NRs. The NR score of 
this Wagner tree could not be lowered when the Wagner tree itself was the starting dendrogram, ı.e. 


uoBawilik; jo Alufj souifior,| jewiuy [eis ) 
usdauılın] Fo 'Arun ‘sanıpoe,] jewruy [erusd 


urey we Jınpyueig anınsuf Sffpneg “AaIvAHOS 'y Iq 
us3aurlın] punose sarypenses orypen “sSyaLag "A A 
9uoZ [euren eweueg ur '[ aa’ 'M Ag P9R23]J09 


"yn S3pııquen ‘430[007 Jo wnasny 'Atun ‘AasAo[ "y 'Iq 
wepisjsuny Jo 'Aruf) “[OOZ "UOXe], su] “aaug NVA 'd IA 


eILIONTED “O3I1Lq UrS “PLIOMEIS “TTANYOD "T IA 

"yPN ‘wayumy “aysgqınnıeN 100A InnIsursyerg “NALavvVd NVA '[ IQ 
usgawlın] Fo 'Atuf) ‘sone jewruy Jenmusg 

weplsjsuny Jo 'Atun “°JOOZ "UoXxe], 'ısu] ‘aaag NVA 'J IAq 

usa3awfın] Fo 'Arun) ‘sonmıpoe,g jewruy Jemusdg 

us3awlın] Fo "Atun ‘sanıpoeg jewruy Jemusdg 


jpeas] *eAays-199g “A93aN >yı Jo "Au “"IOVA N A 
us3aulın] Fo 'Atun ‘sone, jewruy fenusg 

wep4sjsumy JO 'Atufn) “°JOOZ "UOXe], 'Isu] “aaag NVA 'J IA 
"yPN Yoaquazeayıfy “usdıag 9sy99g Yıedııeyes 

usdaulın] Fo 'Aruf) ‘sanıpie,] jewruy fenmusdg 


esLiFV YINOS “VAIISIY JWEH JAnjynId ‘aaa "W "IW 
wep13120% ‘007 dıopliig ‘waog aa "T ıq 
"YPN Aysann Psnoyasıydnegs fedistunyy 


uepı1alsury jo 'Arun ‘A30]007 'UOXe], 'IsuJ “aayg NVA 'q 'Iq pue 
“yaoN “uUapI97 “r1osırq aylıpınnyaen ueA wnasnusyiog ANaansS ID 'ıq 
ÄeMION ‘usdlag “y9ıeasay auLIeN] Fo JınınsuJ “NASNALSIAHO "I 'IW 


07 


N ee 


ın ıNn 


um oa ee semrar 


S33J3DyUcH UIBIUT) 


ey 

MOIUSS91] 

3oya3pay ueadoany 

3eq 3une>-AnAF} ueoreuref 


‚urpodued 991], 
umo3ued uekefeW 


uol] e9s eIUIOFeY 
jess Aeı9, 
yurw uesLIauy 


Je9q YIOIS 
ıe7 
3oq 


Arep9woılg] 
XO 
snuregododdıry 


preitg 
31 


SOIHIOUTYI UA 
aıdeı ueAepeW 
9SI0H 


astodıod uowwoy 


ofeyMa SyUIM 


Fr ae Sarnen. ee 
191907849121 Ra111.SANOILATAT 
rıjnmarin IIELOSII DIESEN YV 


SHABINAOU SN11PY 
enuapoy 

1uagurjaq vırdn], 
enuapueag 

snovdo4na sna9vu117 
BIOATIIOSUJ 


sısuasıvıvl snoguyAy 
eısıdonyg 


sıdsnn4 (snudvıvg]) stur 

vauravl sıupy 
GOpASUd 

snurmusofivs snqdojvz 

sndd43 snı0ogqa1P 7 

uosıa vjoIsny 

snuisan sSnsÄnJaW 

sn4v3 5194] 

sıavımmo[ sıuv) 
eIOATUJED 

smiivpawoıp snjawv) 

snınv Sog 

smiqigduv snwojododdır] 

sıypparvdojawuv> vffpun 

v[045 snS 
ep419eponıvy 

wunwıs wn14941010495 

snaıpun snAdv ] 

snyjpgv> sunb] 
ep419epossuIiag 


vuov9oqd vuav9ogg 
vv4150401n9P vAsıdouarvg 
E99 


391n0g 


s9su9] Jo JaqumN 


9WeU UOWWOT 


9Wweu S1FNU9TDS 


jersoyeun SU9] pare3nsaAur 9yJ Jo s9IA1NOos pue paıpnys u99g 9ALU s3>9uanbas V urpje3sA.19-0 9y) yIyM WOAJ sanads Jo SSWEN 


TAGE 


usdauılın] Fo 'Aruy) ‘sanıpme,g jewnuy jenusy 
usdauılın] Fo 'Aruy) ‘“sanıpe,g jewruy [enusy 


UOZ21Q “purpLlog 
“u03319 Fo 'arun “Ansıursysorg Jo ıdaq “TaZNaLs 'd Ida 
eıpensn\y ‘I9Juny [eIsI9WuIo9 


eureueg “ısuf 'soy "dos] ueruosyirus ANaWOSLNOW I IQ 
auoz [eure eweueg ur '[ aa X, 'A Äq P3129][09 
auoz [eue)) eweueg ur 'f aa’, X fq P2129][09 


esLlyy yınoS UT P9399][09 (MONI) INVHSOHS '[ 'ıyı pue 
“wepisisuny Jo 'Atuf) “°JOOZ "UOXe] 'IsuJ (aaag NVA 'd IA 


zeig ‘sneueg “eruozeury ep 'bsaq ‘sen su] ‘ısag 'y IQ 
estıyy Yanos ‘yıeg JeuoneN Jodnıy ‘soy aa 'A IA 

esLıyy yanog ‘yıeg JeuoneN Jodnıy “avvnalg 'Aaa'N aa 
JuraIpaW Fo [OOY>S usdawulın Fo 'Arum ‘Auroreuy Jo ıdaq 
usdauılın] Fo "Arup) ‘sanıpiseg jewruy [enusd 

umısjog ‘9317 Jo 'Atuf) ‘LNVII09 'W 'JOld 


"D’N ‘weyang] ‘ArfrIe,] Jeumıg and “NOSUaaNnY 'q 'ıW 
D’N ‘weymg ‘Aıfeg ewıg 9ynq, “NOosaaany 'q 'IW 


"opeur us9q 194 10U pey avU0S59] ‘y pue xa]du1o9 pjua]nDsd pvuvy yl U99AI9A 


OSI 


007 


301] 


uaypıy) 


umnssodo 
uesLI9WwNy YLIOoN 
ooJeduey pay 


Jeaq ıuy 
Y20JS P901-39J1U ] 


y2oIs P902-OM], 
NIeAPIeY 
9areuew ueimtzeig 
xeıAy ade) 
yueydaja uesLupy 
ueunp 

Aayuouı snsayy 
onod 


odefed) 


InulIo] UMOIG 


uonsunsıp ay3 s3017 9539y1 Jo UONIS]JO9 Jo Jun aylıy „ — "Poutuntsgsp uaag Ajfenred Auo sey urodued aan ayı Jo ureys yo ayı jo aouanbas pi>e ourure ayy, : 


„yruajn9sa vury 
eıqrydury 

snjv3 sap 
SOAY 


sıpıdnsıvu sıydjapıq 
sn/nı sndosvw 
eıperdnsseyy 


vuvaxau vnpuvuv] 

snivBaııva sndApvig 

zuuvuf[og sndaojogqd 
ereuap 


aafo sSndo1aBK4O 
ereruspipngqn], 

sındunun snq9agqMAaL 
erualıg 

sısuadv3 viavI0ALL 
eaprosesAH 

vuvsııo v1uU0p0xo7 
eapI9sogol] 

suaıdvs owoy 

vaınu v3vV3ovW 

03104 sn21121p042J 

snippnv21ssvo 030109 

snayn[ anwaT 
SOIEUILIT_ 


EEE DEE a TEE 


ERBETEN Eee eR genen 2 ro Set Fans rn SS enernnneh ——n 0 2 ren 


Table I 


Names of species from which the a-crystallin A sequences have been studied, and sources of the investigated lens material 


Scientific name Common name Number ol lenses Source 
Ceracea 
Balaenoptera acutorostrata Minke whale 5 Mr. I. Christunsen, Institute of Marine Research, Bergen, Norway 
Phocaena phocaena Common porpoise 5 Dr. ©. SmEEnK, Rijksmuseum van Natuvrlijke Historie, Leiden, Neth., 
and Dr. P. van Baer, Inst. Taxon. Zoology, Univ. of Amsterdam 
Perissodactyla 
Equus caballus Horse 19 Municipal slaughterhouse, Utrecht, Neth. 
Tapirus indicus Malayan tapir 2 Dr. L. ve Born, Blijdorp Zoo, Rotterdam 
Ceratotherium simum White rhinoceros 6 Mr. M. Keep, Hluhluwe Game Reserve, South Africa 
Artiodactyla 
Sus scrofa Pig so Central Animal Facilities, Univ. of Nijmegen 
Giraffa camelopardalis Giraffe 2 Safaripark Beekse Bergen, Hilvarenbeek, Neth. 
Hippopotamus amphibins Hippopotamus 2 Dr. P, van Brex, Inst. Taxon. Zool., Univ. of Amsterdam 
Bos taurns Ox many Central Animal Facilities, Univ. of Nijmegen 
Camelus dromedarins Dromedary 2 Dr. R. Yacıı, Univ. of the Negev, Beer-Sheva, Israel 
Carnivora 
Canis familiaris Dog 10 Central Animal Facilities, Univ. of Nijmegen 
Felis catus Cat 8 Central Animal Facilities, Univ. of Nijmegen 
Melursus ursinus Sloth bear 2 Dr. P. van Brex, Inst. Taxon. Zool., Univ. of Amsterdam 
Mustela vıson American mink 50 Central Animal Facilities, Univ. of Nijmegen 
Halichoerus grypus Gray seal 3 Dr. J. van Haarten, Rüjksinstituut voor Natuurbeheer, Arnhem, Neth. 
Zalophus calıfornianus California sea lion 2 Dr. L. Corneuı, Seaworld, San Diego, California 
Pholidora { 
Manis Javanica Malayan pangolin 4 Dr. P. van BRe£, Inst. Taxon. Zool., Univ. of Amsterdam 
Manis (Phataginus) tricuspis Tree pangolin! 4 Dr. K. Jovsev, Univ. Museum of Zoology, Cambridge, U.K. 
Chiroptera 
Artıbeus jamaicensis Jamaican fruit-eating bat 20 collected by W. W. ve J. in Panama Canal Zone 
Insectivora 
Erinaceus europaens European hedgehog *2 Dr. W. Perexs, traffic casualties around Nijmegen 
Scandentia 
Tupaia belangeri Treeshrew 20 Dr. A. Schwater, Battelle Institut, Frankfurt am Main 
Rodentia 
Rattus norvegicus Rat 100 Central Animal Facilities, Univ. of Nijmegen 
Mesocricetus anratus Golden hamster 100 Central Animal Facilities, Univ. of Nijmegen 
Meriones unguiculatus Mongolian gerbil 27 Central Animal Facilities, Univ. of Nijmegen 
TOIEHSLonaTprINEeBs TRETEN BI mg eseeat by DI MTMEREN eBloradeT > = we - 
Oryetolagus euniculus Rabbit 12 local butcher, Nijmegen 
Primates 
Lemur fuluns Brown lemur 6 Mr. D. Anperson, Duke Primate Facility, Durham, N.C. 
Galago crassicandatus Galago 4 Mr. D. Anperson, Duke Primate Facility, Durham, N.C. 
Perodicticus potto Porto 4 Prof. M. GorrART, Univ. of Liege, Belgium 
Macaca mulatta Rhesus monkey 10 Central Animal Facilities, Univ. of Nijmegen 
Homo sapiens Human 14 Dept. of Anatomy, Univ. of Nijmegen School of Medicine 
Proboscidea 
Loxodonta africana African elephant 11 Dr. U. pe V. Pıenaar, Kruger National Park, South Africa 
Hyracoidea 
si Procavia capensis Cape hyrax 30 Dr. V. pe Vos, Kruger National Park, South Africa 
irenia 
Trichechus inunguis Brazilian manatee 6 Dr. R. Best, Inst. Nac. Pesq, da Amazonia, Manaus, Brazil 
Tubulidentata 
Orycteropus afer Aardvark 3 Dr. P. van Bret, Inst. Taxon, Zool., Univ. of Amsterdam, 
and Mr. J. SHosHanı (Detroit) collected in South Africa 
Edentata 
Choloepus hoffmanni Two-tocd sloth 12 collected by W. W. oe J. in Panama Canal Zone 
Bradypus varıgatıs Three-toed sloth 16 collected by W. W. ne J. in Panama Canal Zone 
amandna mexicana Ant bear 2 Dr. G. MontGomerr, Smithsonian Trop. Res. Inst., Panama 
Marsupialia 
Macropus rufus Red kangaroo 120 commercial hunter, Australia 
Didelphis marsupialis North American 3 Dr. P. Srenzeu, Dept. of Biochemistry, Univ. of Oregon, 
opossum Portland, Oregon 
Aves 
Gallus gallus Chicken 200 Central Animal Facilities, Univ. of Nijmegen 
‚Amphibia 
Rana esculenta® Frog 150 Central Animal Facilities, Univ. of Nijmegen 


! The amino acid sequence of the aA chain of the tree pangolin has only partially been determined. - ? Atthe time of collection of these frogs the distinction 
between the Rana esculenta complex and R. lessonae had not yet been made. 


262 W. W. de Jong and M. Goodman 


Table 2 
Substitutions among the a-crystallin A chains of 41 mammalian species, chicken (Gallus gallus) and 
frog (Rana esculenta) 


Only those positions in the aA chains are shown at which substitutions have been found in at least two 

species. References to the sequence analyses of the aA chains of whale, horse, rhinoceros, pig, ox, dog, cat, 

hedgehog, treeshrew, rat, rabbit, monkey, human, elephant, hyrax, kangaroo and opossum are given in 
DE Jong et al. (1977). The other sequence determinations will be described elsewhere 


(DE Jong et al., in prep.). 
1 1 
e 5 
E S 
a 
S 
2 
1 
+ 
N 
G 
D L 
DSTL 


G 


POSITION NR 347 


MINKE WHALE IAH 
PORPOISE 

HORSE 

TAPIR 

RHINOCEROS 

PIG 

GIRAFFE, HIPPOPOTAMUS 
0X 

CAMEL 

DOG, CAT 

BEAR Y 
MINK Q M V 
SEAL, SEA LION PaVES 
PANGOLIN Y 
BAT 

HEDGEHOG 

TUPAIA 

RAT, HAMSTER, GERBIL 
GUINEA PIG, SPRINGHAAS 
PIKA 

RABBIT 

LEMUR 

GALAGO, POTTO 

RHESUS MONKEY 

HUMAN 

ELEPHANT 

HYRAX 

MANATEE 

AARDVARK 

2-TOED SLOTH 

3-TOED SLOTH 

TAMANDUA 

KANGAROO 

OPOSSUM 


CHICKEN 
FROG 


Ho 
zZ or 
N nn 
zZ wre 
nr one 
N nen 
N or- 
DM vurr 
zz our 
DD wor 
wN vor 
> ou 


nm sw 

nn u 

nm nu 

< on 

oO ou 

RR on 

< nn 

7 

< 

De] 

m 

[7 

OD one 
44-4 02 sn 

> nur 


ana 
670 
PIRES 


<< —T wwr 
— TU SER 
D—wn on 


nn ou 


<< 


1 
3 
A 
5 
T 
T 


a dd  — — 0fFH 


aD — 


errerrrrePre 


Prrrtrtrtrorrorororm<e<<<—_<<<<<—rm 


44-4090 
<< 
<< 


4-4 
4-4 


(le 
ı 0000000000000 D>o 


<< u << << 
me Wmm 


4-4 -44--------------"---- - 
v —NYNNN9D9D 


<— r — re 


ZZ ı 


2— X 90000 
OO OooOoOoOo9oO0000000 
—: 2 


ww wo 
nm wwwnww— nmuno ı 


IS een 
q1—- HH70Ho4o Ho [ln 
=2— j 

' 

ı 

<<< 

NO m— OO 
OO wmwn-09- 


4-4 4-4 


SlL 
Sk 
5 


a nn mn DR n ] 
— 4-1 
vo zOoı 


mm 


S 
NFIFGF 


The one-letter notation for amino acids has been used: A = alanıne; C = cysteine; D = aspartic acıd; 
E = glutamic acid; F = phenylalanıne; G = glycine; H = histidine; I= isoleucine; K = lysine; 
L = leucine; M = methionine; N = asparagine; P = proline; Q = glutamine; R = arginine; S = se- 
rine; T = threonine; V = valine; Y =tyrosine; (-) means deletion and (-) not determined. - a) 
Substitutions observed in only a single species: hedgehog 23 F> L, 154 H > P; rhesus monkey 162 
S— A; manatee 19 PH, 167 P> A; aardvark 84 D-> E, 170 A > V; two-toed sloth 11K->R; 
three-toed sloth 34 Y > S; opossum 151 D>E;chicken 8 Y> 1,32 F>L,39 F>L,40 L-F, 
86 1T > S,130S->T, 135 A > S; frog 26 F— V, 31 L—M,33 E—D,124V-L,1253D-N, 128 
A-S,138M-1. 


the single step changes in branch arrangement at adjacent nodes made by the branch swapping 
algorithm failed to find a tree of lower NR score. We noticed, however, that bear and pangolin aA 
chains are unique among eutherian mammals in having 74 Tyr (Table 2). Thus by joining bear to 
pangolin we found the tree (Fig. 1) at 152 NRs, which may be the lowest NR score for this set of 
sequences since further searching has not resulted in any lower score. 


Lowest NR+GD+GE length trees 


In order to find the trees of lowest NR+GD+GE length we first specified the a priori cladistic 
relationships for which there is strong classical zoological evidence or sizeable evidence from the 
sequences of other molecules such as a-hemoglobin, ß-hemoglobin, and myoglobin. This search was 
aided by a modification of the branch swapping algorıthm described in section E of Appendix A-1 in 


Mammalian phylogeny and protein sequence analysis 263 


HORSE 


PANGOLIN 
3] TAPIR SEAL, 
RHINOCEROS Sn an 
3 2 4 
PoRPOISE | 3 
BEAR 
2 1 1 
. = ! MINK 
Fig. 1. Lowest NR length tree HHALENG I 4 KT nor 
(152 NRs) of o-crystallin A BIRIEBESSNIDEO _——a 
. 1 1 
sequences obtained on star- CAMEI: \ 
ting the search from the di- u: 5; Re 
BAT 
stance Wagner tree. Unaug- a 3 
mented NR values are given Ar Zu 
for all branches in this tree EN 
and the following ones. A - 
computer algorıthm was used RAT, HAMSTER, LER 
TUPAIA 
which selected a particular set 3 
GALAGO, POTTO MAN 
of most parsimonious ance- 5 ANTEATER 


stral residues, called the A- 
solution set (GooDman et al. 
1974). Wherever possible, 
these A-solution residues had 
the nucleotide replacements 
fall on terminal links to con- 


3 


7 RHESUS MONKEY 


2-TOED SLOTH 


HYRAX 


5 1 : 


ELEPHANT 


1 

1 4 
3-TOED SLOTH 
3 MANATEE 


5 


temporary species rather than EDURS 


between ancestral nodes. By a 5 
this means, the synapomor- 
phous residues characterizing 
each monophyletic grouping 
of the proposed cladograms 2 Q 
are kept to a minimum. 
Springhaas and gerbil are not 
shown in Figs. 1-3, but are 
identical ın position to guinea 
pig and rat, respectively FROG 


KANGAROO 


OPOSSUM 


CHICKEN 


GooDMmAanN et al. (1979). The modification allows one to designate nodes within the tree as though they 

were the roots of monophyletic subtrees; it then prevents branch swaps across each particular edge 

(the link connecting two adjacent nodes) which joins such an hypothesized monophyletic subtree to 

the rest of the tree. This modification helped us find trees of lowest NR score that did not require 

hypothetical GDs and GEs. We could then evaluate whether the NR score of these trees was less than 
the NR+GD+GE scores of the lowest NR length trees (those in Figs 1-3), or conversely whether ıt 
would be more parsimonious to have some paralogous gene lineages rather than have only ortholog- 
ous lineages. 

Our a priori cladıstic assumptions were: 

— all mammals are more closely related to each other than to chicken or frog; 

— the grouping of the investigated species into their respective traditional orders (e.g. rabbit and pika 
in Lagomorpha) is accepted, with tree shrew in its own higher taxonomic group (order Scandentia) 
rather than assigned, as it sometimes has been, to either Primates or Insectivora; 

— the classical intra-order relationship within the Edentata is adhered to (i.e. Bradypodidae and 
Myrmecophagidae are each considered monophyletic groups). 

The main freedom then allowed in the search for parsimonious trees that did not require hypothetical 

GDs and GEs was the reshuffling of the mammalıan orders in relation to each other as well as 

appreciable reshuffling of species within their orders. The two major trees of lowest NR score that 

required no GDs and GEs are shown in Figs4 and 5. Each had a score of 157 NRs. Equally 
parsimonious variants placed rhinoceros first with horse rather than with tapir. These alternative 
positions of rhinoceros were also found for the 152 NR length trees, and are caused by the sharing of 

residue 13 Ihr by tapır and rhinoceros, and residue 146 Ile by horse and rhinoceros (Table 2). 

Before we could conclude that the trees with score 157NR+0GD-+0GE were the lowest 

NR+GD-+GE length trees found we had to determine the numbers of GDs and GEs in the 152 NR 

length trees, i.e. we had to evaluate the score in total genetic events of these trees. By our criteria, the 

trees shown in Figs 2 and 3 are clearly more parsimonious than the tree shown in Fig. 1 because they 
violate fewer of the a priori assumed relationships. Nevertheless, these Figs 2 and 3 trees still each 
require, in addition to the 152 NRs, 5 GDs and 15 GEs. The extra genetic events are needed to account 


saouanbas y urpeısA19-m Jo 991 YN ZSI JAneulayYy 'E "317 


904 
NaNIIHI 
[44 
vi 
WNSSOdO 
L 
N 
S OOUWINVN 3 z 
8 
5 AAN ACANN LNVH4313 
O S 
: s / 
Z EEIRUNGN s 
Ss HLOTS 030L-€ HLOIS 0301-2 XYAH ; 
S v 1} 
[1 L 
80 
S - 
x Y3LV3LNV rn 
=S 90H3903H 8 YaLSWVH 
Wa 
> OllOd_ | een 
09V1Y9 z 
> En >t91d YaNIN9 
> A3NNOW SNSIHY—! v un 
Nun —€ 
v8 = 
NITO9NVd : z 
Z Id 
uv38 en 
I T OddIH 
NOIT v3S € 900° | Tr: 344W819 
1W3S h 3IVHM Io 
YNIW 
3S10440d 
YIdVL 
E 
€ 
3SYOH 


SOU3IONIHY 


264 


suondummnsse ausıpej> LıIouıd e Aq paurensuo9 zou “Burddems 
yaueıg Kuownsıed umuıxeur Aq y21eas ayı ıno ZurArre) pue sureı3 
-oIpuap oigısnegd Afesnsusdo4kyd yım Zuriejs uo paureigo sa) 
-uanbas y urpersA19-w Jo (SYN ZSI) 991 yı3uaj MN 3159407 ‘zZ '31] 


90% 


N3XIIHI 
€Ez 
vr 
WNSSOdO 
L 
OOUVINY 3 € 
YAVAQUYV INVH4313 A 
S (4 
EEIKIN z " 
XYAAH 


HLOIS 0301-2 v 
D 


HLOIS Q30L-€ 


[2 ; 9 
- nen e A 9Id vanınd |? 
Y3LWaLNV AWO yauaı Ifsava _! a 
n \ 5 va! |, 
A3XNOnN SNS3 ı °YIJLSWVH 
NYW 3 1v8 € 
[4 
90H3903H NI109NVd y ı lim and 
avıg N q 
NOIT v3S OddIH 
1435 (2 u) “3334419 
37VHM AD x0 
[} 
35104304 
IN yıdvl 
€ 
€ 
3SYOH 
SOYFIONIHA 


265 


Mammalian phylogeny and protein sequence analysıs 


03,921 
GEN sel 
+1 bel 
Ara 0L 


syuswsse]ds1 apnoappnu 0oMı aımbaı (@) suonninsgns 
auIos ‘paresıpur aIe suonnınsqns (>) ray, De („) [olfereg ‘sau Jusssjfip ur uonnınsgns Jo 
saye1 1us1aJfIp Aqe1apısuo9 ayI s[e9As1 yIIrym “um ur Paupueg jo 19p10 ayı Jo uorssaadun ue apraoıd 
02 SI afeas sum ayı Jo asodınd Afuo Ay], '("wuros ‘sid ‘VNNayOW ‘I 'W) Ode YAW se aq Argegoıd 
pjnoys ırggeı pue eyıd Jo Ze aum ay]L ‘(8/61 Te ?? vAaUUaH-ONIWON ‘6961 SNINAHL, 
‘9961 AAWOY) a1nze1aııf [est3ojosuoafed ayı 1017 Pa9ureIgo 91e J9U9319ALp Jo saum parewmsy "suonefal 
I9PIO-enur Inoge suorurdo Jua11ın9 YIM 39UEPIOHIe ur (sJUaPpo1 pue syueurum! ur) umeıp aıe 10 ‘s99M 
79 + AI + AN 15940] Jo uonsn.nSU09 ayı ur PaAjoAur suondunnsse In9usdofAyd Lıonıd e ay3 1997794 
suonnINsgqns pie ouTUIe UTeIUO9 10U OP YIyM saydueig [eu1sgu] ‘(sıpnsaı paysıigndun) 39usnbas Yo 
(svıqguv»v snjonbs) ysır3op ayı yıım uosuieduro9 Aq Pa11ajur 91e s9J0TuUUIe 3yJ 0) pue 301} ayı 01 9911 
ay2 Jo 1004 U WOAF SOy9ueIg ayı ur suonnıNsqns Sy], 'sapou ayı 1e sosuanbas y urjessA19-n Tensaoue 
P319n.nSU0991 9yı UIO1F P9INPap a1e soydueiqg JUsIaJFIP ayı Ur suonnINSqns piae ouruy 'sJ9 pure sq9 
jesnaysodAy aımbaı 10u pıp seyr (syYN ZSI) 31095 YN 159M0J Jo s991 1olew 0m} ayJ Jo 3uQ 'p 31] 


on 
wdO 
Lpl 1°s 691 SeL,891 
8 \ed 891 
gey 9°d 851 «1'621 
SET 0x3 981 NS TOT 
| MeITE 
us a m osı 
Eis a al u; lan EST 0 16 
Z21.3+0 89 “ ade ‚2sı 
ae LöP 6HT-Zpl , «dd WI—VeL td 
I | Sven Ezı Den 
Sed 91 + 
| u.) 1+S 221 a)ah 
1 1-S\ 121 
[sc elev'el 
u Gl ala Kl 
85 ie IpI «D<1 06 1+S sst "7 
I Teı Lap Est | 
| al-s 8hL EN 
N«Q a El >59 ie 
A 6b 851 alen 91 est 9p1 
>03 16 
nr abs 9-0 Kelg, alu IV I*S 
14 nn alen ! N; : ee1 u Gl en 
n A+J Ir avN >|+S Aasın „N*I 
ei n i 891 aren FOL rn 
hi Ne Sch N Sen Mn Se4 hl A 
“ri IE „De1 1 291 TOT 9-1 €6 19 ost 
UN 06 v8 „D+H “Ies «L+V 06 „Ars > St+L 
QL | | / &2 en vl 2 
[I ' ı [} ' I I 
{m} oO Ss m zZ > nm w — =g zZ [p) [9,) fep) = e=) —o zo u {=} ae] [94] [%4) z oO je) je=} [@) {op} @) SE o u = zZ ee] a 
SEI JS BASS AS SB SISE EN IS EST R EN NEE 3.250055 
x u > _E SIE XxEZ<ZEBSI Em Em aEemE rar = = a = a S = = S rm a zZ = = r Ss ° m m =} Es) 
me rX@>m >= oo > mn oo > (77) SEN mw = au) n-] I ©) 
z z0co©5 z2zm 2 m SS m) o mes m o 7) m 
oO — Bay p=) (9) (9) [p) —o z = z — m =] 
Im 2 > — =} >= 7 
Se a D- = 2 
= = Bas o 
< 


0:73 


052 


001 


09 


Or 


02 


JAN 


266 W. W. de Jong and M. Goodman 


N 
x AIERS 
Sy N 0» SS 
9 SI S SASS NIS 
D S SS N BURN B_I ILL SS 
I x S SS SS < S DRS II 
© S D N S S So IR, VERECHSZSESTESIN DIDI 
S SEE SER I SEOSD EI DETELRERÄEERTD 
EL SD DES “ VS ID OS SS IE MTOMMMIS TI 5 
II R TIER SECHEN REISE IENS OST POKI DOC D INES 
wm FOETEOF SSEEEETSTELTÜFTSTEOTTTILTETDTEITTETTETS 
O\ /O 
ol /1 2\ololal 9 A A „| ol jo 5\ 015 
0| |2 o| lo 1 9 
20 1) 11 0O| |7 
> jS 2\o\ Iı 
5472 ol Ie = 
0 ) 
5 
40 0 R } 
1 1 1 
1 
1 [6] 6) (6) 
60 1 1 (6) (6) 0 
} 6 3 
2 
2 7 () 
80 3 


100 


120 


Fıg. 5. Alternative major tree requiring 157 NRs and no GDs and GEs. Only the numbers of NRs are 
indicated for the different branches; internal O-links are due to a priori phylogenetic assumptions. The 
chicken and frog branches (not shown) are as ın Fig. 4 


for: 1) the marsupial branch being closer to chicken than to the eutherian branch; 2) the three-toed 
sloth being closer to anteater than to the two-toed sloth; 3) rabbit being closer to Primates than to 
pika; 4) guinea pig and springhaas being closer to Lagomorpha and Primates than to Muroidea; 5) bear 
being closer to pangolın than to other carnıvores. These violations of well accepted cladistic 
relationships each cost 1 GD + 3GEs. In contrast the orthologous arrangement can be had for only 
1 NR more in each case. Thus it is indeed more parsimonious (total genetic score 157 compared to 
172) to have the aA sequences in the orthologous branching arrangement to one another, as shown in 
Figs 4 and 5. 


Results 
Phylogenetic inferences 


The most unbiased and therefore strongest inferences obtainable from the o-crystallin A 
sequences come from the lowest NR length trees (Figs 1-3). The constant features in these 
trees provide important indications of possible phylogenetic relationships. There are, 
however, several conspicuous deviations from generally accepted opinions about mamma- 
lian phylogeny. Most disturbing is the joining of chicken to the marsupial branch; ın 
addition several species belonging to the same orders fail to group together, albeit often 
separated by the presence of only a single amino acıd difference. 

These shortcomings are overcome by the introduction of some a priori phylogenetic 
assumptions, resulting in the lowest NR+GD-+GE trees shown in Figs 4 and 5. The 
introduction of such reasonable constraints may result in a more realistic branching pattern 
of those OTUs of which we would like to study the phylogenetic relationships and which 
have been left free in their positioning. 

The differences in the branching patterns of equally parsimonious trees (Figs 1-3 and 
Figs 4-5) reveal the limitations of the method and indicate those regions of the tree where 
no firm phylogenetic conclusions can be made. The relative significance of a branching 
arrangement increases with the length (in NRs) of the branches involved. A 1 NR 
connection between two OTUs is more likely to be due to a parallel or back mutation than 


Mammalian phylogeny and protein sequence analysis 267 


a longer branch. Also the position and type of a substitution has some importance, for 
instance: the substitutions 13 Ala— Thr or 61 Ile— Val occur scattered over four orders 
(Table 2), and thus have less diagnostic value than 13 Pro, which only occurs in prosimians, 
or 127 Thr, which only occurs in Perissodactyla. We therefore have included in Fig. 4 the 
positions and directions of all inferred amino acid substitutions, and indicated all instances 
of supposed parallel or back mutations. 

In the following sections we will successively discuss the major branching characteris- 
tics of the oa-crystallin A cladograms in relation to prevailing ideas on mammalian 
phylogeny; i.e., the phylogenetic information which can be deduced from these cladog- 
rams is judged and weighed in relation to the prevailing opinions about mammalian 
phylogeny. For this purpose we have relied mainly upon some recent authorative works 
(Simpson 1945; ROMER 1966; THENIUS 1969; van VALEN 1971; HOFFSTETTER 1973; 
McKenna 1975; SzarLay 1977). We have also taken into account evidence from other 
comparative biochemical studies of myoglobin (ROMERO-HERRERA et al. 1978), hemoglo- 
bin (Goonbman et al. 1979), pancreatic rıbonuclease (BEINTEMA and LENSTRA 1982), 
cytochrome c (FouLps et al. 1979; Baga et al. 1981) and fibrinopeptides (O’NEIL and 
DooLiItTLeE 1974; DayHorF 1972; GOODMAN 1981). The immunological comparisons of 
mammalian albumins are also relevant (SarıcH 1976, 1982). 


Monophyly of therian mammals 


There seems to be little reason to doubt the monophyletic origin of the eutherian and 
metatherian mammals from an advanced synapsid reptilian stem in the Upper Triassic 
(MARSHALL 1979). Nevertheless our lowest NR trees (Figs 1-3) depict marsupials and birds 
as sister groups. A single additional NR restores the therian monophyly, but it is still 
amazing that no greater number of synapomorphies should have accumulated in the 
estimated 200 million years of supposed evolution of the common therian ancestor. A 
marked reduction in the number of amıno acid substitutions in the mammalian ancestral 
stem has also been observed in myoglobin (ROMERO-HERRERA et al. 1978; GOODMAN 
1981), hemoglobin (GoopMAN 1981) and cytochrome c (BaBa et al. 1981). However, the 
lowest NR length trees for globin sequence data support the monophyly of therian 
mammals. 


Metatherian-eutherian divergence 


Both in the lowest NR and lowest NR+GD-+GE trees the marsupials are well separated 
from the placental mammals, each being characterized by a convincing number of 
apomorphous (unique) substitutions. This agrees well with traditional taxonomic evidence 
that Metatheria and Eutheria are each monophyletic. However, this therian dichotomy is 
not always reflected in the lowest NR length trees constructed for globins and cytochrome 
c sequences. 


Edentate relationships 


The three investigated edentate species (2-toed sloth, 3-toed sloth and tamandua) share 5 or 
6 substitutions and a rare deletion of 3 residues, clearly reflecting a monophyletic origin. 
The phylogenetic unity of the South American edentates (= Xenarthra) has indeed never 
seriously been questioned (ENGELMANN 1982) The aA trees depict the edentates as one of 
the oldest offshoots of the eutherian stem. The trichotomy in Fig. 5 indicates that the aA 
sequences cannot discriminate between three possibilities: the edentates are the oldest 
eutherian branch and the paenungulates the next oldest; the paenungulates are the oldest 
eutherian branch and the edentates the next oldest; the edentates and paenungulates are 
sister groups and together constitute the oldest eutherian branch. 

The relationship of the edentates to the other mammals is obscure, although pangolıns 
and aardvarks have been considered as relatives (reviews by GLass 1982; ENGELMANN 


268 W. W. de Jong and M. Goodman 


1982). Placement of the Edentata within the Eutheria appears to be preferred. It has 
recently been hypothesized that the Edentata are a sister group to all other eutherians 
(McKenna 1975; ENGELMANN 1982), which is compatible with the a-crystallin data. 

The two investigated sloth genera Bradypus and Choloepus are grouped together in the 
Bradypodidae, although it recently has been recognized that many similarities may be 
regarded as convergent features and important differences have been noted (ENGELMANN 
1982). It is therefore not surprising that the two sloth aA chains show no synapomorphies 
ın the lowest NR trees, and that Bradypus aA chain joins the anteater branch on the basis 
of the unique substitution 56 Val— Ala. There is, however, little need to conjecture that 
the sloths represent a grade rather than a clade, because one additional NR restores the 
accepted division between Bradypodidae and Myrmecophagidae. 

The only other edentate protein sequence data are of sloth ribonuclease (BEINTEMA and 
LEnsSTRA 1982) and armadillo ß-hemoglobin (DE Jong et al. 1982). Sloth rıbonuclease, as 
compared with rodent, whale, ungulate and kangaroo ribonucleases, indicates that sloths 
are on the oldest branch among the eutherians. Armadillo Hbß joins in the most 
parsimonious solutions the branch leading to elephant Hbß, but does not specifically 
appear as the oldest eutherian branch. Immunological comparisons of albumin supported 
the traditional grouping of three-toed and two-toed sloths in Bradypodidae and of anteater 
in Myrmecophagidae, but did not reveal the position of edentates among the eutherians, 
although they were especially distant from ungulates and rabbits (SArıcH 1982). The 
immunological distances between sloths, anteaters and armadillos suggests that the period 
of edentate monophyly has been rather brief, and that they have diverged from each other 
at about the same time as bats, primates and carnıvores diverged. Preliminary results did 
not indicate a special relationship of any edentate to either pangolin or aardvark. 


Aardvark-paenungulate relationships 


From both Table 2 and the computer-constructed cladograms it can be seen that elephant, 
hyrax, manatee and aardvark, as compared to all other investigated mammals, share 3 to 4 
apparently derived substitutions, of which 70 Lys— Gln and 72 Val—> Leu are unique for 
this group, while 74Phe—Leu and 142Ser>Cys also occur as supposedly parallel 
substitutions in Tamandua and Homo, respectively. 

Most authors agree on the grouping together of Proboscidea, Sirenia and Hyracoidea ın 
the superorder Paenungulata, although Sımpson (1945) as he proposed this grouping, 
clearly stated that superorders are theoretical constructions, not necessarily reflecting 
common origins. In fact Sirenia and Proboscidea are usually considered to be most closely 
related among the three orders. To bring elephant and manatee together on the same 
branch in Figs. 1-5 would require one additional NR, most lıkely a back substitution 
72 Leu— Val in elephant. It has been proposed that Hyracoidea might be more closely 
related to Perissodactyla than to Proboscidea and Sırenia (McKennA 1975). Removing 
hyrax from the paenungulates (Fig. 4) and placing it with Perissodactyla costs at least 10 
additional NRs, and therefore seems unjustified. Similarly, bringing all paenungulates 
together with Perissodactyla (van VALEN 1971; SzaLay 1977) costs 5 additional NRs and 
therefore also seems improbable. Paenungulate monophyly is further supported by 
immunological crossreactivity between hyrax and elephant (WeEıTz 1953). 

Most significant is the strong connection of the aardvark aA sequence to those of the 
paenungulates. The aardvark is clearly a phylogenetic problem case. The hypothesis that 
aardvark is related to edentates and pangolins seems to have been largely abandoned. 
Indeed, to separate the aardvark aA chain from the paenungulates and connect it, without 
GDs and GESs, to the edentate branch would cost an additional 4 NRs. It is now usually 
suggested that the Tubulidentata originated from a condylarthran stem, and thus should be 
most closely related to the ungulates (PATTERsoN 1978). Because the paenungulate orders 


Mammalıian phylogeny and protein sequence analysıs 269 


Proboscidea, Sırenia, and Hyracoidea are generally thought to be descended like the true 
ungulates from a common condylarth stock, the joining of aardvark and paenungulates is 
not incompatible with the current vague opinion that aardvark ancestry also traces back to 
a common condylarth stock. Morphological resemblances between aardvarks, hyraxes and 
elephants have been noted (LE GRoS CLARK and SONNTAG 1926), and some recent 
immunological and osteological findings further support such a relationship (SHOSHANI et 
al. 1978). Furthermore, both Tubulidentata and the paenungulates apparently originated ın 
Afrıca. Unfortunately no other protein sequence or immunological data from the aardvark 
are as yet available to help further elucidate its phylogenetic relationships. 


Position of the Paenungulata 


In all the most parsimonious trees of the a-crystallin A chains the enlarged Paenungulata 
(including Tubulidentata) are far apart from the true ungulates, in disagreement with the 
view of a common condylarth ancestry. Placing the paenungulates with Perissodactyla, 
which is its most parsimonious position if constrained to be in the ungulate-cetacean region 
of the tree, costs, as already mentioned, 5 NRs more than when it is separated from other 
eutherians as one of the most ancient branches. The early divergence of the paenungulates 
in these trees is mainly caused by the lack of some derived substitutions shared by most of 
the other eutherians: 91 Glu, 150 Leu or Val, and 153 Gly. The ancıent branches of 
paenungulates and edentates can be changed to amore recent and simultaneous radiation of 
eutherian orders, at a cost of 2 NRs, by placing an edentate-paenungulate branch next to 
the branch of primates, insectivores, rodents and lagomorphs (cf. Fig. 4). 

The only other paenungulate protein sequences known are elephant ß hemoglobin, 
myoglobin and fibrinopeptides. In a recent analysis of the combined sequences of up to 
7 proteins in 49 vertebrate taxa, the most parsimonious trees had the Paenungulata 
(represented by Proboscidea) originate as a separate branch in the earliest Eutheria 
(GooODMAN 1981). McKennA and Manniıng (1977) have provided paleontological evi- 
dence for a very early origin of Proboscidea. Such evidence greatly reduces the lıkelihood 
of acommon ungulate-paenungulate origin. 


The insectivore-primate-rodent-lagomorph cluster 


The lowest NR trees (Figs 2 and 3) faıl to resolve the clustering of the aA sequences from 
these orders. This is mainly due to the paucity of substitutions ın rodents, Jagomorphs and 
early primates. To place the investigated rodents and lagomorphs in their respective orders, 
as has been done in Figs 4 and 5, requires 2 more NRs than the shortest trees. The need to 
do so, however, is supported by the combined sequence analysıs (GooDMAn 1981) ın 
which cytochrome c, hemoglobin, and fibrinopeptide A sequences override the aA se- 
quence and cause guinea pig to group with Myomorpha. Similarly the recently completed 
pika myoglobin sequence clearly groups pika and rabbit together (DEne et al. unpublished 
data). 

The few a-crystallin A substitutions which are present ın rodents and lagomorphs and 
which might be used for tree-construction, are unreliable indicators of relationship, 
because the same substitutions are found frequently in other mammalian orders (see 
positions 13, 90, 101 and 150 ın Table 2). 

The frequency of substitutions is greater in later primate evolution and in the hedgehog 
line than in most other mammalıan lines and brings to light the undisputed dichotomy 
between prosimians and Anthropoidea, and agrees with a lorısoıd monophyly. The 
considerable differences between prosimians and Anthropoidea are also reflected in the 
myoglobin sequences (ROMERO-HERRERA et al. 1978). The common substitution 
90 Leu Gin in springhaas and guinea pig would be compatible with a grouping of the 
Pedetidae (springhaas) in the African rodent suborder Hystricomorpha, and the grouping 


270 W. W. de Jong and M. Goodman 


of this suborder with that of the guinea pig, ı.e. with the South American Caviomorpha 
(HoFFSTETTER 1973; LavocAT 1978). Considerable support for a caviomorph-hys- 
tricomorph monophyly stems from rıbonuclease sequence data (BEINTEMA and LENSTRA 
1982). 

Despite the inabiılity to resolve fully the relationships of the lineages among the orders 
Insectivora, Primates, Rodentia, and Lagomorpha, the aA sequences are nevertheless 
useful in indicating that these four orders may have descended either from a common 
ancestor shortly after ıt separated from the major branch leading to carnıvores and 
ungulates (Fig. 4) or separately from the stem of this carnıvore-ungulate branch (Fig. 5), 
and after the earlier separation of edentates and paenungulates. As far as primates and 
insectivores are concerned either pattern of descent is not seriously at odds with the 
prevailing views on eutherian phylogeny. However, the analysis of up to seven combined 
polypeptide chains and of just combined a and ß hemoglobin chains (see respectively Figs. 
4 and 8 of GoopDMAN 1981) supports the more distant ancestral separation of Insectivora 
(hedgehog) from Primates depicted in Fig. 5 rather than the closer one depicted in Fig. 4. 

The prevailing views on eutherian phylogeny do not allow us to choose between the 
positioning of Rodentia near a primate-lagomorph branch depicted in Fig. 4 or the 
somewhat more ancıent origin of Rodentia depicted in Fig. 5. In fact both lagomorphs and 
rodents appear as ısolated groups of largely obscure origin, and their grouping together in 
the cohort Glires (Sımpson 1945) still finds opponents as well as proponents (SZALAY 
1977). 

A considerable body of protein data in addition to that provided by aA crystallin is 
available for investigating the phylogenetic relationships of Primates, Lagomorpha, Roden- 
tia, and Insectivora. As yet, however, no consistent picture of their relationships emerges 
from it. It may well be that the periods of common ancestry between these orders and 
other eutherian branches were too short to have left their traces in the DNA and protein 
sequences so far available. 


The position of Tupaia 


The relationship of the tree shrews to primates or insectivores is a much discussed issue 
(LuckETT 1980), and there are reasons to place them in the separate order Scandentia 
(BUTLER 1972). The Tupaia a-crystallin A sequence is identical to that of the investigated 
muroid rodents. This does not reflect a special relationship, but just the complete lack of 
fixed substitutions in both evolutionary lines. Because of this lack of change the Tupaia aA 
chain can equally parsimoniously be connected to the base of the rodent or insectivore 
lineage as to acommon lagomorph-primate stem. Connecting it at the base of the primate 
line adds only 1 NR. The previously investigated Tupaia myoglobin and hemoglobin 
sequences certainly do not indicate a relationship with Primates, but rather in the most 
recent combined sequence analysis (GooDMAN 1981) cause Tupaia to group with 
Lagomorpha. 
Chiroptera 


The aA chain of the bat Artibens jamaicensis is most parsimoniously connected to the 
branch leading to ungulates and carnivores, mainly due to the presence of residues 3 Ile and 
150 Val. On the other hand the plesiomorphous residues 4 Thr and 147 Gln prevent it from 
being separated too far from the rodent-primate-insectivore-lagomorph cluster. 

On the basis of morphological evidence the bats have been proposed to be derived from 
early insectivores (ROMER 1966), or to be closely related to primates (GREGoRY 1910; 
Szaray 1977). To get a monophyletic bat-hedgehog or bat-primate branch in the 
a-crystallin tree would require an additional 2 NRs. The sequence of bat myoglobin shares 
some derived substitutions with that of the hedgehog (CasTıLLo and LEHMAn 1977) and in 
the most parsimonious phylogenetic trees of myoglobin these species are shown as sister 
groups (GooDMman et al. 1979). The minimal tree for vertebrate cytochrome c joins bat to 


Mammalian phylogeny and protein sequence analysis 271 


the carnıvore branch, not to primates (an insectivore cytochrome c sequence is not yet 
available) (FouLos et al. 1979; Baga et al. 1981). Actually the myoglobin was obtained 
from a representative of the suborder Megachiroptera and cytochrome c and a-crystallin 
from microchiropterans. Although a monophyletic origin of these two suborders seems 
likely (Thenıus 1969), the varying placement of the two suborders in the cytochrome cand 
a-crystallin trees may be explained if these two suborders prove to be of biphyletic origin. 


Pholidotes not close to Edentates 


Apart from the complete sequence of the Malayan pangolin aA chain we have obtained the 
largest part of the aA sequence of the tree pangolin Manis (Phataginus) tricuspis (DE JONG 
et al. 1982). The sequences differ from each other at least at four positions, indicating a 
considerable time of evolutionary divergence. The Malayan pangolin aA sequence is most 
parsimoniously connected to the carnıvore-ungulate region of the tree, on the basis of the 
shared derived substitutions 4 Ala and 147 Pro. On the other hand 3 Ile, 101 Ser (which, 
however, is Asn in the tree pangolin) and 153 Ser could be considered as primitive 
characters, placing the pholidotes at the base of the eutherian radıation. Such a separation 
of the pholidotes as the earliest eutherian offshoot would cost an additional 2 NRs, or 
3 NRs in the case of the tree pangolın. 

The pangolin aA sequence certainly shows no synapomorphies with the edentates, and 
it actually adds 3 NRs to group the pangolins with the edentates. This finding has relevance 
for the continuing discussion about possible pholidote-edentate relationships. Some 
authors consider the pholidotes as the closest relatives of the edentates (van VALEN 1971; 
SzaLay 1977; PATTERSON 1978), or do not exclude this possibility (Simpson 1945; 
ENGELMANN 1982), but others have entirely abandoned this idea (ROMER 1966; THENIUS 
1969; McKenna 1975) and leave their origins completely open. The possibility that the 
Pholidota might be placed within the Edentata, derived from a myrmecophagid-like 
species (SZALAY 1977; ENGELMANN 1982) seems to be excluded by the aA sequence data. 

The palaeanodonts, which may be ancestral to or relatives of early pangolins, have been 
placed by Emry (1970) in the order Pholidota. Rose (1978) has suggested that palaeano- 
donts are possibly close to the Pantolestoidea, which McKenna (1975) grouped with the 
carnıvores in the grandorder Ferae. The similarıties between pholidote and carnıvore aA 
sequences thus support the relationships based on thıs paleontological evidence. Although 
no other protein sequence data are yet available for the pangolins, immunological findings 
with chicken antisera support the grouping of Pholidota with Carnıvora (SHOSHANI, ]. 
unpublished data). 


Carnivore-ungulate relationship 


In all parsimony solutions carnıvores, ungulates, cetaceans and pangolins are grouped on 
the same branch. This ıs due to the shared derived substitutions 4Thr— Ala and 
147 Gln— Pro, which are present in all investigated species from these groups (apart from 
porpoise, which has the autapomorphous substitution 147 Pro— Thr). These residues at 
these positions do not occur in any other mammalian order (Table 2). Furthermore the 
back substitution 3 Val— Ile contributes to the separation of a carnıvore-ungulate branch. 

Apart from the uncertainties concerning the position of the pangolins, the joining of 
ungulates, whale and carnıvore aA sequences on the same branch may well reflect a 
monophyletic origin. VAn VALEN (1966) and McKenna (1969) argue that the cetaceans 
arose within the mesonychid condylarthrans. The placement of the whales among the 
ungulates, strongly supported by neontological studies (THENIUS 1969), is now widely 
accepted. Sımpson (1945) brought together the ungulates (but not the whales) and the 
carnıvores in the cohort Ferungulata. Both LirLEGrRAvEn (1969) and McKenna (1969) 
considered the possibility that late Cretaceous palaeoryctids gave rise to creodonts, 
carnıvores and ungulates. According to SzaraY (1977) no substantive evidence exists to 


DIR W. W. de Jong and M. Goodman 


contradict the concept Ferungulata, although no undisputed shared derived characters have 
clearly been brought forward. A common origin of ungulates and carnivores is not widely 
accepted, and most authors prefer to remain uncommitted on this issue. 

The sequences of cytochrome c, myoglobin and pancreatic ribonuclease of different 
whales are known. They tend to join the Cetacea to the ungulates, although the most 
parsimonious solutions in these cases are not sımilar to those based on traditional 
anatomical evidence. Carnivores and ungulates can be compared by known sequences of 
myoglobin, cytochrome c, a- and ß-hemoglobins, and fibrinopeptides. Cytochrome c 
seems to support a carnıvore-ungulate relationship, whereas myoglobin tends to place 
Carnivora closer to Lagomorpha and Tupaia than to either Primates or ungulates and 
cetaceans while fibrinopeptides and hemoglobin tend to place carnıvora on a branch 
containing Primates, Lagomorpha, Tupaia, and Rodentia. 


Monophyly of Pinnipeda 


The aA sequences show no shared derived substitutions for all six investigated carnivores, 
and are unable to resolve the relationships between dog, cat, bear and mink, which 
represent four different families, and the pinnipeds. The identical sequences of seal and sea 
lıon aA chains, however, contain 3 synapomorphous substitutions. Two of these (51 Pro 
and 52 Val) are unique among all investigated aA chains, and strongly indicate a 
monophyletic pinniped origin. Such a pinniped monphyly is indeed the classical phy- 
logenetic opinion. It has, however, been proposed that the Otariidae (sea lions) might be 
related to the families Canıdae or Ursidae, and the Phocidae (seals) to the family 
Mustelidae (SAvaGE 1957; MCLAREN 1960; TEDFORD 1976). The only other protein 
sequenced both in seal and sea lion is myoglobin which also supports, albeit weakly, a 
monophyletic origin. Albumin immunological evidence likewise indicated a pinniped 
monophyly (PRAGER and Wırson 1978). Preliminary comparative data on hemoglobin 
chains from six fissiped carnivore families revealed close relationships between badger 
(Mustelidae) and raccoon (Procyonidae) chains (HOMBRADOoSs et al. 1978; BRIMHALL et al. 
1979). 


Ungulate interrelationships 


The order of branching of the orders Carnıvora, Pholidota, Cetacea, Artiodactyla and 
Perissodactyla is not really resolved by the aA sequences. The ungulates and whales appear 
to be monophyletic in Figs 2 and 4, but just on the basis of a single substitution (90 Leu— 
GIn), which has occurred repeatedly, back and forth, in different taxa. The same is true for 
the preferred position of the whales as a sister group to the Perissodactyla, based on 
substitution 150 Val—Met. There actually is some immunological and karyological evi- 
dence to consider the whales as most closely related to Artiodactyla among the mammalıan 
orders (THEnıus 1969), but in recent classifications of the mammals this issue is left 
undecided. The improbable, but sometimes discussed possibility of a biphyletic origin of 
Odontoceti and Mysticeti is refuted nor supported by the sequences of the aA chains of 
their respective representatives porpoise and minke whale. 

Among ungulates only the three perissodactyls are joined together by a unique 
synapomorphous substitution 127 Ser— Thr. Within the Perissodactyla it is equally par- 
simonious to show tapir closer to rhinoceros than to horse, or alternatively to bring 
rhinoceros and horse together. However, in a recent analysis involving fibrinopeptides A 
and B of 47 mammals tapir and rhinoceros grouped first before joining Equidae (GooD- 
MAN 1981), in agreement with the taxonomically preferred position. 

Although the Artiodactyla are classically divided in the suborders Suiformes and 
Ruminantia, the hippopotamus which belongs to the first group is placed on the ruminant 
branch of all aA trees, due to the shared derived substitution 146 Val—> Ile. As can be seen 
in Fig. 4, however, this 146 Val > Ile substitution occurs frequently throughout the tree. 


Mammalian phylogeny and protein sequence analysıs 273 


Nevertheless the separation of the pig and hippopotamus aA sequences deserves some 
attention since the ribonuclease sequences also join hippopotamus to the ruminants rather 
than to pig (BEINTEMA and LENSTRA 1982). Pig and hippopotamus cytochrome c differ at 
three positions (THompson et al. 1978), while porcine cytochrome c is identical to the 


bovine and ovine sequence, and hippopotamus is slightly more similar to camel and 
guanaco than to pig. 


Discussion and prospects 


In using protein sequences for the elucidation of phylogenetic problems, one is dependent 
on the numbers and kinds of amino acid substitutions which occurred during descent of 
the proteins under investigation. The a-crystallin A chain has undergone just the “right” 
degree of change in certain evolutionary lines to be very informative. For example, the fact 
that in our most parsımonious trees there are three to four amino acıd substitutions on the 
paenungulate stem which are then retained by all four paenungulate orders, Proboscidea, 
Sirenia, Hyracoidea, and Tubulidentata, provides evidence for the monophyletic origin of 
these orders. Similarly a sufficient number of synapomorphous or shared derived substitu- 
tions occur on the stem to sloths and anteaters as to provide evidence of the monophyly of 
these edentates. This is also the case on the ancestral line to the catarrhine primates man and 
rhesus monkey. In other lines, the rate of change of the aA gene has been so slow as to be 
uninformative, as in the rodents and lagomorphs. In these instances proteins such as 
hemoglobin, myoglobin and pancreatic rıbonuclease, which evolve faster than a-crystallin, 
may be able to help unravel the relationships. However, patterns of relationship between 
taxa involving relatively short periods of common ancestry in the distant past, may turn 
out to be unresolvable by any macromolecular data. 

Hoımauiıst (1978) has shown that the “denseness” of a phylogenetic network bears 
importantly on the accuracy of evolutionary reconstructions from protein sequence data. 
The quality of the evolutionary information derived from many closely related sequences is 
higher than that derived from a few distantly related sequences. A tree is maximally dense 
when the link lengths between nodes correspond to one nucleotide replacement or zero 
replacement. The a-crystallin A tree approaches this maximal density fairly well in certain 
regions. It can indeed be seen that the increased denseness of this tree as compared with the 
one in DE Jon et al. (1977) (in which few regions have lınk lengths of 1 or 0) allows more 
precision in the assignment of substitutions to certain branches. For instance, from the 
present data set the substitution 13 Ala—> Thr is seen to be an autapomorphy of the ox, 
which probably occurs independently in tapır and rhinoceros. The previous limited data 
set showed horse and pig to have 13 Ala, and ox and rhinoceros to have 13 Thr, but this 
information did not allow a decision about location and direction of the substitution. 

Taken as a whole, the present results demonstrate the general usefulness of protein 
sequence data in the study of mammalıan phylogeny, provided that an appropriate choice 
of taxa ıs made, and that the investigated protein shows an appropriate degree of change. 
The suggestion (DE Jong et al. 1977) that the a-crystallin A chain would probably be a 
suitable tool to study the phylogenetic relationships of Edentata, Pholidota, Tubulidentata, 
and Chiroptera has been justified by the present study. Similarly, it can now be suggested 
from the results in Figs 4 and 5 which remaining taxa are likely to further elucidate 
mammalıan phylogeny if added to the aA tree. The occurrence of 5 synapomorphous 
substitutions and a deletion in the aA chain of Macaca mulatta and Homo sapiens make it 
very promising to study Tarsius, in view of its disputed relationship with the 
Anthropoidea. The inclusion of a new-world monkey, preferrably the large-eyed Aotes, ot 
which a few specimens should yield ample material, would obviously make the outcome 
more significant. The considerable numbers of autapomorphies in hedgehog and primate 
aA chains would make it worthwhile to study the Dermoptera and possible relatives of the 


274 W. W. de Jong and M. Goodman 


insectivores such as the elephant shrew. Since only 2 or 3 autapomorphous substitutions 
were detected in the microchiropteran bat Artibeus ıt ıs unlikely that «A sequences can 
reveal whether the origin of Microchiroptera and Megachiroptera was monophyletic or 
biphyletic. Nevertheless, ıt would still be desirable to study a representative of the latter 
suborder because it may strengthen the positioning of the Chiroptera in the aA tree. The 
presence of a number of autapomorphies in several carnivores indicates that useful studies 
could be conducted on the remaining carnıvore superfamilies. Finally, the finding of 9 
synapomorphies ın the marsupial aA chains and 7 autapomorphies in the opossum make it 
likely that the aA chain sequences could help unravel the phylogenetic relationships among 
marsupial higher taxa. 


Acknowledgements 


We thank all persons who have made available eye lenses for our investigations (Table 1), EuG£ENIE 
Nuy-TERWINDT, ANNEKE ZWEERS and MARLIES VERSTEEG for technical assistance with the sequence 
determinations, and Dr. M. C. McKenna for valuable discussions. We also thank Dr. GABRIEL 
LASKER for critically reading an earlier draft of this paper and offering helpful suggestions. This work 
was supported in parts by the Netherlands Foundation for the Advancement of Tropical Research 
(WOTRO) and by the Netherlands Foundation for Chemical Research (SON) with financial aid from 
the Netherlands Organization for Pure Research (ZWO), and by grant DEB 78-10717 from the 
National Science Foundation. 


Zusammenfassung 


Untersuchung der Stammesgeschichte der Sängetiere durch Sequenzanalysen 
des Angenlinsenproteins a-Kristallın 


Die Aminosäuresequenzen des Augenlinsenproteins a-Kristallin A von 41 Säugetierspecies (17 

Ordnungen vertretend) wurden untersucht mit dem Ziel, die Beziehungen zwischen den Säugerord- 

nungen zu klären. Die beobachteten Aminosäuredifferenzen wurden zur Konstruktion von Klado- 

grammen verwendet. Die a-Kristallin-A-Sequenzen weisen darauf hin, daß: 

— die Paenungulatenordnungen Proboscidea, Hyracoıdea und Sirenia eine monophyletische Gruppe 
bilden, zu der auch die Tubulidentata (Erdferkel) gehören 

— die Paenungulaten nicht mit den Ungulaten verwandt sind, sondern gemeinschaftlich mit den 
Zahnarmen die ältesten Abzweigungen des Stammes der Placentalia darstellen 

— die Schuppentiere keine Beziehungen zu den Zahnarmen aufweisen und am besten in die Nähe der 
Raubtiere gestellt werden - 

— die Ungulaten zusammen mit den Walartigen und den Raubtieren eine monophyletische Gruppie- 
rung bilden 

— unter den Raubtieren die Robben und Seelöwen monophyletisch sind 

— die untersuchten Fledermäuse (Microchiroptera) sich nicht als verwandt mit den Insektenfressern 
oder Primaten herausstellten. 

Die a-Kristallin-A-Sequenzen ließen die Nagetiere, Hasenartige, Insektenfresser, Primaten und 

Spitzhörnchen als einen unaufgelösten Cluster von Ordnungen erscheinen, aber innerhalb der 

Primaten konnten die Prosimiae eindeutig von den Anthropoidea unterschieden werden. Die Ergeb- 

nisse werden mit gängigen Auffassungen über die Säugetierphylogenie verglichen und mit anderen 

vergleichenden Proteinsequenzdaten in Beziehung gebracht. 


References 


BaABA, M.; DARGA, L.; GOODMAN, M.; CZELUSNIAKR, J. (1981): Evolution of cytochrome cinvestigated 
by the maximum parsimony method. J. Mol. Evol. 17, 197-213. 

BEINTEMA, ]J. J.; LENSTRA, J. A. (1982): Evolution of mammalian pancreatic ribonucleases. In: 
Macromolecular sequences in systematics and evolutionary biology. Ed. by M. GooDMman, New 
York: Plenum Press (in press). 

BLOEMENDAL, H. (ed.) (1981): Molecular and cellular biology of the eye lens. New York: Wiley- 
Interscience. 

BRIMHALL, B.; STANGLAND, K.; JonEs, R. T.; BECKER, R. R.; Baıtey, T. J. (1979): Tryptic peptide 
compositions of hemoglobins from mink (Mustela vison), and hyena (Hyaenae hyaenae) and Hb 
o-chain of coatimundi (Nasna nasıa). Hemoglobin 3, 271-292. 

BUTLER, P. M. (1972): The problem of insectivore classification. In: Studies in vertebrate evolution. 
Ed. by K. A. Jovysey and F. R. Kemp. Edinburgh: Oliver and Boyd. pp. 253-265. 

CasTILLo, O.; LEHMANN, H. (1977): The myoglobin of the fruit-bat (Rousettus aegypticus). Biochim. 
Biophys. Acta 492, 232-236. 


Mammalian phylogeny and protein sequence analysıs 275 


DayHorr, M. O. (ed.) (1972): Atlas of protein sequence and structure. Vol. 5. Silver Spring, 
Maryland: National Biomedical Research Foundation. pp. 1-124. 

DE Jong, W. W.; TERWINDT, E. C. (1976): The amıno acid sequences of the a-crystallin A chains of 
red kangaroo and virginia opossum. Eur. J. Biochem. 67, 503-510. 

DE Jong, W. W.; GLEAVvES, J. T.; BOULTER, D. (1977): Evolutionary changes of a-crystallin and the 
phylogeny of mammalıan orders. J. Mol. Evol. 10, 123-135. 

DE Jong, W. W.; ZwEERS, A.; GOoODMAn, M. (1980): Trends in the molecular evolution of a- 
erystallin. Prot. Biol. Fluids, 28th Colloq. Oxford: Pergamon Press. pp. 161-164. 

DE Jong, W. W.; ZwEErs, A.; JoysEy, K. A.; GLEAvES, ]J. T.; BOULTER, D. (1982): Protein sequence 
analysis applied to xenarthran and pholidote phylogeny. In: The evolution and ecology of sloths, 
anteaters, and armadıllos. Ed. by G. G. MONTGoMERY. Washington, D. C.: Smithsonian Institu- 
tion Press (in press). 

Emry, R. J. (1970): A North American Oligocene pangolın and other additions to the Pholidota. Bull. 
Amer. Mus. Nat. Hist. 142, 459-510. 

ENGELMANN, G. F. (1982): The phylogeny of the Xenarthra. In: The evolution and ecology of sloths, 
anteaters, and armadıllos. Ed. by G. G. MONTGOMERY. Washington, D. C.: Smithsonian Institu- 
tion Press (in press). 

FARRIS, J. S. (1972): Estimating phylogenetic trees from distance matrices. Amer. Nat. 106, 645-668. 

Fırcan, W. M. (1977): On the problem of discovering the most parsimonious tree. Amer. Nat. 111, 
223257. 

— W.M. (1979): Cautionary remarks on using gene expression events in parsimony procedures. Syst. 
Zool. 28, 375-379. 

FırcH, W. M.; MARGOLIASH, E. (1967): The construction of phylogenetic trees. Science 155, 279-284. 

Fouıps, L. R.; PEnny, D.; Henpy, M. D. (1979): A general approach to proving the minimalıty of 
phylogenetic trees illustrated by an example with a set of 23 vertebrates. J. Mol. Evol. 13, 151-166. 

Grass, B. P. (1982): History of classification and nomenclature in Edentata (= Xenarthra). In: The 
evolution and ecology of sloths, anteaters, and armadıllos. Ed. by G. G. MONTGOMERY. Washing- 
ton, D. C.: Smithsonian Institution Press (in press). 

GoopMan, M. (1981): Decoding the pattern of protein evolution. Progress in Biophysics and 
Molecular Biology 37, 105-164. 

GOODMAn, M.; MOORE, G. W.; BARNABASs, J.; MATSUDA, G. (1974): The phylogeny of human globin 
genes investigated by the maximum parsımony method. J. Mol. Evol. 3, 1-48. 

GOODMAN, M.; CZELUSNIAK, J.; MOORE, G. W.; ROMERO-HERRERA, A. E.; MATSUDA, G. (1979): 
Fitting the gene lineage into its species lineage, a parsimony strategy illustrated by cladograms 
constructed from globin sequences. Syst. Zool. 28, 132-163. 

GREGoRY, W.K. (1910): The orders of mammals. Bull. Amer. Mus. Nat. Hist. 27, 1-524. 

Henpy, M. D.; Penny, D.; Fouıos, L. R. (1978): Identification of phylogenetic trees of minimal 
length. J. Theor. Biol. 71, 441-452. 

HOFFSTETTER, R. (1973): Origine, comprehension et signification des taxons de rang sup£rieur: 
quelques enseignements tires de l’histoire des mammiferes. Ann. Paleont. 59, 137-169. 

Hoımauist, R. (1978): A measure of the denseness of a phylogenetic network. J. Mol. Evol. 11, 
225-231. 

HoMBRADOSs, I.; NEUZIL, L.; DEBUIRE, B.; Han, K. (1978): The amino acıd sequence of the a chain of 
badger (Meles meles) haemoglobin. Biochim. Biophys. Acta 535, 1-10. 

Lavocar, R. (1978): Rodentia and Lagomorpha. In: Evolution of African mammals. Ed. by V. ]. 
Macuio and H. B. S. Cooke, Cambridge, Mass.; Harvard Univ. Press. pp. 69-89. 

LE Gros CLARK, W. E.; SONNTAG, CH. F. (1926): A monograph of Orycteropus afer. III. The skull, 
skeleton of the trunk and limbs. Proc. Zool. Soc. Lond. 1926: I- II, 445-485. 

LILLEGRAVEN, J. A. (1969): Latest Cretaceous mammals of upper part of Edmonton Formation of 
Alberta, Canada, and review of marsupial-placental dichotomy in mammalian evolution. Univ. 
Kansas Paleont. Contr., art. 50 (vert. 12), 1-122. 

LuckETT, W.P. (ed.) (1980): Comparative biology and evolutionary relationships of tree shrews. New 
York; Plenum Press. 

MAEDA, N.; FircH, W. M. (1981): Amino acid sequence of a myoglobin from lace monitor lizard, 
Varanus varius, and its evolutionary implications. J. Biol. Chem. 256, 4301-4309. 

MARSHALL, L. G. (1979): Evolution of metatherian and eutherian mammalıan characters: a review 
based on cladistic methodology. Zool. J. Linn. Soc. 66, 369410. 

McKenna, M. C. (1969): The orıgin and early differentiation of therian mammals. Ann. N. Y. Acad. 
Scı. 167, 217-240. 

— (1975): Toward a phylogenetic classification of the Mammalıa. In: Phylogeny of the primates: a 
multidisciplinary approach. Ed. by W. P. LucketT and F. S. Szaray. New York; Plenum Press. 

21-46. 

IEREN. M. C.; Mannıng, E. (1977): Affinities and palaeobiogeographic significance of the 
Mongolian paleocene genus Phenacolophus. Geobios Mem. spec. 1, 61-85. 

MCLAREN, I. A. (1960): Are the Pinnipedia biphyletic? Syst. Zool. 9, 18-28. 


276 W. W. de Jong and M. Goodman 


MOORE, G. W.; BARNABAS, J.; GooDMAn, M. (1973): A method for constructing maximum 
parsimony ancestral amino acıd sequences on a given network. J. Theor. Biol. 38, 459485. 

O’NeEır, P. B.; DooLITTLE, R. F. (1974): Mammalıan phylogeny based on fibrinopeptide amino acid 
sequences. Syst. Zool. 22, 590-595. 

PATTERson, B. (1978): Pholidota and Tubulidentata. In: Evolution of African mammals. Ed. by N. ]. 
Macuiıo and H. B. $. CookE. Cambridge, Mass.: Harvard Univ. Press. pp. 268-278. 

PEAcock, D.; BOULTER, D. (1975): Use of amino acid sequence data in phylogeny and evaluation of 
methods using computer simulation. J. Mol. Biol. 95, 513-527. 

PRAGER, E. M.; Wırson, A. C. (1978): Construction of phylogenetic trees for proteins and nucleic 
acids: empirical evaluation of alternative matrix methods. J. Mol. Evol. 11, 129-142. 

ROMER, A. $. (1966): Vertebrate paleontology. 3rd ed. University of Chicago Press. 

ROMERO-HERRERA, A. E.; LEHMANN, H.; JOYsEv, K. A.; FrıDay, A. E. (1978): On the evolution of 
myoglobin. Phil. Trans. Roy. Soc. B 283, 61-163. 

Rose, K. D. (1978): A new paleocene epoicotheriid (Mammalia), with comments on the Palae- 
anodonta. J. Paleont. 52, 658-674. 

SARICH, V. M. (1982): Edentate systematics: albumin immunological evidence. In: The evolution and 
ecology of sloths, anteaters, and armadillos. Ed. by G. G. MONTGOMERy. Washington, D. C.; 
Smithsonian Institution Press (in press). 

SARICH, V. M.; CronIn, J. E. (1976): Molecular systematics of the primates. In: Molecular 
anthropology. Ed. by M. Goopman and R. TasHıan. New York: Plenum Press. pp. 141-170. 

SAVAGE, R. J. G. (1957): The anatomy of Potamotherium, an Oligocene lutrine. Proc. Zool. Soc. 
Lond. 129, 151-244. 

SHOSHANI, J.; GOODMAnN, M.; PrRYycHopDKo, W. (1978): Clasdistic analysıs of Paenungulata by 
computer (Abstract). Amer. Zool. 18, 601. 

Sımpson, G. G. (1945): Principles of classification and a classification of mammals. Bull. Amer. Mus. 
Nat. Hist. 85, 1-350. 

SokAL, R. R.; MICHENER, C. D. (1958): A statistical method for evaluating systematic relationships. 
Univ. Kansas Scı. Bull. 38, 1409-1438. 

SzaLay, F. S. (1977): Phylogenetic relationships and a classification of eutherian mammals. In: Major 
patterns in vertebrate evolution. Ed. by. M. K. HEcHT, P. C. Goopy and B. HEcHT. New York: 
NATO ASI Series, Plenum Press. pp. 315-374. 

TEDFORD, R. H. (1976): Relationship of pinnipeds to other carnıvores (Mammalıa). Syst. Zool. 25, 
363-374. 

THEnIUs, E. (1969): Phylogenie der Mammalıa. Berlin; Walter de Gruyter. 

THomPpson, R. B.; BORDEN, D.; TARR, G. E.; MARGOLIASH, E. (1978): Heterogeneity of amino acıd 
sequence ın hippopotamus cytochrome c. J. Biol. Chem. 253, 8957-8961. 

Van DRUTEN, H. A. M.; PEER, N. G. M.; Bos, F. A. B. H.; De Jong, W. W. (1978): Reciprocal 
amino acid substitutions in the evolution of homologous peptides. J. Theor. Biol. 73, 549-561. 

Van VALEN, L. (1971): Adaptive zones and the orders of mammals. Evolution 25, 420-428. 

Weıtz, B. (1953): Serological relationships of hyrax and elephant. Nature 171, 261. 


Authors’ adresses: Dr. WILFRIED W. DE Jong, Laboratorıum voor Biochemie, Geert Grooteplein 
Noord 21, Universiteit van Nijmegen, 6525 EZ Nijmegen, The Netherlands, and 
Prof. MORRIS GOODMAN, Department of Anatomy, Wayne State University 
School of Medicine, 540 East Canfıeld Avenue, Detroit, Michigan 48201, USA 


Lernversuche mit Maulwürfen (Talpa europaea L.) unter 
Anwendung einer speziellen Labyrinthmethode 


Von KrISTINA JOHANNESSON-GROSsS und H. Gross 


Fachbereich Biologie, Abteilung für Didaktik der Biologie, Universität Kassel, und Abteilung für 
Verhaltensbiologie, Institut für Allgemeine Zoologie, Freie Universität Berlin 


Eingang des Ms. 17. 3. 1982 
Abstract 


Learning experiments with moles (Talpa europaea L.) applying a special maze method 


Constructed was a transparent tube maze with a multiple choice of combinations, particularly adapted 
to the living conditions of Talpa europaea. Learning experiments were carried outina Y- and T-maze 
(discrimination of left from right, transfer, and memory). The results were shown in quantity in 
learning curves. It became evident that the mean learning curve (n = 5) of 10 tests per anımal and day 
in the Y-maze reached a level of 96 % of correct runs after 20 days. After changing the angle in order 
to construct a I-maze the moles reacted with an increased number of incorrect runs. Under the 
changed circumstances they have to learn again. This learnıng process, however, starts on a higher 
level of performance, the former level of correct runs being reached quickly. After a period of 41 days 
without training the existence of memory can still be proved. 


Einleitung 


In der Vergangenheit stand der Maulwurf wiederholt im Mittelpunkt experimenteller und 
systematisch beobachtender Untersuchungen. Durch sie ıst das Wissen über viele Aspekte 
der Biologie von Talpa europaea wesentlich erweitert worden (Literaturzusamenstellung 
bei GoDFREY und CRoWcROoFT 1960; QuILLIAM 1966; MELLANBY 1974). Dennoch weisen 
unsere Kenntnisse auf einigen Gebieten Lücken auf. So hat man sich bislang nur wenig mit 
dem Lernverhalten dieser Tiere beschäftigt; erste Ansätze hierzu finden sich bei KrıszAT 
(1940b), GoDET (1951), Lunp und Lunn (1965) sowie CHRZANOWSKI (Zit. in RENSCH 
1973). Erklärlich ist dies einerseits aus der, einer direkten Beobachtung nur schwer 
zugänglichen unterirdischen Lebensweise, andererseits aber auch aus den Problemen, die 
einer erfolgreichen Haltung der Tiere im Wege standen. Es war nicht möglich Maulwürfe 
längere Zeit in Gefangenschaft am Leben zu erhalten (SCHAERFENBERG 1939 u.a.). Erst in 
den letzten Jahren konnten die Haltungsbedingungen tiergerechter gestaltet werden 
(SKOCZEN 1961; RuDGE 1966; Rozmus 1973). Durch die Arbeiten von MÜHLBAUER und 
WITTE (1978) und WıTTE (1981) ist ein Haltungssystem in Form eines Kunstbaues 
entwickelt worden, in dem Maulwürfe mehrere Jahre gesund überleben. Hierdurch 
wurden die Grundlagen für Langzeitversuche mit diesen Tieren geschaffen. 

Ziel der vorliegenden Untersuchung war es, aufbauend auf den Erfahrungen zur 
Maulwurfshaltung von MÜHLBAUER (1978) und WITTE (1981), eine maulwurfsadäquate 
Lernmethode zu entwickeln, diese in Seitendressurexperimenten zu erproben und erste 
quantitative Ergebnisse in Form von Lernkurven zum Thema Seitenlernen, Umdressur 
und Gedächtnis beim Maulwurf zu gewinnen. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4705-0277 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 277-282 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Kristina Johannesson-Gross and H. Gross 


278 


ıyseuradjne Juawa[a[aız (wur ur uadunssawqy) "uaNasumg ap ne yoırq :,p “2 “aypıspny 
um odrualarp pa 3unuyojsq.Aonn,] I9P U9I9TQqUV wınz INN "UISSOTUISI9A :p-e "swo3sAsyruriAge] uadrunIoJusiygN sap YU9uls[9J9Zur7 Mh "aqaV 
sone’ SYUII pusIyea puIs 3u9weT>9pu pun -J9TZ un usdunuyg) U9sJ013 I 
‘uawsfa11odsue1], pun -uU9sneT Ju yurskge]-X sep ne 3y>IsInWV c "aaV 
BunJsyypy4aqaıyaS 
BunuyojJag An) BunuyJo yıw 97} Dpıdsbunpuimg4aN 


= Fr 
F-h m: 
u:: 


w 4 

je} o Pa 

en o = 

3 | a 

o 

3 | 2 

vo o + 

=) o ul 

- o ® 

Fer: oO 3 
EE——H a v 49Q91UIS: 

— 


aqnpıyas 

=sBunpuig 
=4o\ IN} 
Bunuyog... 


(57) 
sıyoy" 


CEEE |EEEEZZERTE 
.n 
[DI 


® 


Lernversuche mit Maulwürfen 279 


Material und Methode 


Der Maulwurf lebt überwiegend unterirdisch in einem selbstgegrabenen Gangsystem. An diese 
Umwelt ist er in seinen Orientierungsleistungen angepaßt (Krıszart 1940a, c). Es liegt daher nahe, eine 
Lerndisposition (vgl. BUCHHOLTZ 1973) zur Raumorientierung anzunehmen. Ein Kunstgangsystem in 
Form eines Röhrenlabyrinthes erscheint deshalb als Lernapparatur besonders geeignet. 


Lernapparatur 


Als Material für das Kunstgangsystem wurden Plexiglasplatten von 2 mm Stärke und durchsichtige 
PVC-Röhren mit einem Innendurchmesser von 45 mm und einer Wandstärke von 2 mm verwendet. 
Das gesamte System besteht aus Einzelelementen, die baukastenartig durch Schraubverbindungen zu 
einem Labyrinth gewünschter Form und Größe zusammengesetzt werden können. Unter den 
Einzelelementen lassen sich 3 verschiedene Typen unterscheiden: Grundelemente (Abb. Ic, c’); 
Winkelelemente, ausgebildet als Y-Stück (Abb. 1a) oder T-Stück (Abb. 1b) und Verschlußelemente 
(Abb. 1d, d’). Letztere können auf beiden Seiten durch einen Schieber geschlossen werden. Entspre- 
chend ihrer Lage und Funktion werden die Verschlußelemente als Start-, End-, Ziel-, Pausen- und 
Transportelement bezeichnet (Abb. 2). 


Versuchsdurchführung 


Das Zähmen der Maulwürfe erwies sich als problemlos (vgl. CRAnBROOoK 1966), alle Tiere waren 
deshalb zu Versuchsbeginn handzahm. Trotzdem wurden sie nicht per Hand, sondern mit Hilfe eines 
Transportelementes aus der Käfiganlage, in der sie sich außerhalb der Versuchszeit aufhalten (WITTE 
1981), abgeholt und ın das Startelement des Labyrinthes eingeführt. Fehlerquellen durch unterschied- 
liches Anfassen der Versuchstiere (VT) können durch diese Verfahrensweise ausgeschlossen werden. 

Vor Beginn eines jeden Versuchsdurchganges befindet sich der Maulwurf in dem verschlossenen 
Startelement (Abb. 2). Nach Hochziehen des Schiebers läuft das Tier bis zur Verzweigungsstelle vor 
und entscheidet sich dort für eine der beiden Labyrinthseiten. Erreicht das VT das Zielelement, dann 
wird hinter ihm der Schieber eingesetzt, das Belohnungsloch geöffnet und eine Mehlkäferlarve als 
positive Verstärkung geboten. Die Futterbelohnung erfolgt nur im Zielelement, die Öffnung im 
Endelement (Abb. 2) ist deshalb immer verschlossen. Falschläufe werden nicht zusätzlich bestraft. 
Sobald die Belohnung gefressen ist, wird dem VT die Möglichkeit zum Übertritt in das Pausenelement 
gegeben. Dort bleibt es, bis es vor Beginn des nächsten Laufes erneut ın das Startelement gelassen 
wird; auch hierbei braucht der Maulwurf nicht angefaßt zu werden. Alle Tiere absolvieren pro Tag 10 
Versuchsdurchgänge. Um eine olfaktorische Orientierung auszuschließen, wird das Labyrinth nach 
jedem Durchgang mit reinem Alkohol ausgewischt. Auferdem erfolgt jedesmal eine Drehung der 
Lernapparatur nach Zufall, damit Außenreize zur Lösung der Aufgabe nicht genutzt werden können. 

Mit der dargestellten Methode wurden 5 labyrinthnaive Maulwürfe zunächst auf die linke Seite 
eines Y-Labyrinthes dressiert, danach wurde der Labyrinthwinkel durch Einsetzen eines T-Stückes 
verändert, und schließlich erfolgte die Überprüfung der Gedächtnisleistung nach einer Pause von 41 
Tagen. Als Lernparameter fand der prozentuale Anteil an Richtigläufen pro 10 Durchgängen sowie die 
Gesamtzeit für 10 Läufe Verwendung. 


Ergebnisse und Diskussion 
Seitendressur 


Alle 5 VT ließen sich erfolgreich auf die linke Seite eines Y-Labyrinthes dressieren. Die 
Mittelwertskurve der Richtigläufe (Abb. 3a) beginnt im Zufallsbereich, nähert sich am 11. 
Tag der Signifikanzgrenze (p<0,05: Prüfung einer Grundwahrscheinlichkeit nach KoL- 
LER 1969), überschreitet diese zwischen dem 13. und 17. Tag und erreicht mit dem 20. Tag 
ein Kannphasenniveau bei 96 % Richtigläufen. 

Diese relativ lange Lernphase steht im Widerspruch zu Befunden von KrıszarT (1940b), 
wonach ein Maulwurf bereits nach einmaligem richtigem Durchlaufen eines T-förmig 
verzweigten Glasrohres in der Lage sein soll, die beiden Seiten sicher zu unterscheiden. 
Diese Diskrepanz erklärt sich aus der speziellen Vorgehensweise von KrıszaT. Methodisch 
war sein Versuch so angelegt, daß in einem auf der Erdoberfläche befindlichem T-Rohr 


280 Kristina Johannesson-Gross and H. Gross 


[%) 


Richtigläufe 


(Imin.) 


Gesamtzeit 


Abb. 3. Mittelwertslernkurven von 5 (4) Maulwürfen für die Seitendressur im Y-Labyrinth, das 

Umlernen im T-Labyrinth und die Gedächtnistests (10 Läufe pro Tier und Tag). - - - - : Signifikanz- 

grenze (p<0,05); + bzw. —: belohnte bzw. nicht belohnte Seite; S: Start. Die eingezeichnete 

Standardabweichung dient allein zur graphischen Veranschaulichung der Streuung der Einzelwerte 
(vgl. Sachs 1978) 


Fleischstücke in geringem Abstand auslagen, die vom Maulwurf sukzessive aus dem Rohr 
geholt und rückwärts unter die Erde in seine Käfigkiste gezogen wurden. Mit jedem 
Durchgang verlängerte sich die zu durchlaufende Strecke. Eine olfaktorische Orientierung 
war für das Tier immer möglich. Der Maulwurf wendete sich deshalb am Entscheidungs- 
punkt sofort in den richtigen Schenkel. Als er danach im T-Rohr mit vertauschten Seiten 
wieder zur ursprünglichen Seite - nun also falsch - lief, wurde diese Reaktion als Beweis 
für eine erfolgte Seitendressur gewertet. Dieser Schluß aufgrund eines Einzelbefundes 
erscheint nach den hier vorgelegten Ergebnissen zweifelhaft. 


Umlernversuch 


Eine Veränderung des Labyrinthwinkels, sobald die Kannphase erreicht ist, wirft die Frage 
auf, ob die Tiere die veränderte Situation neu lernen müssen, oder ob sie im Sinne einer 
Transferleistung handeln können und u.U. den Weg von Anfang an richtig durchlaufen, 
wie es KOEHLER und Dinger (1970) in Labyrinthversuchen mit Hausmäusen heraus- 
fanden. 

Nach Anbieten eines T-Labyrinthes (Abb. 3a) kommt es anfänglich zu einem Absinken 
der Anzahl Richtigläufe, ohne daß aber am 23. bis 25. Tag das Niveau der Tage 1 bis 3 
erreicht wird (p = 0,01: Wilcoxon-Rang-Test). Bereits nach 4 Tagen wird die Signifikanz- 
grenze wieder überschritten, und nach 60 Läufen hat sich die Kannphase erneut eingestellt. 
Der Verlauf der Lernkurve (Abb. 3a) deutet darauf hin, daß die Maulwürfe den richtigen 
Weg ım Labyrinth mit abgewandeltem Winkel nicht spontan beherrschen; sie müssen sich 


Lernversuche mit Maulwürfen 281 


erst auf die veränderten Gegebenheiten einstellen und neu lernen. Dieser Lernvorgang setzt 
jedoch auf einer höheren Leistungsstufe ein und führt schnell zur neuen Kannphase. Die 
Tiere waren also ın der Lage, Information aus dem vorausgegangenen Lernprozeß auf die 
neue Situation zu übertragen, wenn auch nicht in solch einem Ausmaß, daß es ıhnen 
möglich gewesen wäre, das winkelveränderte Labyrinth sofort fehlerfrei zu durchlaufen. 


Gedächtnisleistung 


Nach 41 dressurfreien Tagen unterschreitet die Lernkurve der Richtigläufe sowohl für das 
Y-Labyrinth als auch für das T-Labyrinth jeweils nur an einem Tag die Signifikanzgrenze; 
der gesamte Kurvenverlauf bewegt sich ansonsten oberhalb dieser Grenze (Abb. 3a). Auch 
die Mittelwertskurve der Gesamtzeit pendelt sich nach der Dressurpause rasch auf einem 
abszissenparallelen Niveau ein (Abb. 3b). Noch nach einer Pause von fast 6 Wochen läßt 
sich demnach das Bestehen eines Gedächtnisses nachweisen. 


Danksagung 
Herrn Prof. Dr. G. R. WıTTE (Universität Kassel) möchte ich (K. J.-G.) für die Möglichkeit danken, 


in seinem Labor mit Maulwürfen arbeiten zu können. 


Zusammenfassung 


Zur Untersuchung des Lernverhaltens von Talpa europaea wird eine tieradäquate und vielfältig 
einsetzbare Versuchsapparatur in Form eines durchsichtigen Röhrenlabyrinthes vorgestellt und in 
ersten Experimenten erfolgreich erprobt. Die Ergebnisse von Seitendressuren, Umlernversuchen und 
Gedächtnistests in einem Y- und T-Labyrinth werden quantitativ als Lernkurven dargestellt. Es zeigt 
sich, daß bei 10 Durchgängen pro Tag ım Y-Labyrinth die Mittelwertslernkurve (n= 5) am 20. Tag 
ein Kannphasenniveau von 96% Richtigläufen erreicht. Nach Veränderung des Winkels zum T 
reagieren die Maulwürfe mit einer erhöhten Zahl an Falschläufen. Sie müssen die veränderte Situation 
neu lernen; dieser Lernvorgang setzt aber auf einer höheren Leistungsstufe ein und führt schnell zur 
neuen Kannphase. Das Bestehen eines Gedächtnisses läßt sich noch nach einer Dressurpause von 41 
Tagen nachweisen. 


Literatur 


BUCHHOLTZ, C. (1973): Das Lernen bei Tieren. Verhaltensänderungen durch Erfahrung. Stuttgart: 
Gustav Fischer. 

CRANBROOKR, EARL OF (1966): In: Catching and keeping live moles. Ed. by Runge, A. ]. B. J. Zool. 
149, 42-45. 

GODFREY, G. K.; CROWCROFT, P. (1960): The life of the mole. London: Museum Press. 

KOEHLER, O.; DINGER, W. (1970): Orientierungsvermögen bei Mäusen. Versuche im Hochlabyrinth. 
Begleitveröffentlichung zum Film B 635. Göttingen: Inst. Wiss. Film. 

KoLLEr, $. (1969): Graphische Tafeln zur Beurteilung statistischer Zahlen. 4. Aufl. Darmstadt: 
Steinkopf. 

Krıszat, G. (1940a): Untersuchungen zur Sinnesphysiologie, Biologie und Umwelt des Maulwurfs 
(Talpa europaea L.). Z. Morphol. Ökol. Tiere 36, 446-511. 

KRIsZAT, G. (1940b): Die Orientierung im Raume bei Talpa europaea. Z. Morphol. Ökol. Tiere 36, 
512-556. 

KrıszaT, G. (1940c): Wie orientiert sich der Maulwurf in seinen Gängen? Umschau 44, 561-565. 

Lunp, R. D.; Lunp, J. S. (1965): The visual system of the mole, Talpa europaea. Exptl. Neurol. 13, 
302-316. 

MELLANBY, K. (1974): The mole. London: Collins. 

MÜHLBAUER, $. (1978): Entwicklung und Erprobung unterschiedlicher Käfigtypen für Langzeitversu- 
che mit Talpa europaea L. Neue Erkenntnisse aus der Käfighaltung. Staatsexamensarbeit: Univer- 
sıtät Kassel. 

MÜHLBAUER, $.; WITTE, G.R. (1978): Beiträge zur Käfighaltung von Maulwürfen (Talpa europaea 
L.). I. Käfigtypen und Pflegemaßnahmen. Philippia 3, 423442. 

QuiLLiam, T. A. (Hrsg.) (1966): The mole: its adaptıon to an underground environment. J. Zool. 149, 
31-114. 

ReEnscH, B. (1973): Gedächtnis, Begriffsbildung und Planhandlungen bei Tieren. Berlin und Ham- 
burg: Paul Parey. 

Rozmus, S. (1973): Keeping the mole (Talpa europaea Linnaeus, 1758) in captivity. Acta Theriol. 18, 
489-4. 


282 Kristina Johannesson-Gross and H. Gross 


Runge, A. J. B. (1966): Catching and keeping live moles. J. Zool. 149, 42-45. 

Sachs, L. (1978): Angewandte Statistik. Statistische Methoden und ihre Anwendungen. 5. Aufl. 
Berlin, Heidelberg, New York: Springer. 

SCHAERFENBERG, B. (1939): Haltung und Pflege des Maulwurfs (Talpa europaea L.). Zool. Garten N. 
F. 11, 107-109. 

SKOCZEN, S. (1961): A new keeping arrangement for the mole (Talpa europaea Linnaeus, 1758), in 
captivity. Acta. Theriol. 5, 287-289. 

WITTE, G. R. (1981): Erfahrungen mit der Käfighaltung von Maulwürfen (Talpa europaea L.). Zool. 
Garten N. F. 51, 193-215. 


Anschriften der Verfasser: KrısTINA JOHANNESSON-GROSs, Abt. für Didaktik der Biologie, Universi- 
tät Kassel, Fachbereich Biologie, Postfach 10 13 80, D-3500 Kassel und Dr. 
HERBERT GROoss, Abt. für Verhaltensbiologie, Freie Universität Berlin, 
Institut für Allgemeine Zoologie, Haderslebener Str. 9, D-1000 Berlin 41 


Häufigkeitsschwankungen der Rötelmaus 
(Clethrionomys glareolus Schreber, 1780) 


Von V. WENDLAND 


Eingang des Ms. 15. 1. 1982 
Abstract 
Fluctuations of Bank vole (Clethrionomys glareolus Schreber, 1780) 


Investigated the fluctuations of Bank voles in three forest areas of West Berlin during 28 years 
(1952-1979) by analyzing owl pellets. A total of 43535 small vertebrates from pellets of Tawny owls 
and Long-eared owls were determined. It has been established that the Bank vole in contrast with 
three other species of mice investigated by the author in the same woodlands did not display any cyclic 
fluctuations of density. 


Einleitung 


Im Gegensatz zu Dichteschwankungen innerhalb eines Jahres sind solche über längere 
Perioden selten oder nur unvollkommen erfaßt worden. Alle solche Untersuchungen 
wurden mit Hilfe von Tot- und Lebendfallen durchgeführt. Es fehlen dagegen fast 
vollständig langfristige, d.h. sich über eine Reihe von Jahren erstreckende Untersuchungen 
über Fluktuationen der jährlichen Bestandsdichte. Meine beiden, dieses Thema behandeln- 
den Arbeiten (WENDLAND 1975, 1981) untersuchen die langfristigen, bei manchen Arten 
sich über fast drei Jahrzehnte erstreckenden Dichteschwankungen. 


Material und Methode 


In den beiden oben erwähnten Arbeiten wurden Populationsuntersuchungen zum erstenmal nicht 
durch Fallenfänge, sondern durch Analyse von Eulengewöllen durchgeführt. Im Laufe der Untersu- 
chungen stellte sich heraus, daß bei den häufiger vorkommenden Beutetierarten der jährliche Anteil an 
der Gesamtbeute der betreffenden Beutegreifer (in diesem Fall Strix aluco und Asio otus) jährlichen 
Schwankungen unterliegt, und zwar in der Weise, daß dieser %-Anteil nach einer bei jeder Art sich 
stets gleichbleibenden Reihe von Jahren regelmäßig einen Höhepunkt erreicht, um in den nächsten 
Jahren wieder bis ungefähr auf die Hälfte zurückzugehen. Diese Fluktuationen laufen also zyklisch 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4705-0282 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 282-286 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


} 


Häufigkeitsschwankungen der Rötelmaus 283 


ab. So fand ich, daß die Gelbhalsmaus (Apodemus flavicollis) alle drei Jahre einen Höhepunkt in der 
Beuteliste erreicht, die Feldmaus (Microtus arvalıs) alle vier Jahre und die Erdmaus (Microtus agrestis) 
alle fünf Jahre (WEnDLAnD 1981). Wie aus diesen Ergebnissen der Schluß gezogen werden darf, daß 
diese Kurven auch den Verlauf der Populationsentwicklung wiedergeben, die in für jede Art typischen 
regelmäßigen Zyklen abläuft, ist in beiden Arbeiten dargelegt. Wieweit jedoch die Schwankungen den 
wirklichen Veränderungen der Bestandsdichte auch proportional nahekommen, ist eine andere Frage. 
Vor allem bei der Gelbhalsmaus sprechen mehrere schwerwiegende Gründe dafür, daß die Schwan- 
kungen der Kurven tatsächlich den realen Fluktuationen der Bestandsdichte proportional nahezu- 
kommen scheinen. Für die Feldmaus wird in einem Falle dasselbe angenommen (WENDLAND 1981). 

Von 1952 bis 1979 habe ich in Berlin-West in den Revieren brütender Waldkäuze und Waldohreu- 
len regelmäßig Gewölle gesammelt und darin insgesamt 43 535 kleine Wirbeltiere bestimmt; etwa zwei 
Drittel stammten aus drei Waldgebieten. 

a. Forst Berlin Grunewald (etwa 31 km?). Der Grunewald ist ziemlich trocken. Ungefähr 65 % der 
Bäume sind Kiefern (Pinus silvestris), der Rest Laubholz. Der Wald enthält viel Unterholz und ist zum 
größten Teil stark zugewachsen, wodurch den Waldkäuzen die Jagd sehr erschwert wird. Auf drei 
Seiten ist er vom Häusermeer West-Berlins umschlossen. An der vierten Seite, der Westseite, grenzt er 
an die 500 bis 900 m breiten von der Havel durchflossenen Seen. Einige ziemlich ausgetrocknete 
Waldmoore und mehrere kleine Seen liegen innerhalb des Grunewalds. 

b. Forst Berlin-Spandau (etwa 15 km?). Dieser Wald im Nordwesten von Westberlin grenzt 
unmittelbar an die DDR. Er ist feuchter als der Grunewald und enthält einige größere Waldmoore 
sowie einen bachähnlichen Wasserlauf. 

c. Pfaueninsel (68 ha). Eine Insel im Seengebiet der Havel. Waldstücke mit altem Baumbestand, 
darunter Eichen, Platanen, Kiefern und vielen ausländischen Bäumen, wechseln mit freien Flächen, 
Wiesen, Gärten und Heide. 


Ergebnisse 


Im Grunewald brüten Waldkäuze und Waldohreulen, in Spandau und auf der Pfaueninsel 
nur Waldkäuze. Beim Waldkauz wurden festgestellt (Tab. 1): 


Tabelle 1 


Anzahl gefundener Beutetiere (n) und Prozentsatz darin festgestellter Clethrionomys glareolus 
(%) in den Jahren 1952 bis 1979 in Berlin-West in Gewöllen brütender Eulen 


1953 1954 1955 1956 1957 1958 1960 1961 


a Waldkauz, Grunewald; b Waldkauz, Spandau; c Waldkauz, Pfaueninsel; d Waldohreule, 
Grunewald 


284 V. Wendland 


a. Im Grunewald von 1959 bis 1979 unter 17985 kleinen Wirbeltieren 758 (4,2 %) 
Rötelmäuse. Nach meinen Fallenfängen ist die Siedlungsdichte der Rötelmaus am 
größten ın den hiesigen jüngeren Eichenbeständen, sobald sie Früchte tragen. Auch die 
Gelbhalsmaus erreicht hier eine sehr hohe Dichte. In den Grunewaldmooren ist 
Clethrionomys glareolus der häufigste Kleinsäuger. 

b. Im feuchteren Waldgebiet von Spandau wurden von 1963 bis 1979 unter 13564 
Wirbeltieren 1825 (13,5 %) Rötelmäuse gefunden, also weit mehr als im Grunewald. 

c. Drei Waldkauzpaare auf der Pfaueninsel lieferten von 1964 bis 1979 unter 1799 
Beutetieren 425 (19,8 %) Rötelmäuse. 

d. Von den Waldohreulen im Grunewald stammen aus den Jahren 1952 bis 1961 3018 
Beutetiere, darunter nur 38 (1,3%) Rötelmäuse. Abb. 1 zeigt die Schwankungen der 
Rötelmausanteile in den verschiedenen Jahren aus folgenden Fundserien: a bis c. Kurve 
d wurde nicht abgebildet, da die Rötelmausanteile wegen der geringen Zahlen stark 
zufällig sein können, doch sind die Anteile in Tabelle 1 angeführt. 


Abb. 1. Anteile von Clethrionomys in Ge- 
wöllen des Waldkauzes (Strix aluco) in drei 


b 
| NY 

54 SDR N Waldgebieten in Berlin-West in verschiede- 
| \ nen Jahren. a: Grunewald; b: Spandau; c: 


Pfaueninsel. (Werte s. Tab. 1) 


Die drei Kurven zeigen Spitzen ın den folgenden Jahren: 
a2 1959, 1964,21968,.197155197777 
b. 1965,.1968, 19701977 
c2 21964, 1968, 197071977, 
Die Abstände der Höhepunktjahre (2, 3, 4, 5, 6 und 7 Jahre) sind sehr unterschiedlich 
und regellos verteilt, ein Zyklus ist nicht zu erkennen. 


Diskussion 


Die unregelmäßigen Schwankungen der Rötelmausanteile stehen im Gegensatz zu den 
zyklischen Häufigkeitsänderungen anderer Nager im gleichen Gebiet (Microtus arvalıs, 
Microtus agrestis). Besonders deutlich treten sie in Erscheinung bei dem häufigsten 
Beutetier der Waldkäuze, der Gelbhalsmaus (Apodemus flavicollis). In 28 Jahren wurden 
im Grunewald zwischen den Spitzen der Anteile der Gelbhalsmaus siebenmal drei Jahre 
und einmal vier Jahre beobachtet (Abb. 2 und Wenprann 1975, 1981). Jahre höchster 
Anteile waren 1955, 1958, 1961, 1964, 1967, 1970, 1974 und 1977. Zwei Spitzenjahre (1964, 
1977) fallen mit solchen der Rötelmaus im gleichen Gebiet zusammen, und 1970 erreichte 
die Rötelmaus zumindest in Spandau und auf der Pfaueninsel einen Gipfel. Demnach sieht 
es nicht so aus, als seien die Anteile der Rötelmaus in den Gewöllen von denen der 
Gelbhalsmaus abhängig oder als unterdrücke eine hohe Dichte der Gelbhalsmaus die 
Entwicklung der Rötelmaus (J. NIETHAMMER, brieflich). 

Zwei Jahre heben sich heraus, wo in allen drei Gebieten die Höhepunktjahre zusam- 
menfallen: 1968 und 1977. In zwei anderen Jahren stimmen die Höhepunkte wenigstens 
zweier Gebiete überein: 1970 in Spandau und auf der Pfaueninsel sowie 1964 im Grune- 


Häufigkeitsschwankungen der Rötelmaus 285 


wald und auf der Pfaueninsel. Die Tiefpunkte fallen nur einmal in allen drei Gebieten 
zusammen (1969). 1978 ıst ein gemeinsames Tiefpunktjahr in Spandau und auf der 
Pfaueninsel. 

Zusammenfassend kann gesagt werden: Es gibt wohl gemeinsame Höhe- und Tief- 
punktjahre, die auf einen gemeinsamen Verlauf in den drei Gebieten schließen lassen, aber 
ein Zyklus läßt sich nicht erkennen. 

Erschwerend für die Deutung der Populationsdynamik der Rötelmaus ist die Tatsache, 
daß diese Art auch tagaktıv ist, was man namentlich im Frühjahr und Sommer nicht selten 
beobachten kann. Es wäre daher denkbar, daß sie nachts längere Ruhepausen einlegen 


Apodemus |% 


\ 


Abb. 2. Anteile der Gelbhalsmaus (Apode- \ N I IN | 130 
| So \ / \ / \ y | 
Vaneo 
\/ 


mus flavicollis — rechte Ordinate) und der 
Rötelmaus (Clethrionomys glareolus - linke „1% | 
Ordinate) in Gewöllen des Waldkauzes (Strıx = 
aluco) im Grunewald in Berlin-West von | N If 
1959 bis 1979. (Werte für Clethrionomys s. " en, \ ERS I 
Tab. 1, c, für Apodemus WENDLAND 1975, | Clethrionomys — 

1981) ne 


muß, so daß sie in der Beuteliste einer Eule unterrepräsentiert erscheinen könnte. ToDo- 
ROVIG etal. (1966) haben vier Jahre lang von Juli bis Anfang September die Aktivitätsinten- 
sität der beiden Arten Apodemus flavicollis und Clethrionomys glareolus, die beide den 
gleichen Standort bewohnen, untersucht. Die Fallen wurden alle zwei Stunden kontrol- 
liert. Dabei wurde festgestellt, daß die Rötelmaus im Laufe der 24 Stunden eine Veränder- 
lichkeit ihrer Aktivitätsintensität zeigt, die sowohl von ihrer eigenen Dichte als auch vor 
allem von der jeweiligen Bestandsdichte der Gelbhalsmaus, mit der sie am gleichen 
Standort lebt, abhängig ist. „Obwohl die Rötelmaus 1964 nur recht spärlich vorkam, war 
doch zu erkennen, daß sich ihre Aktivität vor und nach der dominierenden Aktivität der 
sehr zahlreichen Population von Apodemus flavicollis abspielte.‘“ Und weiter heißt es: 
„Clethrionomys glareolus paßt also den Rhythmus ihrer Aktivität der jeweiligen An- und 
Abwesenheit von Apodemus flavicollis an.‘“ Aufgrund dieser Feststellung müßte man 
annehmen, daß eine solche interspezifische Konkurrenz auch ım Ablauf der Kurven zum 
Ausdruck kommt: In den Jahren eines Tiefpunkts der Gelbhalsmausdichte würden die 
Rötelmäuse in größerer Menge von den Waldkäuzen erbeutet werden, und die Kurven 
müßten reziprok (gegenläufig) zueinander verlaufen. Das ist jedoch, wie wir bereits 
gesehen haben, nicht der Fall. 

Daß sich die Rötelmaus in ihrer Dynamik deutlich von anderen Nagerarten unterschei- 
det, wird von mehreren Forschern betont. SOUTHERN (1970), der seine Mäuse- und 
Waldkauzuntersuchungen bei Oxford durchgeführt hat, sagt, daß der Bestandswechsel der 
Rötelmaus unregelmäßiger sei als der der Waldmaus. Nach A. ScHmipr (1970), der in 
einem Gebiet 70 km südöstlich von Berlin beobachtete, ıst die Populationsdynamik der 
Rötelmaus komplizierter, als die der Gelbhalsmaus. Er stellte Minima in den Jahren 1964, 
1966 und 1968 fest sowie Maxima 1965 und 1967. Ein Vergleich seiner Daten mit denen 
meiner Untersuchungen ergab wenig Übereinstimmung, obwohl die beiden Gebiete ja 
nicht weit auseinanderliegen. GLiwicz (1980) fand auf einer Insel im Beldansee (Masuri- 
sche Seenplatte, Polen) einen regelmäßigen Zweijahreszyklus. Inselpopulationen unter- 
scheiden sich jedoch nach der Autorin von Festlandspopulationen durch schwächere 
Fluktuationen. Nach LinkoLA und MyııyMmärt (1969) sollen Rötelmaus und Erdmaus 
(Microtus agrestis) in Finnland synchron fluktuieren. Nach WIGER (1979) ist die Rötelmaus 


286 V. Wendland 


im nördlichen und mittleren Skandinavien eine zyklische Art, die alle drei bis vier Jahre 
einen Höhepunkt ihrer Dichte erreicht. Vom südlichen Schweden an schwankt dagegen die 
Dichte unregelmäßig. Soweit bekannt, fluktuiert also die Rötelmaus in Nordeuropa 
zyklisch, weiter südlich dagegen nicht. Diese Feststellung muß jedoch weiter geprüft 
werden, denn mir scheint die Beobachtungsdauer für das Erkennen langfristiger periodi- 
scher Zyklen oft zu kurz gewählt zu sein. Zur Vorsicht mahnt das Untersuchungsergebnis 
von GarsD und Howaro (1981), die die Frage der zyklischen Dichteschwankungen an 
einem Material von 4737 Wühlmäusen (Microtus californicus) statistisch prüften. Sie 
kommen zu dem Schluß, daß die meisten angeblich zyklischen Populationsschwankungen 


rein zufällig sind. 
Danksagung 


Herrn Prof. Dr. J. NIETHAMMER danke ich bestens für die Durchsicht des Manuskriptes und 
Verbesserungsvorschläge. 


Zusammenfassung 


In drei Waldgebieten von Berlin-West wurden die Rötelmausanteile (Clethrionomys glareolus) in 
Gewöllen von Waldohreule und Waldkauz über Zeiträume von 10, 16, 17 und 21 Jahren untersucht. 
Dazu wurden insgesamt 43535 kleine Wirbeltiere ausgezählt. Im Gegensatz zu drei anderen, im 
gleichen Gebiet von mir untersuchten Mäusearten, läßt die Rötelmaus hier keine zyklischen Bestands- 
schwankungen erkennen. Dies steht im Einklang mit den Befunden anderer Autoren, wonach die 
Rötelmaus im nördlichen und mittleren Skandinavien ın regelmäßigen Zyklen schwankt, weiter ım 
Süden dagegen nur unregelmäßige Dichteänderungen zeigt. 


Literatur 


GARSsD, A.; HowARD, E. (1981): A 19-year study of microtine population fluctuations using time- 
series analysis. Ecology 62, 930-937. 

Gwiwicz, J. (1980): Island populations of rodents: their organization and functioning. Biol. Rev. 55, 
109-138. 

LınkoLa, P.; Myırymäkı, A. (1969): Der Einfluß der Säugerfluktuationen auf das Brüten einiger 
kleinsäugerfressender Vögel ım südlichen Häme Mittelfinnland. Ornis Fennica 46, 45-78. 

SCHMIDT, A. (1975): Populationsdynamik und Ökologie der terrestrischen Kleinsäuger des Natur- 
schutzgebietes Schwarzberge. Naturschutz Berlin Brandenburg 11, 78-93. 

SOUTHERN, H. N. (1970): The natural control of a population of tawny owl-Strix aluco. J. Zool. 
(Lond.) 162, 197-285. 
Toporovi£, M.; Savı6, I.; Mikes, M. (1966): Ritam aktivnosti vrsta Apodemus flavicollis ı Cle- 
thrionomys glareolus u odnosu na brojnost njichovih populacija. Arch. Biol. Scı. 18, 33-34 P. 
WENDLAND, V. (1975): Dreijähriger Rhythmus im Bestandswechsel der Gelbhalsmaus Apodemus 
flavicollis. Oecologia (Berlin) 20, 301-310. 

— (1981): Cyclic population changes in three mouse species in the same woodland. Oecologia 
(Berlin) 48, 7-12. 

WIGER, R. (1979): Demography of a cyclic population of the bank vole Clethrionomys glareolus. 
Oikos 33, 373-385. 


Anschrift des Verfassers: Dr. Vıcror WENDLAND, Ringbahnstr. 79, D-1000 Berlin 42 


Reproductive biology of three species of gerbils 
(Genus Tatera) in East Africa 


By B.R. Near 
Department of Biology, University of Saskatchewan, Saskatoon, Saskatchewan, Canada 


Receipt of Ms. 17. 3. 1982 
Abstract 


The reproductive ecology of Tatera nıgricanda and T. robusta was studied in Meru National Park in 
central Kenya, and T. valıda was studied in the Ruwenzori National Park in western Uganda. In both 
areas seasonal changes in photoperiod were negligible, but there was a bimodal pattern of rainfall. 
Samples of 398 T. nigricanda, 61 T. robusta and 116 T. valıda, collected over periods of at least one 
year, provided information on seasonal breeding activity, litter size, population structure, and 
reproductive capacity. 

Tatera valıda only bred during the rains in western Uganda. Adult females had the capacity to 
produce two litters each year. The average litter size was 4.25 + 0.21, resulting in an average annual 
reproductive capacity of 8.5 young per adult female. 

Tatera nigrıcauda bred throughout most of the year ın central Kenya. Breeding was at amaxımum 
during the rains and immediately afterwards, then declined and ceased towards the end of the dry 
seasons. However, breeding began again before the onset of the rains. Young were only recruited into 
the population during the two months after each rainy season, suggesting that litters conceived during 
the dry seasons died before they were weaned. Adult females had the capacity to produce approxi- 
mately 5 litters each year. The average litter sıze was 5.2 # 0.24, and the annual reproductive capacıty 
was 25-28 young per adult female. 

Tatera robusta appeared to have a breeding periodicity sımilar to that of 7. nigricauda. The average 
litter size was 4.7 # 0.43. 

A comparison of the reproductive characteristics of different species of Tatera did not show any 
significant relationships between body mass, litter size, duration of breeding, and reproductive 
capacity. It is suggested that food may be the most important factor affecting the breeding pattern of 
this genus. 


Introduction 


There have been numerous studies on the breeding biology of members of the genus 
Tatera, both within Afrıca (ArzLanson 1958; DELAnY and NEAL 1969; GAUTUN 1975; 
HAnneyY 1965; HUBERT 1977; HUBERT and Adam 1975; MEAsRocH 1954; PIRLOT 1954; 
SHEPPE 1973; SHORTRIDGE 1934; SMITHERS 1971) and the Indian subcontinent (BEG and 
Ajmau 1977; CHANDRAHAS and KRISHNASWAMI 1974; Jaın 1970; PHıLLıps 1923; PRASAD 
1953, 1956, 1961). The Asıan studies have been compared, but there has been little attempt 
to interrelate the various studies on African Tatera. 

In this paper I document the breeding biology of Tatera nigricanda and present notes 
on the breeding biology of 7. robusta collected in Meru National Park ın central Kenya. To 
data there have been no detailed studies on either of these species. I also present additional 
information, to that previously published (Derany and NEAL 1969), on 7. valıda ın 
western Uganda. Finally, I review the available information on breeding season, body 
mass, litter size and annual reproductive capacity of both African and Asıan Tatera to 
determine what patterns, if any, exist. 

The nomenclature of the African species follows Davıs (1971), but because of disagree- 
ment on the status of some named forms (see, for example, MATTHEY and PETTER 1970) the 
type name is given as a subspecies in some cases. The nomenclature of the Asian types 
follows ELLERMAN and MORRISON-Scott (1951). 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4705-0287 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 287-296 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


288 B. R. Neal 


Material and methods 
Collection and classification of material 


A total of 116 T. valida was collected from the Crater Track region (0° 06°S, 29° 54’E) of the 
Ruwenzori National Park, Uganda; and 61 T. robusta and 398 T. nigricanda were collected from the 
Rojewero Plains (0° 11’N, 38° 10’E) and surrounding areas of the Meru National Park, Kenya. 
Descriptions of the study areas are provided by Near (1970, 1981) and AmEnT (1975). Both areas were 
sampled by standard trap lines, spaced approximately 50 m apart with the traps set at 10 pace intervals. 
Collections were made each month for periods of at least one year. 

Processing and classification of reproductive material was ıdentical to that described by NEAL 
(1981). Briefly, animals were measured, dissected, examıned macroscopically, and a sperm smear was 
taken from the cauda epididymides of the males, soon after capture. The mass of the various 
reproductive organs was measured after fixation ın Bouin’s solution, and the ovaries of the females 
collected in Kenya were also serially sectioned. Based on the examination of the reproductive organs 
the anımals were classed as immature or mature, and mature females were further classified as visibly 
pregnant, reproductively active, or reproductively inactive. 

Head-body length was used as a criterion of age. It was useful for separating the young, immature 
animals from the adults, but different cohorts could not be separated after the anımals reached adult 
size. 


Reproductive capacity 


I calculated the reproductive capacity (1. e. the potential annual production of young per adult female) 
of T. valida and T. nigricanda, and of other species for which information was available in the 
literature, using the method described by Near (1981). In this method the interval between litters is 
considered to be the sum of the non-pregnant period (p,), the pre-implantation period (pı), and the 
post-implantation period (p,) when embryos are macroscopically visible. The gestation period (G) is 
the sum of pı and p,. In most studies females with pre-ımplantation stages of pregnancy were not 
distinguished from non-pregnant adults. In these cases the ratio of p, + pjto p, was assumed to be the 
same as the ratio of non-visibly pregnant adults to visibly pregnant adults. However, MEASROCH 
(1954) detected all stages of pregnancy. In this case I assumed that the ratio of p, to G was the same as 
the ratio of nonpregnant adults to pregnant adults. 

These equations were solved for p, using estimates of G in the literature (BLAnD 1969; HUBBARD 
1970; HUBERT and Apam 1975; MEASROCH 1954), and assuming that pj equaled 4d (see NEAL 1981). 
For those species for which no estimate of gestation period was available a range of values (22-25d) 
was used. The mean interval between litters (p, + pı + p,) was then calculated, and the production of 
young was estimated for a given time period using the average litter size. However, the gestation 
period can be considerably lengthened due to delayed implantation in lactating females (BLanD 1969; 
MEaAsrocH 1954). In those studies where only visible pregnancies were detected the calculation of 
reproductive capacity is not influenced by delayed implantation, because the error in pı is exactly 
compensated by the error in p,. However, in the study of MEAsRocH (1954), ın which all stages of 
pregnancy are detected, the increased length of gestation must be calculated. This was determined 
from her data by assuming that the ratio of p; with delayed implantation to p; with no delayed 
implantation (4d) was equal to the ratio of the percentage of unimplanted embryos in pregnant 
lactating females to the percentage of unimplanted embryos in pregnant non-lactating females. 


Statistics 


Seasonal changes in body mass, mass of the testes and seminal vesicles, and litter size were analyzed by 
one way analysıs of varıance (SOoKAL and ROHLF 1969). 

Variation in litter size in relation to evicerated maternal body mass and season of pregnancy was 
investigated by analysis of covariance (SNEDECOR and CocHRAN 1967). The sample was divided into 
those collected during the rains and immediately afterwards (November to mid-January, mid-April to 
mid-June) and those collected during the dry periods. 

The interrelationships between body mass, litter size, reproductive capacity, and duration of 
breeding of the various species were analyzed by Kendall’s coefficient of rank correlation (SoKAL and 
RoHLr 1969). 


Results 
Body mass 


The body mass of embryos of the heaviest litters ranged from 3.6-3.8 g in 7. valıda and 
3.5-3.9 g in T. nigricanda. The birth mass is estimated to be approximately 4 g in both 


Reproductive biology of three species of gerbils in East Afrıca 289 


Table 1 


Adult body mass (g), mean and range (in parentheses), of various species of Tatera 


Body mass 


Species 
T. robusta 
T. nıgricanda 


T. lencogaster 
T. lencogaster 


T. afra 

T. brantsii 
T. brantsii 
T. valida 


Locality 
Kenya 
Kenya 
Zambia 


Botswana 
Okavanga 
Kasane 

S. Africa 
S. Africa 
Botswana 


Uganda 


Males 


91 (70-130) 
103 (80-152) 
115 (70-170) 
132 (80-195) 


Females 


Authority 


85 (70-100)! 
102 (80-130)? 
95 (70-120)! 
114 (80-161)? 


(51-106)? 


73 (50-96 ) 
58 (46-79 ) 
70 (50-121) 
81 (50-116) 
96 (82-125) 
113 (70-140) 
129 (90-160) 


71 (50-110)? 
54 (46-78 )2 
66 (45-90 )! 
74 (50-106)! 
89 (74-105)? 
92 (70-105)! 
113 (90-130)? 


\ 


This study 

This study 
SHEPPE (1973) 
SMITHERS (1971) 
AıLanson (1958) 
MEASROocH (1954) 


SMITHERS (1971) 
This study 


HUBERT and 
ADAM (1975) 
CHANDRAHAS and 
KRISHNASWAMI 
(1974) 

PRASAD 

(1956, 1961) 

Jaın (1970) 


Senegal (100-180)? 


T. v. gambiana 


T. indıica India 


hardwickei 


(80-200+)2 


T. 1. cuvieri India (90-210) (90-250)? 


T. i. indica India (100-213) (70-193)2 


! evicerated body mass; ? total body mass. 


species, compared to estimates of 4.5 g for 7. afra and 5.5 g for T. brantsii (MEASROCH 
1954), and a recorded birth mass of 4.4 g for 7. brantsı (MEESTER and HALLETT 1970) in 
South Africa. 

The smallest 7. valıda and T. nigricauda trapped had a body mass of approximately 30 g 
(evicerated mass 25 g) which is assumed to be the mass of weanlings. Animals normally 
attained puberty at a body mass of 90-100 g ın 7. valıda, 80-90 g ın T. robusta, and 
80-100 g in 7. nıgricanda. Adult body mass varıed seasonally in 7. nıgricanda, reaching a 
maximum during the rains and decreasing 6-7 percent during the dry seasons. The 
difference was significant in females (F, 1» = 7.1; P < 0.01) and almost significant in males 
(F, 126 = 3.3; P< 0.1). No seasonal changes in adult body mass were observed in 7. valıda 
or T. robusta. 

The adult body mass of the species I studied is compared to those of other species in 
Table 1. The species, ranked in order of increasıng adult body mass, are 7. lencogaster, T. 
afra, T. brantsıi, T. robusta, T. valıda and T. nigricanda, T. v. gambiana, T. indica. It may 
be noted that males are usually a little heavier than females. 


Breeding season 


There were two periods of breeding activity a year in 7. valıida in western Uganda, which 
were correlated with the bimodal pattern of rainfall (Fig. 1 B, D). During the second 
April-May period of study only 3 adult females were collected, which makes it impossible 
to deduce anything from the absence of pregnancies at this time. 

T. nıgriıcanda also had a bimodal pattern of breeding which was correlated with the 
pattern of rainfall (Fig. 2 B, D). Breeding was at amaximum during the rains and the month 
following the rains, then declined during the dry season, and ceased in September and 
March. However, breeding began again before the onset of the rains. One peculiar feature 


eaın] eıod 
-109 YIIM safewoaf ıpnpe Jo »deruss1ad O---O pur “Jueusaıd 
Agısıa sapeursy ıpnpe Jo sderussıad @e — @ 'q 1daaxa [ "Sg 
ur se uoneloN '3ın3e1aduısı pue [jezurei 0) uonejs1 ur vpnv) 
-1uB1u :L 30 Arıanoe aanonpo1daı ur saduey> Jeuoseag 'Z 317 


(90) dwa]L 
(ww) jje}uley 


B. R. Neal 
yıBus] IH 


KU) 


(B) sseyy AS/sasaL anı19y 3 Beud % 


290 


epnesußiu eusye] "80 


[[] paa1>5Jur 10 > paA1asgo sarueugaıd 
pue © —O sapduıes ur speunue ıpnpe Jo sdegusad :L IS XT FE 
sau1f Jesm1aA pue sueaur 1uasaıdaı syurod ereq] 'sseur saj2ısaA Jeuruas 
ıpnpe O---O ‘ssewi s9Jsa} ıInpe @ — @ :\y 'SIZIS opdures zuasaıdaı 
O-V saındıy ur sıaqwnn 'aınessdwsı pue [jegureı 03 uoneja4 
ur vpıyva '] Jo Arıamse aamanpoıdaı ur saduey> [Jeuoseag 'z "317 


y3uoyy 
NESVZENGEISZETEEQgEE NEE Oo SEEEE Ve N EV, 


uw) [je4UulPy 


(ww) yıßue] Ih 


(5) sseyy AS/saIsa] 'poldey $ B INPY % 


epijen eiolej "0% 


Reproductive biology of three species of gerbils in East Africa 291 


of the breeding pattern 
was that breeding activity 
declined and ceased much 
more rapidly during the 
January - mid-April dry 
season compared to the 
June - October dry sea- 
son. This pattern was also 
observed in Arvicanthis 
(NEAL 1981), and may be 
associated with increased 
temperatures at the end of 


AFRICA 


T. v.gambiana 
SENEGAL 14°35'N 


T. valida 
UGANDA 0° 


T. nigricauda 
KENYA 0° 


T. leucogaster 
MALAWI 15-16°S 


T. I. schinzi 
NAMIBIA 18°S 


T. leucogaster 
BOTSWANA 18-20°S 


T. brantsii 
each dry season. S.AFRICA 26°S 
Pregnancies were ob- T. afra 


5 - S. AFRICA 33°S 
served in T. robusta, or in- 


ferred from the presence 
of lactating females, dur- 
ing the periods July-Au- 


ASIA 


T. i. ceylonica 
SRI LANKA 6-10°N 


gust and November-Janu- T. i. cuvieri 
ary. It seems lıkely that the ee 
pattern of breeding is simi- 5 a 
lar to that of T. nigricauda. Tina 
The mass of the adult N. INDIA 28°N 
T. i. indica 


testes and seminal vesicles 
of T. nıgricanda were sig- 
nificantly greater during 
the rains and the month 
following the rains, com- Fig. 3. Period of pregnancies in relation to wet and dry seasons. The 
pared to the dry seasons. shaded areas represent rainy periods which have been given a 
. common length. Where a localıty has two rainy seasons each year 

There was no evidence in 
these are represented by two cycles; if there is only one rainy season 
the three species I studied then the information is duplicated in the two cycles. The open 
that males became sexually blocks represent the periods when pregnancies are particularly 
inactive during the periods Numerous, the solid line periods when few pregnancies are observed 


PAKISTAN 31°N 


Cycle 1 Cycle 2 


of the year when females 

were not breeding. In contrast, the testes regress, and the males become sexually quiescent 
during the periods of the year when females are in anoestrus, in 7. afra (ALLAnson 1958), 
T. leucogaster (HanNnEy 1965), and T.indica (BEG and AjmaıL 1977; CHANDRAHAS and 
KRrISHNASWAMI 1974; PRAsAD 1956). 

There is little consistency in the seasonal pattern of breeding in relation to rainfall in the 
genus 7atera (Fig. 3). In Africa, the delay between the onset of the rains and the start of the 
breeding period, or period of peak breeding activity, increases as one proceeds further from 
the equator. However, the opposite trend occurs ın Asıa (Fig. 3). In 7. afra ın South Africa 
(MEASRocH 1954) and 7. ı. indıica ın Pakistan (BEG and Ajmaı 1977), the cessation of 
breeding is associated with the coldest season of the year. 


Litter size 


Only five pregnant 7. valıda were collected: four of which had 4 embryos each, and one 3 
embryos. The average litter size of 16 anımals that were either pregnant or had a single set 
of placental scars was 4.25 + 0.21 (range 3-6). Derany (1964) records a female with 4 
embryos in Uganda, and HuBERT and Adam (1975) record that 7. v. gambiana has 2-6 
young per litter in Senegal. 

Five pregnant 7. robusta were collected: single individuals of which had 2, 6 and 7 


292 B. R. Neal 


embryos, and two individuals each had 4 implanted embryos which included a single 
resorbing embryo. The average number of implanted embryos of 13 animals that were 
either pregnant or had a single set of placental scars was 4.69 + 0.43 (range 2-7). 
SHORTRIDGE (1934) records an average litter size of 4-5 young in this species. 

The mean number of ımplanted embryos (resorbing embryos included) of 37 pregnan- 
cies in 7. nıgricanda was 5.43 + 0.23 (range 1-8); that of live embryos only was 5.22 + 
0.24 (range 1-8). Litter size increased sıgnificantly with increase in maternal body mass 
(F\34 = 12.9; P< 0.001 for implanted embryos, F, ;; = 7.1; P < 0.025 for live embryos), 
and this relationship was the same in the wet and dry seasons. However, the average litter 
size during the wet seasons (5.7 # 0.21 implanted embryos, 5.54 # 0.22 live embryos) was 
significantly higher (F, 3; = 5.34; P < 0.05 for implanted embryos, F, 3; = 6.37; P < 0.025 
for live embryos) than that recorded during the dry seasons (4.56 + 0.65 implanted 
embryos, 4.22 + 0.64 live embryos). This difference in litter size was largely accounted for 
by the difference in size of the mothers in the two seasons. The average evicerated body 
mass of pregnant females during the wet seasons was 102 g compared to 90 g during the 
dry seasons. Litter size has also been observed to vary seasonally, and be positively 
correlated with maternal body mass ın T. afra (MEAsRocH 1954) and 7. 1. cuvieri (PRASAD 
1961). 

The litter sizes of T. valıda, T. robusta, and T. nigricanda are amongst the largest 
recorded for African Tatera, but are generally smaller than the Asian species. The uswal 
litter size of other African species are 4-5 in T. leucogaster in Botswana (SMITHERS 1971), 


Table 2 


Potential breeding rates and reproductive capacity of various species of Tatera 


Species Period of Interval between No. of litters Litter sıze Prod. young/adult 
reproduction litters (days) 


T. valida Rains 109 
(This study) 


T. nigricanda Wet 
(This study) Dry 


Annual 


T. afra Aug-Mar 38- 47 
(MEAsSROocH 1954) 


T. brantsii Mar-Sep 56- 69 
(MEASRocH 1954) Oct-Feb 109-133 


Annual 


T. 1. cuvieri Aug/Sep- 60 
(Prasan 1953, 1961) Feb/Apr 


T. 1. hardwickei Aug-Mar/Apr 95-103 
(CHANDRAHAS and 
KRISHNASWAMI 1974) 


T. 1. indica All year 67 5.5 
(Jaın 1970) 


T. 1. indica Most of year 42- 52 5.3-6.5 6.3 33—41* 
(BEG and Ajmar 1977) 


Rates vary according to: ! estimate of gestation period; ? delayed implantation; ? length of breeding 
season; * length of breeding season and breeding rate. 


Reproductive biology of three species of gerbils in East Africa 293 


Zambia (SHEPPE 1973), and Malawı (Hanner 1965); 3-5 in T. afra in South Africa 
(MEAsSRocH 1954); 2-4 in T. brantsü in South Africa (MEAsrocH 1954) and Botswana 
(SMITHERS 1971); 2-3 ın T. inclusa pringlei in northern Tanzanıa (HUBBARD 1970). The 
Asian species have litter sizes of 2-4 ın T. indica ceylonica ın Srı Lanka (PHrLLirs 1923); 5-8 
in 7. 1. cuvieri ın Bangalore, India (PrAasap 1961); 4-8 in T. 1. hardwickei in Kolar, India 
(CHANDRAHAS and KRISHNASWAMI 1974); 3-6 in T. 1. indica in the Rajasthan Desert, India 
(Jaın 1970), and 5-7 for this subspecies ın Pakistan (BEG and Ajmar 1977). 


Breeding rates and reproductive capacity 


The reproductive capacity (potential annual production of young per adult female) of 
several different species of Tatera is compared in Table 2. T. valida appears to have a 
particularly low reproductive capacity, mainly because of a low breeding rate and short 
breeding season. T. v. gambiana probably also has a low reproductive capacity in Senegal, 
because breeding was confined to two months each year and there were only 2-6 young per 
litter (HUBERT 1977; HUBERT and ADAM 1975). 

The reproductive capacity of T. nigricanda was similar to that of several other species of 
Tatera, and was much higher than that of T. valıda (Table 2). Adult females had the 
potential to produce about 5 litters a year, but counts of placental scars showed that no 
female had more than three litters. Thus, the observed reproductive capacity must have 
been achieved by having a high turnover rate of adult females. 

Reproductive capacity depends on the intensity of breeding activity, the length of the 
breeding period, and litter size. However, there appeared to be no simple correlation 
between any one of these parameters and reproductive capacity (Table 2). The reproductive 
capacity could be low because of a low intensity of breeding, as in T. valıda and T. ı. 
hardwickei, a short breeding period, as in 7. v. gambiana; or a small litter size, as in 7. 
afra. 


Population structure 


The population structure of 7. valıda and T. nigricauda (Fig. 1 C,2 C) indicated a marked 
seasonal recruitment of young into the population. This was particularly evident ın 7. 
nigricauda, ın which recruitment of young was confined to the periods January-February 
and July-August, even though breeding continued throughout most of the year (Fig. 2 B, 
C). Thus, young only appeared to be successfully weaned in those litters which were 
conceived during the rains. 


Discussion 
Reproductive characteristics 


The observed positive correlation between litter size and maternal body mass in 7. 
nigricauda (this study), 7. afra (MEASRocH 1954), and T. i. cuvieri (PrAsaD 1961) is a trend 
which is probably present in most populations of this genus. The trend is probably related 
to the increased size of the reproductive tract and increased energy reserves of larger 
females. 

In contrast, when different populations and species of Tatera are compared a negative 
correlation is expected between litter size and maternal body mass (size). This is because 
longevity correlates positively with absolute body size, and longer lived species generally 
have lower reproductive rates than shorter lived species (EISENBERG 1981). However, the 
largest species of Tatera had the largest size of litters, although the trend was not 
statistically significant. In addition there was no obvious correlation between litter size and 
latitude, contrary to what has been reported for this genus by BEG and Ajmar (1977). 

A negative correlation is also expected between reproductive capacity (number of 


294 B. R. Neal 


offspring per adult female per year) and adult body mass (size), for similar reasons to that 
stated above. However, no such trend is evident in the genus Tatera (cf Tables 1 and 2). 

Thus, a comparison of the reproductive characteristics of the various species of Tatera 
failed to identify many of the patterns that are normally expected. In this respect, Tatera 
gerbils are similar to the spiny mice of the genus Acomys (NEAL 1982a) but differ from the 
unstriped grass rats of the genus Arvicanthıs (NEAL 1981) and the elephant shrews (NEAL 
1982b). 


Breeding season 


T. valıda, T. nigricauda and probably 7. robusta bred at a maximum during the rains and 
either ceased or reduced breeding activity during the dry seasons. In 7. nigricauda there 
appears to be a strong selection pressure to maintain its seasonal pattern of breeding, 
because only those litters conceived during the rains were recruited into the population. It 
is likely that litters conceived during the dry seasons died because of lactation failure. There 
was a significant reduction in maternal body mass during the dry seasons, so presumably 
the adults were experiencing some kind of stress at these times. 

It is not possible to determine from the data presently available which enivironmental 
factors are responsible for the seasonal breeding rhythm of the three species I studied on 
the equator. Obviously, the most likely factors are rainfall and some aspect of food supply. 
However, the extreme varıability of breeding patterns in relation to rainfall observed in the 
various species of the genus (Fig. 3) seems to rule out the possibility that rainfall ıs the only 
factor controlling the reproductive pattern in this genus. There is some circumstantial 
evidence which suggests that food affects reproduction in this genus. One feature which 
appears to be consistent in the genus ıs a low breeding rate compared to other genera (eg. 
see Derany and NEAL 1969). It is unusual for more than 50 percent of the adult females to 
be visibly pregnant at any one time. One reason for this apparent low breeding rate ıs a 
relatively long delay in implantation when females are lactating (BLAnD 1969; MEASROCH 
1954). The delay ın implantation averages 14 d in T. indica (BLAnD 1969), and I calculate 
the average delay in 7. afra and T. brantsii as 5 d from the data presented in MEASROCH 
(1954). This suggests that the energy requirement of lactation ıs high in Tatera gerbils, and 
that an individual cannot sımultaneously lactate and maintain the later stages of pregnancy. 
If this is the case, then food may be a very important factor governing the pattern of 
breeding in this genus. The increased energy demands in cold weather might also explain 
why breeding ceases during the colder seasons of the year at higher latitudes (MEASROCH 
1954; BEG and Ajmau 1977). 

Acknowledgements 


I am indebted to Prof. M. J. DELAny for initiating and encouraging my interest in african rodents. The 
field work in western Uganda was carried out at the Nuffield Unit of Tropical Animal Ecology in the 
Ruwenzori National Park. I am grateful to the Director of the Unit, Dr. R. M. Laws, for his guidance, 
and to the Trustees of the Uganda National Parks for permission to collect animals. Mr. J. Ocıkı 
helped considerably with the field work. I wish to thank the National Environmental Research 
Council and the University of Southampton for financial support. 

The field work in Kenya was carried out during a sabbatical leave from the University of 
Saskatchewan. I wish to thank the former Director and Trustees of the Kenya National Parks for 
permission to study in Meru National Park, the Park staff for their help and cooperation, and Prof. F. 
MUTERE of Kenyatta University College for making my stay in Kenya possible. It is a pleasure to 
thank the Park Warden, Mr. P. Jenkins, for providing excellent facilities and help for the field and 
initial laboratory work; the Park Rangers, J. NAMERIO and T. NToBgurı, for their help with the field 
work; and my wife, Jenny, for help with the field and laboratory work. Finally, it is a pleasure to 
thank Drs. C. RaıpH and B. WunDEr of the Department of Zoology and Entomology, Colorado 
State University for use of the facilities of the Department during the completion of this work. 


Reproductive biology of three species of gerbils in East Africa 295 


Zusammenfassung 


Fortpflanzungsbiologie von drei Gerbilliden-Arten (Gattung Tatera) in Ostafrika 


Die Fortpflanzungsökologie von Tatera nigricanda und T. robusta wurde im Meru-Nationalpark in 
Mittel-Kenia, die von 7. valıda im Ruwenzori-Nationalpark in West-Uganda untersucht. In beiden 
Gebieten waren die jahreszeitlichen Änderungen der Photoperiode zu vernachlässigen, hingegen 
folgten die Regenfälle einem bimodalen Muster. 

Eine Gesamtzahl von 398 T. nıgricanda, 61 T. robusta und 116 T. valıda, die über einen Zeitraum 
von mindestens einem Jahr gesammelt wurden, gaben Aufschluß über jahreszeitlich bedingte Verän- 
derungen der Fortpflanzungsaktivität, Populationsstruktur sowie Wurfgröße und Fortpflanzungs- 
kapazität. 

er valıda pflanzte sich in West-Uganda nur während der Regenzeit fort. Adulte Weibchen 
konnten zweimal jährlich werfen. Die durchschnittliche Wurfgröße war 4.25 + 0.21, was eine 
durchschnittliche jährliche Fortpflanzungskapazität von 8.5 Jungen pro adultem Weibchen ergibt. 

Tatera nigricauda pflanzte sich in Mittel-Kenia nahezu während des ganzen Jahres fort. Die 
Fortpflanzungsaktivität war maxımal während der Regenzeiten, nahm dann ab und erlosch gegen 
Ende der Trockenzeiten. Allerdings begann die Fortpflanzung vor dem Beginn der Regenzeiten. 
Jungtiere wurden aber nur 2 Monate nach dem Ende jeder Regenzeit gefangen, was nahelegt, daß 
Würfe, die während der Trockenzeit gezeugt wurden, starben, bevor sie entwöhnt waren. Erwachsene 
Weibchen konnten fünfmal pro Jahr werfen. Die Wurfgröße war 5.2 + 0.24 und die jährliche 
Fortpflanzungskapazität lag bei 25-28 Jungen pro adultem Weibchen. 

Tatera robusta schien eine ähnliche Fortpflanzungsperiodik wie 7. nigricauda zu haben. Die 
durchschnittliche Wurfgröße war 4.7 # 0.43. 

Ein Vergleich der Fortpflanzungscharakteristiken der 3 Tatera- Arten zeigte keinerlei signifikante 
Beziehungen zwischen Körpergewicht, Wurfgröße, Zeitraum der Fortpflanzung und Fortpflanzungs- 
kapazität. Es liegt nahe, daf das Nahrungsangebot der wichtigste Faktor ist, der die Fortpflanzungs- 
muster ın dieser Gattung beeinflußt. 


References 


ALLANSoN, M. (1958): Growth and reproduction in the males of two species of gerbil, Tatera brantsii 
(A. Smith) and Tatera afra (Gray). Proc. zool. Lond. 130, 373-396. 

AMENT, J. G. (1975): The vascular plants of Meru National Park, Kenya. J. E. Afr. Nat. Hist. Soc. 
154, 1-33. 

Bes, M. A.; Amar, M. (1977): Reproduction in the Indian gerbil, Tatera indica indica. Mammalia 41, 
213-220. 

BLanp, K.P. (1969): Reproduction in the female Indian Gerbil (Tatera indica). J. Zool. (Lond.) 157, 
47-61. 
CHANDRAHAS, R. K.; KrIsHNASwAMI, A. K. (1974): Studies on the ecology of the Indian gerbil, 
Tatera indica hardwickei (Gray), in Kolar (Mysore State). Indian J. Med. Res. 62, 971-978. 
Davıs, D. H. S. (1971): Genera Tatera and Gerbillurus. In: The mammals of Africa: an identification 
manual. Ed. by J. MEEsTErR; H. W. SETZER. Washington: Smithsonian Institution. 

Derany, M. J. (1964): A study of the ecology and breeding of small mammals in Uganda. Proc. zool. 
Soc. Lond. 142, 347-370. 

Derany, M. J.; Near, B. R. (1969): Breeding seasons in rodents ın Uganda. J. Soc. Reprod. Fert. 
(Suppl.) 6, 229-235. 

EISENBERG, ]J. F. (1981): The Mammalian Radiations. Chicago: University of Chicago Press. 

ELLERMAN, J. R.; MORRISON-SCOTT, T. C. S. (1951): Checklist of Palaearctic and Indian Mammals, 
1758 to 1946. London: Brit. Mus. Natural Hist. 

GAauTun, J. C. (1975): Periodicite de la reproduction de quelques rongeurs d’une savane preforestiere 
du centre de la Cöte d’Ivoire. Terre Vie 29, 265-287. 

Hanney, P. (1965): The Muridae of Malawı (Africa: Nyasaland). J. Zool. (Lond.) 146, 577-633. 

HUBBARD, C. A. (1970): A new species of Tatera from Tanzanıa with a description of its life history 
and habits studied in captıvity. Zool. Afr. 5, 237-247. 

HUBERT, B. (1977): Ecologie des populations de rongeurs de Bandıa (Senegal) en zone 
sahelosoudanienne. Terre Vie 31, 33-100. 

HUBERT, B.; ADam, F. (1975): Reproduction et croissance en Elevage de quatre especes de rongeurs 
Senegalaıs. Mammalia 39, 57-74. 

Jaın, A. P. (1970): Body weights, sex ratio, age structure and some aspects of reproduction in the 
Indian gerbil, Tatera indica ıindica, ın the Rajasthan desert, India. Mammalia 34, 415-432. 

MATTHEy, R.; PETTER, F. (1970): Etude cytogenetique et taxonomique de 40 Tatera et Taterillus 
provenant de Haute-Volta et de Republique Centrafricaine (Rongeurs, Gerbillidae). Mammalıa 34, 
585-597. 

MEASRocH, V. (1954): Growth and reproduction in the females of two species of gerbil, Tatera brantsi 
(A. Smith) and Tatera afra (Gray). Proc. zool. Soc. Lond. 124, 631-658. 


296 B. R. Neal 


MEESTER, J.; HALLETT, A. F. (1970): Notes on early postnatal development in certain southern african 
Muridae and Cricetidae. J. Mammalogy 51, 703-711. 

Near, B. R. (1970): The habitat distribution and activity of a rodent population in Western Uganda, 
with particular reference to the effects of burning. Revue Zool. Bot. afr. 81, 29-50. 

— (1981): Reproductive biology of the unstriped grass rat, Arvicanthis, ın East Africa. Z. Säugetier- 
kunde 46, 174-189. 

— (1982a): The breeding pattern of two species of spiny mice, Acomys (Muridae: Rodentia), in central 
Kenya. Mammalıa (in press). 

— (1982b): Reproductive ecology of the Rufous elephant-shrew, Zlephantulus rufescens (Macros- 
celididae), in Kenya. Z. Säugetierkunde 47, 65-71. 

PrirLırs, W. W. A. (1923): Some notes on the habits of the Ceylon gerbil, Tatera ceylonica. ]. 
Bombay Nat. Hist. Soc. 29, 214-216. 

PırLoT, P. L. (1954): Pourcentages de jeunes et periodes de reproduction chez quelques Rongeurs de 
Congo belge. Annls. Mus. r. Congo belge Ser, 4to Zoologie 1, 41—46. 

Prasanp, M.R.N. (1953): Habits, breeding season and placentation of Tatera (Gerbillus) indica. Proc. 
40th Indian Sci. Congr. 3, 206-207. 

— (1956): Reproductive cycle of the male Indian gerbille Tatera indica cuvieru (Waterhouse). Acta 
Zoologica 37, 87-122. 

— (1961): Reproduction in the female Indian gerbille Tatera indica cuvierii (Waterhouse). Acta 
Zoologica 42, 245-256. 

SHEPPE, W. A. (1973): Notes on Zambian Rodents and Shrews. Puku 7, 167-190. 

SHORTRIDGE, G. C. (1934): The mammals of South West Afrıca. London: Heinemann. 

SMITHERS, R. H. N. (1971): The mammals of Botswana. Salisbury: Nat. Mus. Rhodesia. 

SNEDECOR, G. W.; CoOCHRAN, W.G. (1967): Statistical Methods. 6th Ed. Ames: Iowa State University 
Press. 

SOKAL, R. R.; ROHLF, F. $. (1969): Biometry. San Francisco: W. H. Freeman & Co. 


Author’s address: Prof. Dr. B. R. NEAL, Department of Biology, University of Saskatchewan, 
Saskatoon, Saskatchewan, Canada S7N OWO 


Age determination and population characteristics 
of red foxes from Maryland! 


By B. B. Neıson and J. A. CHAPMAN 


Appalachian Environmental Laboratory, Center for Environmental and Estuarine Studies, 
University of Maryland, Frostburg, USA 


Receipt of Ms. 10. 3. 1982 
Abstract 


Between 1976 and 1979, 210 red foxes (Vulpes vulpes) were collected throughout Maryland to 
determine the age of each fox based on eye lens weights, epiphyseal closure of the humerus, baculum 
weight, tooth cementum annuli and skull measurements. Approximately 50 % of the red foxes 
collected during the trapping season (October-January) were juveniles; 20% were one year olds; 
16 % were 2-year olds; 8% were 3-year olds; and 4 % were 4-years old or older. An ovulation rate of 
6.0 was calculated from the number of corpora lutea per parous female. A mean litter size of 5.0 was 
determined from the mean number of placental scars per parous female. The peak of the breeding takes 
place in mid-January. Sixty-seven percent of the adult males produced sperm in October. Ninety 
percent of the adult males were producing sperm in November compared to 47 % of the juveniles. To 
assess the physical condition of each fox an adrenal index, spleen index, and body fat index were 
developed. Adrenal and spleen indices indicated that 5-year olds may be subjected to more stress than 


' Contribution Number 1321-AEL, University of Maryland, Center for Environmental and 
Estuarine Studies. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4705-0296 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 296-311 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Red fox population characteristics 297 


younger age classes. Body fat indices showed a significant increase from fall to winter suggesting that 
red foxes are in their best condition during the peak of the breeding season. Ten foxes were infected 
with mange and significant variation was found between adrenal indices for absent and severe cases of 
mange. 


Introduction 


During the 1970’s there were significant increases in fox harvest, fox pelt value, and 
decreases in habitat throughout North America. This is a land-wildlife problem that should 
be addressed by North American wildlife ecologists. 

Maryland (USA) fur harvest records clearly show an increased interest ın red fox as a 
furbearer. During the 1970-1971 trapping season an estimated 362 red foxes were trapped 
in Maryland. This number steadily increased to a peak of 4,620 during the 1975-1976 
season (DEEMS and PursLey 1978). The 1976-1977 and 1977-1978 seasons showed a 
decline from the 1975-1976 high with catches of 3,928 and 4,318 red foxes, respectively. In 
1978 the average pelt value climbed to $62, representing a total revenue to Maryland 
trappers of $267,716. Estimates of the 1978-1979 harvest indicated the number of red foxes 
trapped in Maryland approached 5,000 (Maryland Department of Natural Resources 1979 
Fur Harvest Records, Unpublished). The increase in the number of red foxes trapped may 
be explained partly by the increase in the number of trappers in Maryland. For example, 
between 1975 and 1979 the number of Maryland Fur Trapper’s Association district clubs 
alone increased from 2 to 6. 

Additionally, red fox habitat is declining in Maryland as new housing projects continue 
to reduce the amount of farm land. In 1968 there were 19,700 farms in Maryland 
comprising 1,296,000 hectares. By 1977 Maryland had 17,500 farms totaling 1,183,000 
hectares, which represented a loss 113,000 hectares of farm land (Maryland Department of 
Economic and Community Development 1977). 

With harvest levels increasing and suitable habitat decreasing, the Maryland Department 
of Natural Resources recognized the need to effectively manage this valuable furbearer. In 
1978 the red fox was reclassified from a predator to game anımal. This action allowed the 
Maryland Department of Natural Resources to post its first open season on red foxes 
(November 1, 1978 to January 8, 1979). Two important implications of this action were (1) 
pregnant females would be protected, increasing pup production; and (2) pups would be 
protected from indiscriminate shooting while still at the den site. 

However, the only source of information on the red fox ın Maryland was from annual 
fur harvest records. These records are inadequate for obtaining the biological information 
necessary to effectively manage the red fox as a fur resource. In this study reproductive 
parameters, age structure and physical condition of the Maryland red fox population are 
assessed. 


Study area 


The area of collection was broken down into 5 regions as shown in Fig. 1 (p. 298). 


Region 1 


Garrett County, the western most county of Maryland, was designated as Region 1. Part of the 
Allegheny Plateau of the Appalachian Mountain Range, this region consists of several mountains over 
900 meters above sea level. The highest point is Backbone Mountain which is 1,024 meters above sea 
level. Bedrock underlying this region consists of gray and red sandstone, acid and calcareous shale, 
chert, and limestone. Soils are characterized as rocky, shallow, and acidic. Average annual temperature 
ranges from 9 °C to 10 °C (MıLLer 1976). Hardwoods such as sugar maple (Acer saccharum), red oak 
(Quercus rubra), beech (Gagus grandifolia) and yellow birch (Betula alleghaniensis) are the dominant 
forest species. 


298 B. B. Nelson and ]J. A. Chapman 


MARYLAND 


BOARD OF NATURAL RESOURCES 
Annaopolis 


Fig. 1. The 5-regions of Maryland from which red foxes were collected (1976-1979). Numbers indicate 
sample sizes 


Region 2 


Region 2 consists of Allegany and Washington Counties within the Ridge and Valley province of 
Maryland. Soils are varıable, dominated by medium and moderately coarse textured soils on residuum 
of acıd shale and sandstone. Average annual temperature is about 12 °C. Oaks (Quercus sp.) pines 
(Pinus sp.), and hickories (Carya sp.) characterize the forests of this region (MıLLER 1976). 


Region 3 


Region 3 is comprised of Frederick, Carroll, Howard, Montgomery, and Prince Georges counties. 
Situated between the Blue Ridge and the Coastal Plain, this region is called the Piedmont Region. 
Rolling hills and stream valleys are characteristic of this area. Soils are medium textured on residuum 
of acıd and basic igneous rocks and metamorphic rocks. Average annual temperature fluctuates around 
12 °C (MıLLer 1976). 


Region 4 


Region 4 consists of Baltimore, Harford, and Cecil Counties. Although the soils and rock strata are 
sımilar to Region 3, this area was considered separately for several reasons. Trappers were not 
organized into a district club in this area until the second year of this study. Consequently, foxes were 
collected from this region for 1-year only. All 3 counties in this region border the Chesapeake Bay and 
contain areas of sandy and silty soils, characteristic of the coastal plain. Average annual temperature is 
12 °C (MiıLLEr 1976). 


Region 5 
Assateague Island is designated as Region 5. Tidal marshes, sand dunes, and coastal beaches are 


characteristic of this island. Twelve red foxes were collected on this island the first year of the study. 
Average annual temperature ranges from 13° to 15 °C (MıLLEr 1976). 


Red fox population characteristics 299 


Materials and methods 


Red foxes were collected in Maryland counties west of the Chesapeake Bay and on Assateague Island 
on Maryland’s eastern shore. Red foxes were not collected from eastern shore counties because: 1. 
eastern shore trappers were not organized into district clubs with regular monthly meetings, making it 
difficult to solicite their assistance in collecting foxes, and 2. eastern shore trappers concentrate their 
pen effort on muskrats (Ondatra zibethicus), nutria (Myocastor coypus), and raccoons (Procyon 
otor). 

Red foxes were collected by trapping, shooting, and road kills. However, the majority of foxes 
were collected from Maryland trappers during the trapping season. Eye lenses were removed from 
freshly killed foxes and placed in buffered formalin before the carcasses were frozen (Harrıs 1978). 

Each fox was weighed in grams. Total length, tail length, hind foot length, and ear length were 
recorded in millimeters. The spleen, kidneys, liver, heart, and adrenals were weighed to the nearest 
0.001 gram on a top loading Mettler Balance (Model P163). 


Reproduction 


Uterine horn length was measured in millimeters from the junction of the two horns to the ovary. 
Each horn was flushed with distilled water and was examined for sperm under a microscope. The 
number of placental scars or fetuses was recorded. Ovarıes were detached and fixed in Bouin’s fluid 
and later dehydrated in a series of alcohols. Using a microtome, the ovaries were sectioned 10 microns 
thick. The ovaries were stained, permanently mounted on slides and examined microscopically for 
corpora lutea. 

The epididymus was separated from the testes and then both were weighed on a Mettler Balance. 
Each testis and epididymus was examined microscopically for sperm. The weight of the testis was used 
to calculate a testes index. Total body length was divided into the total weight of both testes to adjust 
for body size. The formula for calculating the testes index was: 


FRITNETETNZT, 
10£ 


where TI = testes index; RTWT = right testis weight; LTWT = left testis weight; TL = total length. 


ut 10, 


Age determination 


The right humerus and skull were removed from each fox and placed in boiling water until the muscle 
tissue could be separated easily from the bone. The proximal epiphysis of the humerus was examined 
to determine the degree of ossification. Each skull was placed in bleach for 1-minute, then rinsed with 
water and allowed to air dry overnight. Skull measurements were taken using dial callıpers. 

Baculums were boiled until the muscle tissue could be removed easily from the bone. Each 
baculum was measured to the nearest 0.01 millimeter using a dial callıper, and weighed dry on a 
Mettler Balance. 

Eye lenses were placed in a drying oven at 70°C for 4 days, then weighed on a FPE precision 
balance to within + 0.2 milligram. 

A canıne and premolar from each fox were removed and placed in decalcifying solution (3 % nitric 
acıd) for 48 hours, or until the teeth could be cut easily with a scalpel. Teeth were rinsed in water and 
then sliced 15 to 20 microns thick on a freezing microtome, and immediately stained using a procedure 
similar to that of GRUE and JEnsEN (1973) and Harrıs (1978). Cementum rings were counted and 
recorded immediately after staining. 


Mortality 


A life table was constructed using a format similar to that of Harrıs (1977) and STORM et al. (1976) 
and described by CAucHLe£y (1966, 1967). 

The number of female live births per female was calculated by dividing the mean number of 
placental scars by 2 for each age class. The number of vixens alıve at the start of each age class was 
calculated by dividing the adjusted frequency of age class, x, by the adjusted frequency of juveniles 
multiplied by 1,000. The number dying during each age class was calculated by subtracting the 1x 
values between 2 successive age classes (CAUGHLEY 1967). Since data were available for 2 successive 
years, mortality rates were calculated by substracting the second year frequency, X + 1, from the first 
year frequency, x, and dividing by the first year frequency. In other words, the number of 1-year olds 
alive the second year came from the population of juveniles the first year. Therefore, the difference 
between the 2 frequencies divided by the number of juveniles gives the mortality rate for juveniles 
(CAUGHLEY 1967). 


300 B. B. Nelson and ]J. A. Chapman 


Physiological responses 


The weight of the adrenal glands was used to calculate an adrenal index (AI) similar to that used by 
CHAPMAN et al. (1977). Total length was used to adjust for body size instead of total weight in order to 
eliminate variability associated wıth body weight. This index also allowed the use of foxes which had 
been pelted. The formula for calculating the adrenal index was: 


__ Right Adrenal Wt. + Left Adrenal Wt. Y 
Total Length 


Spleen weight was also used to measure physical condition (WILLNER et al. 1979). The formula for 
calculating the spleen index, SI, was: 


AI 10% 


Spleen Weight. e 
Total Length 
A body fat index was used to determine the condition of each fox. This index developed by RınEy 
(1955) and modified by CHapman et al. (1977) provided a value from 1 to 4: 1 indicated no body fat 
present, 2 indicated a small amount of fat around the kidneys, 3 indicated fat covering more than one- 


third of the kidneys and adrenal glands and 4 indicated fat covering more than half of the kidneys and 
fat present in the mesentaries and throughout the body cavity. 


Statistical analysis 


Statistical tests used to analyze the classification variants were one-way and two-way analysis of 
variance, Duncan’s New Multiple Range Test, and Spearman’s Rank Correlation. Statistical tests were 
considered significant when P <0.05 and highly significant when P<0.01. Statistical abbreviations 
and table formats are similar to those by STEEL and ToRRIE (1960). 


Results and discussion 
Collection of red foxes 


Between December 1976 and January 1979, 210 red foxes were collected throughout 
Maryland. The largest number of foxes was collected ın Region 3 where trapper coopera- 
tion was highest (Fig. 1). Ninety-four red foxes were collected during November and 65 
during December. No foxes were collected in February and May. 


Morphology 


Whole weight, pelted weight, and body measurements were recorded for each sex and age 
class. Pelted weight was the only variable that showed a significant difference between age 
classes for both males and females (Table 1). Heart weight showed a significant difference 
between juveniles and adults (P <0.01). Both adrenal (P <0.01) and kidney weights (P 
<.0.01) showed significant differences between juveniles and adults. 

A significant difference was found between the ovary weights of juveniles and adults 
and the uterine horn lengths of the two age classes (Table 2). There was no significant 
difference between testis weights of juveniles and adults, but epididymus weights were 
significantly different between the age classes (Table 3). 

Many of the red foxes obtained from trappers already had been pelted. Therefore, pelted 
weights were used when whole weights were not available. Although juvenile and adult age 
classes could be separated on the basis of both whole weight and pelted weight, there is a 
considerable overlap in ranges and 95 % confidence intervals in both groups. According to 
STORM et al. (1976) juvenile red foxes are similar in size to adults by October, which may 
explain the overlap in ranges. The mean whole weight of adult male red foxes collected in 
Maryland was 4,483 grams and 3,645 grams for adult females. This compares to 5,250 
grams for adult males and 4,128 grams for adult females collected in Illinois and 4,822 


Red fox population characteristics 301 


Table 1 


Analysis of variance of whole weight (g), pelted weight and body measurements (mm) for sex and 
age of red foxes collected in Maryland (1976-1979) 


95 % Confidence 
Interval 
Group .D. ld in. i For Mean Significance 


Whole Weight 
Male 
Juvenile 3763.86 768.47 3053.1 to 4474.6 
Adult 4482.89 443.07 4142.3 to 4836.2 
Female 
Juvenile 3430.45 531.53 3073.4 to 3787.5 
Adult 3644.67 1042.70 1054 to 6234.0 


Pelted Weight 
Male 
Juvenile 3984.64 562.24 3844.2 to 4125.1 
Adult 4352.58 697.59 4129.5 to 4575.7 
Female 
Juvenile 3414.73 497.73 3257.6 to 3571.8 
Adult 3610.50 405.35 3487.3 to 3733.7 


Total Length 
Male 
Juvenile 968.9 to 
Adult 964.4 to 
Female 
Juvenile 913.6 to 
Adult 928.9 to 


Tail Length 
Male 
Juvenile 65 368.5 to 
Adult 42 371.7 to 
Female 
Juvenile 44 i 351.0 to 
Adult 39 5 356.30 


Hind Foot Length 
Male 
Juvenile 60 i 157.7 to 
Adult 41 | 155.8 to 
Female 
Juvenile 41 3 151.1 to 
Adult 59 A 153.320 


Ear Length 
Male 
Juvenile 
Adult 
Female 
Juvenile 


Adult 


grams for adult males and 3,938 grams for adult females collected in Iowa (STORM et al. 
1976). In Indiana, Horrman and KırkPATRIcK (1954) reported the mean weight of male 
foxes as 5,253 grams and females 4,213 grams. 

Data on liver, heart, spleen and kidney weights for red foxes are lacking in the literature. 
Although both heart and kidney weights show a significant difference between juveniles 
and adults, the overlap between individual juvenile and adult organ weights indicates that 
no single weight is reliable for differentiating between juveniles and adults. 


302 B. B. Nelson and J. A. Chapman 


Table 2 


Analysis of variance of ovary weight and uterine horn length by age of red foxes collected in 
Maryland (1976-1979) 


95 % Confidence 
Interval 
Group N .D. ar in. : For Mean Significance 


Right Ovary Weight 
Juvenile 43 0.177 to 0.244 NS 
Adult 44 0.207 to 0.289 
Left Ovary Weight 
Juvenile 42 0.183 to 0.226 
Adult 44 0.2180 0310 > 
Right Uterine Horn 
Length 
Juvenile 40 96.42 to 100.56 
Adult 43 115.35 to 122.21 Doc 


Left Uterine Horn 
Length 
Juvenile 40 94.74 2.010 70.0 to 98.82 


Adult 43 114.67 3.376 85.0 to 12120002 7 


Adrenal weights of Maryland red foxes range from a low of 0.045 grams in juveniles to a 
high of 0.500 grams in adults. Horrman and KIRKPATRICK (1954) reported heavier adrenals 
for Indiana red foxes (0.270 to 1.455 grams). They also reported high ovary weights (0.315 
to 2.430 grams) and testis weights (2.9 to 15.5 grams). 

LAvyne and McKeon (1956) found that the uterine horns of juveniles are usually shorter 
and narrower than adult foxes. This study also shows a significant difference between the 
uterine horn lengths of juveniles and adults. 


Table 3 


Analysis of variance of testes and epididymides weights by age of red foxes collected in Maryland 
(1976-1979) 


95 % Confidence 
Interval 
Group N .D. JE: in. 5 For Mean Significance 


Right Testis Weight 
Juvenile 34 2.535 2.046 to 3.025 NS 
Adult 24 3.010 2.449 to 3.572 


Left Testis Weight 
Juvenile 34 2.548 2.066 to 3.031 
Adult 23 3.088 2.450 to 3.726 


Right Epididymus Weight 
Juvenile 34 0.676 0.523 to 0.769 
Adult 23 0.826 0.722 to 0.031 


Left Epididymus Weight 
Juvenile 34 0.690 0.594 to 0.786 
Adult 22 0.847 0.736 to 0.958 


Red fox population characteristics 303 


Reproduction 


Only 4 of 96 female red foxes had ovaries containing corpora lutea. Examination of the 
uteri of these foxes showed that 2 were pregnant, 1 contained sperm but no visible signs of 
pregnancy, and 1 contained neither sperm nor fetuses. Since each egg ovulation is followed 
by the formation of a corpus luteum, the mean ovulation rate was 6.0 corpora lutea per 
female. None of the female reproductive tracts examined showed any signs of post- 
implantation mortality. 

Ovulation rate based on the number of corpora lutea per vixen was calculated to be 5.9 
for red foxes in New York (LAyne and McKeon 1956). In Ohio, GIER (1947) reported 7.2 
corpora lutea per female. The ovulation rate of 6.0 corpora lutea per female found in this 
study falls between the rates calculated for New York and Ohio, even though only four 
vixens we collected had ovaries containing corpora lutea. 

Only one vixen which was captured on January 7, 1978 had sperm in the uterus. Two 
foxes were pregnant; one captured on January 27, 1978, the other on January 29, 1978. In 
both pregnant females the fetuses appeared to be about 7 to 10 days old, placing conception 
sometime between 17 and 22 January. In New York, Layne and McKeon (1956) found 


18 
31 
10 
8 
x 
wu 
=) 
= 
6 
& 
on 
2 2 
Fr 
z 4 
= 
Fig. 2. Seasonal varıations in A 
mean testis indices for red 
foxes collected in Maryland 2 


(1976-1979). Vertical bars 
represent standard deviations, 
vertical lines are ranges, and 
numbers are sample sizes 
DEC-FEB MAR-MAY JUN-AUG SEPT-NOV 


8.3 % of the matings occurred the third week of January and 29.2 % occurred the fourth 
week of January. In Illinois and Iowa the peak of the breeding season was estimated to 
occur the third week of January (SToRM et al. 1976). Generally, the peak of the breeding 
appears to be later in northern than in southern latitudes (Layne and McKron 1956; 
Lrovyp and EnsLun 1973). 

SToRM et al. (1976) found 4% of the red foxes in Illinois conceived before mid- 
December, and three of 20 vixens bred between 13 and 29 December. Out of 52 vixens 
examined, HoFFMAn and KırkPATrick (1954) reported 3 red foxes breeding in December. 
Although no fetuses were found, and no sperm was detected in the uterus, 1 fox caught on 
15 December had 6 corpora lutea. It is possible that breeding may have occurred a few days 
prior to capture. 

Sixty-seven percent of the adult males showed sperm production in October. Ninety 
percent of the adult males were producing sperm in November compared to 47% ın 


304 B. B. Nelson and ]J. A. Chapman 


juveniles. By December, 60 % of the juveniles had testes containing sperm. The mean testis 
index increased from September to November and peaked from December to February 
(Fig. 2). 

Testıs weight and spermatogenesis are closely related in the red fox (JOFFRE 1977). From 
October through January, testes weight increases as the percent of red foxes with sperm in 
the testes increases (LLoyD and EnGLUNnD 1973; JOFFRE 1976; STORM et al. 1976). In this 
study the occurrence of sperm in the testis increased from October through December in 
both juvenile and adult red foxes, except for a slıght decrease for December adults. This 
decrease may be attrıbuted to the small sample size of 6 adult males collected during this 
month. 

The literature suggests that juveniles are capable of breeding their first year, but there ıs 
disagreement on the month that spermatogenesis begins. STORM et al. (1976) did not find 
sperm ın juvenile testes until December in midwestern red foxes. In Ireland, FAIRLEY 
(1970) found sperm in juvenile foxes in October. In this study 47% of the juveniles had 
sperm in their testes in November. Whether spermatogenesis occurs in September and 
October in juveniles cannot be determined from this study since no juveniles were 
collected in September and only 1 was collected in October. 

Mean litter size based on two pregnant females was 6.5, and for 41 females whose uteri 
contained placental scars ıt was 5.0. Analysis of varıance of mean number of placental scars 
showed no significant difference between regions. 

Litter size for red foxes has been determined using various techniques (Table 4). In this 
study the mean litter sıze was 6.5 based on 2 pregnant females. An estimate of 5.00 young 
per vixen (n = 41) was obtained from the number of placental scars per female. Both values 
are within the range of that reported in other studies (Table 4). 

Females may breed during their first winter, but the proportion of vixens coming into 
heat their first year varıes by different regions (STORM et al. 1976). LaynE and McKEoN 
(1956) reported as high as 16.6 % of the vixens were barren in northern New York 


Table 4 


Red fox mean litter size from different regions of the world 


Reference Location Method Mean Litter Size 


SHELDON (1949) New York BSSE 
SWITZENBERG (1950) Upper Michigan D 
Lower Michigan D 
RıCHARDS and Hiıne (1953) Wisconsin BSSD 
HoFrFrMman and KırKPATRIcK (1954) Indiana PSYE 
LAayneE and McKeon (1956) New York PS 


SCHOFIELD (1958) Michigan 


STANLEY (1963) 
FAIRLEY (1970) 
EnGLunD (1970) 


Kansas 
Ireland 
Sweden 


STORM et al. (1976) Illinois, Iowa BS 
E 
D 
Ryan (1976) Australia PS 
E 
Inszey (1977) England D 
Pırs and MarTın (1978) Wisconsin PS. EsD 
This Study (1982) Maryland PS 


Abbrev. for methods of determination are: E = embryos or fetuses, D = number of young in 
den, and PS = placental scars 


Red fox population characteristics 305 


compared to only 2.1 % in the southern part of the state. Five of the 6 barren vixens in this 
study were a year old and therefore had lived through only one breeding season. 

The sex ratio of midwestern red foxes was found to significantly favor males (STORM et 
al. 1976). Fifty-four percent of the juveniles collected at dens were males and 56 % of the 
adults were males. In Wisconsin, 58% of the red foxes collected were males (Pırs and 
MAarTIn 1978). However, the higher proportion of males collected suggests that males are 
more vulnerable to trapping, reflecting the tendency to travel more extensively than 
females, particularly during the fall (SHELDoN 1949; FAIRLEY 1970; STORM etal. 1976; Pırs 
and MARTIN 1978). In this study there was no significant deviation from the 50:50 sex 
ratio. 


Age distribution 


Ossification of the proximal epiphysis of the humerus begins at nine months of age and has 
been reported to be a reliable technique for separating red foxes into juvenile, subadult, and 
adult age classes (REıLLy and CuRrREN 1961). 


Table 5 


Age distribution of red foxes collected in Maryland (1976-1979) using epiphyseal closure of the 
humerus 


Relative Frequency Cumulative Frequency 
[e) [0) 


Epiphysis/Age (%) 


Open/Juvenile 22.4 


Closing/Sub-adult 56.7 
Closed/Adult 100.0 
Total ; 100.0 


Examination of the proximal epiphysis of the humerus of 210 red foxes revealed 22.4 % 
had no ossification, 34.3 % showed some ossification and 43.3 % were ossified completely 
(Table 5). 

Eye lens weights were separated by age (Fig. 3). Although sample sizes were small, there 
was no overlap in ranges when compared to age classes determined by cementum annulı. 


5 300 
e 
F 
E 
[) 
m 
3 
w 
(2) 
zZ 
wi 
=200 
wu 
> 
wi 
Fig. 3. Separation of mean eye lens < 
weights by age for red foxes collected in = 
Maryland (1976-1979). Vertical bars rep- 
resent standard deviations, vertical lines 
are ranges, and numbers are sample sizes 100 
JUV 1 2 3 
AGE 


Baculum weights showed a large overlap between age classes when compared to age 
classes determined by cementum annuli. Baculum weight was adjusted for body size by 
dividing it by total body length (X 10). Although overlap between age classes still 
occurred, significant differences existed between the means of juveniles and 1, 2, 3, and 5- 


year olds (Table 6). 


306 B. B. Nelson and J. A. Chapman 


Baculum weights from three juvenile red foxes collected on 10 July, 17 July, and 19 July 
were 28, 44, 89 milligrams, respectively. These three juveniles were the only animals 
collected before puberty began, and were estimated to be three to 3.5 months old if they 
were born the first or second week of April. 


Table 6 


Duncan’s New Multiple Range Test for age variation in adjusted baculum weights for red foxes 
collected in Maryland (1976-1979) 


Age Class (Years) Difference in Means Significance 
2.8406..1.4:.097%— 1251 
4.097 — 4.184 = —0.087 


4.184 — 4.939 = 0.755 


Juvenile — 1 


p < 0.05 
p < 0.05 
PF=0.05% 
p< 0.05 
p< 0.05 
p < 0.05 
p < 0.05 
p < 0.05 
p < 0.05 


Juvenile 
Juvenile 
Juvenile 
Juvenile 


2.846 — 4.939 = — 2.093 
2.846 — 3.508 = —0.662 
2.846 — 6.658 = — 3.812 


Table 7 


Age distribution of red foxes collected during two trapping seasons 
(October-January - 1977-78 and 1978-79) in Maryland 


1977-1978 Season 


Age Frequency Relative 
(Years) Frequency 


1978-1979 Season 


Frequency Relative 
Frequency 


Juvenile 


Table 8 
Analysis of variance of width of brain case by sex and by age class for red foxes collected in 


Maryland (1976-1979) 


95 % Confidence 
Interval 


Group 


Males 
Juveniles 


Adults 


Total Males 


Females 
Juveniles 


Adults 


Total Females 


For Mean 


46.56 to 47.47 
46.56 to 47.27 


46.68 to 47.27 


45.78 to 46.55 
45.71 to 46.51 


45.86 to 46.41 


Significance 


NS 


Red fox population characteristics 307 


Table 9 


Analysis of variance of zygomatic breadth by sex and by age class for red foxes 
collected in Maryland (1976-1979) 


95 % Confidence 
Interval 
Group .D. -E.. in. ; For Mean Significance 


Male 5 i i \ 3 73.69 to 74.95 
Female ; ; \ Ä Ä 69.77 to 70.99 
Juveniles ; 5 . ; & 71.09 to 72.49 
Adults 5 i h ä 5 72:1.9.1074. 22 


Males 
Juveniles 72.61 to 74.14 
Adults i ä k ; 2 64.65 to 76.56 
Females 


Juveniles 5 ; N E Ä 68.53 to 70.39 
Adults ; 70.54 to 71.98 


p< 0.1 


p< 0.01 


p<0.o1 


p< 0.01 


Aging red foxes by cementum annuli has been found to be a reliable aging technique in 
foxes (GRUE and JENSEN 1973; Monson et al. 1975; GEIGER et al. 1977; Harrıs 1978). 
This technique agreed 100 % with the juvenile and adult age classes found using epiphyseal 
closure of the humerus. Sometimes several slices of the canıne had to be made to clearly see 
the annular rings. The results of this technique were used to place eye lenses into age classes 
resulting in complete agreement between the 2 aging techniques (Table 7). 

Of the cranıal measurements taken, only width of brain case and zygomatic breadth 
showed significant differences between sexes and ages. These differences were sufficient to 
be used as an aging or sexing technique (Tables 8 and 9). 


Population characteristics 


The pattern of changing mortality rates with age is best expressed in the form of a life table 
(CAUGHLEY 1966). From the data ın a life table, one may determine the population’s rate of 
increase, mean life expectancy at birth, and percentage of the population that dies each 
year. 

Survivorship (xx) ıs the probability at birth of an individual surviving to any age (x). A 
clearer value ıs obtained by multiplying 1x by 1,000, which gives the number of anımals 
surviving to age out of 1,000. Only 73 out of 1,000 red foxes reached 5-years of age 
(Table 9). The oldest female collected was 6-years old. 

Mortality (dx) is the portion of the population that dies during the age interval x, to 
x +1. This value is usually expressed as 1,000 dy, the number of anımals that die out of 
sample of 1,000 during the age intervalx,tox +1. 

Mortality rate (qx) ıs the number of animals that died during the interval ,tox + 1, 
divided by the number of anımals alıve at age x. The number of anımals out of 1,000 alıve at 
age x which died before x +1 is expressed as 1,000 qx. For Maryland red foxes the 
mortality rate was highest between birth and 1-year of age, and lowest between one year 
and two years of age (Table 10). 

The age distribution remained relatively constant from the 1977-78 trappıng season to 
the 1978-79 season (Table 7). Both years show that over 50% of the population is 
comprised of juveniles and that after the age of 2 years the percent of individuals per year 
class drops. 

When constructing a life table a possible source of bias is behavioral or range differences 
between males and females (CAucHLeY 1966). If one sex is more vulnerable to the 


308 B. B. Nelson and J. A. Chapman 


Table 10 


Life table for female red foxes collected during the trapping season 
(October-January) in Maryland 


1977-1978 1978-1979 No. Females Live 
Age Frequency Adjusted Frequency Adjusted Births Per Female 
(Years) Frequency Frequency at Agem, 1000 1, 1000 q, 


Juveniles 0.00 
2.66 
3.34 
3.13 
3.05 
3.01 
3.00 


L, = fraction alive at start of age class; d, = number of dying during age class; q, = mortality rate 
per 1000 alıve at start of age class. | 


collection method than the other, then the data will be biased. This bias was eliminated in 

this study by considering only females. | 
Not all age classes are equally vulnerable to the collection method, in this case trapping. | 

If juveniles are more vulnerable to the trap, then distortions of each xx and dy value below | 

the juvenile values in the series will result. However, the qx values (mortality rates) are | 

independent of the first age class (juveniles) since qx is the ratio of the interval. The value of 


27 ! 


9 31 13 4 8 | 

Mi ı2 An 4 2 

2 

ar 1 x 6 ® 

x 

z Se 5 

[mm] ja} [=] 

[22 zZ zZ 

a 5 _ = 

= ı 

z z? r 4 

< wu z 

Wu 3 Er m 

= a = Sı 
ik 2ır <2 


1 .— 10 RB 


; AGE i ABSENT MILD SEVERE 
nn 2 o DEGREE OF MANGE 


Fig. 4. Age variation in mean adrenal 
4 8 8 indices for red foxes collected in Mary- 
land (1976-1979). Vertical bars represent 
standard deviations, vertical lines are 


2 x 

S3l Os! x6 z ranges, and numbers are sample sizes 

Fr z 

& Mr ’ Fig. 5. Comparison of adrenal indices, 

w . . . . 

52 24 24 on spleen indices and body fat indices for 

o Re z red foxes from 5-regions of Maryland 
<< 


(1976-1979) 


1 2 2 
ee ren SER Fıg. 6. The effect of mange on adrenal and 


spleen indices of red foxes collected in 
Maryland (1976-1979) 


Fıg. 5 REGION 


Red fox population characteristics 309 


qx ıs not directly dependent on absolute values of 1x, but on the differences between 
successive values (CAUGHLEY 1967). For Maryland red foxes the mortality rate (qx value) 
was highest for juveniles (Table 9). If a red fox reaches one year of age it has agood chance 
of surviving to its second year (mortality rate = 129). At age 2 years mortality rates 
increase and remain high. 


Physiological response 


Spearman’s Rank correlation coefficients were used to determine if there was any signifi- 
cant correlation between condition indices, month of capture, and weather parameters. 
Only month of capture showed a significant correlation with the adrenal index for both 
males and females. A significant correlation was shown between adrenal index and total 
precipitation, and spleen index and total precipitation for females but not males. 

Age variations in mean adrenal indices showed juveniles to have the lowest AI value 
(Fig. 4). One, 2, and 3-year olds show a leveling off, of fairly constant adrenal indexes with 
a rıse seen in 4 and 5-year olds. However, the difference in adrenal indices was significant 
only between juveniles and 5-year olds. Region 5, where the climate is the mildest, had the 
lowest adrenal index. Spleen indices were significantly different between juveniles, 3-year 
olds, and 5-year olds (P<0.05). 

A significant varıation was found between the mean body fat index of summer and 
winter red foxes (males and females combined) (P<0.05) and between fall and winter 
animals (P< 0.05). Significant regional variations in body fat indices were also observed 
(P<.0.05) (Fig. 5). 

Ten red foxes were infected with mange and an attempt was made to assess the effect of 
mange on fox physiology (Fig. 6). Mild cases of mange were defined as having infected 
areas on the tail and sometimes the legs. Severe cases had mange covering most of the body, 
including the tail, legs, back, face, and ears. Adrenal indices were higher with increased 
severity of mange, whereas no trend was apparent for spleen indices. The varıation between 
adrenal indices for absent and severe cases of mange was significant (P< 0.05). 


Acknowledgements 


We are indebted to the Maryland fur trappers, including PauL CHANEY, LLOYD ABBOTT, and PETE 
LEGGETT who supplied most of the foxes for the study. The manuscript was reviewed by J. EDwArD 
GATEs, Dan M. HARMAN, Jay R. STAUFFER, Jr., Appalachıan Environmental Laboratory, and 
THomas F. REDIcK, Frostburg State College. DuAnE PursLey made available Maryland fur harvest 
data and helped in making contacts with the trappers. KENNETH R. Dixon, AEL, assisted with the 
analysıs of the life table data and Kım Tırus and Dan Jacogs provided computer programming 
assistance. FRANCES YOUNGER, scientific illustrator, University of Maryland, Center for Environmen- 
tal and Estuarine Studies, prepared the figures, and Prof. Dr. Joun B. GEnYS the German summary. 


Zusammenfassung 


Altersbestimmung und Populationsmerkmale beim Rotfuchs in Maryland 


In den Jahren von 1976 bis 1979 wurden während der allgemeinen Fangzeit (Oktober bis Januar) im 
Staate Maryland, USA, insgesamt 210 Rotfüchse (Vulpes vulpes) gesammelt und ihr Alter bestimmt. 
Die Altersbestimmungen basierten auf Gewicht der Augenlinsen, Epiphysenschluß am Humerus, 
Baculumgewicht, Jahreszuwachs des Zahnzementes und Schädeldimensionen. Von den Füchsen. 
waren 50 % juvenil, 20 % 1 Jahr alt, 16 % 2 Jahre alt, 8% 3 Jahre alt und 4 % waren 4 Jahre oder älter. 
Aus der Anzahl der Corpora lutea pro Fähe konnte eine Ovulationsrate von 6,0 berechnet werden. 
Eine durchschnittliche Wurfgröße von 5,0 ergab sich aus der durchschnittlichen Anzahl von Plazen- 
tanarben pro Weibchen. Im Maximum findet die Fortpflanzung Mitte Januar statt. Im Oktober 
produzieren 67% der adulten Rüden Spermien und im November 90% der adulten und 47 % der 
juvenilen Männchen. Zur Bewertung der physischen Kondition wurden für jeden Fuchs Adrenalın-, 
Milz- und Körperfett-Indices ermittelt. Adrenalin- und Milz-Index zeigten an, daß 5jährige Tiere 
stärkerem Streß ausgesetzt sind als diejenigen jüngerer Altersklassen. Der Körperfett-Index ergab 
einen signifikanten Anstieg vom Herbst zum Winter und deutet auf beste körperliche Verfassung der 


310 B. B. Nelson and J. A. Chapman 


Tiere während der Fortpflanzungszeit hin. 10 Füchse waren von Räude infiziert. Signifikante 
Unterschiede im Adrenalin-Index ergaben sich zwischen gesunden und kranken Tieren. 


References 


Maryland Department of Economic and Community Development (1977): Maryland Statistical 
Abstract. pp. 137. 

CAUGHLEY, G. (1966): Mortality patterns in mammals. Ecology 47, 906-917. 

— (1967): Parameters for seasonally breeding populations. Ecology 48, 834-839. 

CHAPMAN, J. A.; HARMAN, A. L.; SAMUEL, D. E. (1977): Reproductive and physiological cycles in the 
cottontail complex in western Maryland and nearby West Virginia. Wildl. Monogr. 56, 1-73. 
Deens, E. F., Jr.; PursLey, D. (eds.) (1978): North American furbearers, their management, research, 

and harvest status in 1976. Univ. of Maryland, Univ. Press, College Park. 

EnGLunD, J. (1970): Some aspects of reproduction and mortality rates in Swedish foxes (Vulpes 
vulpes) 1961-63 and 1966-69. Viltrevy 8, 1-82. 

FAIRLEY, J. S. (1970): The food, reproduction, form, growth, and development of the fox, Vulpes 
vulpes (L.) in northeast Ireland. Proc. R. Irish Acad. 69, Section B., No. 5, 103-137. 

GEIGER, G., VON; BROMEL, J.; HABERMEHL, K.H. (1977): Konkordanz verschiedener Methoden der 
Altersbestimmung beim Rotfuchs (Vulpes vulpes L., 1758). Z. Jagdwiss. 23, 57-64. 

GIER, H. T. (1947): Populations and reproductive potentials of foxes in Ohio. Midwest Wildl. Conf. 
9, 3747. 

GRUE, H.; JENSEN, B. (1973): Annular structures in canine tooth cementum in red foxes (Vulpes vulpes 
L.) of known age. Danish Rev. Game Biol. 8, 1-12. 

Harrıs, $. (1977): Distribution, habitat utilization, and age structure of a suburban fox (Vulpes 
vulpes) population. Mammal Rev. 7, 25-30. 

— (1978): Age determination in the red fox (Vulpes vulpes) - an evaluation of technique efficiency as 
applied to a sample of suburban foxes. J. Zool. (London) 184, 91-117. 

HOoFrFMAN, R. A., KIRKPATRICK, C. M. (1954): Red fox weights and reproduction in Tippecanoe 
County, Indiana. J. Mammalogy 35, 504-509. 

InsLey, H. (1977): An estimate of the population density of the red fox (Vulpes vulpes) in the New 
forest, Hampshire. J. Zool. (London) 183, 549-553. 

JOFFRE, M. (1976): Puberty and seasonal sexual cycle of the wild male fox (Vulpes vulpes). Ann. Biol. 
Anım. Bioch. Biophys. 16, 503-520. 

— (1977): Blood circulation in the testes of the adult wild red fox (Vulpes vulpes): involvement of the 
testicular capsule and septa in the seasonal changes of testicular vasculature and capillary blood 
flow. J. Physiol. (Paris) 73, 155-176. 

LAYnE, J. N.; McKeon, W.H. (1956): Some aspects of red fox and gray fox reproduction in New 
York. N. Y. Fish and Game ]. 3, 44-74. 

Lroyp, H. G.; EnGLunD, ]J. (1973): The reproductive cycle of the red fox in Europe. J. Reprod. Fert. 
Suppl. 19, 119-130. 

MILLER, F. P. (1976): Maryland Soils. Univ. of Maryland, Univ. Press, College Park. pp. 42. 

Monson, R. A.; STONE, W. B.; Parks, E. (1975): Aging red foxes (Vulpes vulpes) by counting the 
annular cementum rings in their teeth. N. Y. Fish and Game ]. 20, 54-61. 

Pırs, C. M.; Martın, M. A. (1978): Population dynamics predator-prey relationships and manage- 
ment of the red fox in Wisconsin. Tech. Bull. No. 105. Dept. Nat. Res. Madison, Wisc. 

Reııry, J. R.; CurrEn, W. (1961): Evaluation of certain techniques for judging the age of red foxes 
(Vulpes ulva). N. Y. Fısh and Game ]J. 8, 122-129. 

RıcHarops, $. H.; Hıng, R. L. (1953): Wisconsin fox populations. Tech. Bull. Wisc. Cons. Dept. 6, 
1-78. 

Rıney, T. (1955): Evaluating condition of free-ranging red deer, with special reference to New 
Zealand. New Zealand ]J. Scı. Tech. 36, 428-463. 

Ryan, G. E. (1976): Observations on the reproduction and age structure of the fox, Vulpes vulpesL., 
in New South Wales. Aust. Wildl. Res. 3, 11-20. 

SCHOFIELD, R. D. (1958): Litter size and age ratios of Michigan red foxes. J. Wildl. Manage. 22, 
313-315. 

SHELDON, W. G. (1949): Reproductive behavior of foxes in New York State. J. Mammalogy 30, 
236-246. 
STANLEY, W. C. (1963): Habits of the red fox in northeastern Kansas. Univ. Kansas Mus. Nat. Hist. 

Misc. Pub. 34, 1-31. 

STEEL, R. G. D.; ToRRIE, J. H. (1960): Principles and procedures of statistics. New York: McGraw- 
Hill Book Co., Inc. 

STORM, G. L.; AnDREws, R. D.; PrırLırs, R. L.; BısHoP, R. A.; Sınırr, D. B.; TESTER, J. R. (1976): 
Morphology, reproduction, dispersal, and mortality of midwestern red fox populations. Wildl. 
Monogr. 49, 1-82. 


Red fox population characteristics 3 


SWITZENBERG, D. F. (1950): Breeding productivity in Michigan red foxes. J. Mammalogy 31, 194-195. 
WILLNER, G. R.; CHAPMman, J. A.; PuRsLEy, D. (1979): Reproduction, physiological responses, food 
habits and abundance of nutria on Maryland marshes. Wildl. Monogr. 65, 1-43. 


Authors’ addresses: Brap B. Neıson, U. $. Bureau of Land Management, Laramie, Wyoming, USA, 
and Prof. Dr. JosepH A. CHapman, Appalachian Environmental Laboratory, 
Center for Environmental and Estuarine Studies, University of Maryland, Frost- 
burg, Maryland 21 532, USA 


Home range and movement of sika deer (Cervus nippon) 
in Maryland 


By G. A. FELDHAMER, K.R. Dixon, and J. A. CHAPMAN 


Appalachian Environmental Laboratory, University of Maryland, Frostburg, USA 


Receipt of Ms. 10. 3. 1982 
Abstract 


Studied the home ranges and movements of a radiocollared male and female sika deer. The home range 
area of the male was 182.5 ha; that of the female was 127.8 ha. The male also exhibited significantly 
greater mean movement throughout its home range than did the female. 


Introduction 


Sika deer (Cervus nıppon, Temminck) are native to Japan, and eastern Asıa from Man- 
churia south to Vietnam. Introduced populations currently exist ın England, Ireland, 
several other European countries, Australia and New Zealand. In the United States, feral 
populations occur in Texas, Virginia, Wisconsin, Kansas, Oklahoma and Maryland (FELD- 
HAMER 1982). Sıka deer are highly adaptable and thrive in many types of habitat. Despite 
the numerous introductions and wide distribution of the species, few studies on their home 
range or movement patterns have been conducted. Davıpson (1979) investigated the linear 
distance from the point of tagging to kill site of 54 sıka deer in New Zealand. MARUYAMA 
et. al. (1978) calculated the home range of two radıocollared sika deer in Japan. 

The primary objective of this study was to determine the activity pattern and home 
range of sıka deer relative to sex and season of the year. A secondary objective was to test 
the utility of the harmonic mean measure of home range calculation (Dixon and CHAPMAN 
1980) on a large mammal. 

In Maryland, sıka deer (C. n. nippon) occur throughout the southern portion of 
Dorchester County, where they are sympatric with native white-tailed deer (Odocoileus 
virginianus, Zimmermann) (FELDHAMER et al. 1978). The study area was located south of 
Blackwater National Wildlife Refuge (Fig. 1). It consisted of awooded tract dominated by 
loblolly pine (Pinus taeda, Linnaeus) and oak (Quercus sp., Linnaeus) with a dense 
understory of greenbrier (Smilax sp., Linnaeus), bayberry (Myrica sp., Linnaeus), Ameri- 
can holly (/lex opaca, Aiton) and poison ivy (Rhus radicans, Linnaeus). To the north and 
west ofthe wooded area was marshland, primarily three-square rush (Scirpus olneyı, Gray). 
The study area ranged in elevation from sea level to approximately 2-3 m. Following rain, 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4705-0311 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 311-316 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


312 G. A. Feldhamer, K. R. Dixon, and J. A. Chapman 


ummnmnennnennnd 


SL sr Er Sn die Refuge 
ng re I 
es Area 9. 


Fig. 1. Study area, south of Blackwater National Wildlife Refuge, Dorchester County, Maryland, 
WS“. 


snow melt or high tide, water often inundated portions of the wooded area. The mean 
yearly precipitation in this region is 109 cm, of which 28 cm may be snow. Weather is fairly 
mild, with winter temperatures rarely below — 4°C. Summers are very humid and 
temperatures may exceed 35 °C. for extended periods. 


Material and methods 


Two sıka deer calves, a female and a male both approximately 10 months old, were captured in 
Stephenson box traps on 5 and 7 March 1978, respectively. They were fitted with radıo transmitter 
collars and released at the point of capture. Location bearings were taken from two or three of sıx 
stations 0.4 or 0.8 km apart, using a hand-held, 4-element Yagı antenna and a compass. Bearings for 
each fix were taken within a 5-min time interval. When possible, a series of locational fixes were taken 
at hourly intervals to investigate the extent of circadıan movements. Movements, as determined by the 
linear distance between consecutive hourly locations, were analyzed on a daily and seasonal basıs. 
Seasons were defined as 3-month periods as follows: spring (March-May); summer (June-August); 
fall (September-November); and winter (December-February). Daily actıvity patterns were based on 
four 6-h periods: 0100-0700; 0700-1300; 1300-1900; and 1900-0100. 

Home range and center of activity analyses were done using a harmonic mean method (Dixon and 
CHAPMAN 1980) which was developed using data on brush rabbits (Sylvilagus bachmani Waterhouse). 
The present study ıs the first to apply this technique to a large mammal. With this method, the 
calculated center of activity is located at the point of greatest actıvity and the isopleths of activity are 
directly related to the frequency of occurrence of each deer within its home range. Unlike traditional 
arıthmetic mean center of activity calculations in which a single “center” is established, the harmonic 
mean method may establish several centers of activity which may be associated with shifts in home 
range use. 


Results and discussion 


Between 12 April and 9 September 1978, when the transmitter failed, 66 locational fixes 
were obtained for the male sika deer. His home range encompassed 182.5 ha. This deer was 
killed by a hunter on 27 November 1978, 4.2 km from the point of capture and 4.3 km 
from the final location fix. For the female, a total of 235 locational fixes was obtained 
during the period 12 April 1978 to 25 February 1980. Although this represented 169 more 


Home range and movement of sika deer (Cervus nippon) in Maryland 313 


—i_ 


Fig. 2. Home range isopleths of male and female sıka deer and associated degree of overlap. Numerals 
denote activity isopleth values in meters. Dashed line represents a road 


locational points than for the male, the home of the female was only 127.8 ha, or 42.8 % 
less than that of the male (Fig. 2). Differential seasonal use of the home range area by the 
male was noted during the five months he was monitored, as denoted by the shifts in his 
centers of actvity (Fig. 3). Much smaller shifts were noted for the female, despite the much 
longer period during which she was monitored and the greater number of locational fıxes. 

Considering seasonal activity, as determined by mean linear distances moved hourly, 
the female showed no significant seasonal differences in movement (see Table). The only 
difference occurred when comparing distance moved by the female in the spring of 1978 vs. 
spring of 1979. She moved significantly less as a calf (spring 1978) than she did as a yearling. 
Seasonal comparison of activity was limited to the spring and summer of 1978 for the male, 
and no difference was noted. 


d 


25 Jun 


13 Apr-25 Jun 


17 Aug-24 Aug 
>4 Aug), 225 Jun-28 Jul 


30 Jul-16 Aug] 25 Aug 
25 Aug-9 Sep 


9 Sep* 
12 Apr-25 Ne] 
27 May a 


24, 26 Jun 
8 Jul-16 Aug 


8 Jul 


200 400 


meters 


Fıg. 3. Centrograms of female (left) and male (right) sıka deer from 12 April to 9 September, 1978, and 
associated dates of shifts in centers of activity 


314 G. A. Feldhamer, K. R. Dixon, and J. A. Chapman 


Table 


Seasonal comparisons of the mean distances moved between consecutive hourly location points 


Seasonal comparison Mean distance moved (m) Student t,,ı 


Female 
1978 spring 
vs. 
summer 


summer 
VS. 


fall 
fall 


vs. 
spring 


spring 
vs. 
winter 


spring 


VS. 


spring 


spring 
vs. 
summer 


152.9 (10)! 


237.9 (40) 
237.9 (40) 


197.3 (42) 
197.3 (42) 


201.7 (31) 
201.7 (31) 


179.3 (20) 
152.9 (10) 


201.7 (31) 


456.5 (11) 


338.4 (23) 


1.610 


! Number of fixes in parentheses. — ? Significant difference P < 0.05 


In 1978, when both radiocollared sıka were on the study area, the male exhibited 
significantly greater mean movement (in meters) than the female, both in the spring [male 
X = 456.5 (n= 11); female X = 152.9 (10); P< 0.05] and in the summer [male X = 338.4 
(23); female X = 237.9 (40); P< 0.05]. 

This difference in activity was further illustrated by greater shifts ın the centers of 
activity of the male than those of the female during the spring and summer of 1978 (Fig. 3). 
Thus, the male exhibited both a larger total home range area and greater mean activity and 
movement within that area. 

Seasonally, no statistically significant differences in activity (movement) were seen 
among the four 6-h intervals. This was due in part at least, to the very large varıances 
associated with each mean and the limited samples sızes. For both seasons during which 
data were collected on the male, the most active interval was between 1300-1900 h. This 
was also the most active period for the female in three of the fıve seasons she was 
monitored. 

Numerous environmental factors influence home range size, as do methods used to 
analyze the data. Comparisons between studies may show general patterns or trends, 
however. Maruyama et al. (1978) reported home range sızes for two adult male sıka deer, 
calculated by the “observation-area curve” method, as 6.5 and 19.0 ha, respectively. Data 
were limited to 39 locational fixes during a 4-day period in the first case, and 42 locational 
fixes during a 35-day period in the second. In addition, the study of MArUYAMmA et al. 
(1978) was conducted from September through November, when adult males may have 
used only a portion of their home range, that is, established territories in association with 
the rut. Yearling sika males probably do not establish territories and breed. In the present 
study, the male (a yearling in the fall of 1978) may have been forced out of the study area 


Home range and movement of sika deer (Cervus nippon) in Maryland 315 


by adult males during the rut. As noted previously, it was killed 4.3 km from the final 
location fix. This was similar to the mean linear distance from tagging to kill site off 2.5 km 
for yearling sika males in New Zealand (Davıpson 1979). MıurA (1974) also noted that 
males tended to “wander”’ outside their previous home range areas during the pre-rut 
period. 

Sika deer on the study area were active intermittently throughout each 24-h period. The 
longest distances moved during hourly intervals occurred in the afternoon during most of 
the year. Home range use by the female sika deer was not affected by seasonal environmen- 
tal changes; food was not critical and the weather was generally mild. Intraspecific 
behavioral interactions may have mitigated against similar restriction of home range by the 
young male sika, however. 

The activity of the female was more restricted as a calf than as a yearling. When 
radiocollared as a calf, she probably was still closely associated with her dam, although she 
probably was weaned by this time (KıppıE 1962). Few female sıka deer shot on the study 
area during the hunting season (November-December) were lactating. Thus, in spring 
1978, the dam probably was pregnant, forage was not limited, and she (and her calf) may 
not have been inclined to move great distances. As a yearling, the radiocollared deer may 
not have been as closely associated with her dam, and as a result showed a greater range of 
movement. 

The results of this study show that the harmonic mean measure of home range 
calculation can be applied to large mammals with correspondingly larger home ranges as 
well as to small mammals with smaller home ranges. When the same units of measurement 
are used ın determining the reciprocal distance deviations, the isopleth values are directly 
comparable. Thus, the isopleth value determining the home range boundary of brush 
rabbits was 30 m (Dıxon and CHAapMaAn 1980), whereas with sıka deer the values were 600 
m for the female and 700 m for the male. 


Acknowledgements 


W. BRUCE TALIAFERRO, THOMAS D. BEAVER and numerous students from the Appalachian Environ- 
mental Laboratory aided in trappıng deer and attempts over a 2-year period to capture additional 
animals. Mr. Wırıam SPICER and Mr. P. A. RansoME kindly allowed traps to be placed on their 
property and assisted in many facets of this study. The computer time for this project was supported 
in part through the facilities of the Computer Science Center of the University of Maryland. We also 
thank FRAN YOUNGER, Chesapeake Biological Laboratory, for drawing the figures. The Maryland 
Wildlife Administration provided use of the deer traps. This is Contribution No. 1320-AEL, Appala- 
chian Environmental Laboratory, Center for Environmental and Estuarine Studies, University of 
Maryland, and Federal Aid to Wildlife Restoration W-49-R Maryland. 


Zusammenfassung 


Reviergröße und Ortswechsel beim Sika-Hirsch (Cervus nippon) in Maryland 


Mit telemetrischen Methoden wurden Reviergröße und Ortswechsel eines männlichen und eines 
weiblichen Sika-Hirsches in Maryland untersucht. Der Lebensraum des Hirsches war 182,5 ha groß, 
der des weiblichen Stückes nur 127,8 ha. Außerdem war der Hirsch stärker tätig als die Hirschkuh. 


Literature 


Davınson, M. M. (1979): Movement of marked sika (Cervus nippon) and red deer (Cervus elaphus) ın 
Central North Isand (sic), New Zealand. N. Z. ]J. For. Sci. 9, 77-88. 

Dixon, K. R.; Chapman, J. A., (1980): Harmonic mean measure of anımal activity areas. Ecology 61, 
1040-1044. 

FELDHAMER, G. A. (1982): Sika deer. In: Wild Mammals of North America: Biology, Management, 
and Economics. Ed. by CHAPMmANn, J. A., FELDHAMER, G. A., Baltimore: The Johns Hopkins 
University Press. Pp. 1, 114-1, 123. 

FELDHAMER, G. A.; CHAPMAN, J. A.; MILLER, R. L., (1978): Sika deer and white-tailed deer on 
Maryland’s eastern shore. Wildl. Soc. Bull. 6, 155-157. 


316 G. A. Feldhamer, K. R. Dixon, and J. A. Chapman 


Kıopıe, D. G. (1962): The sika deer (Cervus nippon) in New Zealand. N. Z. For. Serv. Info. Ser. No. 
44. 


MARUYAMA, N.; ITo, T.; TAMURA, M.; Mıyakı, M.; ABE, $.; TAKATsukı, S.; NAITo, T. (1978): 
Application of radiotelemetry to sika deer on Kinkazan Island. J. Mammal. Soc. Jpn. 7, 189-198. 
(In Japanese with English summary.). 

Mıura, $. (1974): On the seasonal movements of sika deer population in Mt. Hinokiboramaru. J. 
Mammal. Soc. Jpn. 6, 51-66. (In Japanese with English summary.) 


Authors’ address: GEORGE A. FELDHAMER, KENNETH R. Dixon, and JOsErH A. CHapman, Appala- 
chian Environmental Laboratory, Center for Environmental and Estuarine Studies, 
University of Maryland, Frostburg State College Campus, Gunter Hall, Frostburg, 
Maryland 21532, USA 


WISSENSCHAFTETEHE KURZMTIILTELEUNIGSEN 


Der Dachs, Meles meles (Linnaeus, 1758), in NW-Syrien! 


Von D. Kock und R. KınZELBACH 
Eingang des Ms. 16. 6. 1982 


Das Vorkommen des Dachses ın Syrien ist bisher allein durch einen Schädel (BM. 30.3.3.1) 
belegt, dessen exakte Herkunft jedoch unbekannt ist (HARRISON 1968: 242; KUMERLOEVE 
1975b). Die Verbreitungskarte (Abb. 1, n. HARRISON 1968, 1972: 628; KUMERLOEVE 1967, 
1975a) zeigt, daß Meles meles zumindest in NW-Syrien zu erwarten ist. 

Am 31. 7. 1978 fand die Nahost-Fxkursion 1978 (RK) des Zoologischen Institutes der 
Universität Mainz einen überfahrenen M. meles etwa 20 km südlich Tartous. Dieser 
Nachweis fällt in die Verbreitungs-,,Lücke‘, die ökologisch (vgl. LEHMANN 1966) oder 
zoogeographisch schwerlich erklärbar wäre. Das erkennbare Farbmuster des nicht einge- 


Verbreitung von Meles meles im Nahen Osten. Stern: syrisches Vorkommen 


! Ergebnisse der Reisen von R. KınzELsacH im Vorderen Orient, Nr. 51. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4705-0316 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 316-317 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Meles meles in NW-Syrien SU 


sammelten Fundes (in Putrifikation) erlaubt keine Äußerungen zur subspezifischen Zuord- 
nung (Merkmale dafür in HArRIson 1968). 


Literatur 


Harrıson, D. L. (1968): The mammals of Arabia, 2. Carnivora, Artiodactyla, Hyracoidea. London: 
E. Benn. 

— (1972): The mammals of Arabia, 3. Lagomorpha, Rodentia. London: E. Benn. 

KUMERLOEVE, H. (1967): Zur Verbreitung kleinasiatischer Raub- und Huftiere sowie einiger Groß- 
nager. Säugetierkdl. Mitt. 15, 337-409. 

— (1975a): Die Säugetiere (Mammalia) der Türkei. Versuch einer kursorischen Übersicht (Stand 
1973/74). Veröff. zool. Staatssamml. München 18, 69-158. 

— (1975b): Die Säugetiere (Mammalia) Syriens und des Libanon. Eine vorläufige Übersicht (Stand 
1974). Veröff. zool. Staatssamml. München 18, 159-225. 

LEHMANN, E. von (1966): Über die Säugetiere im Waldgebiet NW-Syriens. Sitz.-Ber. Ges. natur- 
forsch. Fr. Berlin (N.F.) 5, 22-38. 


Anschriften der Verfasser: Dr. DiETEr Kock, Forschungsinstitut Senckenberg, Senckenberg-Anlage 
25, D-6000 Frankfurt a. M.; Prof. Dr. RAGNAR KINZELBACH, Institut für 
Zoologie, Technische Hochschule, Schnittspahnstr. 3, D-6100 Darmstadt 


Tandem-Fusion als chromosomaler Evolutionsmechanismus 
bei Microtus agrestis (Rodentia, Microtinae) 


Von ROSWITHA GAMPERL 


Institut für Medizinische Biologie und Humangenetik der Universität Graz 


Eingang des Ms. 21. 6. 1982 


Die Chromosomen von Microtus agrestis sınd ein beliebtes Studienobjekt für Zytogeneti- 
ker. Der Grund liegt darin, daß sie ungewöhnlich große, spätreplizierende Heterochroma- 
tinabschnitte haben, die auf den Geschlechtschromosomen lokalisiert sind und die auch im 
Interphasenkern als deutliche Chromozentren nachgewiesen werden können. Nicht nur 
das Heterochromatin der Geschlechtschromosomen ist wiederholt untersucht worden 
(WoLr et al. 1965; Schmp 1967; SCHMID und LEPPERT 1968; ARRIGHI et al. 1970; DE LA 
MazaA and Junis 1974; PERA und ScHoız 1975; PERA und MAaTTIAs 1976; WEGNER und 
SPERLING 1976), sondern auch verschiedene Bändermuster (CooPEr und Hsu 1972; SETH 
et al. 1973) und die Lokalisation der Nukleolusorganisatorregionen (GOODPASTURE und 
Boom 1975) sind bekannt. Vergleichend zytogenetische Untersuchungen sind bisher aber 
nur an ungebändertem Material durchgeführt worden (MATTHEY 1957; KrAr 1972; KrRAL 
et al. 1979). 

Microtus agrestis besitzt — abgesehen vom X-Chromosom und dem Autosomenpaar 
Nr. 24 - ebenso wie die Wühlmausarten mit ursprünglichen Karyotypen (MATTHEY 1975) 
hauptsächlich akrozentrische Chromosomen. Zum Unterschied von diesen hat sie aber 
eine geringere Chromosomenzahl und auch eine geringere Anzahl von Chromosomen- 
armen. Eine Veränderung, die sowohl die Chromosomenzahl als auch die Anzahl der 
Arme reduziert, ıst die sog. Tandem-Fusion, d. ı. eine Fusion zwischen Zentromer und 
Telomer bzw. zwischen zwei Telomeren (Hsu 1979). Ob tatsächlich Tandem-Fusionen für 
die geringere Chromosomenzahl von Microtus agrestis verantwortlich sind, soll durch 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4705-0317 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 317-320 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


MAG 


CRU 


Roswitha Gamperl 
MAG ERU MAG 


CRU 


318 


en 


0! 


MAG 


CRU 


MAG 


CRU 


MAG 


ERU 


MAG 


CRU 


HBEDDR 2 
nn 


FREDÄR 


“ee 


LED 


WRHER 


wu 


8 
12 


MAG 
MAG 
MAG 


CRU 


26112 
CRU 
13 
ERU 


MAG 


\ 
5 


MAG 
MAG 


CRU 


CRU 
CRU 


MAG 


MAG 
14 


0 


CRU 


ERU 


2 


MAG 
Ä 
MAG 


N 


CRU 


[RU 
16 
CRU 


8} 


| 


MAG 


CRU 


MAG 


2% 


(MAG) 


ao 
I 
es 


ERU 


MAG 


CRuU 


CRU 
Vergleich der G-Bänderkaryogramme von Clethrionomys rufocanus (CRU) und Microtus agrestıs 


Tandem-Fusion als chromosomaler Evolutionsmechanismus bei M. agrestis 319 


einen Vergleich der G-Bänder von Microtus agrestis mit denen von Clethrionomys rufoca- 
nus geklärt werden. Dafür wurden Bänderkaryogramme von einigen Exemplaren (229, 
38&) der Art Microtus agrestis, die bei Neumarkt (Steiermark, Österreich) gefangen 
wurden, angefertigt. Sie stimmen mit den aus der Literatur bekannten Karyotypen überein. 
Der Karyotyp von Clethrionomys rufocanus, der einem ursprünglichen Wühlmauskaryo- 
typ entspricht, wurde bereits an anderer Stelle im Vergleich mit dem von Clethrionomys 
glareolus publiziert (GAmPERL 1982). 

Einander entsprechende Chromosomen beider Arten sind in der Abb. gegenüberge- 
stellt. Die G-Bändermuster der meisten Chromosomen weisen weitgehende Übereinstim- 
mung auf. Nur für die Chromosomenpaare Nr. 1, 4, 8 und 9 von Microtus agrestis gibt es 
keine entsprechenden Partner bei Clethrionomys rufocanus. Dafür bleiben bei Clethriono- 
mys rufocanus sechs mittelgroße bis kleine Chromosomenpaare übrig. Davon lassen sich 
die beiden Paare Nr. 4 und 8 durch eine Tandem-Fusion (Verschmelzung des Telomers 
von Chromosom Nr. 8 mit dem Zentromer von Chromosom Nr. 4) in das Chromosom 
Nr. 1 von Microtus agrestis überführen. Die beiden Chromosomenpaare Nr. 8 und 9 
dürften ähnlich entstanden sein. Allerdings lassen sich hier die Homologien wegen der 
Kleinheit der beteiligten Chromosomen und des zum Teil nicht sehr charakteristischen 
Bändermusters nicht so eindeutig nachweisen. Auch die Herkunft des distalen Abschnittes 
am Chromosom Nr. 4 von Microtus agrestis ließ sich mit den hier angewandten Methoden 
nicht klären. 

Wenn sich die unterschiedliche Chromosomenzahl der beiden Arten auch nicht aus- 
schließlich auf Tandem-Fusionen zurückführen läßt, so kann doch festgehalten werden, 
daß Tandem-Fusionen sehr wesentlich an der Reduktion der Chromosomenzahl beteiligt 
sind und mit der Anhäufung von Heterochromatin in den Geschlechtschromosomen 
charakteristisch sind für die Evolution des Karyotyps von Microtus agrestis. Tandem- 
Fusionen gibt es innerhalb der Microtinae nicht nur bei Microtus agrestis, sondern sie 
führten wohl auch zur Entstehung der Chromosomenpaare Nr. 1 von Microtus nivalıs 
(BELCHEvA 1980) und Microtus oeconomus (FREDGA 1980). Sie sind also — wenn auch in 
geringerem Maße als Robertson’sche Fusionen (= Fusion zweier Zentromeren) - an der 
Entstehung von abgeleiteten Wühlmauskaryotypen mit geringerer Chromosomenzahl und 
größeren Chromosomen (MATTHEY 1957) beteiligt. 


Literatur 


ARRIGHI, F. E.; Hsu, T. C.; SAUNDERS, P.; SAUNDERS, G. F. (1970): Localızation of repetitive DNA 
in the chromosomes of Microtus agrestis by means of in situ hybridization. Chromosoma 32, 
224-236. 

BELCHEVA, R. G.; PesHev, T. H.; PEsHev, D. T. (1980): Chromosome C- und G-banding patterns in a 
Bulgarian population of Microtus guentheri Danford and Alston (Microtinae, Rodentia). Genetica 
52/53, 4548. 

CooPER, J. E. K.; Hsu, T. C. (1972): The C-band and G-band patterns of Microtus agrestıs 
chromosomes. Cytogenetics 11, 295-304. 

DE La Maza, L. M.; Yunıs, J. J. (1974): Application of an established diploid Microtus agrestis cell 
line as amodel for the understanding of mammalian heterochromatin. Exp. Cell Res. 84, 175-182. 

FREDGA, K.; Persson, A.; STENSETH, N. C. (1980): Centric fission in Microtus oeconomus. A 
chromosome study of isolated populations ın Fennoscandia. Hereditas 92, 209-216. 

GAMPERL, R. (1982): Chromosomal evolution in the genus Clethrionomys. Genetica 57, 193-197. 

GOODPASTURE, C.; BLoom, $. E. (1975): Visualization of nucleolar organızer regions in mammalıan 
chromosomes using silver staining. Chromosoma 53, 37-50. 

Hsv, T. C. (1979): Human and mammalian cytogenetics. An historical perspective. New York, 
Heidelberg, Berlin: Springer. 

KrAr, B. (1972): Chromosome characteristics of Muridae and Microtinae from Czechoslovakia. Akta 
Sc. Nat. (Brno) 6, 1-78. 

Kräı, B.; Zıma, J.; HERZIG-STRaAscHIL, B.; STERBA, ©. (1979): Karyotypes of certain small mammals 
from Austria. Folia Zool. 28, 5-11. 

MATTHEYy, R. (1957): Cytologie compar&e, systematique et phylogenie des Microtinae (Rodentia- 
Muridae). Rev. Suisse Zoll. 64, 39-71. 


320 Roswitha Gamperl 


NADLER, C. F. (1969): Chromosomal Evolution in rodents. In: Comparative mammalian cytogene- 
tics. Ed. by K. BENIRSCHKE. New York, Heidelberg, Berlin: Springer. 

PErA, F.; ScHoız, P. (1975): Late DNA replication of X chromosomes in female and pseudofemale 
cells of Microtus agrestis. Humangenetik 30, 173-177. 

PErA, F.; MarTıas, P. (1976): Labelling of DNA and differential sister chromatid staining after BrdU 
treatment in vivo. Chromosoma 57, 13-18. 

SCHMID, W. (1967): Heterochromatin ın mammals. Arch. Klaus-Stift. Vererb.-Forsch. 42, 1-62. 
SCHMID, W.; LEPPERT, M. F. (1968): Karyotyp, Heterochromatin und DNS-Werte bei 13 Arten von 
Wühlmäusen (Microtinae, Mammalıia-Rodentia). Arch. Klaus-Stift. Vererb.-Forsch. 43, 88-91. 
SETH, P. K.; Pera, F.; Hırwıc, I.; Gropr, A. (1973): Fluorescence banding pattern of the 
chromosomes of Microtus agrestis with a benzimidazol derivative. Humangenetik 19, 129-134. 
WEGNER, R.-D.; SPERLING, K. (1976): A simple staining technique to demonstrate chromosomal 

DNA replication. Experientia 32, 1220-1221. 
WOLF, U.; FLINSPACH, G.; BÖHM, R.; OHno, S. (1965): DNS-Reduplikationsmuster bei den Riesen- 
Geschlechtschromosomen von Microtus agrestis. Chromosoma 16, 609-617. 


Anschrift der Verfasserin: Dr. ROSWITHA GAMPERL, Institut für Medizinische Biologie und Human- 
genetik der Universität Graz, Harrachgasse 21/8, A-8010 Graz, Österreich 


Ökologie der Tiere 


Ein Lehr- und Handbuch in drei Teilen 
Von Professor Dr. Fritz Schwerdtfeger 


Band I: Autökologie 


Die Beziehungen zwischen Tier und Umwelt. 2., neubearbeitete Auflage. 1977. 460 Seiten mit 
268 Abbildungen und 55 Übersichten. Leinen 120,— DM 


Der erste Band behandelt die Beziehungen des Tieres als Individuum oder Repräsentant seiner 
Art zu den Gegebenheiten der Umwelt. Vorangestellt ist eine kurze Erörterung der Grundlagen 
der Ökologie, insbesondere ihrer Grundbegriffe. Der eingehenden Schilderung der Tier- 
Umwelt-Beziehungen ist die Gliederung der Umweltkomponenten in abiotische, trophische und 
biotische Faktoren zugrunde gelegt. Die Einflüsse der einzelnen Faktoren, beispielsweise des 
Lichtes, der Nahrungsmenge oder der Parasiten, auf den Bau, die Leistungen und das 
Verhalten des Tieres werden unter dem Gesichtspunkt vorgeführt, die Relationen eindeutig 
zahlen- und kurvenmäßig zu erfassen. 


Band Il: Demökologie 


Struktur und Dynamik tierischer Populationen. 2., neubearbeitete Auflage. 1979. 450 Seiten mit 
249 Abbildungen und 55 Übersichten. Leinen 120,- DM 


Der zweite Band behandelt nach einer Einführung zum Wesen der Demökologie und der 
Population in statistischer Sicht die Struktur der Populationen, ihre Zusammensetzung aus 
Elementen wie Dichte, Verteilung oder Altersaufbau sowie deren Beeinflussung durch Um- und 
Mitweltkomponenten. Die zu ihrer Erfassung tauglichen Methoden werden jeweils kurz vor- 
geführt. Gegenstand des zweiten Hauptteils ist die Dynamik der Populationen, die zeitliche 
Veränderung ihrer Strukturelemente; dabei werden die Wandlungen der beiden Grundelemente 
Verteilung und Dichte, also die Dispersions- und Abundanzdynamik, besonders eingehend 
behandelt. 


Band Ill: Synökologie 


Struktur, Funktion und Produktivität mehrartiger Tiergemeinschaften. Mit einem Anhang: 
Mensch und Tiergemeinschaft. 1975. 451 Seiten mit 118 Abbildungen und 125 Übersichten. 
Leinen 98,- DM 


Der Autor erläutert im dritten Band einführend das Wesen der Synökologie im allgemeinen 
sowie speziell das der Synökologie der Tiere. Ihr sind drei Hauptabschnitte gewidmet. Der erste 
befaßt sich mit der Struktur der Tiergemeinschaften, mit ihrer Entstehung, ihren Grundformen, 
Merkmalen und Kennzeichnungen sowie mit ihrer Dynamik. Thema des zweiten Abschnitts ist 
die Funktion der Tiergemeinschaften, ihr Inbetriebsein mit den dafür notwendigen Vorausset- 
zungen und Vorgängen, die in einem Funktionsplan analysiert und in der Schilderung des 
realen Funktionsgeschehens vorgeführt und gewertet werden. Der dritte Teil handelt von der in 
jüngster Zeit in den Vordergrund des Interesses gerückten Produktivität der Tiergemeinschaf- 
ten, namentlich von den Komponenten ihres Stoff- und Energiehaushalts, wie z. B. Konsum- 
tion, Assimilation, Produktion und Respiration; den Abschluß bildet der Versuch einer Stoff- und 
Energiebilanz der Tiergemeinschaften. 


DarEReE PAUL PAREY : HAMBURG UND BERLIN 


Erscheinungsweise und Bezugspreis: Die Zeitschrift erscheint alle 2 Monate; 6 Hefte bilden einen Band 
mit insgesamt 384 Seiten. Der Abonnementspreis beträgt je Band 232,- DM zuzüglich Porto. Das 
Abonnement verpflichtet zur Abnahme eines ganzen Bandes. Es verlängert sich stillschweigend, 
wenn nicht unmittelbar nach Erhalt des letzten Heftes eines Bandes Abbestellung erfolgt. 
Einzelbezugspreis der Hefte: 43,- DM. Die Preise verstehen sich im Inland incl. Mehrwertsteuer. 
Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, 
Spitalerstraße 12, D-2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesell- 
schaft für Säugetierkunde“ erhalten die Zeitschrift unberechnet im Rahmen des Mitgliedsbeitrages. 


Die große Monographie 
iiber das Muffelwild 


hne Übertreibung 
darf man hier von 
einem Standard- 
werk sprechen. Für 
dieses Prädikat bürgen die 
Autoren — zwei führende 
Muffelwildkenner, die in 
engem Kontakt mit Fachleu- 
ten des In- und Auslandes ein 
weitgehend neues Buch 
geschrieben haben. 
Es beschäftigt sich umfas- 
send mit der Natur- 
geschichte, Hege und Jagd 


dieser Wildart und berück- 
sichtigt die Themenbereiche 
Stammesgeschichte, Lebens- 
weise und Verhalten, Verbrei- 
tung, Fütterung, Krankheits- 
vorbeuge und -bekämpfung, 
neue Erkenntnisse der 
Wissenschaft und Erfolge 
der Praxis. * 
Informationswert und Über- 
sicht werden durch viele gute, 
zum Teil farbige Fotos, Zeich- 
nungen, Karten und Tabellen 
erhöht. 


Friedrich Türcke/ 

Herbert Tomiczek 

Das Muffelwild _ 
Naturgeschichte, Okologie, 
Hege und Jago. 2., völlig neu 
bearbeitete Auflage. 1982. 
197 Seiten mit 195 Einzeldar- 
stellungen in 114 Abbildun- 
gen, davon 11 farbig, und 30 
Tabellen. Leinen 78,— DM 


Zu beziehen 
durch Buchhandlungen 
und Fachgeschäfte 


Verlagsbuchhandlung 
Paul Parey 
Spitalerstraße 12 
2000 Hamburg 1 


PAUL 
PREV 


| 47 (6), 321 ml: Dezember 1982 ISSN 0044-3468 C21274F 


_ Mammals 


EITSCHRIFT FÜR 
>ÄUGE TIERKUNDE 


NTERNATIONAL JOURNAL 
IF MAMMALIAN BIOLOGY 


gan der Deutschen Gesellschaft für Säugetierkunde 


<er, Victoria; Hellwing, S.; Terkel, J.: Parental behaviour in male and virgin white-toothed shrews Crocidura 
ssula monacha (Soricidae, Insectivora). — Elterliches Verhalten von Männchen und Jungweibchen bei der 
itzmaus Crocidura russula monacha (Soricidae, Insectivora) 321 


ar, G.; Schmidt, U.: Der Einfluß visueller Information auf die Echoortung bei Phyllostomus discolor (Chiroptera). 
The influence of visual information on echolocation in Phyllostomus discolor (Chiroptera) 324 


„W.; Alonso, Carmen; Langguth, A.: Field observations on Callithrix jacchus jacchus L. — Feldbeobachtungen 
Callithrix jacchus jacchus L. 334 


“Rowe, D. T.; Meester, J.: Population dynamics of small mammals in the Drakensberg of Natal, South Africa. — 
pulationsdynamik kleiner Säugetiere in den Drakensbergen von Natal, Südafrika 347 


yerl, Roswitha: Die Chromosomen von Microtus arvalis (Rodentia, Microtinae). — Chromosomal studies in 
srotus arvalis (Rodentia, Microtinae) 356 


R. S.; Silva L., J. L.; Valdes L., J. A.: Feeding behavior and diets of Neotropical water rats, genus /chthyomys 
omas, 1893. — Ernährungsweise und Nahrung der neotropischen Wasserratte, Genus Ichthyomys Thomas, 


33 364 

ar, D.; Wollnik, Franziska: Untersuchungen zur Kurzzeitperiodik beim Meerschweinchen (Cavia aperea f. 

cellus). Teil I. - Investigations on ultradian rhythms in the guinea pig (Cavia aperea f. porcellus). Part I. 370 

ınschaftliche Kurzmitteilung 

naaker, L. C.: The type locality of the Javan rhinoceros (Rhinoceros sondaicus Desmarest, 1822). — Die Terra 

ica des Java-Nashorns (Rhinoceros sondaicus Desmarest, 1822) 381 
383 
384 


erlag Paul Parey Hamburg und Berlin 


HERAUSGEBER/EDIORS 


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

Kiel - H.-G. Kıös, Berlin - H.-J. Kunn, Göttingen - B. Lanza, Florenz - T. C. S. 

MORRISON-SCOTT, London - J. NIETHAMMER, Bonn - H. REICHSTEIN, Kiel - M. Rönrs, 

Hannover - D. STArRcK, Frankfurt a. M. - F. Strauss, Bern - E. THENIUS, Wien - 
W.VERHEYEN, Antwerpen 


SCHRIETFEITUNG/EDITORITAT-OERI@R 


H. SCHLIEMANN, Hamburg — D. Kruska, Kiel 


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. Harald Schliemann. 


Manuskripte: Manuskriptsendungen sind zu richten an die Schriftleitung, z.Hd. Prof. Dr. Dieter 
Kruska, Institut für Haustierkunde, Biologie-Zentrum, Neue Universität, Olshausenstr. 40-60, 
D-2300 Kiel. Für die Publikation vorgesehene Manuskripte sollen gemäß den „Redaktionellen 
Richtlinien“ abgefaßt werden. Diese Richtlinien sind in deutscher Sprache Bd. 43, H. 1 und in 
englischer Sprache Bd. 43, H. 2 beigefügt; in ihnen finden sich weitere Hinweise zur Annahme von 
Manuskripten, Bedingungen für die Veröffentlichung 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 
Vervielfältigung auf photomechanischem oder ähnlichem Wege oder im Magnettonverfahren 
sowie der Speicherung ın Datenverarbeitungsanlagen, bleiben, auch bei nur auszugsweiser Ver- 
wertung, vorbehalten. Werden von einzelnen Beiträgen oder Beitragsteilen einzelne Vervielfälti- 
gungsstücke in dem nach $54 Absatz 1 UrhG zulässigen Umfang für gewerbliche Zwecke 
hergestellt, ist dafür eine Vergütung gemäß den gleichlautenden Gesamtverträgen zwischen der 
Verwertungsgesellschaft Wort, vereinigt mit der Verwertungsgesellschaft Wissenschaft GmbH, 
rechtsfähiger Verein kraft Verleihung, Goethestr. 49, D-8000 München 2, und dem Bundesver- 
band der Deutschen Industrie e.V., dem Gesamtverband der Versicherungswirtschaft e.V., dem 
Bundesverband deutscher Banken e.V., dem Deutschen Sparkassen- und Giroverband und dem 
Verband der Privaten Bausparkassen e.V., an die Verwertungsgesellschaft zu entrichten. Erfolgt 
die Entrichtung der Gebühren durch Wertmarken der Verwertungsgesellschaft, so ist für jedes 
vervielfältigte Blatt eine Marke im Werte von 0,40 DM zu verwenden. Die Vervielfältigungen sind 
mit einem Vermerk über die Quelle und den Vervielfältiger zu versehen. 


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 internal use 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 
ee 


Fortsetzung 3. Umschlagseite 


© 1982 Verlag Paul Parey, Hamburg und Berlin. - Printed in Germany by Westholsteinische Verlagsdruckerei 
Boyens & Co., Heide/Holst. 


Z. Säugetierkunde 47 (1982) 6, 321-384 


InterCode: ZSAEA 7 47 (6), 321-384 (1982) 


Mit 3 Beilagen des Verlages Paul Parey und 1 Beilage des Verlages Bernd Busse, Bonn 


Parental behaviour in male and virgin white-toothed 
shrews Crocidura russula monacha (Soricidae, Insectivora) 


By VICTORIA SOROKER, $. HELLWING and J. TERKEL 


Department of Zoology, George S. Wise Faculty of Life Sciences, Tel Aviv University, Tel Aviv, Israel 


Receipt of Ms. 10. 3. 1982 
Abstract 


Both the parental behaviour of male shrews toward their own young as well as virgins to foster young 
was examined. In the intact family group, ninety percent of the males crouched over young in the 
family nest (Experiment 1). In the absence of the mother, males only crouched over the young which 
were in the nest and only one out of five retrieved scattered young (Experiment 2). Thus it appears that 
the contributions of fathers to the care of young is limited to bodily contact which may be an 
important source of heat for the young. Virgin shrews did not exhibit spontaneous maternal behaviour 
and could not be induced to become maternal by exposure to 3 pups for 6 hours daily for 10 days 
(Experiment 3). The pups were most commonly ignored by virgins although agonistic behaviour was 
seen ın 25 % of the observations. 


Introduction 


Parental behaviour has been studied ın a varıety of mammalian species, but only little is 
known about this behaviour in shrews (Soricidae; Insectivora). Since the insectivores are 
generally regarded as the remote ancestors of all living eutherian mammals (RoMER 1970) it 
would seem important to learn more about the behaviour of shrews in comparison with 
mammals of higher orders. 

Descriptive studies of maternal care in shrews include general reproductive behaviour ın 
Sorex araneus (CROWCROFT 1957), mutual mouth licking between mother and young in 
Suncus murinus (STINE and DRYDEN 1976) and caravannıng in several species of Soricidae- 
Crocidurinae (GRÜNWALD and MÖHRES 1974). Nursing, caravannıng, retrieving and nest- 
building were studied in lactating white-toothed shrews Crocidura russula monacha 
(©. r. m.) by Herıwıng (1971a). 

In laboratory rats (Rattus norvegicns) prolonged exposure to young induces the onset of 
parental behaviour in males and nulliparous females (ROsSENBLATT 1967). In laboratory 
mice (Mus musculus) parental behaviour occurs spontaneously in both sexes (NOIROT 
1972). Studies of responsiveness to young in shrews other than the mother are very scanty. 
ConAawAY (1958) reported that adult non-lactating females of least shrews (Cryptotis 
parva) cared for alien young and that males also showed considerable “maternal 
behaviour”. In a preliminary study on white-toothed shrews, HeLıwınG (1971a) found 
that virgins were aggressive towards young while four out of ten males retrieved their 
offspring when the mothers were removed from the nests. 

The purpose of the present study was to learn about paternal participation in raising 
pups in Crocidura russula monacha and to determine whether virgin females of this species 
show maternal behaviour and, if not, whether it is possible to induce such behaviour by 
exposing virgin animals to young, as was demonstrated in the rat (ROSENBLATT 1967). 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4706-0321 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 321-324 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


322 Victoria Soroker, S. Hellwing and ]J. Terkel 


Methods 
Animals 


Subjects were white-toothed shrews reared in captivity at the Canadian Center of Ecological Zoology 
of Tel Aviv University. Breeding pairs were housed outdoors on stainless steel cages (49 x 34 x 20 
cm) with bedding consisting of a sand layer covered with wood shavings, and straw provided as 
nesting material. Throughout the year, the duration of daylıght ranged from 10-14 hours, the 
temperature varıed from 8 °C (January) to 30 °C (August) (20 °C annual average) and the mean 
relative humidity was 70 %. Animals were fed (ad libitum) with fly larvae and a mixture of ground 
meat, hard-boiled eggs and milk powder. For details of the maintenance of the shrew colony see 
Heııwıng (1971b). 


Experiment 1 


To examine the interaction between the father and offspring in the intact family unit, the male’s 
location with respect to the young was recorded daily ın 23 pairs, from the day of birth until weaning 
23 days later. Two clay pots for nesting and shelter were placed side by side at the rear of the cage. 
Every morning, the clay pots were lifted and the male’s location (in the pup’s nest or in another nest) 
was recorded. 


Experiment 2 


To check whether the presence of the mother influences the interaction between the male and his 
offspring, the behaviour of 5 fathers was recorded in the absence of their mates. On days 2, 4, 6 and 8 
postpartum, each male was left alone in the cage following removal of the mother and young. Thirty 
minutes later, 3 of the offspring were placed in the middle of the cage in a small depression in the sand 
20 cm away from the nest site. Observations were recorded for 15 minutes after presentation of young 
and 3 spot checks of the location of the male and young were made at one hour intervals. At the end of 
the 3 hour observation period, the mother was returned to the family cage. 


Experiment 3 


In this experiment, the response to young in nulliparous female shrews was tested. Fifteen sexually 
mature virgins were housed individually in glass cages (46 X 23 x 30 cm) wrapped with red paper (to 
reduce the amount of light entering the cage) and provided with paper strips (12 X 1.2 cm) as nesting 
material. After 2-3 days of habituation to the cage, each virgin was exposed to 3 pups (2-8 days old), 
for 6 hours daily, for 10 days. The young were placed in a depression in the middle of the cage and the 
female’s behaviour was recorded (on a standard observation sheet) for 15 minutes followed by hourly 
spot checks for 6 hours. When a pup was detected bitten or killed, the remainder were removed from 
the cage and on the next day the female was exposed to a single pup (usually 6-8 days old). If this pup 
was unharmed, the female again received the standard litter of 3 pups in the following tests. The pups 
were provided by donor mothers and returned to them at the end of the 6-hour test-period. 


Results 
Experiment 1 


All 23 males were observed to maintain body contact with young in the family nest. The 
daily percentage of males found in the nest with the young ranged between 80 % to 100 % 
throughout the 23 days postpartum. However, males were never seen to exhibit more 
active components of parental behaviour, e.g. retrieving, caravanning or nest-building. 


Experiment 2 


Even in the absence of their mates, males did not retrieve young to the nest site, excepting 
on a single occasion. Pups aged 6-8 days crawled to the nest by themselves and crouching 
behaviour occurred in 75 % of 28 cases in which young were present in the nests. Males 
never crouched over young in the sand depression outside the nest. Only on one occasion 
was a male observed to crouch over younger pups. 


Parental behaviour in male and virgin Crocidura russula monacha 323 


Experiment 3 


During the 10 days of tests, most of the 15 virgins ignored the pups on most tests. 
Complete maternal behaviour (crouching, retrieving or caravanning, nest building, ano- 
genital licking) did not occur spontaneously on the first daily test nor did it develop as a 
result of daily exposure to pups. Only 5 females crouched over pups during some spot 
checks (11 % of mean of daily sample means). 


Discussion 


In some mammalıan species, males are prone to killing young, and females usually prevent 
them from having access to their offspring. In many species of rodents males show 
caretaking behaviour which includes nest-building, defence of home range and nest, 
retrieving, licking of young and crouching over young (SPENCER-BOOTH 1970). The 
present study shows that male white-toothed shrews do not normally kill young, and the 
females, who tolerate the males’ presence in the nest, permit them to maintain body contact 
with the offspring. Of the repertoire of parental activities, male white-toothed shrews 
almost exclusively exhibited crouching over young. The absence of other components of 
parental care cannot be attributed to suppression by the females since even in their absence 
the males’ parental repertoire consisted solely of crouching. 

Shrews are among the smallest of mammals and their energetic requirements for 
maintaining homeothermy are extremely high. Conservation of body heat of the young 
may be of primary importance for their successful development, and this would lend great 
ecological and physiological significance to the males’ crouching behaviour over the young. 
In addition, since shrews spend much of their time obtaining food, such parental behaviour 
may permit lactating mothers to spend less time in the nest and more time foraging. In fact, 
males were frequently observed to crouch in the family nest while the females were feeding 
in the food dish. 

Virgin shrews failed to show maternal behaviour either spontaneously or by induction 
during 10 days of exposure to pups (Experiment 3). These females did not kill pups, rather 
they usually ignored them. This pattern differed from the behaviour observed in virgin 
females of domesticated rodent species in which most animals show maternal behaviour 
spontaneously (laboratory Mus musculus and some Rattus norvegicus strains, JAKUBOWSKI 
1979), or after continuous exposure to pups (golden hamsters Mesocricetus auratus, 
SwAnsoN and CAMPBELL 1979; rats Rattus norvegicus, ROSENBLATT 1967). However, naive 
virgin wild Mus musculus show no spontaneous maternal behaviour, nor can they be 
induced to become maternal by exposure to young (JAKUBOWsKI 1979). We speculate that 
spontaneous maternal behaviour may be an artifact of domestication which does not 
ordinarıly occur in wild members of a mammalıan species; consequently virgin shrews 
showed no spontaneous maternal behaviour, although the shrew breeding colony exists for 
about 15 years at the research zoo at Tel Aviv University. 


Zusammenfassung 


Elterliches Verhalten von Männchen und Jungweibchen bei der Spitzmaus Crocidura russula monacha 
(Soricidae, Insectivora) 


Die Bereitschaft zur Jungenfürsorge wurde bei Vätern und jungen Weibchen der Hausspitzmaus 
(Crocidura russula monacha) geprüft. Der Beitrag der Väter beschränkte sich auf anhaltenden 
Körperkontakt, dessen Bedeutung für die Jungenentwicklung diskutiert wird. Weder unerfahrene 
Jungweibchen noch solche, die bereits zehn Tage lang mit Jungtieren zusammen waren, zeigten 
mütterliche Verhaltensweisen. Gewöhnlich kümmerten sie sich nicht um die Jungen. 


324 Victoria Soroker, S. Hellwing and ]. Terkel 


References 


Conaway, C. H. (1958): Maintenance, reproduction and growth of the least shrew in captivity. ]. 
Mammal. 39, 507-512. 

CROWCROFT, P. (1957): The life of the shrew. London: Max Reinhardt. 

GRÜNWALD, A. von; MÖHRES, F. P. (1974): Beobachtungen zur Jugendentwicklung und Karawanen- 
bildung bei Weißzahnspitzmäusen (Soricidae-Crocidurinae). Z. Säugetierkunde 39, 321-339. 
Herıwıng, $. (1971a): Reproduction in the white-toothed shrew, Crocidura russula monacha 

Thomas, 1906 (Soricidae, Insectivora, Mammalıa). Ph. D. thesis, Hebrew Univ. 

— (1971b): Maintenance and reproduction in the white-toothed shrew, Crocidura russula monacha 
Thomas, in captivity. Z. Säugetierkunde 36, 103-113. 

Jaxusowskı, M. (1979): Infanticide and parental care in virgin rats and mice: genetic, social and neural 
control. M. Sc. thesis, Tel Avıv Univ. 

No1RoT, E. (1972): The onset of maternal behaviour in rats, hamsters and mice: A selective review. 
Advances in the Study of Behavior 4, 107-146. 

RoMER, A. S. (1970): The vertebrate body. 4th ed. Philadelphia, London, Toronto: W. B. Saunders 
Company. 

ROSENBLATT, J. A. (1967): Nonhormonal basıs of maternal behaviour in the rat. Science 156, 
1512-1513. 

SPENCER-BOOTH, Y. (1970): The relationships between mammalıan young and conspecifics other than 
mothers and peers: a review. In: Advances in the Study of Behavior 3, 120-194. Ed. by D. S. 
LEHRMAN, R. A. Hınpe and E. Shaw. New York: Academic Press. 

STINE, C. J., DryDEn G. L. (1976): Lip-licking behavior in captive musk shrews, Suncus murinus. 
Behaviour 62, 298-311. 

Swanson, L. H.; CampBELı, C. S. (1979): Induction of maternal behavior ın nulliparous golden 
hamsters (Mesocricetus auratus). Behavioural and Neural Biology 26, 364-371. 


Authors’ address: VICTORIA SOROKER, Dr. SHLOMO HELLwInG and Dr. JOSEPH TERKEL, Department 
of Zoology, George S. Wise Faculty of Life Sciences, Tel Avıv University, Tel Aviv, 
Israel 


Der Einfluß visueller Information auf die Echoortung 
bei Phyllostomus discolor (Chiroptera) 


Von G. ROTHER und U. SCHMIDT 


Zoologisches Institut der Universität Bonn 


Eingang des Ms. 26. 4. 1982 
Abstract 


The influence of visual information on echolocation in Phyllostomus discolor (Chiroptera) 


Investigated was the obstacle avoidance performance in a flight tunnel of the neotropical bat 
Phyllostomus discolor. The orientation sounds resemble the typical phyllostomatid type (duration: 0.3 
to 2.5 ms; 3 downwards modulated harmonics, which cover the frequency range from ca. 45-100 
kHz). In flight the intensity of the sounds reaches 124 dB (calculated to a distance of 10 cm in front of 
the nose); when approaching the landing site the intensity falls to ca. 75 dB. 

Wires down to a diameter of 0.25mm can be detected acoustically; this threshold value is not 
improved by visual information. In the above-threshold-range the repetition rate in front of the 
obstacle is significantly lower, when the wires can be localized visually. In this situation body contacts 
with the obstacle are much reduced compared with flights in the dark tunnel. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4706-0324 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 324-334 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Echoortung bei Phyllostomus discolor 3225 


Einleitung 


Alle Microchiropteren sind in der Lage, allein mit Hilfe des Echoortungssystems sicher zu 
fliegen. Eine Fülle von Untersuchungen deutet jedoch darauf hin, daß bei vielen Fleder- 
mausarten auch der Gesichtssinn eine nicht unwesentliche Rolle spielt. 

Schon frühzeitig gelang der Nachweis, daß Fledermäuse zu einer Hell-Dunkel-Unter- 
scheidung fähig sind (EisENTRAUT 1950; Curris 1952; Davıs und BARBOUR 1965); auch 
die Merkmale optischer Muster können von einigen Arten unterschieden werden (JAsTR- 
ZEBSKI 1958; SUTHERS et al. 1969; MASTERsSoN und ELLins 1974; MANSKE und SCHMIDT 
1979). Verschiedene Kleinfledermäuse sind außerdem in der Lage, zum Teil sehr kleine 
Hindernisse in ihrem Flugweg optisch zu entdecken und ihnen auszuweichen. CHAsE und 
SUTHERS (1969) konnten zeigen, daß sehende, mit Ohrstopfen vertaubte Phyllostomus 
hastatus und Carollia perspicıllata Kollisionen mit 30cm breiten, weißen Stoffstreifen 
signifikant besser vermeiden als geblendete und taube Tiere. Nicht hörende Myotis 
lucifugus kollidieren nicht so häufig mit 2mm dicken Baummwollfäden, wenn eine 
optische Orientierung möglich ist (BRADBURY und NOTTEBOHM 1969). 

Der visuelle Sinn wird in unterschiedlichen Funktionszusammenhängen eingesetzt. In 
einer Reihe von Freiland-Untersuchungen stellte sich heraus, daß Fledermäuse ihre 
Tagesquartiere nicht wiederfinden, wenn sie mit abgedeckten Augen in größerer Entfer- 
nung von ihrer heimatlichen Höhle freigelassen werden (BARBOUR et al. 1966; WILLIAMS et 
al. 1966; WırLıams und WILLıams 1967; Davıs and BARBOUR 1969). Neben der Fernorien- 
tierung dient der Gesichtssinn, besonders bei Fledermäusen, die relativ helle Ruheplätze 
bevorzugen (z.B. Saccopteryx, Artibeus), auch der passiven Überwachung der Umgebung 
(SUTHERS 1970). In starkem Maße werden Aktivität und circadiane Rhythmik vom Licht 
bestimmt (GRIFFIN und WELSH 1937; CHUrcH 1957; DE Coursey und DE Coursey 1964; 
ERKERT 1974; ERKERT et al. 1980), und selbst das Fluchtverhalten unterliegt einem 
visuellen Einfluß (CHase 1981). 

Bei der neotropischen Phyllostomus discolor deuten die sehr gut ausgebildeten Augen 
auf ein leistungsfähiges visuelles System hin. Unsere Untersuchungen sollten klären, ob 
Echoortung und Gesichtssinn bei der Orientierung zusammenwirken. 


Material und Methode 


Als Versuchstiere (VT) dienten 3 männliche adulte ?. discolor (Kleine Lanzennase, Abb. 1); sie wurden 
in einem klimatisierten Raum (23 °C; ca. 70 % relative Luftfeuchtigkeit) gehalten und mit Bananen, 
Haferflocken, „Mehlwürmern‘ und subadulten Mäusen gefüttert. 

Die Flugversuche wurden in einem Flugtunnel (3 X 0,8 x 0,8 m) durchgeführt (Abb. 2), dessen 
Boden, Decke und Seitenwände zur Reflexionsverminderung mit schaumstoffbeschichtetem, schwar- 
zem Stoff verkleidet waren. Die Zielwand bestand aus grauem Kunststoff (Trovidur®); im oberen 
Drittel befand sich ein Landefeld (30 x 30cm) aus schwarzem Stoff. In der Fluganlage waren 15 
Lichtschranken in vier Reihen angeordnet (2,7 m, 2,0 m, 1,2 m und 0,4m von der Zielwand entfernt). 
Die Unterbrechung der Lichtschranken durch das fliegende Tier wurde über die Lichtschranken- 
Markierung mit Signalen unterschiedlicher Amplitude auf dem Magnetbandgerät registriert. Im Boden 
des Flugtunnels waren unmittelbar vor der Zielwand zwei 40-W-Lampen angebracht, durch die die 
Landewand von unten diffus beleuchtet werden konnte (Beleuchtungsstärke ca. 101x) (die Licht- 
schranken allein bewirkten eine Beleuchtungsstärke von 0,03 Ix). Die Leuchtdichte der Zielwand bei 
ein- und ausgeschalteten Lampen ist in der Tabelle mit Werten für Voll- und Neumond verglichen. 

In der Mitte des Flugtunnels konnte ein Hindernis eingeschoben werden. Als Hindernisse dienten 
schwarze Plastikstreifen mit rechteckigem Querschnitt (40, 10 und 5 mm) sowie schwarze Nylonfäden 
mit rundem Querschnitt (2, 1, 0,5, 0,25 und 0,125 mm), die vertikal in einem Rahmen eingespannt 
waren. Ihre seitliche Position innerhalb des Flugtunnels konnte von außen verändert werden. Die 
Hindernisse erschienen bei beleuchteter Zielwand als dunkler vertikaler Streifen vor einem relativ 
hellen Hintergrund, bei ausgeschalteten Lampen hoben sie sich optisch nicht von der Zielwand ab. 

In der Mitte der Zielwand befanden sich ein geeichtes Ultraschall-Kondensatormikrophon (B & K 


326 G. Rother und U. Schmidt 


Typ 4135 mit Vorverstärker B & K 2618) und das Mikrophon eines Bat-Detectors (QMC Modell 
S 100). Die Signale beider Mikrophone wurden auf einem Magnetbandgerät (Lennartz 6000/800) 
aufgezeichnet (Bandgeschwindigkeit 76cm/s), das gleichzeitig die Lichtschranken-Markierung und 
einen Kommentar aufnahm. Die Auswertung der Lautaufnahmen erfolgte mittels Speicheroszillo- 
graph (Tektronix 5113), Sonagraph (Kay 7029 A) und Frequenzanalysator (Nicolet UA 500 A). 


& 


mÜO)OQO-n 


Meßver- 
Tonband - 
gerat 


Bat 


1m 


Abb. 2. Schema des Flugtunnels und der Lautaufnahme-Apparatur. A-D = Abschnitte der Flugbahn; 

E = Ultraschall-Kondensatormikrophon; F = Mikrophon des Überlagerungsdetektors; G = 

Mikrophon für Kommentare; H = Hindernisposition; ] = Zielwand-Beleuchtung; LM = Licht- 
schranken-Markierung; $S = Startwand; Z = Zielwand 


Echoortung bei Phyllostomus discolor 927, 


Anhand der Lichtschranken-Markierung konnte zu jedem Laut der Ort der Lautaussendung berech- 

net werden. Der Schalldruck der Laute wurde auf eine Distanz von 10cm vor der Nase des Tieres 

bezogen, bei Annahme einer quadratischen Reduzierung der Energie mit der Entfernung und einer 

atmosphärischen Abschwächung um 4,5 dB/m (GRIFFIN 1971; JOERMANN und SCHMIDT 1981a). 
Während einer zweimonatigen 

Eingewöhnungsphase, in der die VT 

im Flugtunnel lebten und an der Ziel- Tabelle 

wand ihr Futter erhielten, wurden sıe } : S 

daran gewöhnt, Mehlkäferlarven aus Leuchtdichte der Zielwand (Reflexionsgrad 0,0073 asb/lx) 

einer Pinzette anzunehmen. Danach bei verschiedenen Beleuchtungsstärken 

wurden die Tiere trainiert, den Flug- 

tunnel in direktem, ununterbroche- 


nem Flug zu durchqueren, die als Be- Bmneen 2 x 10° cd/m? 
lohnung durch eine Öffnung im 2 ar 
Landefeld gebotene Mehlkäferlarve Lampe aus 6 x 10% cd/m? 
abzunehmen und sofort zur Start- 


wand zurückzufliegen. Nach 5mona- vollmond 6 x 105 cd/m? 


tigem Training wurden die Kontroll- 
flüge (ohne Hindernis) aufgezeichnet. | Neumond X 102m? | 
Die Versuchsflüge begannen mit dem 
breitesten Hindernis; in den folgen- 


den Tagen fand eine schrittweise Re- 

duzierung der Hindernisbreite statt. 

Um eine Gewöhnung der VT an den Ort der Hindernisse nach Möglichkeit auszuschließen (GRIFFIN 
et al. 1963; NEUWEILER und MÖHRES 1967), erfolgte eine unregelmäßige seitliche Verschiebung der 
Hindernisposition (+ 10cm). Unmittelbar im Anschluß an die letzten Flüge mit dem 0,125-mm- 
Hindernis wurden wiederum Kontrollflüge durchgeführt. Bei allen Experimenten wechselten Hell- 
und Dunkelflüge in unregelmäßiger Folge. 


Ergebnisse 
Ruhelaute 


Phyllostomus discolor gehört zu den nasal emittierenden Fledermäusen. Beim Orten 
befindet sich das prominente Nasenblatt in ständiger Bewegung, das Maul bleibt stets 
geschlossen. Die allgemeine Charakterisierung der Ortungslaute erfolgt anhand der Ruhe- 
laute, die von ın der Hand gehaltenen Tieren gewonnen wurden. Je VT wurden 60 
Ruhelaute analysiert; es fanden sich keine Unterschiede zwischen ım Hellen und ım 
Dunkeln aufgezeichneten Lauten. 

Zu Beginn der jeweils etwa dreiminütigen Aufnahmezeit zeigten alle Tiere eine rege 
Ortungsaktivität. Die Laute wurden vorwiegend in Lautgruppen mit zwei oder drei 
Lauten abgegeben. Bei den 180 untersuchten Lauten traten Einzellaute nur fünfmal und 


n 1207a D 

E 

= 80 \ | 
5 \ N 
Sn 

ae) 


Klammern Rn Br Br 
Abb. 3. Sonagramme eines Ruhe- (a) und eines OmmEZE u 6 O2 I 6 
Fluglautes (b) Zeit [ms] 


Lautgruppen mit mehr als drei Lauten nur zweimal auf. Die Gruppenintervalle lagen 
zwischen 70 und 410 ms (x = 160 ms), die Intervalle innerhalb der Gruppen zwischen 23 
und 5lms (x =38 +#8ms, =26 Laute/s). Etwa eine halbe Minute nach Beginn der 
Aufnahme wurde die Ortungsaktivität der Tiere deutlich geringer. Es traten Pausen bis zu 
einer Sekunde und länger auf, abgelöst von kurzen Ortungssalven und in längeren 


328 G. Rother und U. Schmidt 


Abständen abgegebenen Einzellauten. Die Ruhelaute von Phyllostomus discolor entspre- 
chen dem allgemeinen Phyllostomatidentyp. Es sind sehr kurze Laute (0,4-1,4ms; 
x = 0,75 # 0,2 ms), die eine geringe Intensität besitzen. Der maximale Schalldruck betrug 
83-86 dB (alle dB-Angaben sind bezogen auf 2 X 10° N/m?). Die Laute setzen sich aus 
drei abwärts modulierten Frequenzkomponenten zusammen. Die untere Komponente 
überstreicht den Frequenzbereich von ca. 73 bis 45 kHz. Nach ca. einem Drittel des Lautes 
beginnt die mittlere Komponente, die von ca. 90 kHz auf 61 kHz abfällt. Die obere 


= 50 
210 
pn} 
© 
0 
Pe \ 
3 | 

20 | 

| 
HM ik 
: Abb. 4. Frequenzspektrogramm eines 
Ruhelautes 
0 20 40 60 80 100 120 


Frequenz [kHz] 


Komponente überstreicht den Bereich von ca. 100 bis 78 kHz und findet sich im letzten 
Drittel des Lautes (Abb. 3a). Diese drei Anteile stellen 3. bis 5. Harmonische dar; die 1. 
und 2. Harmonische treten im Laut nicht auf. Ca. 100 us nach Ende des Lautes läßt sich 
häufig eine extrem kurze Komponente registrieren, die besonders in Frequenzspektro- 
grammen auffällt (Abb. 4). Sie setzt sich aus zwei schmalbandigen Frequenztanteilen 
zusammen, die Maxıma bei 16,6 # 1,4 kHz und 34,9 # 2,5 kHz besitzen. Der niederfre- 
quente Gipfel ist nur etwa halb so intensiv wie der höherfrequente und tritt daher im 
Sonagramm nicht in Erscheinung. 


Lautaussendung im freien Flug 


In der ersten Versuchsphase waren keine Hindernisse im Flugtunnel plaziert. Bei diesen 
freien Flügen durch den Tunnel erfolgte die Lautaussendung sehr stereotyp; es zeigten sich 
auch keinerlei Unterschiede zwischen Hell- und Dunkelflügen. Zu Beginn des Fluges 
wurden vorwiegend Einzellaute abgegeben. Später folgten Lautgruppen mit erst 2, dann 3, 
4 und mehr Lauten (Abb. 5a). Die Intervalle zwischen den Einzellauten betrugen 85-95 


a b c 


(ur Br mean‘ 


0.5s 


Abb. 5. Oszillogramme von Lautserien bei drei Flügen. a: ohne Hindernis und Beleuchtung; b: 
0,5 mm breites Hindernis und Beleuchtung; c: 0,5 mm breites Hindernis ohne Beleuchtung. Die Pfeile 
markieren die Hindernisposition; Flugrichtung von links nach rechts 


ms, was einer Wiederholrate von 10,5-11,8 Lauten/s entspricht. Innerhalb der Gruppen 
fanden sich Wiederholraten von 50 bis zu 90 Lauten/s. Die Zeitabstände zwischen den 
Lautgruppen verkürzten sich im Laufe des Fluges von im Mittel 63 ms auf 30 ms. Die 
stetige Verkürzung dieser Gruppenintervalle und der Intervalle innerhalb der Lautgruppen 


Echoortung bei Phyllostomus discolor 329 


bewirkt, daß die Zeit vom ersten Laut einer Gruppe bis zum ersten Laut der folgenden 
Gruppe konstant 85-95 ms beträgt. Die Dauer der Laute nimmt während des Fluges 
sukzessive ab (Abb. 6); die größte gemessene Dauer betrug 2,5 ms, die kleinste 0,3 ms. 
Neben der Dauer verändert sich auch die Intensität der Laute im Fluge. Bei Annäherung an 
die Zielwand verringert sıch der Schalldruck (bezogen auf einen Abstand von 10cm vor der 
Nase) in zunehmendem Maße. Bei einer Mikrophonentfernung von 2,7-2,0 m (Abschnitt 
A, Abb. 2) beträgt die Intensität im Mittel etwa 108 dB (n = 40 Flüge/Tier). Der maximale 
Schalldruck, der hierbei ermittelt wurde, lag bei 113 dB. In den beiden folgenden 
Abschnitten sinkt der mittlere Schalldruck zunächst auf etwa 102 dB (Abschnitt B), dann 


z 10 110 
Oo rs Eee: 
= 
5 100 100 
3 
3 
3 90 90 
80 80 
ll, 2.4 
Abb. 6. Schalldruck (umgerechnet auf 10 & u 50 
cm Abstand vor der Nase des Tieres), &° 
Lautdauer und Orte der Lautaussendung 3 16 16 
bei fünf Flügen. Für Schalldruck und 3, r 
Lautdauer sind Annäherungskurven an- * 2 
gegeben; senkrechte Balken bezeichnen * 
die Orte der Lautaussendung, waage- 04 04 
rechte fassen Lautgruppen zusammen. en nennen umge 
Die letzten 0,2 m Flugstrecke blieben a esch 
unberücksichtigt BEE NER BEL EN ar ee 
30 25 20 15 10 05 0 


Entfernung zum Mikrophon [m] 


_ 


00 


Frequenz [kHz] 


Abb. 7. Mittlere Frequenzbereiche 
der 3. (eng schraffiert), 4. (punktiert) 
und 5. (weit schraffiert) Harmoni- 
schen von Fluglauten der drei Tiere 
A,B und €. Senkrechte Balken: Stan- 
dardabweichung (A: n=26; B: 
„IN REN" TEE n = 23; C: n = 19) 


auf 92 dB (Abschnitt C), wobei im Endbereich Minimalwerte bis ca. 75 dB registriert 
werden konnten (Abb. 6). Auch die Fluglaute setzen sich im wesentlichen aus drei abwärts 
modulierten Frequenzkomponenten zusammen (Abb. 3b). Die Frequenzbereiche dieser 
drei Anteile decken sich mit denen der 3. bis 5. Harmonischen der Ruhelaute und bleiben 
im Verlauf des Fluges weitgehend konstant. Die Frequenzgrenzen dieser drei Harmoni- 
schen sind individuell unterschiedlich (Abb. 7). In den Fluglauten ist häufig die 2. und 6. 
Hamonische registrierbar, die die Frequenzbereiche von ca. 51 bis 35 kHz bzw. von ca. 102 
bis 86 kHz überstreichen. 


330 G. Rother und U. Schmidt 


Hindernisflüge 


Lautparameter 


Um Gewöhnungseffekte auszuschalten, wurden pro Hindernisbreite und Tier nur die 
ersten 5 Hell- bzw. Dunkelflüge ausgewertet; Hindernisposition und Beleuchtungsstatus 
waren zufällig verteilt. 

Die auffälligste Reaktion auf das Hindernis und die größten Unterschiede zwischen 
Hell- und Dunkelflügen fanden sich ın der Wiederholrate. In Abb. 5b, c fällt sofort auf, 
daf die Ortungsaktivität vor der markierten Hindernisposition gegenüber dem Kontroll- 
flug (Abb.5a) erhöht ist. Fast immer treten Lautgruppen an die Stelle von Einzellauten. 
Nach dem Passieren des Hindernisses tritt in vielen Flügen - immer jedoch bei den 
breiteren Hindernissen von 40 bis 5mm - eine „Pause“ auf. Dieses besonders lange 
Intervall zwischen zwei Lauten oder Lautgruppen ist mit 160-190 ms stets gerade etwa 
doppelt so lang wie die Intervalle zwischen Einzellauten bzw. Lautgruppen in den Flügen 
ohne Hindernis (85-95 ms). 

Auf der 1m langen Flugstrecke vor der Hindernisposition (bzw. bei den Kontrollflügen 
zwischen 2,5 und 1,5 m Mikrophonentfernung) wurden oszillographisch die Intervallzei- 
ten der Laute bestimmt und daraus die Wiederholrate berechnet (Abb. 8). Der drastische 
Anstieg der Ortungsaktivität bei der Einbringung des ersten, 40 mm breiten Hindernisses 
in den Flugweg gegenüber den Kontrollflügen ohne Hindernis ist besonders auffällig, auch 
wird hier der Einfluß der visuellen Information sehr deutlich: in den Dunkelflügen findet 
sich eine um durchschnittlich 10 Laute/s höhere Wiederholrate. Diese überproportionale 


Wiederholrate [s-!) 


20 


OGCHCHHCE 


Bas 
< 
rn 
&) 
oO 
un 
”m 
—S 
oO 
in 
oO 
8) 
u 
PS 
=) 
Bas 
< 
> 
(2) 
(5) 
un 
”m 
N 
(&) 
un 
() 
”m 
un 
Br 
=) 


Wiederholrate [s-'] 


GW: 


on or S i oe 
Hindernisbreite [mm] Hindernisbreite [mm] 


20 0 80,2 (nos X K 


Abb. 8. Wiederholraten bei unterschiedlicher Hindernisbreite und Beleuchtung auf der 1 m langen 

Flugstrecke vor dem Hindernis. A, Bund C zeigen Mittelwerte aus je fünf Flügen der drei Einzeltiere; 

D: Mittelwerte aller Tiere. Punktierte Säulen = mit Beleuchtung; schraffierte Säulen = ohne 

Beleuchtung; senkrechte Balken = Standardabweichung; K, = Kontrollflüge vor den Hindernisflü- 
gen; K, = Kontrollflüge nach den Hindernisversuchen 


Echoortung bei Phyllostomus discolor 331 


Reaktion ist auf eine Art „Überraschungseffekt‘“ zurückzuführen, der auftritt, wenn in 
einem bekannten Raum plötzlich Veränderungen auftreten (JOERMANN und SCHMIDT 
1981b). Bei den folgenden Flügen mit 10 und 5 mm breiten Hindernissen ist sowohl mit als 
auch ohne Zielwand-Beleuchtung ein Rückgang der Wiederholraten zu verzeichnen. Mit 
abnehmender Hindernisbreite (2 mm bis 0,25 mm) steigt die Wiederholrate wieder an. Die 
Unterschiede zwischen Hell- und Dunkelflügen bleiben erhalten. Außer bei Tier B, das 
sehr unregelmäßig flog, lassen sich die Differenzen bei dem 40-mm- und den 2- bis 0,25- 
mm-Hindernissen statistisch sichern (t-Test, p<0.01). In den nach Beendigung der 
Experimente mit dem 0,125-mm-Hindernis aufgezeichneten Kontrollflügen ist die Wie- 
derholrate niedriger als in den letzten Hindernisflügen, aber noch wesentlich höher als in 
den Kontrollen vor Beginn der Hindernisversuche, was wahrscheinlich auf eine Erwar- 
tungshaltung der Tiere gegenüber dem Hindernis zurückzuführen ist, wie sie auch für 
Myotıs, Eptesicus und Lasiurus (WEBSTER und BRAZIER 1968) und Desmodus (JOERMANN 
und SCHMIDT 1981b) beschrieben wurde. Das Fehlen helligkeitsbedingter Unterschiede der 
Wiederholrate bei hindernisfreien Flügen läßt sich bestätigen. 

Gegenüber den freien Flügen steigt die Intensität der Laute bei den Hindernisflügen an. 
Auf der 1m langen Flugstrecke vor dem Hindernis betrug die mittlere Intensität bei den 
Kontrollflügen ca. 103 dB, bei den Hindernisflügen zwischen 108 und 113 dB. Mit abneh- 
mender Hindernisbreite war ein leichter Anstieg der Intensität zu verzeichnen. Die 
maximale Intensität von Einzellauten betrug 124 dB. Signifikante Unterschiede zwischen 
Hell- und Dunkelflügen ließen sich nicht aufzeigen. 


Hinderniskontakte 


Bei den Flügen mit den schmaleren Hindernissen — ab 2mm Hindernisbreite - wurden 
Kontakte der VT mit dem Hindernis während des Vorbeifluges festgestellt. Dabei handelte 
es sich zunächst um mehr oder weniger leichte Anstöße mit dem Flügel, die die Tiere nicht 
an der Fortsetzung des Fluges hinderten. Solche Hinderniskontakte waren ın der Regel 
bereits während der Versuchsdurchführung gut hörbar und wurden entsprechend auf dem 
Kommentar-Kanal des Magnetbandgerätes angemerkt. Die oszillographische Auswertung 
der Hindernisflüge ermöglichte eine Kontrolle der kommentierten Hinderniskontakte mit 
Hilfe des von den Lautaufnahme-Mikrophonen ebenfalls registrierten Anstoßgeräusches. 

Bei Hindernissen ab 0,5 mm Breite traten auch massivere Hinderniskontakte der Tiere 
mit Oberarm oder Rumpf auf, die zu einem Absturz führten. Bei sehr schmalen Hindernis- 
sen (0,25 mm, 0,125 mm), bei denen die Zahl der Abstürze drastisch anstieg, wurde es 
zunehmend schwieriger, die Fledermäuse zu einem Abflug von der Startwand zu bewegen. 


% 100 


> Abb. 9. Verteilung der Hindernis- 
kontakte bei Flügen mit (offene Sym- 

bole) und ohne (geschlossene Sym- 

bole) Beleuchtung. Kreissymbole = 

0 . Flügelkontakte; Dreiecke = Abstür- 
ze; Ordinate = prozentuale Häufig- 

keit der Kontakte; Abszisse = Hin- 


) = dernisbreite 
10 5 2 1 05 025 0125 


Hindernisbreite [mm] 


332 G. Rother und U. Schmidt 


Tier B war nicht in der Lage, an dem 0,25-mm-Hindernis vorbeizufliegen: alle Durchflug- 
versuche endeten mit einem Absturz. Dieses Tier wurde hier nicht in die Auswertung 
einbezogen. Bis zu einer Hindernisbreite von 0,5 mm sind bei den Tieren A und © die 
Flügelberührungen in den Hellflügen signifikant seltener als in den Dunkelflügen (x?-Test, 
p<0.025). Bei dem 0,25-mm-Hindernis stieg die Anzahl der Abstürze sprunghaft an. Dies 
wird offensichtlich durch die Möglichkeit der optischen Orientierung nicht beeinflußt 
(Abb. 9). 


Diskussion 


Die kleine Lanzennase, Phyllostomus discolor, die vom südlichen Mexiko bis zum nördli- 
chen Argentinien verbreitet ist (JONES und CARTER 1976), findet sich vorwiegend in 
Waldgebieten (LA Var 1970; FLEminc et al. 1972). Beringungsexperimente ın Costa Rica 
haben gezeigt, daß diese Fledermausspezies saisonal die Tagesquartiere wechselt und dabei 
größere Entfernungen zurücklegt (WILKınson mdl.). P. discolor muß sich in einer Umge- 
bung orientieren, die sich durch dichten Bewuchs und somit mannigfaltige Hindernisse 
auszeichnet. Nahrungssuche und -erwerb sind durch die vielfältige Echostruktur einer 
solchen Umgebung gegenüber freiem Gelände bedeutend erschwert. Es ist zu erwarten, 
daß die Ultraschall-Laute der Tiere dieser komplexen Umgebung und der omnivoren 
Lebensweise angepaßt sind. 

Die herausragenden Charakteristika der Ortungslaute aller bisher untersuchten Phyllo- 
stomatiden sind eine geringe Lautdauer, niedrige Intensität sowie mehrere abwärts modu- 
lierte Harmonische, die einen großen Frequenzbereich überdecken. Die Blattnasenfleder- 
mäuse werden von GRIFFIN (1958) als ‚„flüsternde Fledermäuse‘ bezeichnet, da die 
Intensität ihrer Ortungslaute mit ca. 80 dB weit unter der Lautstärke der damals weitaus 
besser bekannten Rhinolophiden und Vespertilioniden lag. Die Ruhelaute von ?. discolor 
passen auch in dieses Schema. Wie die Flugversuche zeigen, können diese Fledermäuse die 
Intensität ihrer Laute im Bedarfsfall jedoch beträchtlich steigern. Bei der Detektion von 
Hindernissen oder, wenn sie ein entferntes Ziel anpeilen, wıe es das Landefeld ım 
Flugtunnel darstellt, werden Laute mit Schalldrucken von 110-125 dB erzeugt, d. h. Laute, 
deren Schallenergie das 25-35fache der Ruhelaute ausmacht. Eine ähnliche Erhöhung des 
Schalldrucks ist bei Hindernisversuchen mit Glossophaga soricina, Anoura geoffroy: und 
Choeronycteris mexicana beobachtet worden (HowELL 1974). Glossophaga zeigte hier eine 
Verdoppelung des Schalldrucks, wenn die Hindernisfäden dünner wurden. Auch ?. 
dıscolor reagiert in dieser Weise. 

Der Informationsgehalt eines Echos hängt von der Bandbreite des ausgesandten Signals 
ab, wobei breitbandige Signale mehr Informationen tragen als engbandige (Simmons et al. 
1977). Bei P. discolor beträgt die Bandbreite mehr als eine Oktave (45-100 kHz). Vesperti- 
lioniden, die in der Regel ähnlich breitbandige Laute mit nur einer Frequenzkomponente 
besitzen, erreichen die Abdeckung weiter Frequenzbereiche durch die zeitliche Ausdeh- 
nung des „sweeps“, P. discolor durch die Kombination von drei frequenzmodulierten 
Harmonischen. Dieser Lautaufbau erlaubt die Bildung extrem kurzer und trotzdem 
breitbandiger Ortungslaute. Dieser Lauttyp ist in einer komplexen Umgebung besonders 
leistungfähig. 

Obwohl die Echoortung bei allen Microchiropteren das dominierende Orientierungs- 
system darstellt, sind visuelle Einflüsse nicht ohne Bedeutung. In den Hindernisversuchen 
wurde das Ortungsverhalten von ?. discolor deutlich von der Umgebungshelligkeit beein- 
flußt. Bei erstmaligem Auftreten eines Hindernisses in ihrem Flugweg zeigten die VT eine 
„überproportionale Reaktion“ bezüglich der Wiederholrate der Laute, die jedoch nur in 
den Dunkel-Flügen zu beobachten war. Dieses Verhalten, das auch bei Desmodus rotundus 
auftritt (JOERMANN und SCHMIDT 1981b), ist um so erstaunlicher, als die Detektion des mit 
40 mm relativ breiten Hindernisses für die Tiere kein Problem darstellt und da sie das 


Echoortung bei Phyllostomus discolor 333 


gleiche Hindernis bereits einige Male bei eingeschalteter Beleuchtung passiert hatten. Wie 
die Abnahme der Wiederholrate bei den folgenden, schmaleren Hindernissen nahelegt, ist 
die „überproportionale Reaktion‘ vermutlich nicht zur Vermeidung eines Hinderniskon- 
taktes erforderlich, sondern auf eine Art „Überraschungseffekt“ zurückzuführen. Bei 
visuell lokalisierbarem Hindernis unterstützt das Auge die Detektion, so daß ein Gegen- 
stand ın der Flugbahn mit weniger Lauten erfaßt werden kann. 

Durch Auswertung der Kollisionen mit dem Hindernis läßt sich zeigen, daß P. discolor 
fähig ist, Fäden bis 0,25 mm Dicke im Flugweg akustisch zu lokalisieren. Dieser Schwel- 
lenwert läßt sich durch visuelle Unterstützung nicht verbessern; im überschwelligen 
Bereich jedoch hilft eine visuelle Lokalisation, Kontakte mit dem Hindernis drastisch zu 
reduzieren. 


Danksagung 


Herrn Dr. H. ERKERT, Zoophysiologisches Institut der Universität Tübingen, danken wir für die 
freundliche Überlassung der Versuchstiere; der Deutschen Forschungsgemeinschaft für finanzielle 
Unterstützung. 


Zusammenfassung 


Bei der neotropischen Fledermaus Phyllostomus discolor wurde in einem Flugtunnel das Orientie- 
rungsverhalten beim Vermeiden von Hindernissen (senkrechte Fäden) untersucht. Die Ortungslaute 
entsprechen dem Phyllostomatidentyp (Dauer: 0,3-2,5 ms; 3 abwärts modulierte Harmonische, die 
den Frequenzbereich von ca. 45-100 kHz überdecken). Im Flug erreicht die Intensität der Laute bis zu 
124 dB (10 cm vor der Nase); sie fällt bei Annäherung an den Landeplatz auf ca. 75 dB ab. 

Fäden bis 0,25 mm Durchmeser können akustisch lokalisiert werden; optische Einflüße verbessern 
diesen Schwellenwert nicht. Im überschwelligen Bereich ist die Wiederholrate der Laute vor dem 
Hindernis im Hellen jedoch signifikant kleiner. Auch Kontakte mit dem Hindernis treten hier seltener 
auf. 


Literatur 


BARBOUR, R. W.; Davıs, W.H.; HasseLı, M.D. (1966): The need of vision in homing by Myotıs 
sodalıs. J. Mammalogy 47, 356-357. 

BRADBURY, J. W.; NOTTEBOHM, F. (1969): The use of vision by the Little Brown Bat, Myotis lucifugus 
under controlled conditions. Anım. Behav. 17, 480-485. 

CHAsE, J. (1981): Visually guided escape responses of microchiropteran bats. Anım. Behav. 29, 
708-713. 

CHASE, J.; SUTHERS, R. A. (1969): Visual obstacle avoidance by echolocating bats. Anım. Behav. 17, 
201-207. 

CHURCH, H.F. (1957): The times of emergence of the Pipistrelle. Proc. Zool. Soc. London 128, 
600-602. 

Curtis, W.E. (1952): Quantitative studies of echolocation in bats (Myotis I. lucifugus), studies of 
vision of bats (Myotis I. Iucifugus and Eptesicus f. fuscus), and quantitative studies of vision of owls 
(Tyto alba practincola). Ph. D. Thesis, Cornell University, Ichaca, New York. 

Davıs, W. H.; BARBOUR, R. W. (1965): The use of vision in flight by the bat Myotis sodalis. The Am. 
Midland Nat. 74, 497-499. 

— (1969): Homing in blinded bats (Myotis sodalıs). J. Mammalogy 51, 182-184. 

DE Cours£y, G.; DE Course£y, P. J. (1964): Adaptive aspects of activity rhythms in bats. Biol. 
Bulletin 126, 14-27. 

EISENTRAUT, M. (1950): Dressurversuche zur Feststellung eines optischen Orientierungsvermögens 
der Fledermäuse. Jh. Ver. Vaterl. Naturk. Württ. 109, 34-45. 

ERKERT, H. G. (1974): Der Einfluß des Mondlichtes auf die Aktivitätsperiodik nachtaktiver Säuge- 
tiere. Oecologia (Berl.) 14, 269-287. 

ERKERT, H. G.; KracHT, S.; HÄussLEr, U. (1980): Characteristics of circadıan activity systems in 
neotropical bats. Proc. 5th Int. Bat Res. Conf. 95-103. 

Freming, T. H.; Hooper, E.T.; Wırson, D. E. (1972): Three Central American bat communities: 
structure, reproductive cycles and movement patterns. Ecology 53, 555-569. 

GRIFFIN, D.R. (1958): Listening in the dark. New Haven: Yale University Press. 

— (1971): The importance of atmospheric attenuation for the echolocation of bats (Chiroptera). 
Anım. Behav. 19, 55-61. 

GRIFFIN, D. R.; WELsH, J. H. (1937): Activity rhythms in bats under constant external conditions. ]J. 
Mammalogy 18, 337-342. 


334 G. Rother und U. Schmidt 


GRIFFIN, D. R.; McCue, ]J. J. G.; GRINNELL, A. D. (1963): Resistance of bats to jamming. J. Exp. 
Zool. 152, 229-250. 

Howeıı, D.J. (1974): Acoustic behavior and feeding in glossophagine bats. J. Mammalogy 55, 
293-308. 

JASTRZEBsKI, M. (1958): Preliminary investigations on the ability of the visual discrimination of flat 
figures in the bat Myotis myotis Borkhausen. Zeszyty Nankowe Uniwersytetu, Jagiellonskiego, 
Zoologica Z. 3, 191-202. 

JOERMANN, G.; SCHMIDT, U. (1981a): Echoortung bei der Vampirfledermaus, Desmodus rotundus. 11. 
Lautaussendung im Flug und Korrelation zum Flügelschlag. Z. Säugetierkunde 46, 136-146. 

— (1981b): Obstacle avoidance in the vampire bat (Desmodus rotundus). Proc. Ist Europ. Symp. Bat 
Res. In: Myotis 18-19, 142-148. 

Jones, J. K.; CARTER, D.C. (1976): Annotated checklist, with keys to subfamilies and genera. In: 
Biology of bats of the New World family Phyllostomatidae, T. 1. -— Spec. Publ. The Museum. 
Texas Techn. Univ. 

LA Var, R. K. (1970): Banding returns and activity periods of some Costa Rican bats. Southwest Nat. 
15, 1-10. 

MANSKE, U.; SCHMIDT, U. (1979): Untersuchungen zur optischen Musterunterscheidung bei der 
Vampirfledermaus, Desmodus rotundus. Z. Tierpsychol. 49, 120-131. 

MASTERSON, F. A.; Errıns, S. R. (1974): The role of vision in the orientation of the echolocating bat, 
Myotis lucifugus. Behaviour 51, 88-98. 

NEUWEILER, G.; MÖHRES, F. P. (1967): Die Rolle des Ortsgedächtnisses bei der Orientierung der 
Großblatt-Fledermaus Megaderma lyra. Z. vergl. Physiol. 57, 147-171. 

Sımmons, J. A. (Group report) (1977): Localization and identification of acoustic signals, with 
reference to echolocation. In: Recognition of complex acoustic signals. Ed. by T. BurLock. 
Dahlem Workshop, Berlin, 239-277. 

SUTHERS, R. A. (1970): Vision, olfaction and taste. In: Biology of bats, Vol. II. Ed. by W.A. 
WIMSATT. Academic Press; New York, London, 265-309. 

SUTHERS, R. A.; CHASE, ]J.; BRAFORD, B. (1969): Visual form dicrimination by echolocating bats. Biol. 
Bulletin 137, 535-546. 

WEBSTER, F. A.; BRAZIER, ©. G. (1968): Experimental studies on echolocation mechanisms in bats. 
Aerospace Medical Research Laboratories, Wright-Patterson Air Force Base, Ohio. 

WiLLıams, T. C., WILLIAMS, J. M. (1967): Radio tracking of homing bats. Science 155, 1435-1436. 

WILLıams, T. C., WILLIAMS, J. M.; GRIFFIN, D. R. (1966): The homing ability of the neotropical bat 
Phyllostomus hastatus, with evidence for visual orientation. Anım. Behav. 14, 468473. 


Anschrift der Verfasser: GEORG ROTHER, Prof. Dr. UwE SCHMIDT, Zoologisches Institut der Univer- 
sität, Poppelsdorfer Schloß, D-5300 Bonn 


Field observations on Callithrix jacchus jacchusL. 


By W. MAIER, CARMEN ALONSO and A. LANGGUTH 


Receipt of Ms. 25. 5. 1982 
Abstract 


Field observations on three groups of Callithrix jacchus jacchus L. living in patches of secondary 
coastal rain forest at Joao Pessoa (Paraiba, Brazil) are reported. The groups comprised 3-7 anımals 
which occupied partly overlapping home ranges of 2-5 ha. In the main study group, a smaller territory 
around a “home tree” was actively defended. This group was observed on four whole days. On an 
average about 30 % of the 11.5 hours of daily activity was spent on sap feeding on that “home tree” 
(Tapirira guianensis). The functional adaptations to and the biological importance of sap feeding are 
discussed. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4706-0334 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 334-346 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Field observations on Callıthrix jacchus jacchns L. 335 


Introduction 


The Tufted-ear or Common Marmoset Callıthrix jacchus jacchus L. of Northeast Brazil 
has been one of the best known pet-primates for centuries, and it is acommon member of 
zoological gardens and of primate research laboratories. There exist numerous studies on 
the ethology and physiology of the Common Marmoset under captive conditions (EPPLE 
1967, 1968; ROTHE 1975; STEVENSON and PooLE 1976; ROTHE et al. 1978), but there ıs not 
too much information available on the biology of wild living marmosets (CoIMBRA-FILHO 
1972; CoIMBRA-FILHO and MITTERMEIER 1977; STEvENSON 1978). Therefore, HErRsH- 
KOVITZ (1977) had good reason to state that “the home range or territory of wild living 
marmosets has not been studied. Even the nest or living quarters in the wild has not been 
described” (p. 545). 

This situation did encourage us to publish some data collected on three groups of wild 
living marmosets at Joao Pessoa, Northeast Brazil. Most observations were made during a 
visit of 3 weeks ın August 1979 by W. M.; alltogether, about 100 hours were spent on field 
observation, of which about 70 hours were concentrated on group III. C. A. and A.L. 
continued these observations irregularly during parts of 1979 and 1980. It ıs planned to 
extend these studies which can be performed under quite favourable conditions. 

According to HERSHKOVITZ (1977), the marmosets at Joao Pessoa belong to the 
subspecies of Callıthrix 7. jacchus L. Formerly, the marmosets of the eastcoast-region of 
Brazil were subdivided into a number of good species, but since all of them seem to show 
an allopatric distribution it is more convincing to consider them as local populations of one 
superspecies C©. jacchus. MÜLLER (1973) and Kınzey (ms.) derive the distribution pattern of 
these and other neotropical primates from a model of Pleistocene forest refugia. 

Within recent years, the number of field studies on South American primates, including 
some Callitrichidae, has increased considerably (s. KLEIMAn 1977; CoIMBRA-FILHO and 
MITTERMEIER 1981, in press). Comparative data from these studies will be referred to 
subsequently at appropriate places. 


Locality and habitat 


The study area consisted to two patches of forest situated within the campus of the Federal University 
of Paraiba at Joao Pessoa, Brazil (Fig. 1). Area A measures about 5 ha and area B about 7 ha. These 
forests may be classified as a secondary type of the coastal rain forest of East Brazil (HuEcK 1966). 
They have suffered selective lumbering, but a number of older emergent trees are left. A fairly dense 
and differentiated canopy extends between 5 and 15 m height, roughly corresponding to the layer C of 
RıcHARDS (1976). The shrub layer is not very dense. The marmosets appear to spend most of their 
time within that middle layer and only rarely were seen climbing into the crowns of the emergent trees 
or descending to the ground. On some occasions they were seen directly at the edges of the forests. 
According to Dawson (1977) “callitrichids inhabit ‘edge’ vegetation” (p. 32), and Ryranns (1981) 
states for Callithrix humeralifer intermedius that “the majority of the groups activity time is spent.... 
in old and young second growth forest (as well as in) tall forest with a broken canopy and a dense 
middle layer and abundant lianas” (p. 48). 

The forested areas are closely situated by university buildings and are separated by a much 
frequented road. The patches themselves are crossed by a few trails which are occasionally used by 
peoples. It is planned to establish these small forests as reserves for study purposes. 


Territoriality and group structure 


It is well known from both captive and wild living marmosets that they are socially 
organized in small family groups with an adult pair as the social nucleus (Eppre 1975; 
RoTHE 1975; HERSHKoVITZz 1977). However, in Callıthrix humeralifer intermedius 
(RyLanns 1981) and in Saguinus oedipus geoffroyi (Dawson 1977) two and more pairs of 


336 W. Maier, Carmen Alonso and A. Langguth 


adult anımals appear to be typical. In our study area, we were able to distinguish three 
group units occupying three different home ranges. It was not possible to clarıfy all 
problems of territoriality within such a short study period, but some results did emerge 
from our observations: 

The whole forest area north of the road (Fig. 1) was occupied by a large unit (group I) 
consisting of 7 anımals. There were a pair of adult animals, two infants of about 3—4 
months, an older young of perhaps 8-9 months, and two young adults. These two young 
adults seemed to be only loosely attached to the core group; often they stayed behind 
during traveling and sat apart during feeding, but all 7 anımals were always seen sleeping 
together at night. Most often, this group was observed in the northern parts of the forest, 
but on two occasions it was seen in the trees next to the cages of the biotherium, where 9 
animals of Callıthrix jacchus were kept. These encounters, which lasted about one hour 
were accompanied by exited vocal and visual displays by both groups. The wardens of the 
cages confirmed that this kind of confrontation takes place once or twice a week quite 
regularly. Group I was disturbed by man quite often, and it was so shy that it proved 
unsuitable for regular observation. Although the area of forest A did not seem to be 
frequented quite evenly, the home range may be given as 4 to 5 ha. A number of “copiuba 


wer 
u 
. 


----- forest border 


"=... oo Tapırira guianansıs 
oeron routes of group III 
® nest 


© sleepıng tree 
® biotherıum 


Fig. 1. Map of the study area at the university campus of Joao Pessoa (Paraiba, Northeast Brazil). The 
forests are marked by broken lines. For explanation of the other symbols see the legend. The roman 
figures indicate the home ranges of the three groups of Callıthrix j. jacchns 


Field observations on Callıthrix jacchus jacchns L. 337 


trees” (Irapirira guianensis) with biting marks were noted in that area (small circles), but 
one tree (bigger circle) obviously played a major role for this group. 

The forest south of the main road which in its southern and eastern parts lies at a steep 
slope, seemed to be more diversified ecologically. It was inhabited by two distinct social 
units. The eastern part of the forest most probably housed a group of 5 or 6 animals 
(group II), two of which were infants of about 1 month. This group was extremely shy and 
was only seen on a few occasions. 

The western part of this forest area was occupied by a small group of 3 animals, which 
became used to the observers quite quickly even though an adult female had been shot 
there two months ago. This group III consisted of an adult pair and a juvenile animal of 
about 6-7 months of age. Most of our observations were concentrated on this small unit, 
mainly for reasons of convenience. The sex of the adult individuals could not be established 
for sure before the end of the study; therefore, we are presently unable to tell much about 
the roles of the different sexes. 

It became clear after a short time that one tree of Tapirira guianensis formed the center 
of the daily activities of group III. This “home tree” (Castro and Sormnı 1977) apparently 
was the core area (territory) of the home range as well. We observed this group within an 
area of about 140 X 120 m (Fig. 1) and a home range size of about 2 ha should be a fairly 
accurate estimate. On two separate occasions it was observed that the central “home tree” 
was aggressively defended against intruders from group II. During a longer phase of 
absence of the residental group one or two members of group II entered the tree to feed on 
the sap. The strangers were easily recognized by their pronounced uneasiness, and at one 
time they were chased away without difficulties by a returning anımal of group III. On 
another occasıon, a vocal and visual confrontation (reminding the the encounters at the 
biotherium) lasting some minutes only occurred between both groups about 30-40 m east 
of the central tree before group II retreated. 

The road separating the two forests seems to act as a natural barrier between the highly 
arboreal marmosets, but on several occasions anımals were observed crossing this road in 
both directions. The territorial dynamics could only become clarıfied by long-term 
-observations. At irregular intervals, the groups were observed since 1979, but a number of 
significant changes are not understood properly. The table provides some comparative data 
on group size and home range of other Callitrichidae. 


Table 


Compiled data on group size and home range size in some callitrichid taxa 


It becomes evident that the sapivorous taxa Callıthrix jacchus and Cebuella pygmaea are able to subsist 
on only small areas due to their special feeding habits 


Species group size home range authors 


Callıthrix j. jacchus 3-7 2-5 ha present study 
Callithrix j. jacchus 5-13 casıleha STEvENSoN 1978 
Callıthrix j. penicıllata 4-5 Fonsecä et al. 1981 

C. humeralifer intermedius 4-13 ca4 13, ha Ryranns 1981 
Cebuella pygmaea 5-10 0.8-1.3 ha CaAsTRo and Soını 1977 
Cebuella pygmaea 7 0.3 ha RAMIREZ et al. 1977 


Sagninus midas 2- 6 ; THORINGTON 1968 
Saguinus mystax 2- 6 Castro and Soını 1977 
Sagninus fuscicollis 2-10 CaAsTRo and Soını 1977 
Sagninus nigricollis 4-8 IzawA 1978 
S. oedipus geoffroyi 1-14 Dawson 1977 
S. oedipus oedipus 3-12 NEYMAn 1977 
Callimico goeldii 3- 8 HELTNE et al. 1981 

8- 9 Pook and Pook 1981 


1x9] 9yı UT Passnasıp 9ae erep ayı pur “Jındıy ayı Jo puada] ayı ur pourepdxa 
91e s[oquuÄs au], ‘(sısusuvınd vuudv]) 391 3urpaaz} ureur 9yı jo 9uer1odun ayy jo uorssaadun ue Suraıd ‘skep ınoF asayy jo erep ayı 
jo uns je103 ayı saussaıd uumjo9 saddn ay,L, ‘(soep yıım suunn]o9 19M0]J) skep anoy 1e JJJ dno13 Jo uramed KAırause ayı jo weisser] 'z 'd17 


W. Maier, Carmen Alonso and A. Langguth 


338 


ron 


DDR) 


ooHl 


DS DIDI DIDIDSDEHEN | 


7 


3344 buisayua |} 


3334 buada] ) 


syaqwaw dno,b ub13u0, 


burysas ajıuaanf 
6uıpa34 -dos ajıuaanf 
bur4sas 4npo 
buıpaa,j-dos yInpD 


BsoRBBED 


oo0l 


006 


[I] | 
DRKDDDADL DR 


DODPDTLTTI DODDDDI 


Lean 


ron 
DAD°D<D 


BC 


BLz 


862 


Ble 


Field observations on Callithrix jacchus jacchus L. 339 


Daily activity pattern 


Since usually only one observer was working at a time, it proved impossible to follow the 
study group throughout the day. Allthough a considerable part of the daily activity was 
spent feeding on one tree, the anımals dispersed during the rest of the day throughout the 
home range. Therefore it was decided that at the beginning mainly the activity of the 
animals of group III at the most important feeding tree (Tapirira guianensis) should be 
recorded. Whenever possible, the activities at this tree were protocolled every five minutes 
during four whole days (Fig. 2). When the whole group left the tree together, it was 
followed in the forest as long as possible. 

The anımals of group III normally stayed over night in a distinct sleeping place which 
seemed to consist of tangle of vines and dry leaves (Fig. 3). Since this tangle has not been 
removed, it ıs not clear 
whether it should be con- 
sidered as a true nest. The 
large groupI was many 
times observed to rest over 
night in a fork of big 
branches of a Byrsonima 
sericeca some 12-15 m 
above the ground. Very 
diverse sleeping habits are 
reported for Sagnınus o. 
oedipus (NEYMANn 1977) 
and for S. o. geoffroyi 
(Dawson 1979); for the 
latter, sleeping ın con- 
structed nests is con- 
firmed. 

Whereas sunrise was 
around=05205t020525:h8 
the anımals did appear 
from their “nest’”’ between 
05.35 and 06.00 h. With- 
out any delay the anımals 


Fıg. 3. Nest-like tangle, where 
group III stayed over night 
during the study period. This 
“nest” was only about 5m 
above the ground; it was ente- 
red and left from above. 
Members of group I used to 
sleep in a fork of branches of 
an emergent tree however 


ran directly the 40-50 m toward the “copiuba -tree” and started sap-feeding with great 
intensity. This first feeding period on an average lasted about 40 minutes. Then, the animals 
used to climb to the top branches of the feeding tree which are about 15 m high. Here they 
started both auto- and allogrooming, and they exposed themselves to the sun for about half 
an hour. Subsequently, sap-feeding was resumed for another half an hour or so, before one 
or all animals moved away for a longer excursion within the forest. Several times they were 


340 W. Maier, Carmen Alonso and A. Langguth 


observed visiting a number of different fruit trees (not yet determined), but only taking 
relatively few fruits from each. After this fruit eating phase which lasted from a few 
minutes up to about half an hour, they rapidly dispersed whithin the canopy and were 
difficult to follow. Most probably they started hunting for animal food separately from one 
another. Saguinus nigricollis shows very different activities in the morning, for example 
(IzawA 1978). 

It was usually at about 10.00 h when one or more of the anımals returned to the “home 
tree” to feed on the sap and to rest occasionally. There seems to exist a small peak of 
feeding activity on the “copiuba-tree”” between 10.00 and 12.00 h. The following two 
hours show a reduced activity at the feeding tree, but it could not be ascertained whether 
they rested somewhere during this time. This sap-eating activity gradually increased in the 
course of the afternoon. Quite regularly, between 16.00 and 17.00 h two or three animals 
were actively feeding at the “home tree’”’ or went hunting in the vicinity. One day (27. 8.), 
heavy rainfall stopped all feeding activities after 15.00 h. Shortly after 17.00 h the animals 
left the feeding tree and moved slowly to their “nest”’ nearby. Thus their daily activity 
stopped shortly before sunset. Once they were disturbed by the observers and slipped out 
of the “nest”’ well after sunset to find another sleeping place. 

Of course, four full days of observation cannot supply a representative picture of the 
typical activity pattern which appears to vary considerably from day to day and may even 
vary more during the course of the year. But at least a preliminary picture becomes visible: 
During this part of the year the length of the daily activity of Callıthrix j. jacchus amounted 
to about 11.30 hours, whereas Ryranns (1981) records 10 hours only for C. humeralifer 
intermedius. The most remarkable feature appears to be the great importance of sap- 
feeding on one tree of Tapirira guianensis. If we calculate from the upper column, where all 
feeding activities of four days are summed up, the relation of the sap-feeding with the total 
amount of daily activity time, we receive about 29 %. This figure certainly underrates the 
real importance of sap-eating, since only positive sight records were noted, and since some 
less important “copiuba-trees” existed within the home range. Our data compare well with 
those of RAMIREZ et al. (1977), who noticed that troops of Cebuella pygmaea ““spent 32 
percent of observed daily activity eating exudate and excavating exudate holes” (p. 95). 
However, we were unable to make as clear a distinction between sap-feeding and hole 
digging as these authors did. Assuming that the Pygmy Marmosets spend most of the late 
afternoon for only excavating holes includes a connotation of purposeful action which 
would seem to be problematical to us. 

It will be gathered from Fig. 2 that during considerable parts of the day only one or two 
animals were present at the central “‘“home-tree”’, wheras the others were striving through 
the forest. In a number of times it was noted that the animal staying in the tree moved away 
as soon as another came in; normally the animal slipped away just at the opposite side from 
where the newcomer entered the tree. We suspect that some behavioural mechanısm for 
protecting the core area of the territory may be responsible for this phenomenon; as 
mentioned above, two times members of the neighbouring group II had entered the “home 
tree” during a longer absence of the residential group III. Since much work is invested to 
open up and to maintain the sap-providing grooves, it would seem to be a selective 
advantage for a group to protect such an important source of food. In order to minimize 
the locomotory efforts, it would also seem advantageous to concentrate sap-feeding on one 
or few trees for a certain period of time. 


Feeding biology 


As shown in the last chapter, a highly differentiated system of sap-feeding is a very 
conspicuous feature of the feeding biology of Callithrix jacchus. By far the most important 


Field observations on Callithrix jacchus jacchus L. 341 


tree species with respect to this habit is Tapirira guianensis, of which about 40 specimens 
were located in the study area. Occasional feeding on Thrysodium schomburgkianum 
(caboatao de leite) was recorded as well. Meanwhile, a considerable number of South 
American tree species are reported as providing exudates for most observed taxa of 
callitrichids to various degrees (CoıMBRA-FILHo 1972; KınzEy et al. 1975; Castro and 
Soını 1977; RAMIREZ et al. 1977; CoIMBRA-FILHO and MITTERMEIER 1977; Izawa 1981). 
Subsequent observations made on the study groups at Joao Pessoa seem to indicate that 
feeding on the sap of “copiuba-trees” may be less intensiv during other times of the year. 

All of the specimens of Tapırira guianensis within the two forests showed at least some 
scars from biting by Callithrix jacchus. However, in most trees the scars were dry or 
healed. Some of the trees were partly destroyed by the extensive gnawing of the bark layer. 
Both in group I and III, we were able to identify one tree that was most intensively worked 


Fıg. 4A (left). Trunk and lower branches of the central ‘copiuba-tree’ (Tapirira gnianensis) of group 

III. Note the numerous transversal grooves which were gouged into the bark by the marmosets for sap 

feeding. — Fig. 4B. (right). This picture shows a close-up view of some grooves with exudate oozing 

out in some places. The scoop-like lower dentition of Callithrix jacchus exactly fits into these grooves, 
whereas the upper jaws with their divergent canines are too broad 


upon at the time of observation (large circles ın Fig. 1). The stem and the larger branches of 
the main “copiuba” of group III are covered with grooves of about 10-15 mm width and 
9-12 mm depth. They are always orientated perpendicular to the long axis of the stem and 
the branches (Fig. 4). An area of 0.25 m? about 1 m above the ground was covered by 33 
grooves varying in length from 2-20 cm. 18 of these grooves showed signs of fresh biting 
after the first feeding period in the morning. The animals tend to feed at the fresh end of the 
scars, where the exudates of the phloem flow more rapidly, and thus the grooves are 
gradually elongated. 

Although the feeding activity could be observed from as near as about 5 m, the exact 
technique of gouging was not clearly recognized. But experiments with a cleaned skull 


342 W. Maier, Carmen Alonso and A. Langguth 


indicated that the lower jaw of Callıthrix jacchus fıts exactly into the grooves, whereas the 
upper jaw is too broad, mainly due to the prominent and divergent canines. Thus it appears 
most likely that the upper canines are fıxing the skull, and that the scoop-like front teeth of 
the mandible are scraping off the bark layers and the exudates —- much as described by 
COIMBRA-FILHO and MITTERMEIER (1977). It ıs likely that the tongue is helping to ingest 
the soft exudates as well. 

In the early morning, at three intervals of half an hour the positions of the body during 
feeding was recorded whenever visible. In group III, from at total of 117 observations, the 
body was directed upward in 57 %, transversal in 28 %, and downward in 15 %. We have 
also good evidence that during sap-feeding a spacing mechanism keeps apart the group 
members; for this we divided the “home tree” arbitrarily into 6 sectors and checked the 
positions of the anımals every 5 minutes. In only 5 out of 32 records two anımals were 
observed feeding within one sector. Quite often aggressive chasing was occuring, and we 
have preliminary evidence that the adult female was dominating the male, and this again 
was superior to the juvenile. The most favourite and productive feeding place seemed to be 
the lower trunc region; this observation is in accordance with the results of RAMIREZ et al. 
(1977). In most cases the anımals worked upward slowly and were feeding or jumping 
downward more rapidly. 

The exudate of Tapırıra gnianensis at first ıs colourless and transparent, but becomes 
opaque, brownish and jelly-like after some hours. To man it is tasteless and odourless, but 
evidently it is highly attractive to the marmosets. Scraped-off exudate caused great 
excitement when offered to the captive anımals which always were very keen on getting 
this substance. Branches of Tapırıra placed in the cage were gnawed immediately by the 
marmosets. MACHADO and LEITE (1957) analyzed the exudate of a related tree species, 
Anacardium occidentale (Anacardiaceae) and found a high content of polymerous sugars 
(arabın, cerasin etc.) as well as mineral salts. CoiMBRA-FILHO and MITTERMEIER (1977) 
thus conclude that this sap and resin “may serve as a source of energy and of trace 
elements. In addition, gums and other exudates are probably a more dependable food 
source than seasonal fruits’”’ (p. 109). In any case, tree-gouging and sap-feeding appears to 
be an important and deeply-rooted biological adaptation of the marmosets. However, the 
sapivorous habits vary within the genus Callıthrix, and even more so within the other 
Callitrichidae. With reference to Callthrix humeralifer intermedins RyLanns (1981) for 
instance states that “a large part of each day is spent foraging for insects’ (p. 51) and that 
only “infrequently C. h. intermedius was observed eating tree exudates” (p. 52). 

Callıthrix jacchus ıs characterized dentally by the peculiar lower front teeth that form a 
kind of scraper. This feature has been noted for a long time but was mainly used for 
diagnostic purposes only. Since the sap-feeding habits of these primates have become better 
known, these dental structures are understood as to their biological role. The lower front 
teeth look slightly different in the Callıthrix argentata-group and even more so in the 
tamarins (cf. HERSHKOVITZ 1977), whereas Cebuella pygmaea presents typical scoop-like 
front teeth. 

The biological importance of sapivory has been documented in recent years for a 
number of prosimian and sımian primates (CHARLES-DOMINIQUE 1977; SELIGSOHN 1977; 
COIMBRA-FILHO and MITTERMEIER 1977; MAIER 1980). Gouging the bark of suitable trees 
certainly is an energy-consuming activity, but it provides the animals with a stable food 
source and renders them more independent of the fruiting seasons. Apart from the adaptive 
modifications of the front teeth, it proved necessary to develop claw-like nails to find hold 
at the thick and often smooth trunks and branches of the food trees (CArTMILL 1974). The 
claw-like nails of the Callitrichidae most probably are to be considered as secondary 
adaptations to this feeding habit (BLuntscHLı 1929), but some authors tend to interpret 
them as primitive retentions (CLARK 1962; HERSHKOVITZ 1977). However, it is certainly 
premature to speculate whether Callitrichidae in general are primitive simian morphotypes. 


Field observations on Callıthrix jacchus jacchus L. 343 


As far as the functional morphology of the dentition of Callithrix (Fig. 5) is concerned, it 
would be more plausible to derive it from anything like Saimiri rather than reverse (MAIER 
1977). 

Besides a number of anatomical adaptations to sap-feeding, there exist behavioural 
adaptations to this mode of feeding as well. First of all, the territorial behaviour of the 
family groups appears to be organızed around one or a few sap-producing trees. This 
would seem favourable for the anımals, because preparation of a feeding tree and 


facets in 
whereas 


1977). It is concluded that the molar morphology is simplified secondarily, 


Callithrix jacchns. The different hatchings indicate the opposing shearing 
the front dentition is specialized for sapıvory 


Callıthrix jacchus 


Fig. 5. Functional morphology of the dentition of 
upper and lower teeth (For more details see MAIER, 


344 W. Maier, Carmen Alonso and A. Langguth 


maintaining this “plantation’” means a great energy investment which justifies aggressive 
defense. As documented by a number of heavily damaged trees, this treatment of the food 
source must be limited, and the centers of activity will have to be shifted. We have evidence 
that some shifting did occur since 1979, but the consequences for the structure of the home 
ranges have yet to be established. 

Apparently, there exists some competition for the exudates elicited by the marmosets. 
Quite regularly ants are seen feeding on the sap. One day at 15.30 h a marsupial, 
Caluromys philander, was observed entering the “home tree” while all three members of 
group III were sap-feeding. The Woolly Opossum was seen investigating the grooves and 
licking some of the exudates. After some minutes the juvenile marmoset did attack the 
marsupial, which did not appear to be intimidated at all but threatened back. Only after the 
following attacks of an adult marmoset it slowly retreated. Alltogether, this interspecific 
confrontation lasted about 5 or 6 minutes. Two hours later (17.40 h) when it was nearly 
dark, probably the same marsupial was seen again running and jumping in the trees near 
the “home tree”, finally disappearing in direction of the sleeping nest of the marmosets 
which had settled there half an hour before. 

Apart from sap- and fruit-eating, the omnivorous marmosets showed interesting 
hunting activities. On several occasions it was observed that one animal detached from the 
group at the “copiuba-tree”’ and went for a hunting excursion. Then, the anımals appeared 
to display a special hunting behaviour, looking very concentrated in their search for anımal 
prey. They sıt crouched and check the near surrounding carefully; then they either rush 
ahead or take a new watching position. Sometimes they enter thick tangles and clımb 
around quite noisily, probably to hunt up and chase small anımals. They may even descend 
to the ground on such hunting expeditions. Once, an anımal was seen catching a locust at 
the forest floor and then eating it on a bush of 1.5 m height. On another occasion, an 
anımal was observed catching a big, some 20 cm long Phasmid-locust sitting on a trunk; 
the insect fırst fell down to the ground some 6 or 7 m, but was quickly followed and caught 
again. Subsequently it was ingested and chewed for about 15 minutes on a higher branch. 
Several times the juvenile was observed accompanyıng an adult anımal at hunting; 
however, it kept away some 5 to 10 meters, only intensively watching the hunter. 

There existed an opportunity to examine the stomach content of an adult female 
collected some months before this study in the area of group III. Most of the contents 
consisted of bulks of dirty-whitish jelly, most likely corresponding with the exudates of 
the “copiuba-trees”; of other plant material we noted unidentifyable remains of varıous 
fruits. In addition, the stomach contained th remains of a small lizard, a big buprestid 
beetle, parts of spiders and grasshoppers. Some feeding experiments have been made with 
the captive marmosets at the “biotherium”, by offering them insect catches. Most insects 
and spiders are quickly seized with one or two hands and then ingested by mastication 
(Kay and HITEmAE 1974). Bigger grasshoppess are pressed to the substrate with both hands 
before being bitten and chewed from the head. A fairly big lizard was easily overwhelmed 
by an adult female which seemed to be dominant within the caged group, and which was 
usually the first anımal to obtain preferred food items. The lizard was ingested from the 
head and no remains were left for the other anımals which were intensively trying to catch 
some parts. Some small beetles, some ants and some black juveniles of an unidentified 
species of locust were not accepted by the animals. Other species of ants were easily taken, 
and in summer, when cicadas become abundant in the late afternoon, anımals of the 
biotherium used to retire for rest later in order to catch them at the grid. 

It may thus be summarized that Callithrix jacchus is organızed in family groups which 
occupy a home range of a few hectars, but only defend actively acore area. At least during 
certain parts of the year, a “home tree” providing considerable amounts of nutritive 
exudates seems to be an important part of the central territory. Much time and effort is 
spent by each group to feed on these exudates and to keep the wounds of these trees 


Field observations on Callithrix jacchus jacchus L. 345 


opened. Numerous structural and behavioural characteristics of the marmosets have to be 
interpreted as adaptations to this special feeding habit. It is likely that the relatively small 
home ranges of Callıthrix jacchus and Cebuella pygmaea are a result of the availability of 
exudates as a stable food source. In addition, fruits, flowers, and small anımals form an 
important portion of the wide dietary spectrum. It is to be expected that a better 
knowledge of the feeding habits of the Common Marmoset under natural conditions will 
further improve their maintenance ın captivity as well. 


Zusammenfassung 


Feldbeobachtungen an Callıthrix jacchus jacchus L. 


Feldbeobachtungen an drei Gruppen von Callıthrix jacchus jacchus L. werden mitgeteilt. Die Tiere 
lebten in kleinen Stücken von sekundärem Küstenregenwald in unmittelbarer Nähe des Universitäts- 
Campus von Joao Pessoa (Paraiba, Brasilien). Die Familiengruppen umfaßten 3-7 Tiere, die teilweise 
sich überlappende Reviere von etwa 2-5 ha Größe einnahmen. Für die hauptsächlich studierte Gruppe 
wurde ein aktıv verteidigtes Territorium festgestellt, in dessen Zentrum der wichtigste Nahrungsbaum 
lag. Diese Gruppe wurde an vier ganzen Tagen beobachtet; durchschnittlich wurden dabei von den 
11,5 Stunden Tagesaktivität etwa 30 % mit der Aufnahme von Exsudaten der Baumspecies Tapirira 
gmianensis verbracht. Anpassungen an diese sapivore Ernährungsweise und weitere biologische Daten 
werden diskutiert. 


References 


Bruntschui, H. (1929): Ein eigenartiges an Prosimierbefunde erinnerndes Nagelverhalten am Fuß 
von platyrrhinen Affen. Roux’ Archiv f. Entwicklungsmechanik 118, 1-10. 

CARTMILL, M. (1974): Pads and claws in arboreal locomotion. In: Primate locomotion. Ed. by F. A. 
JEenkıns. New York: Academic Press. pp. 45-83. 

CASTRO, R.; SoıInı, P. (1977): Field studies on Saguinus mystax and other callıtrichids in Amazonian 
Peru. In: The biology and conservation of the Callitrichidae. Ed. by D. G. KreEıman. Washington: 
Smithsonian Institution. 

CHARLES-DOMINIQUE, P. (1977): Ecology and behaviour of nocturnal primates. Prosimians of 
Equatorial West Africa. New York: Columbia University Press. 

CLARK, W. E. LE Gros (1962): The antecedents of man. Edinburgh: University Press. 

COIMBRA-FILHO, A. F. (1972): Aspectos ineditos do comportamento de sagüis do genero Callıthrix 
(Callithricidae Primates). Rev. Brasil. Biol. 32, 505-512. 

COIMBRA-FILHO, A. F.; MITTERMEIER, R. A. (1977): Tree-gouging, exudate-eating and the “short- 
tusked”’ condition in Callhithrix and Cebuella. In: The biology and conservation of the Callıtrichi- 
dae. Ed. by D. G. KıEıman. Washington: Smithsonian Institution. pp. 105-115. 

— (eds.) (1981): Ecology and behavior of neotropical primates. Rio de Janeiro: Academia Brasileira 
de Ciencias. 

Dawson, G. A. (1977): Composition and stability of social groups of the tamarin, Saguinus oedipus 
geoffroyi, in Panama: ecological and behavioral implications. In: The biology and conservation of 
the Callitrichidae. Ed. by D. G. KıEıman. Washington: Smithsonian Institution. pp. 23-37. 

— (1979): The use of time and space in the Panamanian Tamarin, Saguinus oedipus. Folia primatol. 
31, 253-284. 

EpPpL£, G. (1967): Vergleichende Untersuchungen über Sexual- und Sozialverhalten der Krallenaffen 
(Hapalıdae). Folia primatol. 7, 37-65. 

— (1968): Comparative studies on vocalization in marmoset monkeys (Hapalidae). Folia primatol. 8, 
1-40. 

— (1975): The behavior of marmoset monkeys (Callithricidae). In: Primate behavior: developments 
in field and laboratory research. Ed. by L. A. RosengLum. Vol. 4. New York: Academic Press. pp. 
195-239. 

Fonseca, G. A. B. (1981): Utilizacao de exudatos de especies arboreas do cerradao pelo mico-estrela 
(Callithrix jacchus penicillata). Res. Conf. VIII Congr. Brasil. Zool. 1981, 171-172. 

FonseEca, G. A. B.; LACHER, T.; MAGALHAES-CASTRO, B.; Auvzs, C. (1980): Observacoes prelimina- 
res em Callithrix jacchus penicillata. Res. Conf. VII Congr. Brasil. Zool. 1980, 109-110. 

HELTNE, P. G., Wojcık, J. F.; Poox, A. G. (1981): Goeldi’s monkey, genus Callimico. In: Ecology 
and behavior of neotropical primates. Ed. by A. F. CoımBrA-FıLHo and R. A. MITTERMEIER. Rio 
de Janeiro: Academia Brasileira de Ciencias. pp. 169-209. 

HERSHKOVITZ, P. (1977): Living New World monkeys. Chicago: University Press. 

Huck, K. (1966): Die Wälder Südamerikas. Ökologie, Zusammensetzung und wirtschaftliche 
Bedeutung. Stuttgart: G. Fischer Verlag. 


346 W. Maier, Carmen Alonso and A. Langguth 


Izawa, K. (1975): Foods and feeding behavior of monkeys in the upper Amazon Basin. Primates 16, 
295-316. 

— (1978): A field study of the ecology and behavior of the Black-mantle Tamarın (Saguinus 
nigricollis). Primates 19, 241-274. 

Kay, R. F.; Hriemaz, K. M. (1974): Jaw movement and tooth use in recent and fossil primates. Am. ]. 
Phys. Anthrop. 40, 227-256. 

Kınzey, W. G. (1979): Distribution of some neotropical primates and the model of Pleistocene forest 
refugia (in press). 

Kınzev, W. G.; ROSENBERGER, A. L.; RAMIREZ, M. (1975): Vertical clinging and leaping in a 
neotropical anthropoid. Nature 255, 327-328. 

Kreıman, D. G. (ed.) (1977): The biology and conservation of the Callitrichidae. Washington: 
Smithsonian Institution. 

MACHADO, A.; LEITE, ©. C. (1957): Goma de cajueiro. Bolet. Inst. Quimica Agrıcola 50, 7-15. 

MAIER, W. (1977): Die bilophodonten Molaren der Indriidae (Primates) — ein evolutionsmorphologi- 
scher Modellfall. Z. Morph. Anthrop. 68, 307-344. 

— (1980): Konstruktionsmorphologische Studien am Gebiß der rezenten Prosimiae. Abh. senckenb. 
naturf. Ges. 538, 1-158. 

MÜLLER, P. (1973): The dispersal centres of terrestrial vertebrates in the neotropical realm. The 
Hague: W. Junk Publ. 

NEyMman, P. F. (1977): Aspects of the ecology and social organization of free-ranging Cotton-top 
Tamarins (Saguinus oedıpus) and the conservation status of the species. In: The biology and 
conservation of the Callıitrichidae. Ed. by. D. G. KLEıman. Washington: Smithsonian Institution. 

. 39-71. 

Bose A. G.; Poox, G. (1981): A field study of the socio-ecology of the Goeldi’s monkey (Callimico 
goeldi:) ın Northern Bolivia. Folia primatol. 35, 288-312. 

RAMIREZ, M. F.; FREEsE C. H.; Revırra, C. (1977): Feeding ecology of the Pygmy Marmoset, 
Cebuella pygmaea, ın Northeastern Peru. In: The biology and conservation of the Callitrichidae. 
Ed. by D. G. Kreıman. Washington: Smithsonian Institution. pp. 91-104. 

RıcHarDs, P. W. (1976): The tropical rain forest. An ecological study. Cambridge: University Press. 

ROTHE, H. (1975): Some aspects of sexuality and reproduction in groups of captive marmosets 
(Callıthrix jacchus). Z. Tierpsychol. 37, 255-273. 

ROTHE, H.; WOLTERs, H. ]J.; HEARn, ]J. P. (eds.) (1978): Biology and behaviour of marmosets. 
Göttingen: Eigenverlag Rothe. 

Ryranos, A. B. (1981): Preliminary field observations on the marmoset, Callıthrıx humeralifer 
intermedius (HERSHKOVITZ, 1977) at Dardanelos, Rio Arıpuana, Mato Grosso. Primates 22, 46-59. 

SELIGSOHN, D. (1977): Analysıs of species-specific molar adaptations in strepsirhine primates. 
Contrib. Primatol. 11, 1-116. 

STEVENSON, M. F. (1978): The behaviour and ecology of the Common Marmoset (Calhthrix jacchus 
jacchns) in its natural environment. In: Biology and behaviour of marmosets. Ed. by H. RoTHE, 
H. J. WOLTERs and J. P. Hearn. Göttingen: Eigenverlag Rothe. p. 298. 

STEVENSON, M. F.; PooLE, T. B. (1976): An ethogram of the Common Marmoset (Callıthrix jacchus 
jacchus): General behavioural repertoire. Anımal Behav. 24, 428-451. 

THORINGTON, R. W. (1968): Observations on the tamarın Saguınus midas. Folia pimatol. 9, 95-98. 


Author’s address: Prof. Dr. WOLFGANG MAIER, Dr. Senckenbergische Anatomie, Theodor-Stern- 
Kaı 7, D-6000 Frankfurt/Main 70, FRG 


Population dynamics of small mammals in the 
Drakensberg of Natal, South Africa 


By D. T. RowE-Rowe and J. MEESTER 


Natal Parks Board, Pietermaritzburg and University of Natal, Durban, South Africa 


Receipt of Ms. 18. 6. 1982 
Abstract 


Studied two shrews and sıx rodents which occur in Giant’s Castle Game Reserve, a mountainous, 
predominantly grassland, high rainfall area in South Africa. Sex ratios of all species are given and the 
population structure cycles of the two most abundantly-trapped animals (Myosorex varıus and 
Rhabdomys pumilio) are detailed, indicating cycles typical of seasonal breeders and lifespans of 12 to 
16 months. Breeding season coincided with the warm wet season (spring and summer). Analyses of the 
large M. varius and R. pumilio samples indicated that breeding started about a month later at higher 
altitudes where temperatures were lower. 


Introduction 


Population dynamics and reproduction of some of the African small mammals dealt with in 
this paper have been studied in tropical grassland (CHEEsEMAN and DeLany 1979; DELANY 
1964; TAyLoR and GREEN 1976). PERRIN (1979; 1980a) studied the rodents Rhabdomys 
pumilio and Otomys irroratus ın a semi-arid area receiving erratic rainfall, and individual 
studies on each of these two species were conducted in a South African highveld region 
(high-lying grassland plains) by Brooks (1974) and Davıs and MEESTER (1981). Prior to 
this study no published accounts existed concerning the population structure and repro- 
duction of small mammals in an African montane region. In this paper, therefore, we 
report on the population dynamics of the small mammals which occur in the Drakensberg, 
the highest mountain range in South Africa. 


Study area and methods 


This study was conducted in Giant’s Castle Game Reserve (GCGR), 29° 08’ to 29° 23’ S and 20° 23’ to 
29° 37' E. The most prominent topographic feature ıs the Drakensberg escarpment which runs along 
the western boundary of the reserve at an average altitude of 3000 m, dropping sharply to 2200 m. 
Below 2200 m the terrain is very broken - numerous rivers flow approximately west to east from the 
escarpment and have incised steep-sided valleys between which high ridges have remained. GCGR is 
vegetated mainly by grassland, with limited patches of forest, scrub, and woodland. 

The rainfall is seasonal, occurring mainly from October to April (Fig. 1). Mean annual rainfall 
measured at 1760 m during the study period (1978 to 1980) was 1092 mm which fell on an average of 
147 raindays. Rainfall increases with altitude to ca. 1700 to 1800 mm on the escarpment. Snow falls 
mainly between April and September, but can precipitate during any month above 2500 m. Summers 
are mild to cool and winters cool to cold (Fig. 1). Temperatures at grass level are considerably lower 
than those recorded in the Stevenson screen, the mean minimum being below freezing point for five 
months of the year. Whereas precipitation increases with altitude in the Drakensberg, temperature 
decreases. 

Small mammal sampling was done in all vegetation types between 1500 and 2700 m, using folding 
aluminium live traps or plywood-based snap traps baited with a mixture of rolled oats and peanut 
butter. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4706-0347 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 347-356 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


348 D. T. Rowe-Rowe and ]J. Meester 


xı 


max 


Results and discussion au 


Two shrews (Myosorex va- 

rıns and Crocidura flaves- X min 
cens) and six small rodents 
(Rhabdomys pumilio, Oto- 
mys irroratus, Dendromus 
melanotis, D. mesomelas, 
Mus minutoides and Gra- 
phiurus murinus) were col- 
lected. All grassland, scrub, 
or woodland habitats were 
dominated by R. pumilio or 
M. varıus, which contrıbu- 


Fıg. 1. Mean daily maximum 
and minimum temperatures, 100 
and mean minimum tempera- 


°C 


[o) 


Temperature 


200 


E 

E 
ture at grass level, recorded at = 
1860 m in the Drakensberg = al 
(KıLLıck 1963), and mean = = 
monthly rainfall recorded at RL TE I ulrSsuleg 
1760 m during the study period Month 


ted 44.1 % and 38.7 % respectively to the total number of small mammals caught. In forest 
G. murinus was dominant. 
Sex ratios 


Sex ratios of trapped small mammals are listed in the table. Deviation from parity was 
significant only in the two shrew species, in which a preponderance of females was 
recorded. 


Table 


Sex ratios of small mammals trapped in GCGR and Chi-square evalution for 
deviation from parity 


Species 


R. pumilio 
M. varıns 
O. ırroratus 
D. melanotis 


G. murinus 
M. minutoides 
C. flavescens 
D. mesomelas 


NS = P> 0.05 


Detailed analyses of data on the two most abundantly-caught species were done to test 
for significant differences ın the sex ratios of adults, sub-adults, snap-trapped and live- 
trapped anımals overall, snap- and live-trapped adults, and snap- and lıve-trapped sub- 
adults. No significant departure from a 1:1 ratio was observed for any of the different 
categories of R. pumilio. The adult male: female ratio of live-trapped M. varıus did not 


Population dynamics of small mammals in the Drakensberg 349 


differ significantly from parity, but in all other categories there was a significant prepon- 
derance of females (P < 0.05). Greatest differences were observed amongst sub-adults 
(Chi-square 14.36; P < 0.001) and in the overall snap-trapped sample (Chi-square 12.91; 
P < 0.001). | 

Brooks (1974) pointed out that trap-based sex ratios do not necessarily reflect the true 
composition of the population, as they may result from differential trap response or 
movement patterns. In his study on R. pumilio BRooks recorded a non-significant 
preponderance of males in live-trapped animals but a highly significant 1.7 : 1 ratio in snap- 
trapped mice, while a significant 0.8 :1 ratio in favour of females was recorded at birth in a 
laboratory colony. 

The 1.2:1 ratio of R. pumilio collected in GCGR is similar to the ratios reported by DE 
Wır (1972), HAnney (1965), PERRIN (1980a), and SMITHERS (1971), all of whom recorded a 
non-significant preponderance of males. The total collection of 435:336 made by RAUTEN- 
BACH (1978) departs significantly from 1:1 (P < 0.01). 

In a study of the activity patterns of the European common shrew (Sorex araneus) 
VLAsAK (1980) recorded sexual differences in the activity of immature animals only, but 
found that males were more active than females. A similar situation may have prevailed 
amongst M. varıus in GCGR, but with young females being more active. In at least two 
other instances (RAUTENBACH 1976, 1978) females exceeded males in M. varins collections, 
but differences were not significant. 

Similar sex ratios to those recorded for two other species collected in GCGR have been 
reported, vız for O. irroratus (Davıs 1973; PERRIN 1980a, RAUTENBACH 1978) in which the 
ratios were ca 1:1, and G. murinus (RAUTENBACH 1978; SMITHERS 1971) in which a 
preponderance of females was recorded. RAUTENBACH (1978) and SMITHERS (1971) 
recorded ratios in favour of males in D. melanotis, but the non-significant departures from 
parity do not differ greatly from the GCGR ratio which favoured females. Sample sizes of 
other species collected are too small for consideration. 


Population structure 


It was possible to examine the population structure of M. varius and R. pumilio only, as 
insufficient data were collected on the other species. 

The monthly catch of each of the two species was divided into mass classes, using 2 g 
intervals for M. varins and 5 g intervals for R. pumilio. The number of animals in each mass 


J F M A M J J A S (0) N 


Months 


Fig. 2. Mass classes of M. varinstrapped during each month. The width of the histogram represents the 
percentage of the total monthly sample which falls into that mass class 


Mass in grams 


350 D. T. Rowe-Rowe and J. Meester 


class was then expressed as a percentage of the total monthly catch and presented 
diagrammatıcally (Figs 2 and 3). 

From November to April there was a wide representation of mass classes in the M. 
varins samples, but from May to September fewer mass classes were present. During May 
and June the catches were made up almost entirely of small anımals and no large adults 
were caught. The pattern strongly suggests that a die-off in adult shrews takes place at the 
end of the breeding season and that during the next season the population builds up from 
young animals. From July to September it is obvious that the small May-June animals 
grow and reach full adult size. The first young anımals appear to enter the trappable 
population in about October. 

BAxTER (1977) quoted a record of M. varius living for as long as two years ın captivity, 
but the GCGR data suggest that under natural conditions most anımals survive for only 12 
to 16 mo, and that there ıs a die-off amongst adults by the end of the breeding season. 

The population dynamics of M. varins have features in common with those previously 
recorded in some other shrews. It has been shown that European and North American 
shrews in the wild have a lifespan which does not exceed 18 mo (VoGEL 1980). BUCKNER 
(1966) found that only 20% of the new generation of Sorex cinerens and S. arcticus reached 
sexual maturity and that almost all surviving anımals die before the age of 15 mo. 
CHURCHFIELD (1980) observed similar dynamics in $. araneus — 20 to 30 % survived to 
breed and maximum life expectancy was 13 mo. Generally shrews appear to survive no 
longer than the end of the breeding season following the one in which they were born. 


J F M A M J J A Ss (0) N 
Months 


Fig. 3. Mass classes of R. pumilio trapped during each month (see Fig. 2 for explanation) 


In the R. pumilio samples there was a wide range of mass classes from January to April 
(Fig. 3). From May to September the catches were made up of what appeared to be some 
adults and young anımals from the previous breeding season. The first young of the season 
entered the trappable population in October, but no small animals were captured during 
November. It was mainly during November each year that wet season sampling of the high 
altitude habitats (2200 to 2700 m) was done, and the absence of young from these samples 
suggests that at the higher altitudes the breeding season commences later than at the lower 
altitudes where trapping was done in October (see “Reproduction”). 

The population structure cycle of R. pumilio in GCGR is similar to what BRooks 
(1974) found near Pretoria and that reported by PErrın (1979) from a study area near 
Grahamstown, in that the greatest range of age classes was recorded in months following 


Population dynamics of small mammals in the Drakensberg 351 


the breeding season, and the smallest range was observed in the non-breeding season, 
followed by an upward shift in population age structure. PERRIN (1979), working in an area 
which experiences erratic rainfall, recorded the greatest range of age classes ın R. pumilio 
during April and May, following a breeding peak ın February. In GCGR the wide size 
range of animals collected from January to April indicates that the breeding peak occurred 
earlier and extended over a longer period, probably as a result of the high regular rainfall 
and the long wet season (see also “Reproduction”’). 

In captivity Brooks (1974) found that R. pumilio had a lifespan which just exceeds two 
years, but he believed this to be considerably less in the wild, where the oldest individual in 
his study population was 16 mo old and only 2.3 % of the animals were older than 12 mo. 
De Wır (1972) also estimated a similar longevity for R. pumilio (17 mo for males and 12 mo 
for females) but was unable to indicate what proportion of the population lived longer than 
12 mo. In the population studied by PERRIN (1979) the age structure suggests that the 
lifespan exceeds 12 mo in only a small percentage of the mice. 

The population structure cycles of M. varius and R. pumilio are similar in that they 
exhibit the patterns typical of seasonal breeders. A difference between the two is that a 
greater range of mass classes was recorded earlier in the season in M. varius than was the 
case in R. pumilio, aphenomenon which is related to the onset of breeding at an earlier date 
in M. varıns (see also “Reproduction’”’). 


Reproduction 


Individual species accounts 

M. varıns 

Breeding season ın the Drakensberg extended from August to March: pregnant females 
were collected during these months and lactating females from September to April, and ın 
males an increase in mean testis length was observed in July with a decrease in April 
(Fig. 4). The percentage of anımals in breeding condition (Fig. 4) was highest from 
September to November (no data for December) and lowest during April to August. 
Although the percentage of males with inguinal testes examined during August was low 
there was an increase in mean testis length even in males with abdominal testes. 

Mean number of foetuses was 2.8 (N = 19, range 1 to 4). 

RAUTENBACH (1978) reported a sımilar breeding season in the Transvaal where pregnant 
females were collected from September to March, and similar litter sizes to those from 
GCGR were recorded by BAxTER and Lroyvp (1980): 3.0 ın captivity and 2.9 in trapped 
anımals. 


C. flavescens 
Pregnant females were collected during October and February. Each of the three pregnant 
females contained five foetuses. The only adult male collected had abdominal testes during 
December. 

In Uganda DeLAanY (1964)-recorded an average of three foetuses (N = 6, range 20 4) ın 
animals trapped just prior to the start of the wet season. BAxTER (1977) observed a breeding 
season from August to April and a litter size of 3.7 in captive anımals. 


R. pumilio 

Pregnant females were collected during all months from September to March, and no 
lactating animals were recorded outside this period. Testes began to increase in length 
during July and decreased during April (Fig. 5). The percentage of anımals in breeding 
condition was lowest from April to June and reached a peak during November (Fig. 5). 
The data suggest that males become reproductively active about a month earlier than do 
females. As the breeding season is determined by the female, however, it may be described 
as extending from August, when mating takes place, to March. 


Testis length (mm) 


Per cent 


352 D. T. Rowe-Rowe and ]J. Meester 


Mean testis length (mm) 


100 


Months 


Per cent 


Months 


Fig. 4 (left). Reproductive condition of M. varins, showing mean testis length (graph), percentage males 
with inguinal testes and percentage perforate females (shaded bars). Unshaded portions represent 
males with abdominal testes and imperforate females. Sample sizes are indicated above the bars. - 
Fig. 5 (right). Reproductive condition of R. pumilio, showing mean testis length (graph), percentage 
males with scrotal testes and percentage perforate females (shaded bars) (see Fig. 4 for explanation) 


Mean number of foetuses was 4.6 (N = 22, range 3 to 7). 

Breeding in R. pumilio appears to coincide with the wet season (BROooKs 1974; HANNEY 
1965; PERRIN 1980a; RAUTENBACH 1978; TAyLoR and GREEN 1976), which in GCGR 
corresponds with the summer months. SMITHERs (1971) is the only author to report some 
winter pregnancies in animals collected in an arid region (Botswana). BRooks (1974), 
PErRIN (1980a), and TAYLOR und GREEN (1976) found too that males became reproduc- _ 
tively active about a month prior to females. The onset of reproductive activity in females 
usually coincides with the first significant rains of the season (cf. GCGR data, BROokS 
1974; PERRIN 1980a; TAYLoR and GREEN 1976). 

Mean litter size has been given as between 4.5 and 5.9 in studies of wild populations 
(Brooks 1974; DE Wır 1972; HannEy 1965; PERRIN 1980a; SMITHERS 1971; TAyLoRr and 


GREEN 1976) and 6.5 to 7.0 in captive anımals (BROoKs 1974; MEESTER and HALLETT 
1970). 


Population dynamics of small mammals in the Drakensberg 353 


O. irroratus 

Although the sample size ıs small the data suggest that breeding extended over a longer 
period than was the case for R. pumilio: a lactating female was collected ın May, pregnant 
females were recorded during October, November, and January; and males with scrotal 
testes were recorded during July, September to November, and January. 

Mean number of foetuses was 1.6 (N = 5, range 1 to 2). 

Both Davıs (1973) and RAUTENBACH (1978) concluded that O. zrroratus bred during the 
warm wet season in the Transvaal, but in the eastern Cape Province PERRIN (1980a) found 
that the species bred throughout the year and that rainfall did not have a significant effect 
on reproduction. PERRIN (1980a) reported a mean litter sıze of 1.48, which is similar to the 
mean for the GCGR |litters, but the mean |itter sizes of 2.3 and 3.2 recorded in the 
Transvaal by Davıs (1973) and RAUTENBACH (1978) respectively were greater. 


D. melanotis 
No pregnant females were recorded and the only lactating anımal was collected during 
January. Perforate females were collected during May and November, and males with 
scrotal testes during April, October, and November. Most immature anımals were 
collected during April and May, which suggests that the young are born during summer. 
Very little has been published on the reproduction of D. melanotis. SMITHERS (1971) 
collected a female containing four foetuses during December in Botswana and recorded 
pregnancies from December to March in Rhodesia. DErany (1964) recorded a pregnant 
female with three foetuses during the wet season (July) in Uganda. 


D. mesomelas 
Males with abdominal testes were collected during September. A lactating female and a 
male with scrotal testes were recorded during November at Cathedral Peak, 40 km north 
of GCGR in the Drakensberg (D. T. RowE-RowE, unpub!.). 

In Malawı Hanney (1965) took two pregnant females at the end of the wet season 
(March and May) and one at the beginning of the dry season (June). Foetuses numbered 
two to sıx. 


G. murınus 

No pregnant or lactating anımals were collected. Perforate females were collected during 
October and November, and from March to May; males with scrotal testes during 
September to November; and immature anımals during March. 

Neither RAUTENBACH (1978) nor SMITHERS (1971) recorded any breeding data. PIENAAR 
et al. (1980) reported one pregnant female taken during February. The times at which 
adults in breeding condition and ımmature anımals were taken in GCGR suggest that this 
species too breeds during summer. 


M. minutoides 
As the sample size is small and as no pregnant or lactating animals were collected, 
interpretation of the data ıs difficult. Perforate females were collected during May and 
June, males with scrotal testes during March, October and November, and immature 
anımals during January, August, and October. 

SMITHERS (1971) suggested that the species breeds throughout the year, and ın Malawi 
HANnNEY (1965) collected pregnant or lactating females during summer, autumn, and 
winter. The data here presented suggest that breeding occurs throughout the year. 


Conclusions 


The overall pattern which emerges from the reproductive data collected in GCGR is that 
most small mammals breed during the wet season. In the Kenya Highlands TayLor and 
GREEN (1976) concluded that food was the primary factor regulating rodent reproduction, 


354 D. T. Rowe-Rowe and J. Meester 


and rainfall the ultimate factor, by controlling food supply. A similar situation occurred in 
GCGR. Good rains fell during early spring each year, herbage growth following the rains 
was rapid, and the nutritional value of the veld (particularly green grass, inflorescences, and 
forbs) was highest during spring and summer; crude protein content, for example, is as 
high as 15 % during spring but drops to 2 or 3 % in winter (SCOTCHER et al. 1980). 

A number of other authors have reported small mammal reproduction coinciding with 
rainfall in regions having a distinct wet season (BRAMBELL and Davis 1941; CHAPMAN et al. 
1959; CHEESEMAN and DELANY 1979; COoETZEE 1975; DELANY 1964; DE WırT 1972). 

PERRIN (1980a) observed a bimodal pattern in the breeding of R. pumilio, with a 
recession in mid-summer, which he positively correlated with the erratic rainfall of the 
study area. Peaks in the number of births were recorded in February or March, following 
months of high rainfall. In areas ın which rainfall is more predictable and even, R. pumilio 
breeds throughout spring and summer, e.g. ın GCGR, study areas near Pretoria (BROOKS 
1974; DE Wır 1972), and in the Kenya Highlands (TAyLor and GREEN 1976). 

M. varius reached a breeding peak (Fig. 4) about a month earlier than did R. pumilio 
(Fig. 5). The differences are probably related to the availability of nutritious food: M. 
varıns lıves entirely on invertebrates (ROWE-ROWE and MEESTER, in press), high in 
nutritive value, which might be available earlier in the season than is suitable plant material, 
which forms the bulk of the diet of R. pumilio in GCGR. 

PERRIN (1980b) reported seasonal changes in the nutrient content of the diet of R. 
pumilio. Breeding in the species followed peaks in nutrient content of the diet. Reproduc- 
tion of O. irroratus ın the eastern Cape Province, however, was not influenced by rainfall 
(PERRIN 1980b). The species produces small litters throughout the year (PERRIN 1979), and 
as it lives on large quantities of poor quality herbage during all seasons, PERRIN (1980a) 
concluded that reproduction in O. irroratus is not directly dependent on rainfall and food 
with high nutritive value, and is therefore K-selected (PrankA 1970). R. pumilio on the 
other hand produces large lıtters seasonally, being dependent on rainfall and nutritious 
food, and may be considered r-selected. 

Brooks (1974) stated that the breeding season of R. pumilio was restricted to the wet 
season, but in considering the factors which controlled reproduction, he rejected rainfall 
and photoperiod, and stated that only temperature and nutritive value of the diet could be 
involved — temperature being the most likely factor. I do not see, however, how BROOKS 
(1974) could have completely eliminated rainfall from his consideration, as new growth ın 
plants and seed production (1.e. increase ın nutritive value of the food) are dependent on 
rainfall in the known range of R. pumilio. 

Data relating to all female M. varınsand R. pumilio collected in GCGR during October 
and November were examined to see whether there were any differences in the pregnancy 
rates of anımals caught at lower altitudes (< 1900 m) and those trapped at high altitudes 
(> 2200 m). In M. varıns (N = 38) 83 % of the low altitude females were pregnant or 
lactating whereas only 35 % of those from above 2200 m were. Of the R. pumilio (N = 35) 
pregnant or lactating females comprised 45 % of the females collected at low altitudes and a 
mere 8% at high altitudes. These percentages not only illustrate the phenomenon that the 
breeding season starts later at higher altitudes, but further emphasises the conclusion that 
onset of breeding occurs earlier in the season in M. varins than in R. pumilio. 

Rainfall at the higher altitudes is greater than at the lower altitudes in the Drakensberg 
(Kırrıck 1963), but temperature decreases with altitude (KırLıck 1963). The later onset of 
breeding amongst the two small mammal species at higher altitudes in GCGR is therefore 
probably due to the effect of lower temperatures. 

In considering the findings of PERRIN (1980a, b), BRooks (1974) and this study, it 
appears that rainfall, food, and temperature all play a part in determining the start of the 
breeding season. Rainfall and food are closely linked, and provided these two factors are 
present, temperature can further influence the timing of reproduction. 


Population dynamics of small mammals in the Drakensberg 953 


Acknowledgements 


We thank Par Lowry who assisted with some of the field work, Mrs. T. AnDERson who typed the 
manuscript, CATHY CALDER for assistance in preparing art work, and Prof. G. L. Macıean for 
translating the abstract. 


Zusammenfassung 


Populationsdynamik kleiner Säugetiere in den Drakensbergen von Natal, Südafrika 


Giant’s Castle ist ein in überwiegend grasbewachsenem Bergland mit hohen Niederschlägen gelegenes 
Wildreservat in Südafrika. Kleinsäuger — zwei Spitzmaus- und sechs Nagerarten - wurden hier über 
das ganze Jahr verteilt im Höhenbereich zwischen 1500 und 2700 m gesammelt. Für alle acht Arten 
wird das Geschlechterverhältnis angegeben (Tabelle). Die jahreszeitliche Änderung der Gewichts- 
struktur der beiden am häufigsten gesammelten Arten, Myosorex varins und Rhabdomys pumilio 
(Abb. 2, 3), lassen auf jahreszeitlich beschränkte Fortpflanzung und eine Lebensdauer von 12 bis 16 
Monaten schließen. Auch nach dem Reproduktionszustand der Tiere ist die Fortpflanzung auf 
Frühjahr und Sommer mit warmer und feuchter Witterung begrenzt. Bei den niedrigeren Temperatu- 
ren höherer Lagen setzte die Vermehrung von Myosorex und Rhabdomys erst ungefähr einen Monat 
später ein. 


References 


BAXTER, R. M. (1977): The biology of captive Crocidura flavescens with reference to other southern 
African soricids. M. Sc. thesis, Univ. of Natal, Pietermaritzburg. 

BAXTER, R. M.; LroyD, C. N. V. (1980): Notes on the reproduction and postnatal development of the 
forest shrew. Acta Theriol. 25, 31-38. 

Brooks, P. M. (1974): The ecology of the four-striped field mouse, Rhabdomys pumilio (Sparrman, 
1784), with particular reference to a population on the van Riebeeck Nature Reserve, Pretoria. D. 
Sc. thesis, Univ. of Pretoria, Pretoria. 

BUCKNER, C. H. (1966): Population and ecological relationships of shrews in Tamarack bogs of 
southeastern Manitoba. J. Mammalogy 47, 181-194. 

CHAPMAN, B. M.; CHAPMmaAn, R. F.; ROBERTSON, 1. A. D. (1959): The growth and breeding of the 
multi-mammate rat, Rattus (Mastomys) natalensis Smith in Tanganyika Territory. Proc. zool. Soc. 
Lond. 133, 1-9. 

CHEESEMAN, C. L.; DeLany, M. J. (1979): The population dynamics of small rodents ın a tropical 
African grassland. J. Zool., Lond. 188, 451-475. 

CHURCHEIELD, $. (1980): Population dynamics and the seasonal fluctuations in numbers of the 
common shrew in Britain. Acta Theriol. 25, 415424. 

COETZEE, C. G. (1975): The biology, behaviour, and ecology of Mastomys natalensis in southern 
Africa. Bull. World Health Organ. 52, 637-644. 

Davıs, R. M. (1973): The ecology and life history of the vlei rat, Otormys irroratus (Brants, 1827), on 
the van Riebeeck Nature Reserve, Pretoria. D. Sc. thesis, Univ. of Pretoria, Pretoria. 

Davıs, R. M.; MEESTER, J. (1981): Reproduction and postnatal development in the vlei rat, Otomys 
irroratus, on the van Riebeeck Nature Reserve, Pretoria. Mammalia 45, 99-116. £ 
Derany, M. J. (1964): An ecological study of the small mammals in the Queen Elizabeth Park, 

Uganda. Rev. Zool. Bot. Afr. 52, 129-147. 

DE Wir, C. (1972): An ecological study of a small mammal community with emphasis on the status of 
Praomys (Mastomys) natalensis. M. Sc. thesis, Univ. Pretoria, Pretoria. 

Hanney, P. (1965): The Muridae of Malawı (Africa: Nyasaland). J. Zool., Lond. 146, 577-633. 

Kırrıck, D. J. B. (1963): An account of the plant ecology of the Cathedral Peak area of the Natal 
Drakensberg. Bot. Surv. S. Afr. Mem. 34, Govt Printer, Pretoria. 

MEESTER, J.; HALLETT, A. F. (1970): Notes on early postnatal development in certain southern 
Afrıcan Muridae and Cricetidae. J. Mammalogy 51, 703-711. 

PERRIN, M. R. (1979): Ageing criteria and population structure of co-existing populations of 
Rhabdomys pumilio and Otomys irroratus. S. Afr. J. Wildl. Res. 9, 84-95. 

PERRIN, M.R. (1980a): The breeding strategies of two co-existing rodents, Rhabdomys pumilio and 
Otomys irroratus: with a brief review of some pertinent life history ideas. Acta Oecologica. Oecol. 
Gener. 1, 383—410. 

PErRIn, M.R. (1980b): The feeding habits of two coexisting rodents, Rhabdomys pumilio (Sparrman, 
1784) and Otomys irroratus Brants, 1827 in relation to rainfall and reproduction. Acta Oecologica. 
Oecol. Gener. 1, 71-89. 

PrankA, E. R. (1970): On r- and K-selection. Am. Nat. 104, 592-597. 

PIENAAR, U. DE V.; RAUTENBACH, I. L.; DE GRAAFF, G. (1980): The small mammals of the Kruger 
National Park. Nat. Parks Board, Pretorıa. 

RAUTENBACH, I. L. (1976): A survey of the mammals occurring in the Golden Gate Highlands 
National Park. Koedoe 19, 133-144. 


356 D. T. Rowe-Rowe and ]J. Meester 


RAUTENBACH, 1. L. (1978): The mammals of the Transvaal. Ph. D. thesis, Univ. of Natal, Pietermaritz- 
burg. 

RowE-ROWE, D. T.; MEESTER, J. (In press): Biology of Myosorex varius ın an African montane 
region. Acta Zool. Fennica. 

SCOTCHER, J. $S. B.; CLARKE, J. C.; Lowrry, P. B. (1980): The effect of fire on herbage production and 
quality in Giant’s Castle Game Reserve. Report to Nat. Prog. Env. Sciences, CSIR. 

SMITHERS, R. H.N. (1971): The mammals of Botswana. Nat. Mus. Rhod., Mus. Mem. 4, 1-340. 

TayLor, K. D.; GREEN, M. G. (1976): The influence of rainfall on diet and reproduction in four 
African rodent species. J. Zool., Lond. 180, 367-389. 

VLAsAK, P. (1980): Seasonal changes in the surface activity of the common shrew, Sorex araneus, 
(Insectivora). Vest. &s. Spolec. zool. 44, 306-319. 

Voceı, P. (1980): Metabolic levels and biological strategies in shrews. In: Comparative physiology: 
primitive mammals. Ed. by K. SchMIDT-NIELSEN, L. Borıs and C. R. TAyLor, Cambridge: Univ. 
Press. 


Authors’ addresses: D. T. RowE-Rowe, Natal Parks Board, Postbox 662, Pietermaritzburg, 3200 
South Africa; and Prof. J. MEESTER, Dept. of Biological Sciences, University of 
Natal, Durban, 4001 South Africa 


Die Chromosomen von Microtus arvalis (Rodentia, Microtinae) 


Von ROSWITHA GAMPERL 


Institut für Medizinische Biologie und Humangenetik der Universität Graz 


Eingang des Ms. 21. 6. 1982 
Abstract 


Chromosomal studies in Microtus arvalıs (Rodentia, Microtinae) 


Studied C- and G-banded karyotypes of Microtus arvalıs s. str. (2n = 46) from different localities in 
Europe. Analysis of C-bands revealed large heterochromatic blocks in the centromeric region of 
several small metacentric autosomes, ranging in numbers from 18 to 22. Further polymorphisms due 
to pericentric inversions could be demonstrated by G-banding in chromosome pairs nos. 5, 9, 17 and 
18. Chromosome homologies with other Microtinae species are discussed. 


Einleitung 


Die Superspecies „Microtus arvalis“ umfaßt eine morphologisch ziemlich einheitliche 
Gruppe mit einem weiten Verbreitungsareal in Europa und Asien (CoRBET 1978). Da 
wegen der geringen Unterschiede morphologische Kriterien zu einer Klassifizierung nicht 
immer ausreichten, wurden häufig Karyotypanalysen zur Abklärung taxonomischer Pro- 
bleme herangezogen. Dabei zeigte sich, daß - im Gegensatz zu anderen Merkmalen - die 
Karyotypen nicht nur zwischen, sondern auch innerhalb der einzelnen Subspecies stark 
variieren. Während die diploide Chromosomenzahl der asiatischen und osteuropäischen 
Vertreter zwischen 2n = 46 und 2n = 54 schwankt (Savıc et al. 1971; Marycın 1973; 
Marysın und OrLov 1974; ZıvkovI6 und PETRoOv 1974; Zıvkovid etal. 1974; KrAL 1976; 
BELCHEVA et al. 1977), wurden bisher aus Mittel- und Westeuropa nur Formen mit 2n = 
46 beschrieben (NIETHAMMER und WınkıngG 1971; KrAL et al. 1979) Es waren vor allem 
die unterschiedliche Chromosomenzahl und Kreuzungsversuche mit sterilen Hybriden, 
die MEJER et al. (1973) zu einer Abgrenzung der Form mit 2n = 54 als Microtus subarvalıs 
von Microtus arvalıs s. str. veranlaßten. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4706-0356 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 356-363 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Die Chromosomen von Microtus arvalıs (Rodentia, Microtinae) 357 


Auch innerhalb von Microtus arvalıs s. str. ist der Karyotyp nicht stabil. Die Anzahl der 
akrozentrischen Chromosomenpaare variiert von 1 bis 12, wobei für europäische Vertreter 
hauptsächlich 4 Paare akrozentrischer Chromosomen angegeben werden (OrLov und 
Marvin 1969; MAryGin 1974; KrAL und LyarpunovA 1975). Als Ursache für diesen 
Polymorphismus werden Umwandlungen von ursprünglich akrozentrischen zu meta- bis 
submetazentrischen Chromosomen durch perizentrische Inversionen vermutet. MATTHEY 
(1957, 1973) betrachtet innerhalb der Microtinae Karyotypen von 2n =54-56 mit akrozen- 
trischen Chromosomen als ursprünglich. Nach seinen Vorstellungen waren perizentrische 
Inversionen und Robertsonsche Fusionen die wesentlichsten Chromosomenveränderun- 
gen, die zu einer Reihe von Chromosomenformeln mit geringerer Chromosomenzahl und 
geringerer Anzahl von akrozentrischen Chromosomen führten. Diese beiden Haupttrends 
in der Evolution der Wühlmauschromosomen bzw. der Chromosomen überhaupt lassen 
sich also auch innerhalb der Superspecies ‚Microtus arvalıs“ nachweisen. Nach Einführung 
der Bänderungsmethoden im letzten Jahrzehnt wurde neben der Robertsonschen Fusion 
und der perizentrischen Inversion ein weiterer häufig vorkommender Mechanismus der 
Chromosomenevolution erfaßt: die Veränderung in Gehalt und Verteilung von konstituti- 
vem Heterochromatın, d.s. Chromosomenabschnitte, die sich mit Hilfe der C-Bänderme- 
thode durch stärkere Anfärbung darstellen lassen und im allgemeinen genetisch inaktiv 
sind, d.h. nicht transkribiert werden. Daraus ergibt sich, daß Veränderungen ım Hetero- 
chromatingehalt kaum Auswirkungen auf die Lebensfähigkeit haben, zum Unterschied 
von Veränderungen im Euchromatingehalt. Die Folge davon ist, daß im Laufe der 
Evolution das Euchromatin zwar umverteilt, sonst aber relativ konstant erhalten bleibt, 
während es im Heterochromatin nicht nur zu Umverteilungen, sondern auch zu beträchtli- 
cher Vermehrung oder Reduktion kommen kann. 

Ziel der vorliegenden Arbeit war es, die Chromosomen europäischer Vertreter von 
Microtus arvalıs (2n = 46) mit Hilfe der Bänderungsmethoden genauer zu charakterisie- 
ren, um sie mit den Chromosomen anderer Wühlmausarten vergleichen zu können und 
dadurch die Chromosomenumbauten, die zur Bildung des Feldmausgenoms geführt 
haben, genauer bestimmen zu können. 


Material und Methoden 


Die Chromosomenuntersuchungen wurden an Tieren von folgenden Fundorten durchgeführt: Micro- 
tus arvalıs arvalis aus Neumarkt (Österreich) und Eifel bzw. Osnabrück (Deutschland), Microtus 
arvalıs asturianus aus Spanien. 

Für die Chromosomenpräparation wurden Fibroblastenkulturen aus Ohrbiopsien angelegt. Die 
Zellkulturen wurden in Eagle’s Medium Earle BSS, das mit 20 % fötalem Kälberserum versetzt wurde, 
gehalten und nach kurzer Kulturdauer mit Colcemid unterbrochen. Nach 2stündiger Colcemidein- 
wirkung wurden die Zellen einer Hypotoniebehandlung mit 0,28%iger KClI-Lösung unterzogen, 
anschließend mit einem Gemisch aus Methanol und Eisessig im Verhältnis 3 :1 fixiert und auf nasse, 
fettfreie Objektträger aufgetropft. Getrocknete Präparate wurden entweder der G-Bänderfärbung 
(nach SumneEr et al. 1971) oder der C-Bänderfärbung (nach SUMNER 1972) unterzogen. 


Ergebnisse und Diskussion 


Mit Hilfe der genannten Bänderungstechniken konnten mehrere Chromosomenvariatio- 
nen nachgewiesen werden. Diese Unterschiede umfassen: perizentrische Inversionen der 
Chromosomenpaare Nr. 17 und 18 als Ursache der verminderten Anzahl von akrozentri- 
schen Chromosomen bei Microtus arvalis arvalis aus Osnabrück, einen Polymorphismus 
der Chromosomen Nr. 5 und 9, der ebenfalls auf perizentrische Inversionen zurückzufüh- 
ren ist und zu einer Änderung des Armverhältnisses führt, und Unterschiede in Anzahl 
und Verteilung der großen Heterochromatinblöcke im Zentromerenbereich. Einen Über- 


358 Roswitha Gamperl 


blick über diese Unterschiede gibt die Tabelle, aus der für jeden der angeführten Fundorte 
ein eigenes Chromosomenmuster abgeleitet werden kann, während die drei Tiere aus 
Osnabrück und die beiden Exemplare aus Spanien untereinander übereinstimmende 
Karyogramme aufweisen. Daher könnte man - trotz der geringen Anzahl der zur Verfü- 
gung stehenden Tiere - annehmen, daß es sich dabei tatsächlich um populationsspezifische 
Merkmale handelt. Allerdings wäre es auch möglich, daß dieselben Variationen innerhalb 
einer Population, und zwar mit geringerer Häufigkeit, auftreten. Diese Frage kann aber 
erst durch weitere Untersuchungen an einer größeren Anzahl von Tieren abgeklärt 
werden. An dieser Stelle soll lediglich festgehalten werden, daß es solche Polymorphismen 
gibt. 


Tabelle 
Übersicht der Karyotypunterschiede 


Fundort Anzahl Anzahl Anzahl d. Polymorphismus der Chromosomen Nr. 
der d. Hetero- akrozentr. 
Tiere chromatin- Chrom. ohne 
blöcke Heterochr. 9 17 18 


. arvalıs Osnabrück 20 sm/sm m/m m/m m/m 
. arvalis Neumarkt 18 sm/sm ’/? ?/? ?/? 
. arvalıs Eifel 22 sm/st /? a/a a/a 
. asturianus Spanien 20 s/st  sm/sm a/a a/a 


akrozentrisch; m = metazentrisch; sm = submetazentrisch; st = subtelozentrisch 


Abb. 1 zeigt einen Vergleich des G-Bändermusters von Microtus arvalıs arvalıs aus 
Osnabrück mit dem von Microtus arvalıs asturianus aus Spanien. Es besteht weitgehende 
Übereinstimmung mit Ausnahme der Chromosomenpaare Nr. 5, 9, 17 und 18. Die beiden 
metazentrischen Chromosomenpaare Nr. 17 und 18 der Exemplare aus Osnabrück lassen 
sich durch perizentrische Inversionen von den entsprechenden akrozentrischen Chromo- 
somenpaaren der ssp. asturianus ableiten. Das Chromosomenpaar Nr. 5 unterscheidet sich 
durch Umlagerung eines dunkel gefärbten Bandes in Zentromerennähe. Dieses Band 
befindet sich bei den Osnabrücker Tieren auf dem kurzen Arm, bei der ssp. asturianus 
dagegen auf dem langen Arm. Ein dem langen Arm der submetazentrischen Form von 
Chromosom Nr. 5 entsprechendes G-Bändermuster findet man z.B. auch im Chromosom 
Nr. 1 von Clethrionomys rufocanns (GAMPERL 1982) oder im Chromosom Nr. 2 von 
Microtus agrestis (COOPER und Hsu 1972). Ein Chromosom, das dem gesamten submeta- 
zentrischen Chromosom entspricht, ist bei Microtus subterraneus vorhanden, während sich 
bei Arvicola terrestris der kurze Arm von einem anderen Chromosom ableitet (eigene 
Ergebnisse). Daher erscheint es wahrscheinlicher, daß die submetazentrische Form dieses 
Chromosoms die ursprünglichere ist und die subtelozentrische Form, wie sie bei Microtus 
arvalıs asturianus vorkommt, durch perizentrische Inversion daraus entstanden ist. 

Ein weiterer Polymorphismus betrifft die großen Heterochromatinblöcke im Zentro- 
merenbereich der kleinen Autosomen (Abb. 2-5). Während bei allen untersuchten Tieren 
die 5 großen Autosomenpaare und auch eine Reihe von den kleinen kein Heterochromatin 
zeigen, weisen die übrigen besonders große und intensiv gefärbte Heterochromatinblöcke 
in unterschiedlicher Zahl auf. Die Zahl variiert von 22 bei der ssp. arvalıs aus der Eifel über 
20 (ssp. arvalıs aus Osnabrück und ssp. asturianus) bis 18 (ssp. arvalıs aus Neumarkt). 
Microtus arvalıs asturianus hat zwei akrozentrische Chromosomenpaare (Nr. 11 und 12) 
ohne Zentromerenheterochromatin, Microtus arvalis arvalıs aus der Eifel und aus Neu- 
markt je eines (Nr. 11 bzw. Nr. 13) und Microtus arvalıs arvalıs aus Osnabrück keines. Da 
die Heterochromatinblöcke sowohl bei den akrozentrischen als auch bei den metazentri- 
schen Chromosomen immer in gerader Zahl vorhanden sind, kann man annehmen, daß sie 


13 13 14 14 s S 16 16 


21 S e2 22 xı 


Abb. 1. Vergleich der G-Bänder von Microtus arvalıs asturianus (A) und Microtus arvalıs arvalıs aus 


Osnabrück (B) 


360 Roswitha Gamperl 


jeweils auf homologen Chromosomen liegen. Diese ausgedehnten Heterochromatin- 
blöcke, die an akrozentrische Chromosomen angelagert kurze Arme bilden, können 
ebenfalls zu einer Veränderung der ursprünglich akrozentrischen Chromosomen bzw. zu 


Problemen bei der Zuordnung führen. 


WR 


I RN N 
x S N N 
oT SS 
UN SS 
NL ar 
iV m. 
. sn 


N Nr S ® 
6 1 8 9 10 41 2 
ar mm u 
S 14 15 16 17 18 49 


xx 


20 21 22 


Abb. 3. C-Bänderkaryogramm von Microtus arvalis arvalis (?) aus Osnabrück 


Die Chromosomen von Microtus arvalis (Rodentia, Microtinae) 361 


Ein detaillierter Vergleich aller Chromosomen von Microtus arvalıs mit denen anderer 
Wühlmausarten liegt nicht in der Absicht dieser Arbeit. Er würde sich auch wegen der 
vielen kleinen metazentrischen Chromosomen schwierig gestalten, da ihre G-Bänder durch 
die perizentrischen Inversionen gegenüber den ursprünglich akrozentrischen Chromoso- 
men nicht nur stark verändert worden sind, sondern auch noch große Ähnlichkeiten 
zwischen einzelnen Chromosomenpaaren bestehen. Der Vergleich beschränkt sich daher 
auf die 5 großen Chromosomenpaare von Microtus arvalıs s. str., die sich durch Robert- 
sonsche Fusionen aus ursprünglich 20 akrozentrischen Chromosomen, wie sie noch im 
Karyotyp von Clethrionomys rufocanus vorhanden sind (GAMPERL 1982), ableiten lassen. 
Auch die Entstehung größerer meta- oder submetazentrischer Chromosomen bei einigen 
anderen Wühlmausarten, z.B. Microtus subterraneus (2n = 52, 1 großes metazentrisches 
und 1 großes submetazentrisches Chromosomenpaar) und Microtus oeconomus (FREDGA 
1980) (2n = 30, hauptsächlich metazentrische Fusionschromosomen), läßt sich so erklären. 
Allerdings lassen sich die 4 großen metazentrischen Chromosomenpaare der Feldmaus 
nicht vollständig mit metazentrischen Chromosomen von Microtus oeconomus oder Micro- 
tus subterraneus homologisieren, da sich die neuentstandenen Chromosomen jeweils aus 
verschiedenen akrozentrischen Ausgangschromosomen zusammensetzen, eine Tatsache, 
die auf eine weitgehend unabhängige Entstehung des Feldmauskaryotyps vor allem 
gegenüber dem von Microtus oeconomus hinweist. Mit Microtus subterraneus hat die 
Feldmaus zumindest das submetazentrische Chromosom Nr. 5 gemeinsam, wobei es — wie 
schon oben erwähnt - intraspezifisch bei einigen der vorgestellten Tiere eine weitere 
Abänderung durch eine perizentrische Inversion erfahren hat. Auch die Entstehung des 
Chromosoms Nr. 3 kann nicht durch Robertsonsche Fusion allein erklärt werden. Es muß 
mindestens noch einen zusätzlichen Umbau durchgemacht haben. Der Karyotyp von 
Microtus arvalıs s. str. setzt sich also nicht nur fast ausschließlich aus Chromosomen 
zusammen, die gegenüber den Ausgangschromosomen verändert wurden, sondern weist 
darüber hinaus an einigen Chromosomen (Nr. 3 und 5) noch zusätzliche Umstrukturie- 
rungen auf. Eine ähnliche Situation besteht auch bei anderen stark abgeleiteten Karyo- 
typen, die hauptsächlich aus zweiarmigen Chromosomen bestehen (ELDER 1980). 


N “x ; \ 
1 2 3 u 5 
; N W r . n - 
6 \ 8 9 10 Ss 12 
13 14 15 16 17 18 19 


a» on. .e 
20 a 22 x 


Abb. 4. C-Bänderkaryogramm von Microtus arvalıs astuarianus (6) 


362 Roswitha Gamperl 


SR 


% 3 L 5 
6 \ nr n 
‚k ”w >.“ “.» “ä 
N 
she .. »-. 
_ . z “c 


Abb. 5. C-Bänderkaryogramm von Microtus arvalıs arvalıs (3) aus Neumarkt 


Betrachtet man das C-Bändermuster der in dieser Arbeit vorgestellten Vertreter von 
Microtus arvalıs, so wird auch hier ein deutlicher Unterschied gegenüber anderen Wühl- 
mausarten offenbar (GAMPERL et al. 1982). C-Bänderstudien von weiteren Vertretern der 
Superspecies „Microtus arvalıs“ wären notwendig, um herauszufinden, ob dieses bisher für 
Wühlmäuse einmalige Heterochromatinmuster auch bei anderen Vertretern vorkommt. Es 
wäre vor allem interessant zu wissen, ob das C-Bändermuster von Microtus subarvalıs mit 
2n = 54 Chromosomen ähnlich aussieht wıe das anderer ‚‚primitiver‘“ Wühlmäuse, z.B. 
der Schneemaus (BELCHEVvA et al. 1980) oder der Rötelmäuse (GAMPERL 1982), d.h. mit 
kleinen Heterochromatinblöcken an mehr oder weniger allen Chromosomen versehen ist, 
oder ob es ähnlich massıve Blöcke an einigen kleinen Chromosomen aufweist wie bei den 
hier vorgestellten Individuen von Microtus arvalıs. Weitere G- und C-Bänderuntersuchun- 
gen von Vertretern der Superspecies „Microtus arvalıs‘“ könnten daher nicht nur dazu 
beitragen, eine lückenlose Abfolge der Chromosomenumbauten zu konstruieren, die zur 
Entstehung des Karyotyps von Microtus arvalıs s. str. geführt haben, sondern auch 
zusätzliche Daten zur Veränderung im Gehalt und der Verteilung von Heterochromatin im 
Verlauf der Evolution liefern. 


Danksagung 


Für die Überlassung und Bestimmung der Tiere möchte ich mich besonders herzlich bedanken bei 
J- NIETHAMMER, Bonn, R. SCHRÖPFER, Osnabrück, und W. SıxL, Graz. 


Zusammenfassung 


G- und C-Bänderkaryogramme einiger europäischer Vertreter von Microtus arvalıs s. str. (2n = 46) 
werden beschrieben. Nach C-Bänderfärbung lassen sich ausgedehnte Heterochromatinblöcke in der 
Zentromerenregion einiger kleiner Autosomen darstellen. Die Zahl dieser Heterochromatinblöcke 
varııert bei den untersuchten Tieren von 18 bis 22. Weitere Chromosomenvariationen, die mit 
perizentrischen Inversionen erklärt werden können, liegen in den Autosomenpaaren Nr. 5, 9, 17 und 
18 vor. - Homologien mit Chromosomen anderer Wühlmausarten werden diskutiert. 


Die Chromosomen von Microtus arvalıs (Rodentia, Microtinae) 363 


Literatur 


BELCHEVA, R. G.; PEsHEv, T. H.; PEshev, D. T. (1980): Chromosome C- and G-banding patterns in a 
Bulgarian population of Microtus guentheri Danford et Alston (Microtinae, Rodentia). Genetica 
52/53, 45-48. 

BELCHEvA, R. G.; PEsHEv, T. H.; Rapjagıı, S. I. (1977): Analysis of the chromosome set in the 
Bulgarian population of Microtus arvalıs. Zoll. ZH 56, 315-317. (Russ.) 

CooBER, J. E. K.; Hsu, T. C. (1972): The C-band and G-band patterns of Microtus agrestis 
chromosomes. Cytogenetics 11, 295-403. 

CoRBET, G. B. (1978): The mammals of the palaearctic region. A Taxonomic review. Brit. Mus. Nat. 
Hist. 

ELDER, F. F. B. (1980): Tandem fusion, centric fusion, and chromosomal evolution in the cotton rats, 
genus Sıgmodon. Cytogenet. Cell Genet. 26, 199-210. 

FREDGA, K.; PERSSON, A.; STENSETH, N. C. (1980): Centric fission in Microtus oeconomus. A 
chromosome study of isolated populations in Fennoscandia. Hereditas 92, 209-216. 

GAMPERL, R. (1982): Chromosomal evolution in the genus Clethrionomys. Genetica 57, 193-197. 

GAMPERL, R.; EHMANN, C.; BACHMAnN, K. (1982): Genome size and heterochromatin variation in 
rodents. Genetica 58, 199-231. 

KrAr, B. (1976): A species of the genus Microtus (Microtidae, Mammalia) new for the fauna of 
Bulgaria. Zool. Listy 24, 353-360. 

KrAL, B.; Lyapunova, E. A. (1975): Karyotypes of 46-chromosome Microtus arvalis (Microtidae, 
Rodentia). Zool. Listy 24, 1-11. 

KrAı, B.; Zıma, J.; HERZIG-STRAscCHIL, B.; STERBA, O. (1979): Karyotypes of certain small mammals 
from Austria. Folia Zool. 28, 5-11. 

Marvcın, V. M. (1973): Karyotypes of Microtus transcaspicus and the role of karyological method in 
estimation of species independence of voles from the group Microtus arvalis (Rodentia, Criceti- 
dae). Zool. ZH 52, 791-794. (Russ.) 

Marycin, V. M. (1974): A comparative morphometric analysis of karyotypes in two geographical 
forms of 46-chromosome common vole, Microtus arvalıs. Zool. ZH 53,, 769-778. (Russ.) 

Marycın, V. M.; Orıov, V.N. (1974): Areas of 4 species of voles (superspecies Microtus arvalıs) by 
karyological data. Zool. ZH 53, 616-622. (Russ.) 

MATTHEY, R. (1957): Cytologie comparee, systematique et phylog&nie des Microtinae (Rodentia- 
Muridae). Rev. Suisse Zool. 64, 39-71. 

MATTHEY, R. (1973): The chromosome formulae of eutherian mammals. In: Cytotaxonomy and 
vertebrate evolution. Ed. by A. B. CHrarELLI and E. Carpanna. London-New York: Academic 
Press. 

MEJER, M. N.; MoRroz, I. M.; OrLov, V. N.; ScHoLt, E. D. (1973): Zwillingsarten der Feldmaus, 
Microtus arvalıs (Pallas). Mitt. Zool. Mus. Berlin 49, 387-402. 

NIETHAMMER, J.; WInking, H. (1971): Die spanische Feldmaus (Microtus arvalis asturianus Miller, 
1908). Bonn. Zool. Beitr. 22, 220-235. 

OrLov, V. N.; Marysın, V. M. (1969): Two forms of the 46-chromosome Microtus arvalıs Pall. In: 
The mammals: Evolution, karyology, taxonomy, fauna. Ed. by N. N. VoRoNTsov. Novosibirsk: 
Acad. Sci. USSR. (Russ.) 

Savı6, I.; SoLDATovIE, B.; Duriıc, B. (1971): Prilog poznavanju kariotipa vrste Microtus arvalıs 
Pallas, 1779 (Rodentia, Microtidae) iz Vojvodine. Arh. Biol. Nauka, Beograd 23, 27P-28P. 

SUMNER, A.T. (1972): A sımple technique for demonstrating centromeric heterochromatin. Exp. Cell 
Res. 75, 304-306. 

SUMNER, A. T.; Evans, H. J.; BuckrLan, R. (1971): New technique for distinguishing between 
human chromosomes. Nature New Biol. 232, 31-32. 

Zıvkovi6, S.; PETRov, B. (1974): Record of a vole of the Microtus arvalis group with 54 chromosomes 
(Microtus subarvalıs Meyer, Orlov et Skholl, 1972) in Yugoslavia and comparison of its karyotype 
with that of Pitymys felteni Malec et Storch, 1963 (Rodentia, Mammalıa). Genetika 6, 283-288. 

Zıvkovi6, S.; Rımsa, D.; Ruzıd, A.; PETRov, B. (1974): Cytogenetical characteristics, taxonomic 
status and distribution of the voles with 46 and 54 chromosomes of the Microtus arvalıs group ın 
Yugoslavia (Rodentia, Mammalia). Arh. Biol. Nauka, Beograd 26, 123-124. 


Anschrift der Verfasserin: Dr. ROSWITHA GAMPERL, Institut für Medizinische Biologie und Human- 
genetik der Universität Graz, Harrachgasse 21/8, A-8010 Graz, Österreich 


Feeding behavior and diets of Neotropical water rats, 
genus Ichthyomys Thomas, 1893 


By R. S. Voss, J. L. SıLva L. and J. A. Vanpes L. 


Museum of Zoology, University of Michigan, U.S.A. and Instituto de Zoologia Tropical, 
Universidad Central de Venezuela 


Receipt of Ms. 18. 8. 1982 
Abstract 


Observations of the feeding behavior of captive animals and dissections of the digestive tracts of kill- 
trapped specimens suggest that Neotropical water rats of the genus /chthyomys regularly consume a 
wide variety of aquatic organisms, chiefly arthropods, and are not so piscivorous as their name 
implies. Behavioral and dietary data further indicate that crabs of the family Pseudothelphusidae may 
be an important food source for lowland populations of water rats; highland populations of 
Ichthyomys, however, may be largely insectivorous. 


Introduction 


Semiaquatic rats of the genus /chthyomys are among the most unusual of Neotropical 
mammals, but few reliable data on their ecology or habits are available. A single fish 
contained in the stomach of the holotype of /. stolzmannı, a Peruvian form, led Tuomas 
(1893a, b) to infer that the rodent was piscivorous; subsequent authors have therefore 
referred to species of /chthyomys as Fish-Eating Rats, and have even attrıbuted the 
alimentary habits implied by this dubious sobriquet to other, related cricetine genera 
whose diets are wholly unknown. In this note we report observations on the feeding 
behavior of two captıve specimens of /chthyomys, present data on stomach contents of 
wild-caught examples, and discuss the probable trophic role of these rats in Neotropical 
streams and rivers. 


Feeding behavior 


The two captive animals from which our behavioral observations derive are examples of 
Ichthyomys pittieri (HanpLev and Monporrı 1963). The holotype of 7. pittieri was 
collected in 1958 from a small, clear mountain stream (Fig. 1) flowing through evergreen 
rainforest in the Cordillera de la Costa of Estado Aragua, Venezuela (BEEBE and CRANE 
1947 describe the local ecology), but efforts by numerous collectors over the following 
22 years failed to secure any additional specimens. In 1979 Voss revisited the type locality 
and took a single adult male water rat (MBUCV [= Museo de Biologia, Universidad 
Central de Venezuela; Caracas] I-2776), the second known specimen, in a National live 
trap set in shallow water beneath an overhanging rootmat at 925 m elevation. The animal 
was kept alive for five days in the laboratories of the Universidad Central de Venezuela at 
Estacion Biologica Rancho Grande. That its behavior might be observed, the rat was placed 
nightly in a glass 50 gallon terrarium filled to a depth of several centimeters with local soil 
and provided with a large, shallow enamel pan containing 1-2 cm of tap water. Food items, 
collected from local streams, were placed in the water or were offered the rodent with 
forceps. Observations totalling about 10 hours were made shortly after dusk by artificial 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4706-0364 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 364-369 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Feeding behavior and diets of Ichthyomys 365 


light on five consecutive nights. Movements and noises outside the tank elicited no 
apparent response from the rat. Behaviors described below were exhibited repeatedly by 
the animal during its period of confinement, and are here reported as though comprising a 
typical feeding sequence. 


Upon entering the water- 
filled pan, the rat invarıably 
lowered its muzzle in such a 
way that the tips of the mysta- 
cıal vibrissae were brought in 
contact with the bottom of 
the container and began to 
move swiftly about in the wa- 
ter with the ventrally-directed 
whiskers sweeping a broad 
path before it. This evidently 
purposive searching behavior, 
characterized by direct and 
close application of vibrissae 
to substrate, was never ob- 
served while the rat was out- 
side of water. Live prey items 
were attacked when they 
moved after having been 
touched by vibrissae; strug- 
gling items underfoot, how- 
ever, were usually ignored as 
were most items (dead or 
alive) that did not move when 
touched by whiskers. Small 
prey (15-20 mm or less in 
length) were seized with the 
forepaws and bitten repeated- 
ly until subdued. Usually, 
such diminutive items were 
eaten zn sıtn, the rat typically 
crouching on its hindquarters 
Fig. 1. The type locality of /chthyomys pittieri ın the Cordillera in the shallow water with the 
de la Costa, Estado Aragua, Venezuala 10° 21’ N, 67°41'’W; prey held between its fore- 


the portion of the stream shown in the photograph is at approxi- paws. Pieces of the morsel 
mately 900 m elevation (photogr. by P. KAARAKKA) 


were bitten off rapıdly, 
chewed and swallowed until 
no more remained. Caddisfly (Trichoptera: Hydropsychidae), stonefly (Plecoptera: Per- 
lidae), dragonfly (Odonata: Libellulidae) and small dobsonfly (Megaloptera: Corydalıdae) 
larvae were eaten in this way. 

Larger prey elicited other, conspicuously different responses from the rat. Tadpoles 
(unidentified, to 50 mm), fishes (Characidae: Creagrutus beni, to 55 mm standard length) 
and large dobsonfly larvae (to 65 mm) were pinned to the bottom of the pan with the 
forepaws, bitten repeatedly until morıbund, and then carried out of the water to be 
consumed on dry land. Securing and eating large prey were often lengthy processes. 
Slippery anımals (fish and tadpoles) appeared difficult to capture, even in the very shallow 
water of the pan. Dobsonfly larvae were difficult to subdue, and the rat acted several times 


366 R. S. Voss, J. L. Silva L. and J. A. Valdes L. 


as thought it had been bitten while attacking these large and formidable insects. Once 
carried out of the water, large prey items were held pressed to the substrate with the 
forepaws, seized with the incisors, and torn apart by powerful, backward jerks of the head 
and neck. Prey of all kınds were thoroughly comminuted by the incisors and only briefly 
chewed. | 
Behaviors described above seemed to vary chiefly with prey sıze inasmuch as similar 
responses were elicited by similar-sized prey of several species. Freshwater crabs 
(Pseudothelphusidae, to 50 mm carapace width), however, were attacked, subdued and 


Fıg. 2. Ichthyomys pittieri (the 
Rancho Grande specimen) 
eating a crab. Fragments of 


a previous meals litter the fore- 
ground (photogr. by P. Kaa- 
RAKKA) 


eaten in a distinctive, stereotyped manner. When encountered by the rat, crabs usually 
reacted by raising and spreading their open chelae. The rodent always attacked immedi- 
ately, darting at the crab with swift, lunging movements of forepaws and head until one of 
the claws and then the other were pinned to the substrate and bitten off at the base. The rat 
then seized the carapace with its forepaws (Fig. 2) and, tipping it from side to side, bit off 
the remaining walking legs; afterwards, the crustacean was turned upside down and the 
exoskeleton opened with powerful bites through the posterior margins of the carapace and 
abdomen. With lower incisors and tongue, the rat scooped out gills, muscle and viscera and 
then discarded the empty shell (Fig. 3). If the crab was a large one, the rat would frequently 
bite open the proximal segments of each of the detached walking legs and eat the enclosed 
muscle. Because the rat was sacrificed shortly after it had eaten several crabs, Voss was able 
to confirm by dissection that only gills, muscle and viscera had been consumed; the few 
small exoskeletal fragments contained in the gut would not have been identifiable had their 
source been unknown. 

The Rancho Grande water rat evinced marked preferences for some of the aquatic 
organısms offered it but completely refused others. Crabs were always accepted and were 
eaten immediately, as were dobsonfly and dragonfly larvae. Fishes and tadpoles, however, 
while invarıiably killed and carried out of the water, were seldom eaten completely. 
Sometimes the rat would eat a small portion, from fishes near the caudal peduncle, from 
tadpoles the tail tip. Often, however, the fish or tadpole was simply carried outside the pan, 
dropped, and subsequently ignored. The rodent appeared unable to capture live Colos- 
tethus collaris (Dendrobatidae), a locally ubiquitous streamside frog, but would accept and 
eat dead frogs when these were provided. By contrast, Pachychilus laevissimns 
(Pachychilidae), an abundant local freshwater snail, was never eaten by the rat even when 
shells had been crushed prior to presentation. 

The Rancho Grande rat fed voraciously; during one typical 75 minute foraging bout the 


rodent consumed a total of two frogs, six crabs, five dobsonfly and two stonefly larvae, and _ 


one caddisfly and one dragonfly larva. 


Feeding behavior and diets of Ichthyomys 367 


Fig. 3. Carapaces of pseudo- 
thelphusid crabs eaten by the 
Rancho Grande water rat; 
scale in cm (photogr. by P. 
KAARAKKA) 


The second live example of /chthyomys (MBUCV I-2803) for which we have some 
behavioral information was collected by SıLva and VALDEs at an elevation of 700 m on the 
northern slopes of the Cordillera de la Costa above Naiguata, Distrito Federal, Venezuela. 
The capture site occupies a zone of transition between the dry forests of the lower slopes 
and the wet forests of the higher slopes of the sierra (see STEYERMARK and HUBEr 1978 for 
a description of local phytophysiognomy). The rat was captured by hand in the early 
afternoon as it was eating a crab (Neopsendothelphusa fossor) at the edge of a cement 
overflow canal that drained a nearby water collecting tank. The animal was maintained 
alive in a small wire cage for two days in the field, but died of unknown causes shortly after 
being transported to Caracas. 

The Naiguata rat was offered local frogs (Colostethus sp.) and crabs (N. fossor and 
Eundaniella ranchograndensis). Only the crustaceans were accepted, and the sequence of 
rat/crab behavioral interactions observed by Sırva and VALDESs were strikingly similar to 
those already described for the Rancho Grande rat: the rodent pinned the crab’s chelae 
with its forepaws and then severed the limbs near their articulation to the carapace with its 
incisors; afterwards, the remaining four pairs of walking legs were detached, and the 
carapace inverted and opened with bites through the caudal margin; the soft contents both 
of the carapace and of the proximal limb segments were eaten. Over the two day period 
during which it was maintained in the field, the Naiguata rat killed and ate a total of 13 N. 
fossor, each of about 40 mm carapace width, and 1 FE. ranchograndensis 54 mm in width. 
The rodent was frozen soon after its death; dissection of the digestive tract revealed that, 
like the Rancho Grande rat, the Naiguata specimen had ingested only a few small 
exoskeletal fragments. 


Dissections 


Additional dietary data are available from ten kill-trapped specimens: /chthyomys hydro- 
bates (Venezuela, Merida; UMMZ [= University of Michigan Museum of Zoology; Ann 
Arbor] 156375), 7. soderstromi (Ecuador, Pichincha; AMNH [= American Museum of 
Naural History; New York] 64624), 7. tweedi (Ecuador, Pichincha; UMMZ 126300, 
155782-155788). Complete alimentary tracts were dissected from fluid-preserved carcasses 
and their contents removed to 70 % ethanol. Macroscopically distinctive prey fragments 
(tergites, antennae, mouthparts, legs, gills, etc.) were sorted and ıdentified by matching 
them with corresponding parts of intact specimens collected from streams; invertebrates 
thus identified as prey were sent to specialists for taxonomic determination. After removal 
of larger fragments, the finely-masticated residuum in each gut was cleared in potassium 
hydroxide and hydrogen peroxide and stained with alızarine ın order to heighten optical 
contrasts between vertebrate bone fragments (if present) and the non-osseous tissues of 
invertebrates. 

All digestive tracts examined contained abundant arthropod exoskeletal fragments. 
Eight of ten tracts contained Arthropoda exclusively, but two (of /. tweedii from western 
Ecuador) contained also teeth and bone fragments of small fishes (Teleostei: Characidae). 
The single specimen of /. hydrobates from Venezuela contained stonfly (Perlidae), cad- 


368 R. S. Voss, J. L. Silva L. and J. A. Valdes I. 


disfly (Hydropsychidae), and dragonfly (Aeshnidae) larvae ın its gut, while the stomach 
and intestines of one /. soderstromi specimen from Quito contained only unidentifiable 
insect fragment. Dobsonfly larvae (Corydalidae), crabs (Pseudothelphusidae) and semi- 
aquatic cockroaches (Orthoptera: Blattidae) together comprised 95 % of the pooled 
stomach contents of eight 7. tweedii from the Andean foothills of western Ecuador. While 
occasional, isolated plant fragments were encountered in the digestive tracts of some 
specimens, no obviously masticated boli of vegetable material or other evidences of 
deliberate herbivory were discovered. 


Discussion 


Fishes are eaten, at least occasionally, by some species of Ichthyomys, but data reported in 
this note suggest that these water rats regularly consume a wide variety of other aquatic 
organisms, chiefly arthropods, and are not so exclusively piscivorous as their name implies. 
In fact, these rodents might, in lowland situations, as appropriately be regarded as 
cancrivorous. That crabs constitute a regular item in the diet of 7. pittieri ıs strongly 
suggested by the very similar, stereotyped behavioral responses elicited by crabs from rats 
of this species collected over 100 km apart. That crabs are also eaten by other lowland 
forms of Ichthyomys is demonstrated by their occurence in the digestive tracts of 
I. tweedıi from western Ecuador. 

The lotice waters of the Neotropical lowlands are rife with decapod Crustacea, and it 
would be remarkable indeed if these invertebrates were not often included in the diets of 
carnivorous semiaquatic mammals such as water rats. Shrimps and crayfishes as well as 
crabs are abundant in the swamps and slower-moving watercourses of the coastal plains 
and river basins, but ın the rocky brooks and streams of the Andean foorthills and 
associated sierras from which most examples of /chthyomys have been taken, the dominant 
decapods are crabs of the family Pseudothelphusidae (PEARSE 1911; HorTtHuiıs 1959; 
Firrkau 1964; PATRICK 1966; RODRIGUEZ 1980). Pseudothelphusids are conspicuous 
elements of the faunas of streams in which they occur. Small crabs frequent the wet 
streamsides and shallow riffles where they can be observed foraging at all hours; adults 
appear to be more nocturnal in their habits, and can be discovered at the bottoms of deep 
pools, beneath rocks and in excavated burrows along the water’s edge or even wandering 
far from streams (CRANE 1949; WEIBEZAHN 1952; Voss, unpubl. field notes). Armed with 
formidable crushing or cutting chelae, even small pseudothelphusids can deliver a painful 
pinch, and it is not surprising that the live /chthyomys specimens we observed exhibited 
distinctive behaviors for despatching these crustaceans. 

While crabs may comprise an important item in the diets of /chthyomys populations ın 
neotropical lowland and foothill sıtuations, however, it is unlikely that they do so at very 
high elevations from which decapods may regularly be absent. Of 58 streams sampled by 
the senior author in the course of fieldwork from 1978 to 1980 in Ecuador and Venezuela, 
decapod crustaceans were found to be present in 26 watercourses sampled below 1500 m 
elevation, but were not encountered in 29 streams sampled above 1700 m; the highest 
elevation at which decapods were taken was at 1643 m in western Ecuador. Sımilarly, 
PEARSE (1916) recorded four species of Psendothelphusa from streams in the foothills of the 
Sierra Nevada de Santa Marta, Colombia between 770 and 1540 m, but found no decapods 
in the cold brooks at 2550 m in the same mountains. In Venezuela, Pseudothelphusid crabs 
are the only decapods that occur regularly at elevations above 500 m, but records from 
altitudes exceeding 2000 m are apparently unknown (RODRIGUEZ 1980). Some highland 
populations of Ichthyomys as, for example, /. soderstromi, might therefore be almost 
exclusively insectivorous, and in this respect may resemble species of related cricetine 
genera that also inhabit streams at high elevations (HooPEr 1968; Voss, in prep.). 


Feeding behavior and diets of Ichthyomys 369 


Acknowledgements 


For advice and assistance we express our thanks and appreciation to PEDRO DURANT, FRANCISCO 
FERNANDEZ, FRANCISCO MAGO, EDGARDO MONDOLFI, GONZALO MEDINA, JUHANI OJASTE and 
ROGER PEREZ ın Venezuela, and to CARLOS AGUIRRE, SERGIO FIGUERROA, MIGUEL MORENO and 
FERNANDO ORTIZ in Ecuador. The Instituto de Zoologia Tropical (Universidad Central de Ve- 
nezuela), the Groupo de Ecologia Anımal (Universidad de los Andes, Merida) and the Museo 
Ecuatoriano de Ciencias Naturales (Quito) extended the senior author support and hospitality during 
fieldwork from 1978 to 1980. 

We are grateful to R. M. BaıL£y, J. B. BurcH, A. G. Kruse, R. W. Baumann, O. S. FLinT and 
L. K. GLoyD for identifying non-mammalian specimens reported here. Partial financial support for 
the fieldwork was provided by grants from the National Science Foundation (DEB 7910189 to R. S. 
Voss and P. Myers), the University of Michigan, the American Museum of Natural History, the 
Explorer’s Club and the Society of Sigma Xi. 


Zusammenfassung 


Ernährungsweise und Nahrung der neotropischen Wasserratte, Genus Ichthyomys Thomas, 1893 


Beobachtungen an gefangenen Tieren und Untersuchungen am Verdauungstrakt von Museumsmate- 
rial ergaben, daß neotropische Wasserratten der Gattung Ichthyomys gewöhnlich eine große Vielfalt 
aquatischer Organismen verzehren, hauptsächlich Arthropoden. Sie sind demnach nicht in dem Maße 
piscivor, wie aus dem Namen zu schließen ist. Die Beobachtungen und Nahrungsanalysen weisen 
außerdem darauf hin, daß für die Flachland-Populationen Crustaceen der Familie Pseudothelphusidae 
eine bedeutende Nahrungsquelle darstellen. Hochland-Populationen von /chthyomys scheinen dem- 
gegenüber stärker insectivor zu sein. 


References 


BEEBE, W.; CRANE, J. (1947): Ecology of Rancho Grande, a subtropical cloud forest in northern 
Venezuela. Zoologica 32, 43-60. 

CRANE, J. (1949): Fresh-water crabs of the genus Pseudothelphnsa from Rancho Grande, Venezuela. 
Zoologica 34, 25-30. 

Fırtkau, E. J. (1964): Remarks on limnology of central Amazon rain-forest streams. Verh. Internat. 
Verein. Limnol. 15, 1092-1096. 

Hanpıey, C. ©. Jr.; MonvoLrı, E. (1963): A new species of fish-eating rat, /chthyomys, from 
Venezuela (Rodentia :Cricetidae). Acta Biol. Venezuelica 3, 417-419. 

Horrnuıss, L. B. (1959): The Crustacea Decapoda of Suriname (Dutch Guiana). Zool. Verh. Leiden 
44, 296 p. 

Hooper, E. T. (1968): Habitats and food of amphibious mice of the genus Rheomys. J. Mammalogy 
49, 550-553. 

Patrick, R. (1966): Limnological observations and discussion of results. In: The Catherwood 
Foundation Peruvian-Amazon Expedition: Limnological and Systematic Studies. Ed. by R. PA- 
TRICK. Acad. Nat. Sci. Philadelphia Monog. 14, 5-40. 

PEARSsE, A. S. (1916): An account of the Crustacea collected by the Walker Expedition to Santa Marta, 
Colombia. Proc. U. S. Nat. Mus. 49, 531-556. 

RODRIGUEZ, G. (1980): Los Crustaceos Decapodos de Venezuela. Caracas: Instituto Venozolano de 
Investigaciones Cientificas. 

STEYERMARK, J. A.; HUBER, ©. (1978): Flora del Avila. Caracas: Publicacion especial de la SVCN 
Vollmer Foundation y MARNR. 

Thomas, O. (1893a): A fish eating rodent. Natur. Scı. 2, 286. 

Thomas, ©. (1893b): On some mammals from central Peru. Proc. Zool. Soc. London 23, 333-341. 

WEIBEZAHN, F. H. (1952): Contribucion al estudio de los cangrejos dulceacuicolas de la region Baruta- 
El Hatillo. Mem. Soc. Cienc. Nat. La Salle 12, 67-70. 


Authors’ addresses: ROBERT S. Voss, Museum of Zoology, University of Michigan, Ann Arbor, MI 
48109, U.S.A.; Jose LoREnZzo Sırva LuGo and JurLıo ANTONIO VALDES 
LuEnco, Instituto de Zoologia Tropical, Universidad Central de Venezuela, 
Caracas, Venezuela 


Untersuchungen zur Kurzzeitperiodik beim Meerschweinchen 
(Cavia aperea f. porcellus)' 


1. Bewegungsaktivität unter Licht-Dunkelwechsel, 
Dauerdunkel und Dauerlicht 


Von D. BÜTTNER und FRANZISKA WOLLNIK 


Institut für Versuchstierkunde und Zentrales Tierlaboratorium 
der Medizinischen Hochschule Hannover 


Eingang des Ms. 10. 6. 1982 
Abstract 


Investigations on ultradian rhythms in the guinea pıg (Cavia aperea f. porcellus) 1. Locomotor activity 
under light-dark, continuous dark and continuous light 


Studied locomotor activity of the male guinea pig (Cavia aperea f. porcellus), exposed to different 
light conditions. 

Under a light-dark cycle (LD 12:12) the guinea pig was found to be dark-active. The main 
maximum of activity occurs at the beginning of the dark period and a ancillary maximum follows 8 h 
later. 

The pattern of locomotor activity ıs characterized by the alternation of short activity bursts and 
short resting periods, mainly during the dark period. The total time of rest within 24 h summed up on 
an average of 6 h 11 min (+ 1 h 27 min) under LD, whereby 4 h 06 min (#1 h 14 min) of rest were 
spent in the light period. 

After exposure to continuous dark or light the cırcadıan rhythms were markly reduced and the 
short activity bursts manifested themselves during the whole 24-h-period. The total time of rest within 
24 h decreases to 4 h 45 min (#1 h 42 min) (DD) respectively 2 h 58 min (+1 h 50 min) (LL). Now the 
time of rest is evenly divided between the previous dark and light period. 

Analysıs of power spectra confirm the reduction of the 24-h-rhythms and the manifestation of 
ultradian rhythms of 1-2 h period length after exposure to continuous dark or light. Furthermore the 
power spectra show a 8-h-rhythm which is nearly constant under different light conditions. 


Einleitung 


Bisherige Beschreibungen zur Tagesperiodik der motorischen Aktivität beim 
Meerschweinchen beruhen fast ausschließlich auf Freilandbeobachtungen domestizierter 
Tiere Cavia aperea f. porcellus (Kıng 1956) oder freilebender Arten der Gattungen Cavıa 
(Fuchs 1980), Microcavıa und Galea (Roop 1972; PEARsoN 1951). 

Unter Freilandbedingungen werden diese Spezies als überwiegend dämmerungsaktıiv 
beschrieben; jeweils mit dem Lichtwechsel ist eine allgemeine Aktivitätszunahme zu 
beobachten. Starke Sonneneinstrahlungen während der Mittagsstunden werden besonders 
gemieden. 

Eine Registrierung der Bewegungsaktivität von Cavia aperea f. porcellus unter Labor- 
bedingungen und an Einzeltieren, jedoch unter konstanten Lichtverhältnissen, wurde von 
NicHoıuts (1926) durchgeführt. Sie beschreibt die Tiere als nahezu ständig aktiv und stellt 
keine tageszeitlichen Veränderungen der Aktivität fest. Dieser Befund erscheint uns 


! Unterstützt durch die Deutsche Forschungsgemeinschaft SFB 146. 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4706-0370 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 370-380 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


Untersuchungen zur Kurzzeitperiodik bei Cavia aperea f. porcellus 371 


bemerkenswert, da andere Nagerspezies unter standardisierten Laborbedingungen eine 
ausgeprägte Circadianperiodik aufweisen (AscHorr 1962). 

Ziel dieser Arbeit war es daher, die Bewegungsaktivität des Hausmeerschweinchens 
kontinuierlich zu erfassen und nach verschiedenen Verfahren zu analysieren. Die Unter- 
suchungen beschränken sich auf einzeln gehaltene männliche Tiere, um soziale Inter- 
aktionen und Einflüsse durch den Geschlechtszyklus auszuschließen. 


Material und Methode 


Für die Untersuchungen standen insgesamt 12 männliche Meerschweinchen (Cavia aperea f. porcel- 
lus) im Alter von 2-5 Monaten aus der hauseigenen Kolonie zur Verfügung. Die Tiere waren vor 
Beginn der Messungen in Gruppenhaltung (Haltungsboxen 50 X 75 X 20 cm) unter standardisierten 
Laborbedingungen (Raumtemperatur 20 # 1 °C, 55 #5 % relative Luftfeuchtigkeit, Futter (Herilan- 
MS-Haltung) und Wasser aus Trinkflaschen ad lib., Weichholzgranulat und Heu als Einstreu, Licht- 
Dunkelwechsel [12:12; 6.00-18.00]) aufgewachsen. 

Für die Messungen wurden die Tiere in eine Klimakammer von 2 X 2 m Grundfläche mit gleichen 
klimatischen Verhältnissen überführt und einzeln in Makrolonkäfigen (Typ IV, 34 x 55 x 20 cm) 
gehalten. Die Beleuchtung erfolgte bei Haltung im Licht-Dunkelwechsel und Dauerlicht durch eine 
65-W-Leuchtstoffröhre, bei Haltung im Dauerdunkel durch eine 60-W-Rotlichtlampe. 

Die Messungen begannen erst nach ausreichender Adaptation der Tiere (3-4 Wochen) an die neue 
Umgebung. Die Bewegungsaktivität wurde registriert. Da immer nur 4 Tiere zur gleichen Zeit erfaßt 
werden konnten, erfolgten die Messungen nacheinander an 3 Gruppen mit jeweils 4 Tieren. 

Zwei Gruppen setzten sich aus Tieren im Alter von 4-5 Monaten zusammen. Die dritte Gruppe 
enthielt etwas jüngere Tiere im Alter von 2 Monaten. Da sich zwischen den beiden Altersgruppen 
keine Unterschiede ın der Bewegungsaktivität feststellen ließen, werden sie in der weiteren Beschrei- 
bung zusammengefaßt. 

Alle drei 4er-Gruppen wurde bei Haltung unter Licht-Dunkelwechsel (LD 12:12) und unter 
Dauerdunkel gemessen, eine der Gruppen zusätzlich unter Dauerlicht. Die Meßperioden umfaßten 
jeweils 5 Tage. Wegen eines technischen Ausfalls konnte eine Gruppe unter Licht-Dunkelwechsel nur 
4 Tage gemessen werden. Die Messungen erfolgten nicht unmittelbar nach dem Beleuchtungswechsel, 
sondern erst nach einer Adaptationszeit von 10-12 Tagen. 

Die Bewegungsaktivität wurde durch einen kapazitiven Bewegungsaufnehmer erfaßt, der sich 
unter dem Kunststoffkäfig befindet und ein der Bewegung proportionales Gleichspannungssignal 
liefert. Dieses Signal wird über einen A/D-Wandler von einem Kleinrechner ın 10-Intervallen 
abgefragt, zu 5-min-Mittelwerten verrechnet und abgespeichert. Parallel dazu wurde die Bewegungs- 
aktivität auch über einen Papierschreiber aufgezeichnet, um eventuelle Störungen schon während der 
Meßperiode bemerken zu können. 

Die verwendeten kapazitiven Bewegungsaufnehmer sind zwar eichbar, jedoch hängt das Gleich- 
spannungssignal auch stark vom Körpergewicht der Tiere ab. Um die Tiere dennoch direkt miteinan- 
der vergleichen zu können, wurden die Meßwerte jedes Tieres vor der weiteren Auswertung bezüglich 
seiner mittleren Aktivität über die gesamte Meßperiode (100%-Wert) normiert. 


Ergebnisse 
Tagesprofil der Bewegungsaktivität unter Licht-Dunkelwechsel 


Abb. 1 zeigt den tageszeitlichen Verlauf der Bewegungsaktivität beim Meerschweinchen 
unter Licht-Dunkelwechsel (LD 12:12). Für das Tagesprofil wurden aus den abgespeicher- 
ten 5-min-Werten jeweils 30-min-Mittelwerte berechnet und die zeitgleichen Werte von 12 
Tieren und 5 bzw. 4 Meßtagen gemittelt. Die senkrechten Linien geben die Standardabwei- 
chung zwischen den Tieren an. Der 100%-Wert der Skala entspricht der mittleren 
Aktivität über die gesamte Meßperiode. 

Während der Dunkelphase ist allgemein eine höhere Aktivität zu beobachten, die mit 
einem auffälligen Aktivitätsgipfel beim Ausschalten des Lichtes einsetzt. Ein zweiter, 
weniger ausgeprägter Gipfel ist etwa 3-4 h vor dem Einschalten des Lichtes zu erkennen. 
Nach Einschalten des Lichtes ist die Aktivität am geringsten, erst zum Abend hin nimmt 
sie langsam wieder zu. Im Mittel entfallen bei den 12 Tieren 63% der Gesamtaktivität auf 
die Dunkelphase und 36% auf die Lichtphase (Standardabweichung s = # 8%). 


BAR D. Büttner und Franziska Wollnik 


500 BA (%) 
400 
300 


200 


0 EEE Tu 
18 6 18 


Abb. 1. Tageszeitlicher Verlauf der Bewegungsaktivität beim Meerschweinchen. Jeder 30-min-Wert 

stellt das Mittel aus 4 bzw. 5 Meßtagen mit 12 Tieren dar. Die senkrechten Linien geben die 

Standardabweichung zwischen den Tieren an. Der 100%-Wert der Skala entspricht der mittleren 
Aktivität über die gesamte Meßperiode 


BA (%) DEE | 
100 
Tag I 
200 
0 u a ‚ll re Mu 
Tag 2 | | 
Bere: breih..be 
en | |; 
0 Ian is 11 ; . Mi 
400 | 
Tag 4 En | | : | 4: 
oJ iM. Arm " ei Ha FRA UBR u HL EE 


18 6 18 


Abb. 2. Bewegungsaktivität eines Tieres über 4 Tage bei Licht-Dunkelverhältnissen. Das Histogramm 

zeigt 5-min-Mittelwerte; der 100%-Wert der Skala entspricht der mittleren Aktivität des Tieres 

innerhalb der Meßperiode. Die Dunkelphasen sind durch waagerechte schwarze Balken gekenn- 
zeichnet 


Untersuchungen zur Kurzzeitperiodik bei Cavia aperea f. porcellus 373 


Einzelregistrierungen der Bewegungsaktivität unter verschiedenen Lichtverhältnissen 


Bei Verwendung feiner Zeitraster wird deutlich, mit welcher Regelmäßigkeit sich beim 
Meerschweinchen Perioden der Ruhe und der Aktivität kurzfristig abwechseln. An den 
Einzelregistrierungen eines Tieres unter den verschiedenen Lichtverhältnissen lassen sich 
auffällige Unterschiede im Muster der Bewegungsaktivität beobachten. 


Licht-Dunkelwechsel 


Abb. 2 zeigt die Einzelregistrierungen eines Tieres über 4 Tage mit einem Licht-Dunkel- 
wechsel von 12:12 Stunden, wobei jeder Histogrammbalken einen 5-min-Mittelwert 
darstellt. Der 100%-Wert entspricht der mittleren Aktivität des Tieres über die jeweilige 
Meßperiode. Auch hier ist die allgemeine Steigerung der Aktivität während der Dunkel- 
phase zu bemerken. Einzelne Aktivitätsschübe, die durch kurze Ruhephasen voneinander 


BA (%) DZ) 
Tag 1 . | | 
| la 


lage2 


Mi a } 


LEITETE TI TTTT Y LLILTTITTIT TEL III Y LELIITTLTEIITTTTTTI IT TTT r DILIITIITT TI TI ITT I ana le I 727 17-7 
| 


a ® | | 
i | In il - | ü ku . L 


Te pn LI. 150007, /7. 7777 gi LT [77 77 ei: nn) 7 Il 77 7 7 71 Einen Ten merIemer lose: lee 


u " mi . Ä Li a | ih. L | . | A Lid. 
BET EITTTT TERN EMO 


Abb. 3. Bewegungsaktivität des Tieres von Abb. 2 über 5 Tage bei Dauerdunkel. Das Histogramm 

zeigt 5-min-Mittelwerte; der 100%-Wert der Skala entspricht der mittleren Aktivität des Tieres 

innerhalb der Meßperiode. Die Dunkelphasen des ursprünglichen Lichtwechsels sind durch waage- 
rechte schraffierte Balken gekennzeichnet 


374 D. Büttner und Franziska Wollnik 


getrennt sind, folgen im Abstand von etwa 1-2 h aufeinander. Der dritte Tag zeigt diese 
Aktivitätsschübe besonders deutlich. Während der Lichtphase ist die Bewegungsaktivität 
stark gedämpft und steigt erst zur Dunkelphase hin wieder an. Kurz vor den Lichtwechseln 
fällt eine Verdichtung der Aktivität auf. 


Dauerdunkel 


Bereits nach zweiwöchiger Adaptation an die Haltung unter Dauerdunkel (D/D) ergeben 
sich wesentliche Änderungen im Aktivitätsmuster. Abb. 3 zeigt eine Einzelregistrierung 
über 5 Tage vom selben Tier wie in Abb. 2. Die einzelnen Aktivitätsschübe treten 
regelmäßiger auf und sind von den Phasen, in denen nur Kleinaktivitäten registriert 
werden, deutlich abgesetzt. Der Abstand zwischen den Aktivitätsphasen beträgt 1-2 h. 
Lediglich in den ersten drei Tagen fallen in der Mitte der früheren Licht- bzw. Dunkel- 
phase einzelne Aktivitätsschübe aus. Die frühere 24-h-Periode ist weitgehend aufgelöst. Im 
3. bis 5. Tag ist eine weitere schwache, rhythmische Komponente erkennbar. Sie kommt in 


BA (%) BL 
4001. 
Tag I > 
DER erbeten Mr eiihh Hl. ul hl il. in la L 
Tag 2 
0 Min ul ART ET eM 
Tag 3 
Tag 4 
400 
Tag 5 
200 . 
" Nm Mi ll ln vn i 


TB BG BE OR ER Di DR OR LI TI TIELTIITTTIIT IT TITTIT y PILLITLTITTTITTTTTTTI 1} KULT IL EIT Mi 


= 18 


Abb. 4. Bewegungsaktivität des Tieres von Abb. 2 und 3 über 5 Tage bei Dauerlicht. Das Histogramm 

zeigt 5-min-Mittelwerte; der 100%-Wert der Skala entspricht der mittleren Aktivität des Tieres 

innerhalb der Meßperiode. Die Dunkelphasen des ursprünglichen Lichtwechsels sind durch waage- 
rechte schraffierte Balken gekennzeichnet 


Untersuchungen zur Kurzzeitperiodik bei Cavia aperea f. porcellus 375 


der Verstärkung eines Aktivitätsschubes und gleichzeitigen Abschwächung des nachfol- 
genden Schubes im Abstand von 8 h zum Ausdruck. 


Dauerlicht 


Bei Haltung im Dauerlicht (L/L) bleiben die Aktivitätsschübe ähnlich regelmäßig wie im 
Dauerdunkel. In Abb. 4 sind wiederum die Einzelregistrierungen desselben Tieres wie in 
Abb. 2 und 3 wiedergegeben. Die einzelnen Aktivitätsphasen sind etwas breiter und nicht 
so deutlich voneinander abgesetzt wie bei der Haltung im Dauerdunkel. Zwischen den 
Aktivitätsschüben liegen häufig kleinere Aktivitäten. Eine Circadianperiodik ist nicht 
erkennbar. 


Gesamtruhezeit über 24 h 


Als Ruhephase wird ein 5-min-Mittelwert definiert, der unter 20 % des Gesamtmittels der 
Bewegungsaktivität liegt. Dies entspricht einer Einheit in den unter Abb. 2, 3, 4 gezeigten 
Einzelregistrierungen. Bei der Festlegung der Ruheschwelle auf den 20%-Wert werden 
einzelne Mikroaktivitäten noch nicht erfaßt, kurz nacheinanderfolgende Körperbewegun- 
gen ohne Ortsveränderung überschreiten jedoch im allgemeinen diesen Wert. 

Um das Verhältnis Ruhezeit zu Aktivitätszeit unter den verschiedenen Lichtbedingun- 
gen zu vergleichen, wurde von allen Tieren die mittlere Dauer der Gesamtruhezeit über 
24 h bestimmt. Zusätzlich wurde bei Haltung unter Lichtwechsel die mittlere Ruhezeit 
getrennt für die Licht- und Dunkelphase berechnet bzw. unter konstanten Lichtverhältnis- 
sen getrennt für die entsprechenden Tageshälften. Die Ergebnisse sind in der Tabelle 
zusammengefaßt. Die Standardabweichungen beziehen sich auf die Streuungen zwischen 
den Tieren. 

Unter allen Lichtverhältnissen sind die Ruhezeiten stets kürzer als die Aktivitätszeiten. 
Die längste Gesamtruhezeit innerhalb von 24 h ist unter Licht-Dunkelwechsel (6 h 11 min) 
zu finden. Die getrennte Berechnung der Ruhezeiten für die beiden Tageshälften zeigt, daß 
dies ın erster Linie auf die besonders lange Ruhezeit während der Lichtphase zurückzufüh- 
ren ist. Diese ist mit 4 h 06 min fast doppelt so lang wie die Ruhezeit in der Dunkelphase 


Tabelle 


Mittlere Gesamtruhezeit aller Tiere innerhalb von 12 bzw. 24 Stunden 


Bei Licht-Dunkelwechsel getrennt berechnet für Licht- (L) und Dunkelphase (D), bei konstanter 
Beleuchtung getrennt für die entsprechenden Tageshälften. Die Standardabweichungen beziehen sich 
auf die Streuungen zwischen den Tieren 


24 Stunden 12 Stunden 
18.00-6.00 (D) 6.00-18.00 (L) 


LD | 6h 11 min (# 1h27 mın) 2 h05 min (+ 39 min) —:#++—4 h 06 min (# 1h 14 min) 


ananen 
ArAr 


DD| 4h45 min (+ 1h42 min) ° 2h41 min (+ 58 min) —ns— 2 h 03 min (# 53 min) 


LL | 2h58 min (# 1h50 = 1 h23 min (+ 21 min) —ns— 1 h 35 min (# 36 min) 


376 


D. Büttner und Franziska Wollnik 


mit 2 h 05 min. Ein T-Test ergibt hochsignifikante Unterschiede (p < 0.001) zwischen den 
Ruhezeiten während der Tag- und denen während der Nachthälfte. Unter Dauerdunkel 
bestehen keine signifikanten Unterschiede zwischen den Ruhezeiten in der ursprünglichen 
Nachthälfte (2 h 41 min) und in der ursprünglichen Taghälfte (2 h 03 min). Die Werte 
beider 12-h-Abschnitte unterscheiden sich auch nicht signifikant von den Werten aus der 
Nachthälfte unter Licht-Dunkelwechsel. 
Bei Haltung im Dauerlicht sind die Ruhezeiten mit 1 h 23 min (Nacht) und 1 h 35 min 
(Tag) etwas kürzer als im Dauerdunkel. Diese Ruhezeiten entsprechen in der Größenord- 
nung eher den Werten der Nachthälfte als denen der Taghälfte unter Licht-Dunkelwechsel. 


N 


6 


[0 1] 


nm 


m 


| 
5 
[A 
3 
2 
1 


1800 - 600 


5 15225 
ion 225 
5 1502.25 


kn 


D 

35 
D 

ss ns 
L 


I 
ii 


Ruhezeit (min) 


N 


6 


S 


nn 


w 


_ 


[e)] 


(0%) 


”m 


6.00 - 1800 


25 


Sorm 4a, 55 


35r La 399 


Ruhezeit (mın) 


Für die mittleren Ge- 
samtruhezeiten über 24 h 
ergeben sich jedoch be- 
trächtliche Unterschiede 
zwischen der Haltung un- 
ter Licht-Dunkelwechsel 
(6 h 11 min), Dauerdunkel 
(4 h 45 min) oder Dauer- 
licht (2 h 58 min). 


Durchschnittliche Länge 
dereinzelnenRuhephasen 


Aus den Registrierungen 
der 12 Tiere unter den ver- 
schiedenen Lichtverhält- 
nissen wurden Häufigkeits- 
verteilungen für die Länge 
der einzelnen Ruhezeiten je 
Tier berechnet. Abb. 5 zeigt 
diese Verteilungen getrennt 
für die Licht- und Dunkel- 
phase bzw. bei Haltungun- 
ter Dauerlicht oder Dauer- 
dunkel getrennt für die ur- 
sprünglichen Tageshältften. 
Für die Dauer einer einzel- 
nen Ruhephase wurde die 
kleinstmögliche Klassen- 
breite von 5 mın gewählt. 
Die Angaben sind Durch- 
schnittswerte aus allen 12- 


h-Abschnitten aller Tiere. 


Abb. 5. Häufigkeitsverteilun- 
gen der Länge einzelner Ru- 
hephasen, gemittelt über alle 
12-h-Abschnitte aller Tiere. 
Die Verteilungen wurden ge- 
trennt für die Licht- (L) und 
Dunkelphase (D) bzw. bei 
Haltung unter Dauerdunkel 
(DD) oder Dauerlicht (LL) 
getrennt für die ursprüngli- 
chen Tageshälften berechnet 


Untersuchungen zur Kurzzeitperiodik bei Cavia aperea f. porcellus 377 


Am häufigsten sind unter allen Lichtverhältnissen Ruhephasen von nur 5 min Dauer. 
Ununterbrochene Ruhephasen von mehr als 20 mın Dauer sind selten, sie treten höchstens 
zwei- bis dreimal innerhalb von 12 h auf. Die Unterschiede zwischen den Häufigkeiten bei 
den verschiedenen Lichtverhältnissen sind gering. Sie sind am deutlichsten noch zwischen 
der Licht- und Dunkelphase unter Licht-Dunkelwechsel vorhanden. Hier bewirkt das 
Licht längere Ruhephasen, was bereits an der Gesamtaktivität festzustellen war. Ebenso 
bestätigt sich die allgemeine Verkürzung der Ruhephasen unter Dauerlicht in der betref- 
fenden Häufigkeitsverteilung. 


Frequenzanalyse 


Zur genaueren Bestimmung der Periodizität der Aktivitätsmuster eignen sich verschiedene 
Verfahren, von denen das von HALBERG und PanorskYy (1961; PAnorskY und HALBERG 
1961) ausgewählt wurde, weil es durch die frequenzlineare Darstellung eine hohe Auflö- 
sung im Bereich kurzer Periodenlängen ermöglicht, aber gleichzeitig die 24-h-Periodik mit 
erfaßt. Dabei muß die schlechte Auflösung längerer Perioden in Kauf genommen werden. 
Ferner bietet dieses Verfahren die Möglichkeit, einen Mittelwert für alle Tiere und einen 
statistischen Vertrauensbereich zu berechnen. 

Abb. 6 zeigt die so gewonnenen Frequenzspektren für alle 12 Tiere unter Licht- 
Dunkelwechsel und Dauerdunkel und für 4 Tiere unter Dauerlicht. Zu jedem Spektral- 
schätzwert ist der 95%-Vertrauensbereich angegeben. 


400 LD 
(12 Tiere x 5 Tage ) 


DD 
(12 Tiere x 5 Tage ) 


10) nn nn 2 nn Do m ae Sa era ee e 


300 LL 
(4 Tiere x 5 Tage) 


Bam B Be 2 15 1 0,8 
Periodenlänge (h) 


Abb. 6. Frequenzspektren der Bewegungsaktivität, berechnet aus den Meßwerten von allen 12 Tieren 
unter Licht-Dunkelwechsel (LD) und Dauerdunkel (DD) und von 4 Tieren unter Dauerlicht (LL). Zu 
jedem Spektralschätzwert ist der 95%-Vertrauensbereich als senkrechter Balken angegeben 


378 D. Büttner und Franziska Wollnik 


Unter Licht-Dunkelwechsel dominiert die 24-h-Periodik. Daneben ist noch eine signifi- 
kante 8-h-Periodik sichtbar, im Bereich von 1-2 h sind periodische Anteile nur schwach 
angedeutet. 

Unter Dauerdunkel wird die 24-h-Periodik erheblich reduziert. Ihr Anteil im Fre- 
quenzspektrum entspricht jetzt dem der 8-h-Periodik. Im Bereich von 1-2 h Periodenlänge 
liegt ein breites Band signifikanter, ultradianer Rhythmen. Die Breite dieses Bandes ist auf 
die große Variabilität dieser Rhythmen zwischen den einzelnen Tieren zurückzuführen. 

Frequenzanalysen einzelner Tiere zeigen, daß bei einem Tier z. B. ultradiane Rhythmen 
von 1,6 h, bei einem anderen Tier von 1,2 h Periodenlänge vorherrschen oder zwei 
verschiedene Periodenlängen in diesem Frequenzbereich gemeinsam auftreten. Auf die 
Darstellung dieser individuellen Unterschiede wird hier verzichtet. 

Bei Haltung unter Dauerlicht werden die 24-h-Periodik und die 8-h-Periodik noch 
weiter reduziert. Ihre Anteile am Frequenzspektrum sind hier sogar kleiner als der Anteil 
der 1- bis 2-h-Periodik. Die einzelnen Spektralschätzwerte sind jedoch mit einer größeren 
statistischen Unsicherheit behaftet, da für die Erstellung des Frequenzspektrums unter 
Dauerlicht nur die Meßwerte von 4 Tieren zur Verfügung standen. 


Diskussion 


AscHoFF (1957) beschreibt den Bigeminus, d. h. das Auftreten von zwei Aktivitätsgipfeln 
im Tagesverlauf als ein typisches Muster der Tagesperiodik. Dabei kann der erste Gipfel zu 
Beginn der Aktivitätsphase größer und regelmäßiger auftreten als der zweite. 

Die größere Aktivität während der Nachtstunden (63 %) sowie der scharfe Aktivitäts- 
gipfel zum Wechsel von der Licht- zur Dunkelphase sprechen dafür, Cavia aperea f. 
porcellus den dunkel- bzw. dämmerungsaktiven Tieren zuzuordnen. Hinweise für die 
Dämmerungsaktivität verschiedener Meerschweinchenspezies unter natürlichen Lichtver- 
hältnissen finden sich auch bei Roop (1972) und Fuchs (1980), während PEARson (1951) 
Cavia musteloides in seinem natürlichen Habitat als tagaktıv beschreibt. 

Die weitgehende Übereinstimmung in der Aktivitätsverteilung zwischen Meerschwein- 
chenkolonien unter Freilandhaltung (Fuchs 1980) und einzeln gehaltenen Tieren läßt 
vermuten, daß soziale Interaktionen die Circadianperiodik nicht wesentlich beeinflussen 
und somit Beobachtungen an Einzeltieren ein durchaus repräsentatives Bild der Aktivitäts- 
verteilung ergeben. 

Fuchs (1980) beschreibt jeweils zum Lichtwechsel zwei annähernd gleich große Aktivi- 
tätsgipfel. Allerdings war bei ihm die natürliche Nachtzeit nur 8 h lang und somit kürzer 
als unsere Dunkelphase unter Laborbedingungen. Es bleibt offen, ob der an unseren Tieren 
einige Stunden vor dem Einschalten des Lichtes zu beobachtende abgeflachte, zweite 
Gipfel unter einer verkürzten Dunkelphase verstärkt werden kann, oder ob diese Abwei- 
chung in den Ergebnissen von Fuchs auf methodische Unterschiede in der Bewertung der 
Aktivität zurückzuführen ist. 

Wegen der häufig und kurzfristig auftretenden motorischen Aktivität ohne Ortsverän- 
derung ergeben sich beim Meerschweinchen besondere Schwierigkeiten bei der Ermittlung 
von Ruhe- oder Aktivitätszeiten. Die Berechnung der Ruhe- und Aktivitätszeiten hängt im 
starken Maße von ihrer Definition und den Möglichkeiten ihrer quantitativen Erfassung 
ab. 

Nach unserer Definition der Schwelle zwischen Aktivität und Ruhe wurde bei Haltung 
unter Licht-Dunkelwechsel eine Gesamtruhezeit von 6 h 11 min über 24 h ermittelt, davon 
entfallen 4 h 06 min Ruhe auf die Lichtphase und nur 2 h 05 min auf die Dunkelphase. Dies 
entspricht den von PELLET und B£RAUD (1967) elektrophysiologisch ermittelten Ruhezei- 
ten. Unter Dauerdunkel wird die Gesamtruhezeit durch das Auffüllen der ursprünglichen 
Lichtphase mit Aktivitätsschüben auf 4 h 45 min reduziert. Die Ruhezeiten unter Dauer- 


Untersuchungen zur Kurzzeitperiodik bei Cavia aperea f. porcellus 379 


licht sind ebenfalls nicht mit der auffallend langen Ruhezeit in der früheren Lichtphase 
vergleichbar. Unter Dauerlicht wird die Gesamtruhezeit innerhalb von 24 h sogar noch 
weiter auf 2 h 58 min reduziert, was in erster Linie durch eine Verbreiterung der 
Aktivitätsschübe bedingt ist. Die Gesamtruhezeit setzt sich aus auffallend kurzen Phasen 
absoluter Ruhe zusammen. Ruhephasen von mehr als 15 min Länge sind nur selten. 

Übereinstimmend mit unseren Ergebnissen beschreibt NıcHoLıs (1926) das Meer- 
schweinchen als nahezu ständig aktiv. Bei Haltung unter Dauerlicht oder Dauerdunkel 
konnte sie keine circadiane Rhythmik feststellen, unterscheidet aber zwischen zwei 
verschiedenen Formen der Aktivität: einer kontinuierlichen Aktivität ohne Ruhephasen 
und einer „intermittent activity‘ mit Ruheperioden von wenigen Sekunden bis zu 10 min. 
Hierin ist eine Parallele zu sehen zu den von uns registrierten Aktivitätsschüben und den 
dazwischenliegenden Ruhephasen, die von nur kurzen Aktivitäten unterbrochen sind. 

Eine genaue Aussage über die Periodizität dieser Aktivitätsmuster ist nur über eine 
Frequenzanalyse möglich. Diese ergibt, daß das Aktivitätsmuster von Cavia aperea f. 
porcellus im wesentlichen durch eine Kurzzeitperiodik von 1-2 h Periodenlänge bestimmt 
‘wird. Diese Kurzzeitperiodik wurde für das Meerschweinchen noch nicht explizit 
beschrieben, tritt aber mit unterschiedlichen Periodenlängen bei vielen Spezies auf 
(AscHorrF 1962). 

Auch für das Meerschweinchen scheint ein enger Zusammenhang zwischen Futterauf- 
nahme und Kurzzeitperiodik zu bestehen. Aus Beobachtungen von NIcHoLtLs (1926) geht 
hervor, daß es - wie viele andere Säuger — nach vorangegangener Bewegung zur Futterauf- 
nahme ruht. Somit könnte beim Meerschweinchen in der Kurzzeitperiodik der Bewe- 
gungsaktivität auch eine ungewöhnliche kurze Periodik des Energieverbrauchs und der 
Energieaufnahme zum Ausdruck kommen. 

Unter Licht-Dunkelwechsel ist die Kurzzeitperiodik mit Periodenlängen um 1,5 h ın 
der Frequenzanalyse nur andeutungsweise vorhanden, obwohl sie in den Einzelaufzeich- 
nungen durchaus sichtbar ist. Sie läßt sich aber bereits nach etwa 14 Tagen unter 
Dauerdunkel eindeutig nachweisen. 

Hierbei ist zu beachten, daß die Variabilität zwischen den Tieren in der Zusammenfas- 
sung zu einem breiten Band signifikaner Periodenlängen führt, während Einzelanalysen 
der Tiere schärfer abgesetzte Periodenlängen ergeben. 

Die Bedeutung der 8-h-Komponente ist bislang unklar. Sie tritt in den Einzelaufzeich- 
nungen als verstärkter Aktivitätsgipfel im Abstand von 8 h mit anschließender Reduktion 
des nachfolgenden Aktivitätsschubes auf. Die Frequenzanalysen zeigen unter allen Licht- 
bedingungen eine Stabilität der 8-h-Rhythmik, die eine biologische Bedeutung vermuten 
läßt. Dafür spricht auch die Tatsache, daß der Abstand zwischen den beiden Aktivitätsgip- 
feln unter Licht-Dunkelwechsel etwa 8 h beträgt. Gegen eine Interpretation als Oberwelle 
der 24-h-Periodik spricht, daß weitere harmonische Oberwellen unterhalb von 8 h fehlen. 
Lediglich unter Dauerlicht heben sich ultradiane Rhythmen von 4 und 2 h vom Grundrau- 
schen ab. Der direkte Vergleich mit den anderen Lichtbedingungen ist allerdings proble- 
matisch, da hier nur die Ergebnisse von 4 Tieren und dadurch größerere Vertrauensberei- 
che der Spektralschätzwerte vorliegen 

Als normale Reaktion auf das Fehlen des Lichtes als ‚„‚Zeitgeber“ ist bei dunkelaktiven 
Tieren unter Dauerdunkel eine Verkürzung und unter Dauerlicht eine Verlängerung der 
Circadianperiodik zu erwarten (AscHorFr 1958). Die hier gezeigten Einzelregistrierungen 
der Bewegungsaktivität lassen jedoch nach Wegfall des Lichtwechsels kein Driften der 
Circadianperiodik erkennen, sondern zeigen eher ein Auffüllen der früheren Lichtphase 
mit periodischen Aktivitätsschüben. 

Die Reduktion der Circadianperiodik unter Dauerdunkel oder Dauerlicht wird auch in 
der Frequenzanalyse sichtbar. Homna und HırosHIGE (1978) beschreiben ähnliche Vor- 
gänge bei der Ratte. Allerdings benötigen diese Tiere etwa 3 Monate bis zur Auflösung der 
Circadianperiodik. 


380 D. Büttner und Franziska Wollnik 


Bei unseren Untersuchungen kann nach der kurzen Adaptation an Dauerlicht oder 
Dauerdunkel noch nicht von einer Stabilisierung des Bewegungsmusters ausgegangen 
werden. Es ist daher zu vermuten, daß die Circadianperiodik nach längerer Haltung ohne 
Lichtwechsel noch weiter reduziert oder vollständig aufgelöst wird, wie es auch von 
NiıcHouts (1926) beschrieben wird. Damit wird die Frage aufgeworfen, ob die Circadian- 
periodik beim Meerschweinchen überhaupt einen endogenen Charakter besitzt oder ob sie 
durch die vorherigen Haltungsbedingungen nur aufgeprägt wurde. Die Verstärkung der 
ultradianen Rhythmen bei konstanten Lichtverhältnissen spricht dafür, die Kurzzeitperio- 
dik als den wesentlichen Bestandteil der endogenen Periodik anzusehen. 


Zusammenfassung 


Die Bewegungsaktivität männlicher Meerschweinchen (Cavia aperea f. porcellus) wurde unter 
verschiedenen Lichtverhältnissen registriert. 

Unter Licht-Dunkelwechsel (LD 12:12) ıst das Meerschweinchen nachtaktiv. Besonders während 
der Dunkelphase ist das Aktivitätsmuster durch einen ständigen Wechsel von kurzen Aktivitäts- und 
Ruhephasen bestimmt. Das Hauptaktivitätsmaximum liegt zu Beginn der Dunkelphase, ein weiteres 
Maximum folgt nach 8 Stunden. 

Die Gesamtruhezeit innerhalb von 24 h beträgt durchschnittlich 6 h 11 mın (+ 1 h 27 min), wovon 
4 h 06 min (+ 1 h 14 min) Ruhe auf die Lichtphase entfallen. 

Bei Haltung unter Dauerdunkel (DD) oder Dauerlicht (LL) wird die Circadianperiodik reduziert, 
während sich die kurzen Aktivitätsschübe über die gesamten 24 h verteilen. Die Gesamtruhezeit 
verkürzt sich auf 4 h 45 mın (+ 1 h 42 min) (DD) bzw. 2 h 58 min (+ 1 h 50 min) (LL), wobei nun auf 
die beiden ursprünglichen Tageshälften etwa gleiche Ruhezeiten entfallen. 

Frequenzanalysen bestätigten die Reduktion der Circadianperiodik und die Manifestierung der 1- 
bis 2-h-Periodik unter Dauerdunkel oder Dauerlicht. Eine weitere Spektralkomponente, die 8-h- 
Periodik, bleibt unter den verschiedenen Lichtverhältnissen nahezu konstant. 


Literatur 


ASCHOFF, J. (1957): Aktivitätsmuster der Tagesperiodik. Naturwissenschaften 13, 361-367. 

— (1958): Tierische Periodik unter dem Einfluß von Zeitgebern. Z. Tierpsychol. 15, 1-30. 

— (1962): Spontane okomotorische Aktivität. In: Handbuch der Zoologie. Ed. G. HELMCKE und H. 
v. LENGERKEN, Berlin. Bd. 8, Lfg. 30, Teil 11, 1-76. 

Fuchs, $. (1980): Spacing patterns in a colony of guinea pigs: Predictability from environmental and 
social factors. Behav. Ecol. Sociobiol. 6, 265-276. 

HALBERG, F.; Panorsky, H. (1961): Thermo-variance spectra; Method and clinical ıllustrations. Exp. 
Med. Surg. 19, 285-321. 

Honma, K.; HırosHIGE, T. (1978): Endogenous ultradian rhythms in rats exposed to prolonged 
continuous light. Am. J. Physiol. 235, R250-R256. 

Kıng, J. A. (1956): Social relations of the domestic guinea pig living under semi-natural conditions. 
Ecology 37, 221-228. 

NicHouts, E. E. (1926): A study of the spontaneous activity of the guinea pig. Comparative 
Psychology 2, 303-330. 

Panorsky, H.; HALBERG, F. (1961): Thermo-variance spectra; Simplified computational example and 
other methodology. Exp. Med. Surg. 19, 323-338. 

PEARson, ©. P. (1951): Mammals in the highlands of southern Peru. Bull. Mus. Comp. Zool. 106, 
117-174. 

PELLET, J.; BERAUD, G. (1967): Organisation nycth&m£rale de la veille et du sommeil chez le cobaye 
(Cavia procellus); Comparaisons intersp&cifiques avec le rat et le chat. Physiol Behav. 2, 131-137. 

Roop, J. P. (1972): Ecological and behavioural comparisons of three genera of argentine cavies. 
Anımal. Behav. Monogr. 5, 1-83. 


Anschrift der Verfasser: Dr. rer. nat. DiETMAR BÜTTNER und Dipl.-Biol. FRANZISKA WOLLNIK, 
Institut für Versuchstierkunde und Zentrales Tierlaboratorium, Medizini- 
sche Hochschule Hannover, Konstanty-Gutschow-Str. 8, D-3000 Hanno- 
ver 61 


WISSENSCHAFTLICHE KURZMITTEILUNG 


The type locality of the Javan rhinoceros (Rhinoceros sondaicus 
Desmarest, 1822) 


By L. C. ROOKMAAKER 


Receipt of Ms. 29. 9. 1982 


There has always been uncertainty about the type locality of Rhinoceros sondaicus. In 1821, 
a hide and skeleton of a young adult single-horned rhinoceros was received in Paris. 
DesmaARrEST (1822: 399-400) described it as “Rhinoceros sondaicus ..... Espece nouvelle, 
decouverte par MM. DiarD et DuvAuceL, envoyee au Museum d’Histoire Naturelle en 
1821””. He first stated that the anımal was found in “Sumatra” (p. 400), but corrected this in 
the supplement to his book, published simultaneously, into “trouve ä Java, et non ä 
Sumatra, comme nous l’avons indique par erreur” (p. 547). 

Sopy (1941: 61; 1946, 1959: 133, 157) doubted the accuracy of this emendation 
primarily because Dıarp and Duvaucer never collected together in Java, and he consi- 
dered Sumatra as the correct type locality of R. sondaicus. Although Sopy’s historical 
arguments have not received any comment, most recent authors give the type locality as 
“probably Java” (e.g. Hoo1jER 1946: 6; Grovss 1967: 233; Honackt et al. 1982: 311; 
ROOKMAAKER in press). It is possible to settle this problem more definitively. 

STRESEMANN (1951: 146) summarized the more important biographical details of 
ALFRED DuvAuceL (1793-1824) and PIERRE-MEDARD DIARD (1794-1863). DuUvAuceL, the 
stepson Of GEORGES CUVIER, had been sent to India as “naturaliste du Roi” in 1817. DiARD 
went to India on his own accord to collect plants and animals. They met near Calcutta in 
May 1818. Shortly after, they were engaged by STAMFORD RArFFLES (1781-1826) and 
travelled with him from India to Sumatra arrıving in 1819. On the way, they collected a 
large number of anımals, some of which they sent to Paris. Because this was a breach of 
their contract, RAFFLES was very annoyed when he learned about it and he sent both 
Frenchmen away. Duvaucer returned to India and probably never visited Java. DIARD 
was emprisoned and taken to Batavia (Jakarta) where he was released. While it is unknown 
if he travelled outside the city, he was able to make a large collection of Javan animals. 
DiaArD left Java in 1821 never to return. 

DESMAREST (1822) was not the only one to describe the rhinoceros received in Paris in 
1821. Cuvier (1822), too, discussed its osteological characteristics. First, he listed a 
“squelette d’un rhinoceros unicorne de cette espece, de Java, adulte, que M. DiArD vient de 
nous envoyer de cette ile”” accompanied by a “Memoire manuscrit de MM. DiArD et 
DuvaAucet, sur ce rhinoceros [the species described by BeELr in 1793] et sur celui de Java” 
(Cuvier 1822: 4). After comparing the skeleton with specimens from Africa and India and 
with the description of the Sumatran rhinoceros by BErı (1793), CuviEr (1822: 20-42) 
accepted its specific distinction without providing a binomial name. This conclusion was 
considerably confirmed by “les observations de deux de mes eleves, MM. DiarD et 
DuvaAuceL, consignees dans un m&moire qu’ils ont presente ä la Societe des Sciences de 
Batavia, et par l’envoi qu’ils nous ont fait d’un squelette adulte, et d’une peau de cette 
espece” (CuvIEr 1822: 26). On the next page, CuvVIEr (1822: 27) again said that “les m&mes 
jeunes naturalistes se sont assur&s que ce rhinoceros particulier jusqu’ä present ä l’ile de 
Java, n’est point une simple varıete du bicorne du Sumatra.” 


U.S. Copyright Clearance Center Code Statement: 0044-3468/82/4706-0381 $ 02.50/0 
Z. Säugetierkunde 47 (1982) 381-382 

© 1982 Verlag Paul Parey, Hamburg und Berlin 

ISSN 0044-3468 / InterCode: ZSAEA 7 


382 L. C. Rookmaaker 


The “Memoire sur le rhinoceros” mentioned above is still in the central library of the 
Museum National d’Histoire Naturelle of Parıs (MNHN, MS. 625 no. 13). There are one 
rough and one fair copy, both 18 pages long, unsigned and undated. DiArD mentioned its 
despatch in a letter to CuvIEr dated “Batavia 9 ayrıl 1821” (MNHN, MS. 625 no. 11). The 
Memoıre essentially contains a description of the external morphology, anatomy, osteol- 
ogy and dentition of the rhinoceros which “parait Etre particulier ä l’ile de Java”. It is 
stated that a copy was sent to the “societe des recherches asiatiques” (CUVIER, above, 
mentioned the Societe des Sciences de Batavia), but it was never published. 

It appears clear that the type specimen of R. sondaicus Desmarest, 1822 was collected in 
Java by DıarD alone. The exact locality is undeterminable because DıArD’s travels or 
transactions in Java are unknown. The note on this species and the Sumatran rhinoceros 
probably was written by DiıAarD although it may have included observations by 
Duvaucer. The type locality of the Javan rhinoceros (Rhinoceros sondaicus Desmarest, 
1822) ıs Java. 


References 


BELr, W. (1793): Description of the double-horned Rhinoceros of Sumatra. Phil. Trans. R. Soc.1793, 
3-6, pls. 2-4. 

CuVIER, G. L. C. F. (1822): Recherches sur les ossemens fossiles, oü l’on retablit les caracteres de 
plusieurs anımaux dont les revolutions du globe ont d£truit les especes. Nouvelle edition, vol. 2, 
part 1. Paris, Amsterdam: G. Dufour et E. D’Ocagne. 

DEsMmAREST, A. G. (1822): Mammalogie ou description des especes de Mammiferes. Seconde partie, 
contenant les ordres des rongeurs, des edentes, des pachydermes, des ruminans et des c£taces. 
Paris: Veuve Agasse. 

Grovss, C. P. (1967): On the rhinoceroses of South-East Asia. Säugetierkundl. Mitt. 15, 221-237. 

Honackı, J. H.; Kınman, K. E.; Koeppı, J. W. (Eds.) (1982): Mammal species of the world: a 
taxonomic and geographic reference. Lawrence: Allen Press and the Association of Systematic 
Collections. 

HOoo1JER, D. A. (1946): Prehistoric and fossil rhinoceroses from the Malay Archipelago and India. 
Zool. Meded. Leiden, 26, 1-138. 

ROOKMAARER, L. C. (in press): Bibliography of the rhinoceros: an analysis of the literature on the 
recent rhinoceroses in culture, history and biology. Rotterdam: Balkema. 

Sopy, H. J. V. (1941): De Javaansche neushoorn Rhinoceros sondaicus historisch en biologisch. 
Buitenzorg: ’s Lands Plantentuin. 

— (1946): [Review of] “Prehistoric and fossil rhinoceroses of the Malay Archipelago and India”, door 
D. A. Hooijer. Diss. Leiden, 1946. Natuurwet. Tijdschr. Ned. Indie 102, 151. 

— (1959): Das javanische Nashorn, Rhinoceros sondaicus, historisch und biologisch. Z. Säugetier- 
kunde 24, 109-240. 

STRESEMANN, E. (1951): Die Entwicklung der Ornithologie von Aristoteles bis zur Gegenwart. Berlin: 
EaWs Peters: 


Author’s address: L. C. ROOKMAAKER, Dokter Guepinlaan 23, 4032 NH Ommeren, Holland 


BEKANNTMACHUNG 


Protokoll über die Mitgliederversammlung der Deutschen Gesellschaft für 
Säugetierkunde e. V. am 28. September 1982 im großen Hörsaal des Zoologi- 


schen Instituts der Universität, Akademiestraße 26, in Salzburg 


Der 1. Vorsitzende, Herr Kunn, eröffnet die Versammlung um 16.45 Uhr. 


I 
2. 


Die Tagesordnung wird angenommen. 

Herr Kunn verliest den Bericht über das abgelaufene Geschäftsjahr 1981: 

Im Berichtsjahr erschien der 46. Band der ‚‚Zeitschrift für Säugetierkunde“ in 6 Heften 
mit zusammen 400 Seiten. Neu aufgenommen wurden seit der letzten Tagung 13 
Mitglieder. Ihren Austritt zum Jahresende 1981 erklärten 2 Mitglieder. Durch den Tod 
verlor die Gesellschaft ihr Ehrenmitglied Prof. Dr. HERMANN POHLE sowie 4 weitere 
Mitglieder: FrıTz BoLLE, Dr. ANDRIENNE PRELL, OTTO SCHULZ, OTTO STRECK. Damit 
hat die Gesellschaft gegenwärtig 720 Mitglieder. 

Auf Einladung von Prof. H. F. MoELıEr (Zoologisches Institut der Universität 
Heidelberg) und von Dr. D. Por£y (Tiergarten Heidelberg) fand die 55. Hauptver- 
sammlung der Gesellschaft vom 7. bis 11. September 1981 in Heidelberg statt. Über 100 
Mitglieder und Gäste versammelten sich zu 38 Vorträgen. Ein gemeinsamer Besuch des 
Heidelberger Tiergartens und eine ganztägige Exkursion zur Deutschen Greifenwarte 
Burg Gutenberg beendeten die Tagung. 

Herr Kunn berichtet, in Vertretung von Herrn EVERTS, über die derzeitige Finanzlage. 
Im Geschäftsjahr 1981 deckten die Mitgliedsbeiträge die Ausgaben nicht. 


. Herr Kunn verliest den Bericht der Kassenprüfer, der Herren DITTRICH und WÄCHT- 


LER, die keinen Anlaß zur Beanstandung gefunden hatten. 


. Die Anträge auf Entlastung des Schatzmeisters und des Vorstandes werden bei drei 


Enthaltungen angenommen. 


. Für das Geschäftsjahr 1982 werden als Kassenprüfer die Herren BECKER, DITTRICH und 


WÄCHTLER gewählt. 


. Herr GEoRrcı berichtet über die Preisentwicklung bei der Zeitschrift für Säugetier- 


kunde. Da die Mitgliedsbeiträge die Kosten für die Zeitschrift nicht mehr decken, 
müssen die Beiträge ab 1983 erhöht werden. Der Vorstand beantragt, den Beitrag für 
Vollmitglieder auf 85,- DM, für Studenten auf 60,- DM und für Mitglieder ohne 
Zeitschriftenbezug auf 10,- DM festzusetzen. Nach längerer Diskussion wird der 
Antrag des Vorstandes bei einer Gegenstimme und 4 Enthaltungen angenommen. 


. Die Einladung von Herrn Nosıs, die nächstjährige Hauptversammlung vom 


25.-29. September 1983 ın Bonn abzuhalten, wird einstimmig angenommen. 

Für 1984 laden die Herren FesterTics, KuHn und VoGEL nach Göttingen ein. Die 

Mitgliederversammlung stimmt diesem Vorschlag vorläufig zu. 

Für 1987 liegt eine feste Einladung von Herrn Krös nach Berlin vor. 

a. Herr NIETHAMMER berichtet über den 3. Internationalen Theriologenkongreß in 
Helsinki. Der nächste Kongreß findet in 3 Jahren in Kanada statt, der übernächste ın 
einem westeuropäischen Land. Herr NIETHAMMER wird gebeten, die Verbindung 
zum Internationalen Theriologenkongreß aufrecht zu erhalten und als Vertreter der 
Gesellschaft aufzutreten. 

b. Herr Novare (Bonn-Bad Godesberg) erstellt eine neue „Rote Liste der gefährdeten 
Tiere Deutschlands“. Es wird um begründete Änderungswünsche gebeten. 


384 Buchbesprechung 


c. Herr Krapp (Bonn) berichtet, daß in Rom eine italienische Gesellschaft für Säuge- 
tierkunde gegründet wurde. 

d. Herr DRESCHER berichtet über die Aufgaben der internationalen Walfangkommis- 
sion. Alle Untersuchungen über Wale sollen dieser Kommission gemeldet werden. 

e. Herr SCHMIDT weist auf das europäische Fledermaus-Symposium im September 1983 
in Bonn hin. 


Prof. Dr. H.-J. KuHnn Prof. Dr. E. KuLzEr Prof. Dr. U. SchMIDT 
1. Vorsitzender Geschäftsführer Schriftführer 
BUCHBESPRECHUNG 


SIEWING, R. (Hrsg.): Evolution. 2. Aufl. Stuttgart, New York: Gustav Fischer Verlag 
1982. UTB 748. 466 S., 142 Abb., DM 26,80. ISBN 3-437-20265-0. 


Das vorliegende Buch ist das Ergebnis einer Ringvorlesung, die an der Universität Erlangen für Hörer 
aller Fakultäten abgehalten wurde. In 25. Kapiteln sind 20 Fachgelehrte bemüht, entsprechend dem 
Konzept des Vorhabens dem naturwissenschaftlich Interessierten die neuesten Ergebnisse auf dem 
Gebiet der Evolution nahezubringen. Ein Blick in das Inhaltsverzeichnis macht deutlich, daß ım 
vorliegenden Falle „Evolution“ nicht im herkömmlichen Sinne nur als „organic evolution“, als 
Evolution der Organismen verstanden werden will, sondern als Entwicklung in einer umfassenderen 
Bedeutung, die die kosmischen Vorgänge ebenso einschließt wie die kulturgeschichtliche und techni- 
sche Entwicklung des Menschen bis hin zur Erfindung des Computers. Gleichwohl herrscht die 
biologische Thematik vor, denn 17 Kapitel sind Problemen gewidmet, die von der Entstehung des 
frühesten Lebens, der Evolution des Stoffwechsels, der Entstehung der eukaryotischen Zelle, der 
Evolution der Tiere und Pflanzen, der Paläontologie und dem Verhalten über die Ökologie, die 
Embryologie, die Zoogeographie, die Selektionstheorie, die Genetik und Populationsgenetik bis hin 
zur Artentstehung, zur Evolution der Hominiden und zur Sonderstellung des Menschen in biologi- 
scher Sicht reichen. 

Das Buch wird eingeleitet mit Ausführungen über die Entwicklung des Kosmos, der Sterne und 
der Elementsynthese. Ein Kapitel befaßt sich dann mit der chemischen Entwicklung, eines mit der 
Entstehung der Erdkruste, ein weiteres mit der Evolution des irdischen Sauerstoffbudgets. Dann 
folgen die oben genannten biologischen Themen. Den Abschluß bilden Erörterungen über die 
Stellung des Menschen im Evolutionsprozeß unter theologischen Aspekten, über die Evolution der 
Kulturen und schließlich noch über die Entwicklung im Bereich der Technik. 

Jedes Kapitel ist in sich abgeschlossen, die Kenntnis des vorausgegangenen also nicht für das 
Verständnis des nachfolgenden notwendig. Daß für das Verständnis mancher Kapitel - besonders für 
die mathematisch-physikalisch ausgerichteten kosmologischen Darstellungen - gewisse Grundkennt- 
nisse vorauszusetzen sind, versteht sich von selbst. 

Obwohl sich das Buch an einen breiten Leserkreis wendet, sollten sich besonders der Biologie- 
student, aber auch der Biologielehrer und andere an biologischen Problemen Interessierte angespro- 
chen fühlen, da in übersichtlicher, wenn auch knapper Form alle Grundprobleme der Biologie 
beleuchtet und die neuesten Forschungsergebnisse klar dargestellt werden. Zahlreiche Abbildungen 
verdeutlichen die textlichen Ausführungen. Den einzelnen Kapiteln ist ein Literaturverzeichnis 
angefügt, so daß eine Vertiefung in die jeweilige Materie möglich ist. Dem raschen Auffinden 
bestimmter Sachverhalte dient ein das Buch beschließendes Stichwortverzeichnis. H. REICHSTEIN, Kiel 


Vergleichende Untersuchungen zur Nahrungswahl 
von Gams- und Rotwild 


unter besonderer Berücksichtigung der mikrobiellen Besiedlung und der Verdauungs- 
vorgänge ım Pansen. Von Ute Drescher-Kaden. Heft 12 der „Fortschritte in der Tier- 
physiologie und Tierernährung‘“; Beihefte zur „Zeitschrift für Tierphysiologie, Tierernäh- 
rung und Futtermittelkunde“. 1981. 108 Seiten mit 29 Abbildungen und 30 Tabellen. 
Kartoniert 54,- DM 


Bei weiterem Rückgang der Viehhaltung im Alpengebiet gilt es zu prüfen, ob ein der trag- 
baren Wilddichte entsprechendes Vorkommen von wildlebenden Wiederkäuern auf auf- 
gegebenen Almen und Ödland zur alternativen Nutzung geeignet wäre. Um hierbei alle 
Ansprüche des in den betreffenden Biotopen lebenden Wildes zu erfassen, sind neben 
Studien über Wohn- und Aktionsbereiche, Aktivitätszyklen und Sozialverhalten speziell 
auch ernährungsphysiologische Untersuchungen erforderlich. 

Diesem Ziel dienen vergleichende, in sieben Gebieten des Alpenraumes an Gams- und 
Rotwild durchgeführte ökophysiologische Untersuchungen sowie Vergleichsmessungen 
am Rotwild eines Gebietes der Voralpenregion. Die in Feldversuchen gewonnenen Ergeb- 
nisse verdeutlichen die Zusammenhänge zwischen Ausbildung des Verdauungstraktes und 
artspezifischer Nahrungswahl unter wechselnden Ernährungsbedingungen, geben Einblick 
in mikrobielle Besiedlung und Fermentationsleistungen des Pansens und erlauben die 
Zuordnung von Gams und Rotwild als ‚‚intermediäre Ernährungstypen“. 

Angesprochen sind Tier- und Ernährungsphysiologen, ferner Ökologen, Wildbiologen, 
Forstwirte, Landschaftspfleger und Gehegetierärzte, die ihre Wildkenntnisse vertiefen 
möchten und Kriterien zur Beurteilung von Biotopkapazitäten suchen. 


Palatability and Flavor Use in Animal Feeds. 
Schmackhaftigkeit des Futters und Flavor-Anwendung 


First International Symposium on Palatability and Flavor Use in Anımal Feeds, 10.-11. 
October 1978 in Zurich. 18 Vorträge, zusammengestellt von Hans Bickel. Heft 11 der 
„Fortschritte in der Tierphysiologie und Tierernährung‘“; Beihefte zur „Zeitschrift für 
Tierphysiologie, Tierernährung und Futtermittelkunde‘“. 1980. 148 Seiten mit 27 Abbil- 
dungen und 77 Tabellen. Texte englisch, deutsch oder französisch, mit je einer englischen 
und deutschen, in einem Fall auch französischen Zusammenfassung. Kartoniert 58,- DM 


Wesentliche Voraussetzung der heutigen, auf hohe Leistung ausgerichteten Tierproduk- 
tion ist eine bedarfsgerechte Fütterung. Welche Möglichkeiten sich Wissenschaft und 
Praxis bieten, um durch Verwendung von Flavor den Geschmack des Futters zu verbes- 
sern, wird in 18 Beiträgen dieses Heftes geprüft. 


Ber PAUL PAREY - HAMBURG UND. BERLIN 


Erscheinungsweise und Bezugspreis: Die Zeitschrift erscheint alle 2 Monate; 6 Hefte bilden einen Band 
mit insgesamt 384 Seiten. Der Abonnementspreis beträgt je Band 232,- DM zuzüglich Porto. Das 
Abonnement verpflichtet zur Abnahme eines ganzen Bandes. Es verlängert sich stillschweigend, 
wenn nicht unmittelbar nach Erhalt des letzten Heftes eines Bandes Abbestellung erfolgt. 
Einzelbezugspreis der Hefte: 43,-— DM. Die Preise verstehen sich im Inland ıncl. Mehrwertsteuer. 
Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, 
Spitalerstraße 12, D-2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesell- 
schaft für Säugetierkunde“ erhalten die Zeitschrift unberechnet im Rahmen des Mitgliedsbeitrages. 


Kleintierchirurgie 


Ein Atlas chirurgischer Techniken 


Von Prof. Dr. Wayne E. Wingfield, Athens/Georgia und Prof. Dr. Clarence 
A. Rawlings, Fort Collins/Colorado. Aus dem Amerikanischen von 
Gabriele König. 1983. 208 Seiten mit 80 Abbildungen in 266 Einzeldarstel- 
lungen und 1 Tabelle. Balacron geb. DM 68,-. 


In diesem topographisch-anatomischen Atlas beschreiben die Autoren, elf 
Chirurgen, mit Unterstützung zweier medizinischer Zeichner und auf der 
Basis ihrer Erfahrungen sowohl in der Universitätsklinik als auch in der 
privaten Kleintierpraxis, die von ihnen ausgewählten, bevorzugten und 
erfolgreichen Operationsmethoden in Wort und Bild. Sie umfassen ortho- 
pädische Eingriffe, die Chirurgie an den Organen, die Neurochirurgie und 
die Ophthalmochirurgie. Weitere Kapitel über allgemeine Chirurgie, 
Anästhesie, Vorbereitung und Nachbehandlung des Patienten, Instru- 
mente und Materialien ergänzen den Atlas zu einer abgerundeten Darstel- 
lung der Chirurgie an Hund und Katze. 


Die naturgetreuen oder schematischen Abbildungen ermöglichen es dem 
Operateur, sich einen exakten Überblick über die operativen Techniken zu 
verschaffen, um auch auf während eines Eingriffs eventuell auftretende 
Komplikationen besser vorbereitet zu sein und gleichzeitig sein weiteres 
Vorgehen seinem Können und den Erfordernissen des Augenblicks ent- 
sprechend zu modifizieren. Die Erklärungen zu den Abbildungen finden 
sich meist auf der jeweils benachbarten Seite. 


Für die Studierenden der Veterinärmedizin, für alle Kleintierpraktiker 
sowie für alle Tierkliniken ist dieser hervorragend ausgestattete Atlas ein 
unentbehrliches, praktisches Arbeits- und Hilfsmittel und für alle veterinär- 
medizinischen Institute und Bibliotheken ein wichtiges Nachschlagewerk. 


PAUL 
PREV Berlin ven Hamburg 


4 18489 


ww) E ad) 2 NV E u 7 u) 5.077 2 u 
EN z Sun? a Ss; 2 x Wan m rosa 2 KW on enosn Sa 
pe SE 107) pen (47) | 
yyygı7 LIBRARI ES „SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31YYyg. 
I ZN z En \ Q z n Be 
: < = < = | G < = y, ul 
= A ZN zZ WÜlm. & I ZQ\ 
ni br. 
m 2 > N 2, +” z' = 2 
> a : > 
rırurion NOILNLILSNI_ NYINOSHLINS 53 IUVYgI1_LIBRARI ES SMITHSONIAN _INSTITUTIN 
wu 7 % 2 5 Lu z EL) l 
=> = x | = 4 /[S 3, | 
& | Q = & = [2 ) N 
= c B \ = e = = (7 I 
= A, 88 2 =) = 3% j.)\ 
au 9 N A — (=) 2a < oO Verno ' | 
an 2 ar zZ N 
Yvuyg ni. IBRARIES „ SMITHSONIAN, INSTITUTION „NOILNLILSNI _NVINOSHLIWS S3l uvuye 
Is 2 urn | 
4 = (ev) = 
und BR? vR 2 = 2, 2 } 
Pe) San | MD. pn) > Pe) % 5 b) 
> 6: 7 ©; = = = > | 
_ 7} = > a = 
m > TÜR m 2 m Q 
w = = 77) ee 
„ NOLLNLILSNI NVINOSHLINS, 53 ıIuvyg ı1_L1 BRAR!I ES „SMITHSONIAN 
Z < WE = N: = < 
nur! z - ze N, — zZ 
SE a oO Bu 29 ., oO N N, = (®) N 
(97) 102) 102) 2] NN 162] (42) 
7 SE } ©: L oO T: 
= E z = = = 
5 = > = Me = 
& => 7) 2 77) ER 7) 
yvyglı_ SMITHSONIAN _ INSTITUTION NOILNLILSNI NVINOSHLIWS 
= o == 4 De — Ra, u 
102) er PL) ur [07] u 
= X. | = «© = X 
<% | < 
c x DVG? c & c x 
4 
= 2 Z, ° QN.D 2 5 2 ü 
< pn} 4 en zZ - BR) ER 
FITUTION NOILNLILSNI _NVINOSHLINS _S3IUVYg17 LIBRARIES SMITHSONIAN _INSTITUTI 
= el, = u | = a E a 
E z = 5 = I 
5 5 I | = i 
F > > = > 
= = > 2 E, = 
2 m 2 Wasn m z 3 g h 
LIBRARIES INSTITUTION NOILNLILSNI NVINOSHLINS SIIYVY8 
zZ % 2] ze (4) r4 2] „a 
< = < = .e. = = DM, 
zZ 4 z = = - 
2 ® 7 u ‘9 & 
ED zZ = 2 = = 
= > = >’ = >' 
a ” = 72) ne 2 (72) r 
NOILNLILSNI _ s3luV4glI7 LIBRARIES SMITHSONIAN 
u — 2 2 . 2 | = e \ 
w EuLN, u 7 = 77) N 
a m a2 en x a u 
< = .<£ ze. < u) - 
x = = 3 = = .) 
- = = ° = ö >” 
=; = - 2 2 = fi 
yvvuyg Ki El BRARI ES _ SMITEISONJAN INSEITUTION NOILNLILSNI NVINOSHLINS S31IUVYA 
Ei nl = D, m Se E4 JE 
>) ame les je e) N = 
> =) > 
> Lam Dr > = > = 
- n = r . 5 
| © = o = on u 
"ITUTION en NOILNLILSNI NYINOSHLIWS, S3 ıIuvyg NL BRARI ES „SMITHSONIAN 
RN S < SEE / “SZ N SS _ m I < 
NN Erw se AI IN, NT > A 8% A = 
IN er BEP E O I = GG 974 OUSSSUN 2 IE O 


wi 2 RI EU 7 u) 5.9772 u) € u 
Pe SI = x Was m rosa 2 “ | a nos 2 X Wwasr 
N m 2“ m 2 
‚RARIES, „SMITHSONIAN_ INSTITUTION NOILNLILSNI_ NVINOSHLINS Sal uvugıı LIBRA 
“ x ” « 
x i 


NVINOSHLINS S314V 


S 


7) Ze 77) z 
z S < Ss < N 
N Dr N er = = z i 
RN RN: 8 z EN 
NS ET NN © 2 [@) IS 
N EIS 2 = 2, EN 
r an = > Me 
INLILSNI_ „53 IUVYSIT_LIBRARI ES SMITHSONIAN INSTITUTION „NIE 
z 2 ke = 
77) ud IM) i MM 
3 E = = Gy? = 
=. .<£ —_ < ZA < 
= [e< = Be Pr 
o© Er o = ö = 
— — r 2 r pe 
3RARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IUVYHIII_LIBRA 
: NE: Ra = 
2 gr 2 Y oO AMSVM 
= o — 5 oJ u 
= 2) = e- E E) 
e > > r >; = > 
Fe a 2 a r > 
c n m EL m Q m 
= (77) = 7) El 7) 
JLNLILSNI NVINOSHLINS S3IUVYSI1 LIBRARIES SMITHSONIAN 
2: [ep] & SE: (02) 2. [07] 
Z = .< = mn 2 = 
rn 4 rn — 2 Br! 
- z z - = 
= > 2 = WS = > 
/ a9 2 77) Se m) > zZ 
3RARIES SMITHSONIAN _ INSTITUTION NOILNLILSNI _NVINOSHLINS s3lyvygıl_ 
2 2 2 z A 
„0 = (47) == (07) 
x : =. & ee 4 a 
<<, =. < 4% < - 
e° S x = & S 
wo: 0:07; ©: - 
LINLILSNIT NVINOSHLIWS S313vV4g13 LIBRARIES SMITHSONIAN 
N * z ae ui 0 z 
D 5 = = u = 
> > > > “ > = 
= F- 2 r,. 2 r 
2 ae z m z = z 
pe ( Fe} === 
3RARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS 
* oa Rx 12] = o 
= = u: = 
4 I ER = par’ 
m m GL; > 2 2 
N N 2 2,4 a = 2, 
N u > 
ö z 2 177) r 
ILNLILSNI LIBRARI ES SMITHSONIAN INSTITUTION 


NOILNLILSNI 


NOILMLILSNI 
NOILNLILSNI 


S3IUVYHIT LIBRARIES SMITHSONIAN 


_ SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS_ S31UVU 81T LIBRA 
° - =) N& =) 
2 Er = 
=. =) > 
7 = = 
m Da, m - 
= 0 2 de 
er neue S3IUVYGIT_ LIBRARIES, SMITHSONIAN 
= 8 SON \ SE 


SHLUNS S3IUVAAIT LIBRARIES SMITHSONIAN 


SHLINS, S31UVUAII_LIBRARIE 


— 
= 
A zZ NÖ 
Fu 2 mn 


SON 
SYM 
&) 
re 
SHLINS 
N 
N 
SONIAN 
9 
2 


LRLFEITE/ y 
DER FE £ NEN 


NEO RR STEFAN HERNE TE 


ruht ZnageN, 


BE TRRRRLER TER REN! Ä a ’ ı sl TOR fe 
BOWET TITTEN uig ALIETL RN A u 5 7 >. 
au at PAR N 2 SURrs nord But 
FRISCH ENTE 2 .. A En 2 ö . 


DRAER ae 


ana r eg > U PR p r vd { 5 r 
Kesncn,drt RE EDER arte r N Rh! f 
BRSEReTI EN 5 viate N Ars re BR 
DEE RE SE TEIRTR v ale ir } Page PR Eh ale 
£ Pan e: Pk a dr r ’ 
KR IR 2 4 Dan, x n “ f 
: 3 Li GER i 


PRLETEN SEHE 


LARA 


ae 


ir 


hr 


6 


TE 


ur. 


war ea m era EIER 
OBEN SEN un 

EEE SEE Er ZN AS 2 ? 2 : : 5 c 5 y 2 x D Er NR 3 

DER or LE Rn Br . ok ' L ’ 


ES 


Bacar 


Gene 


Be i 
LLC > EN Br 
ran hr we; RE 
Brian > \ 1-3} m 5 
EL ze unser ar - 5 .. HS 
EURE E IE { 5 i . 
er SMETEPU ru & 3 x 5 % 4 i z j 
epialdentz ie kei J i 4 x : Bst ; 
Aura nein, ! 2 FI RIE 2 
Er ara r A} hi 
Re ae Er RE j 2 ö 
Ra Er N Fr 
ER Fr B