DEN hen ERUIEFERTET: yes Ang IT Beinlan, LP Fran. ur chz N ne erse m PR er PAR ENN x NR “ L$ % 2 eu uräne are er LER Ele IE . BER SEE E ru ji EN En VS ER DR re & res BehcH wer dr weh De 4 ae ER RER er TEn Er © SEE LIEE BE, Te Een Re ERIRASEET LIES RE u Eee ee ; SSR ELEH ELSE ih ' ; - : ; $ ie‘, pa ; : r : ' > NS . nn Bet h A \ A 5 ? ; “ { BR R r 5 & NETTE 4 “ ar 7 147 ep ga ML © EN zz WR ADDE © 7 . Oo FR 5 5 25/73 = 2 ‘zz O T AN oO DER Fr oO N = = m N N ZEN m < ja) 2 A > = N 2° = » r = 177) | < 192) ee v4 (77) = n TION NOILNLILSNI NYINOSHLIWS S31UV4I17 INSTITUTION N dp) zZ ü 2 2 N 2 or = o ae 2 en, o “ x Er = = x a wi < = N < = << “5 = > ng PR Tıon NOILNLILSN! NVINOSHLINS S31UYYBIT LIBRARIES SMITHSONIAN N 0) zZ = RRE n = on = << = SE = < = = zZ A ed Be en zZ 3. TE. 4 ©: = '# u - 2 DH 2 8 = 8; = = ap = 2 E 22 Rn, a (dp) >= (47) 5 >=: 77) a = 4 g17 SMITHSONIAN _ INSTITUTION NOILNLILSNI NVINOSHLIWS S3183y4gı7 LE z m = u ze En N == „0 = an CR = x. = X ER & SI < En E E = = =. e) = # [e) = fe) en o'% = 2 z _ 2 N, 2 = ION _ NOILNLILSNI _NVINOSHLIWNS _S31UVY48171 LIBRARIES SMITHSONIAN INSTITUTION NO 5 Zr = TC zZ f Je . = 2 N = En = De: a = zZ 0 E > = “> = = > > 7 2 m. a = | z te ” z 5 2 4 81T _LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NYVINOSHLINS SIIYYVYEI1 LI zZ % a Zee 22) 2 [2] z © 5 = < = 2 < = = Du. = e= zZ 4 Hmm. Z = = 3 3 3 322: z S _ 2 E RAZE 2 : >= N» = > = > = | (97) = 77) 2 72) 2 Tu ION nal) NVINOSHLINS _ LIBRARIES SMITHSONIAN INSTITUTION at Kan oO = Me) U. w z - E I = LER 2 = > | N = zur = G7, ENT x < = y = = < Va PA < < = 1 EN = x rn x = ses S =: z = = Ei zZ Sr = _ & 7 Di BRARI ES _SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31y9yY4g17 LIE zZ 4 o@ = Fe =) — 1 > € es) a co N bez XD = 5 D = 2 ER r= =) 7 2x0 4 > E > N\ 2 S er > DE, a = Be) = B) 2 u: = 5 2 m =; ® = . 2 on 4 Ion, NOILMILLSNI_ NVINOSHLIWS SI UV 81 1_LIBRARI ES_SMITHSONIAN NO < < z BR & o Fe =< - a A \ = = = I oO E MB, zZ 4 2 2 DH 3 z 5 E en = = 277 = N E e Be / = > = AR = =>: BRARIES" = “ N 2 7 Lit gı7 LI „„MITHSONIAN _ INSTITUTION, NOILNLILSNI NVINOSHUNS S31ıyvygın LIE < z = Be 105) 3. \ 43) en ten = & E PR?) ES “ MEY AM & = N \ 2 u MSYM 7 Na EN 2 2 BED Da SR En = ° 2 DBPE? 2 20 GR 3 z [ = 2 B DE 2 2 PL? z zZ = = N 3 +7 et = = £ AL, E 2 a 3 E 3 z z : ISNI_NVINOSHLINS _ WEBBRARILES r Br 5 ws : wu 5 u z =. &x Zi x ke X = = SR: er = = = = < = = 5 3 x = S = ö N = ö za ° = 9 = Fe = a 2 = z ttES SMITHSONIAN INSTITUTION NOILNAILSNI SI = Ki = 1 = ER zZ oO eo) oO ae io) = oO E 5 = = = = je 2 NR > = N = 2. BE; > > N = > m; - : TR; = > D) Rn N ze Pr AA“: 2 = : { RE K IE ne? 7 A" . u A - Rn z a a z ILSNI_NVINOSHLIWS S31UVU AI „MITHSONIAN INSTITUTION, NOILNLILSNI_ N .< zZ IE = za z se - EB ANE en Z 5 =r Sr Wade % N AH + AN: fe) Gy TE oO Z oO DmT T 7a; >= EG? m 2 = u; = 7 = U. — = = a >= » i = u => > = 75) v4 “N k zZ 77) > (4) TIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS S319Y3g17 LIBRARIES SI 2 2 n = Da 02) > 72) iu .o u = : N S Wu = w x. a x =. NN rei = x. x = x en N x = Er sy 5 a = > om - = IRRE | = = = Er z er LSNI NVINOSHLINS _S3183V4817 LIBRARIES SMITHSONIAN_ INSTITUTION _ NOILNAILSNI _N' En» Z = oe 3 = je) S = jev) = g 0 = 60) 2 Ne 5 = = @m2 = 5 2 N = > = HL? = > Oo INA je) —_ 9 es) Be) 5 > = 2 r, z = Z = a 2 = = = o = 1%) = 9) = 72) = un TIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS S314VYE8I1_LIBRARIES S 1%) ze (N Pr 2 > zZ - je] E < S 3 < = a, S Ne: 5 2 GM, = NG: 38 2 8 DRM: 2 ZN S = E 2, FE z N = 2 z = z . ne ILSNI_NVINOSHLIWS LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_N =, “ z 2 z % zZ 7) 77) 14 77) IM o = x = zZ Mn a - = rn =; = = = Zi fe) 2 fe) zu ö 2 ° = i —_ = 8 > er tIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHUNS S3IyY4g17 LIBRARIES S | — fr = rm z > = = os oO = \ oO = e = = = 3 = = = 2. > =) Zah =D} > I, 23 = > = = =” 2 = 2 = ze = ö [07] 333 - n z 2 B o = ILSNI SJIMVWYSI7_LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_N E zZ 122} = EI (2 Ne. ep] - = di: 2. Ne: ar 5 > Y zZ oO 3% T N @) TE oO Z a” E = = N 2 = z Z = > P= ER > = > = nr u Dans % ya f ' a g B‘ ef r ” ER; N En 7 m N Y r ZEITSCHRIFT FÜR SAÄUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Organ der Deutschen Gesellschaft für Säugetierkunde Volume 352, 1987 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 153 Abbildungen Verlag Paul Parey Hamburg und Berlin Wissenschaftliche Originalarbeiten BAUER, ]J. J.: Factors determining the onset of sexual maturity in New Zealand chamois (Rupicapra rupicapra L.). - Geschlechtsreifebestimmende Faktoren bei neuseeländischen Gemsen (Rupıcapra rupicapna 1.) a 2 ee rer BöER, M.: Beobachtungen zur Fortpflanzung und zum Verhalten des Drill (Mandhrillus leuco- phaeus Ritgen, 1824) im Zoo Hannover. — Observations on reproduction and behaviour of captive drills (Mandhrillus leucophaeus Rıtgen, 1824) in HannoverZoo . ............ BREITENMOSER, U.; HALLER, H.: Zur Nahrungsökologie des Luchses Lynx lynx in den schweizerischen Nordalpen. — Feeding ecology of the lynx (Lynx lynx) ın the Swiss Alps . . . Corrı, M.; MERANI, MARIA SUSANA; de VILLAFANE, GLORIA: Multivarıiate morphometrics of Vesper mice (Calomys): preliminary assessment of species, population and strain divergence. — Multivariate Morphometrie an Vespermäusen (Calomys): vorläufige Abschätzung von Arten, Populationentund Zuchtstammunterschreden re... a GEMMEKE, H.; RADTKE, MARGOT; NIETHAMMER, J.: Zur innerartlichen Proteinvariation bei der Waldmaus (Apodemus sylvaticus). - On the intraspecific varıation of some proteins in the wood mouse (Apodemus'sylvatıcus) 2.2 m GERELL, R.: Distribution of Myotis mystacinus and Myotis brandtüi (Chiroptera) in Sweden. — Verbreitung von Myotis mystacinus und Myotis brandtü (Chiroptera) inSchweden ...... GiLL, AyzsHA; PETRov, B.; Zıvkovid, $.; Rımsa, Desanka: Biochemical comparisons in Yugoslavian rodents of the families Arvicolidae and Muridae. — Biochemische Vergleiche von jugoslawischen Rodentia der Familien Arvicolidae und Muridae. ............. GRIMMBERGER, E.; HACKETHAL, H.; URBANCZYK, Z.: Beitrag zum Paarungsverhalten der Wasserfledermaus, Myotis daubentoni (Kuhl, 1819), im Winterquartier. - Contribution to the mating behaviour of the Daubenton’s bat, Myotis daubentoni (Kuhl, 1819) in a winter quarter. u... de me een ee Haar, T.; Weıs, H.; SCHMID, M.: A comparative cytogenetic study on the mitotic and meiotic chromosomes in hamster species of the genus Phodopus (Rodentia, Cricetinae). — Verglei- chende zytogenetische Untersuchungen an den Hamstern Phodopus sungorus und P. robo- rovsku (Rodentia, Cricetinae) au u. . nun Ge ea Hamann, UTE: Zu Aktivität und Verhalten von drei Taxa der Zwerghamster der Gattung Phodopus Miller, 1910. — Studies on activity and ethology of three taxa of dwarf hamster, genus Phodopas Miller, 1910 . ......2....... 2 2 sen. ag PIE Krrma, M.; BanGmA, G. C.: Unpublished drawings of marsupial embryos from the Hill Collection and some problems of marsupial ontogeny. —- Unveröffentlichte Zeichnungen von Marsupialia-Embryonen aus der Hill-Sammlung und einige Probleme der Ontogenese von Marsupiala . 2 a.0 20 0 ade nn a a eye Re KoRrPIMÄRT, E.; NORRDAHL, K.: Low proportion of shrews in the diet of small mustelids ın western Finland. - Geringer Anteil von Spitzmäusen in der Nahrung kleiner Musteliden aus West-Finnland .... »2.2.0 2 una sea a nun ae 6 A MADDALENA, T.; MEHMETI, A.-M.; BRONNER, G.; VOGEL, P.: The karyotype of Crocıdura flavescens (Mammalıa, Insectivora) in South Africa. - Der Karyotyp von Crocıdura flaves- cens (Mammalıa, Insectivora)ın Südafrika © 2 2.222 MAIER, W.; SCHRENK, F.: The hystricomorphy of the Bathyergidae, as determined from ontogenetic evidence. - Die Hystricomorphie der Bathyergidae, nachgewiesen durch onto- genetische Befunde... 2... cu een MATTHIAS, U.; WITTE, G. R.: Untersuchungen zur Genese des Maulwurfshaufens bei Talpa europaea (L.). - Studies on the formation of the molehill of Talpa europaea (L.)........ Owen, R. D.; QuUMSIYEH, M. B.: The subspecies problem in the Trident leafnosed bat, Asellia tridens: homomorphism in widely separated populations. - Die Unterarten-Problematik bei der Dreizack-Blattnase, Asellia tridens, Homomorphie in räumlich weit getrennten Popula- GONEen u nee ee ee P£pın, DOMINIQUE; SPITZ, F.; JANEAU, G.; VALET, G.: Dynamics of reproduction and development of weight in the Wild boar (Sus scrofa) in South-west France. - Fortpflanzung und Gewichtsentwicklung bei Wildschweinen (Ss scrofa) in Südwestfrankreich ....... RıcHARDsoN, P. R. K.: Food consumption and seasonal variation in the diet of the aardwolf Proteles cristatus in southern Africa. -— Nahrungsbedarf und saisonale Variabilität in der Ernährung des Frdwoltes Proteles’enistatus m Sudatuka. 2.2.2... Vs Röper, R.: Alter und Linsengewicht von Meriones shawi grandis (Cabrera, 1907). - Age and eye lens weight in Meriones shawi grandıs (Gabrera, 1907) 2.2... u. 116 265 168 236 242 338 247 133 281 65 201 257 129 156 212 329 zzil 307 ROTHE, H.; König, A.; Darms, K.; Stess, MARGARETHA: Analysıs of litter size in a colony of the common marmoset (Callithrix jacchus). - Analyse der Wurfgröße in einer Kolonie von Weiffbüschelatichent(@zlhchrusgaeechus) nn onen une. sonne uain une SALDERN, ALICE VON; SCHLIEMANN, H.; KayanjJa, F. I. B.; Jacos, J.: Die Analbeutel von Civettictis civetta (Schreber, 1776) (Mammalıa, Viverridae). Über ihren Bau, die chemische Zusammensetzung ihrer Sekrete und ihre biologische Bedeutung. — The anal sacs of Civettictis civetta (Schreber, 1776) (Mammalıa, Viverridae). Morphology, chemical compo- SIEIOnloL cheisecretionsfandkiunettonalssıenitieaneer a ee SCHANZ, F.; ERKERT, H. G.: Resynchronisationsverhalten der circadianen Aktivitätsperiodik von Galagos (Galago senegalensis, Galago crassicandatus garnettii). — Re-entrainment of circadian activity rhythms in galagos (Galago senegalensis, Galago crassicandatus garnettü). . SCHRÖDER, W.; MENSAH, G. A.: Reproductive biology of Thryonomys swinderianus (Tem- minck). - Zur Fortpflanzungsbiologie von Thryonomys swinderianus (Temminck) .......... STADLER, $. G.; HENDRICHS, H.: Zum Lautrepertoire eines ursprünglichen Cerviden, Muntia- cus reevesi micrurns (Sclater, 1875). - On the sound repertoire of the Formosan muntjac, DiRzenSreevesumierunns(Sclatersl 87) m ee nee STAHNKE, ÄADELHEID: Verhaltensunterschiede zwischen Wild- und Hausmeerschweinchen. - Bebssnoralleomparısonoryildandidomestiecayiess a. run... STARK, R.; ROPER, T. J.; Mac LArnon, A. M.; CHivErs, D. ]J.: Gastrointestinal anatomy of the European badger Meles meles L. A comparative study. — Der Gastrointestinaltrakt des EusopaischenDachsesMelesznelesieEimesyereleichendesStudie . 2... 2... 2220... STEIN, BARBARA R.: Phylogenetic relationships among four arvicolid genera. — Phylogenetische BertchunzenzwischenwierArvieolıden Gattungen a nun Eye urn. VENTURA, ].; GOSÄLBEZ, J.: Reproductive biology of Arvicola sapidus (Rodentia, Arvicolidae) in the Ebro delta (Spain). - Zur Fortpflanzungsbiologie von Arvicola sapıdus (Rodentia, auteolidao)rımWElpro-DeltaınıSpantene ray 000 ee VıcAr, C. R.; MACHORDOM, ANNIE: Dental and skull anomalies in the Spanish wild goat, Capra pyrenaica Schinz, 1838. - Zahn- und Schädelanomalien beim Iberiensteinbock, Capra Dyremsuea Scharen VıRro, P.: Size varıation in cohorts of the Bank vole, Clethrionomys glareolus (Schreber, 1780), in Northern Finland. - Größenschwankurgen bei der Rötelmaus Clethrionomys glareolus (Schreber, 1780) in verschiedenen Altersklassen, untersucht in Nordfinnland ......... WAUTERS, L. A.; DHONDT, A.: Activity budget and foraging behaviour of the Red squirrel (Scinrus vulgaris Linnaeus, 1758) in a coniferous habitat. - Aktivität und Verhalten bei der Nahrungssuche von Eichhörnchen (Sciurus vulgaris Linnaeus, 1758) im Nadelwaldhabitat . . WENHOLD, B. A.; Rogınson, T. J.: Comparative karyology of three species of Elephant-shrew (Insectivora: Macroscelididae). —- Vergleichende Untersuchungen an Chromosomen von drei gienidergEllelantenspitzmause (Insectivora: Maeroseelididae) „2. cr. uo nun nn.. YLÖNEN, H.; VIITALA, J.: Social organızation and habitat use of introduced populations of the Vole Clethrionomys rufocanus (Sund) in Central Finland. — Soziale Organisation und Habitatpräferenz von Clethrionomys rufocanus — Populationen in Mittelfinnland....... ZACHARIAE, G.; ELSTRODT, W.; HUCHT-CioRGA, InGrip: Aktionsräume und Verteilung erwachsener Luchse, Lynx Iynx (L.), ım Hinteren Bayerischen Wald. - Home ranges and dispersion of adult lynxes, /ynx Iynx (L.), in the Bayerischer Wald, West Germany... .... ZUMBACH, SyLVvIA; Lüps, P: Hinweise zur Kondition des Steinwildes Capra ı. ibex im Berner Oberland (Schweizer Alpen). - Information about the condition of Alpine Ibex Capra ı. ibex in ine Swrs Als me ee re RE N Wissenschaftliche Kurzmitteilungen AıIcH, HELGA; ZIMMERMANN, ELKE; RAHMANN, H.: Social position retlected by contact call emission ın Gelada baboons (Theropithecus gelada). - Kontaktlaute von Dschelada-Pavianen (überopirhecusigelada) weisen auf ıhresozualeStellunghin „2... 2... 2.222022. FISCHER, M. S.: Die Oberlippe der Elefanten. - Theupperlipofelephants............. Fuchs, W.: Haselmäuse (Muscardinus avellanarins) ın Nestern freibrütender Singvögel. — Dormice (Muscardinus avellanarıus) in nests of open nesting song-birds............ GEWALT, W.: Pferdeantilopen (Hippotragus equinus) äsen unter Wasser. -— Roan antilopes ER BBoraeustegumus))proyssesunder water 2277; 96 218 164 372 294 88 140 364 38 76 341 354 30 58 262 261 Kock, D.: Tadarida teniotis (Rafınesque, 1814): Zweiter Nachweis für Marokko, w-paläarkti- sche Arealgrenzen und taxonomische Anmerkung (Chiroptera: Molossidae). — Tadarida teniotis (Rafinesque, 1814): Second record for Marocco, w-paläarctic distribution limits and taxonomie note (Chiroptera>Molossıd ae)y Sr 194 Korn, H.; Braack, L. E. O.: Survival of wild gerbils (Mammalia: Rodentia) parasitized by larvae of the blowfly Cordylobia anthropophaga (Insecta: Diptera). — Überlebensraten bei wildlebenden Gerbils (Mammalıa: Rodentia), die mit Larven der Schmeißfliege Cordylobia anthropophaga (Insecta- Diptera))parasitient sind 56 KrarochHvit, Z.: Oligodontie bei Hasenartigen (Lagomorpha). - Oligodonty ın Lagomorpha . . 196 PErRIn, L. P. A.: Zum Morgenflug von Nyctalus noctula (Schreber, 1774) (Mammalia, Chiroptera). - On the morning flight of Nyctalus noctula (Schreber, 1774) (Chiroptera: Mammalia) «aus 2 rt ee TE Wäl 50 POGLAYEN-NEUWALL, INGEBORG; POGLAYEN-NEUWALL, 1.: A rare nıpple anomaly in Ringtails, Bassariscus astutus (Procyonidae). — Eine seltene Zıtzenanomalie beim Katzentfrett, Bassaris- cusastutus (Proeyoonidae)) .: u 2 er Ws N A ee 54 RıEGER, 1.; WALZTHÖNY, Dorıs: Hair brush organs ın Viscachas, Lagostomus maximus. — Haarbürstenorgan bei Viscachas, 1 2g0stomusımaxımus 2 2 222. 125 TEGELSTRÖM, H.; Hansson, L: Evidence of long distance dispersal in the Common shrew (Sorex araneus). - Nachweis von Ausbreitung über größere Entfernungen bei der Waldspitz- maus (Sörexsaraneus)nins. SE: er ra er er 52 VAN WAEREBEER, K.; REyESs, J. C.; LuscoMBE, B. A.: A second record of the Pygmy sperm whale Kogia breviceps (de Blainville, 1838) (Cetacea, Physeteridae) from the Peruvian coast. — Der zweite Fund eines Zwergpottwals, Kogia breviceps (de Blainville, 1838) (Cetacea, Physeteridae), an der.peruanıschen Ruste, . 2 0.2.....2 2 192 Bekanntmachungen Seiten un. n tar neh ke de nee he u an 2 En 61, 128, 19953832 Buchbesprechungen Seiten... wait east EN 62,.128,200,26358259384 This journal is covered by Biosciences Information Service of Biological Abstracts, and by Current Con- tents (Series Agriculture, Biology, and Environmental Sciences) of Institute for Scientific Information Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugsweiser Verwertung, vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch im Einzelfall grundsätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines Beitrages ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheberrechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflich- tig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen. © 1987 Paul Parey. 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., 2240 Heide/Holstein [er ISSN 0044-3468 52 (1-6) 1-384 (1987) i) Vol.52 (1), 1-64, Februar 1987 ISSN 0044-3468 C 21274 FF EL. ITSCHRIFT FÜR AUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Irgan der Deutschen Gesellschaft für Säugetierkunde nhold, B. A.; Robinson, T. J.: Comparative karyology of three species of Elephant-shrew (Insectivora: Macrosce- ıididae). — Vergleichende Untersuchungen an Chromosomen von drei Arten der Elefantenspitzmäuse (Insecti- vora: Macroscelididae) Zachariae, G.; Elstrodt, W.; Hucht-Ciorga, Ingrid: Aktionsräume und Verteilung erwachsener Luchse, Lynx Iynx (L.), im Hinteren Bayerischen Wald. — Home ranges and dispersion of adult Iynxes, Lynx Iynx (L.), in the Bayerischer —_ Wald, West Germany 9 » pin, Dominique; Spitz, F.; Janeau, G.; Valet, G.: Dynamics of reproduction and development of weight in the Wild boar (Sus scrofa) in South-west France. — Fortpflanzung und Gewichtsentwicklung bei Wildschweinen (Sus scrofa) in Südwestfrankreich 21 ıbach, Sylvia; Lüps, P.: Hinweise zur Kondition des Steinwildes Capra i. ibex im Berner Oberland (Schweizer Ipen). — Information about the condition of Alpine Ibex Capra i. ibex in the Swiss Alps 30 ıl, C. R.; Machordom, Annie: Dental and skull anomalies in the Spanish wild goat, Capra pyrenaica Schinz, 338. — Zahn- und Schädelanomalien beim Iberiensteinbock, Capra pyrenaica Schinz, 1838 38 senschaftliche Kurzmitteilungen >’errin, L. P. A.: Zum Morgenflug von Nyctalus noctula (Schreber, 1774) (Mammalia, Chiroptera). - On the morning ight of Nyctalus noctula (Schreber, 1774) (Chiroptera: Mammalia) 50 ‘ alström, H.; Hansson, L.: Evidence of long distance dispersal in the Common shrew (Sorex araneus). — “ „Nachweis von Ausbreitung über größere Entfernungen bei der Waldspitzmaus (Sorex araneus) 52 oglayen-Neuwall, Ingeborg; Poglayen-Neuwall, I.: A rare nipple anomaly in Ringtails, Bassariscus astutus (Procyonidae). — Eine seltene Zitzenanomalie beim Katzenfrett, Bassariscus astutus (Procyonidae) 54 orn, H.; Braack, L. E. O.: Survival of wild gerbils (Mammalia: Rodentia) parasitized by larvae of the blowifly Coraylobia anthropophaga (Insecta: Diptera). - Überlebensraten bei wildlebenden Gerbils (Mammalia: Roden- tia), die mit Larven der Schmeißfliege Cordylobia anthropophaga (Insecta: Diptera) parasitiert sind 56 ich, Helga; Zimmermann, Elke; Rahmann, H.: Social position reflected by contact call emission in Gelada baboons (Theropithecus gelada). — — Kontaktlaute von Dschelada-Pavianen (Theropithecus gelada) weisen auf ihre soziale Stellung hin | 58 akanntmachung DESSEN 61 ıchbesprechungen LTIHSONIAN 62 [ Mrz | \ j z N LiBr AA RIES, ı erlag Paul Parey kur und Berlin V HERAUSGEBER/JEDITORS P.J. H. van BREE, Amsterdam — W. FIEDLER, Wien — H. Frick, München - W. HERRE, Kiel - H.-G. Krös, Berlin - H.-J. Kuhn, Göttingen - B. Lanza, Florenz - T. C. $. MORRISON-SCOTT, London - J. NIETHAMMER, Bonn — H. REICHSTEIN, Kiel -— M. Rönrs, Hannover - D. STArRcK, Frankfurt a. M. - F. Strauss, Bern - E. THenIıus, Wien — W.VERHEYEN, Antwerpen SCHRIETLEITUNG/EDITORIAT OEEIGE 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 eh 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 ae und dem Verlag erhältlich. Sonderdrucke: Anstelle einer Unkostenvergütung erhalten die Verfasser von Originalbeiträgen und Wissenschaftlichen Kurzmitteilungen 50 unberechnete Sonderdrucke. Mehrbedarf steht gegen Berechnung zur Verfügung, jedoch muß die Bestellung spätestens mit der Rücksendung der Korrekturfahnen erfolgen. Vorbehalt aller Rechte: Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei- ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ıst auch ım Einzelfall grund- sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines Beitrages ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber- rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen. Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of the first page of an article in this journal indicates the copyright owner’s consent that copies of the article may be made for personal or internal use, or for the personal or 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 CC= Dieses Heft enthält einen Einhefter sowie zwei Beilagen des Verlages Paul Parey, die wir Ihrer besonderen Aufmerksamkeit empfehlen. Fortsetzung 3. Umschlagseite © 1987 Paul Parey. Verlag: Paul Parey, Hamburg und Berlin. Anschriften: Pe 12% D-2000 Hamburg 1; Lindenstr. 44-47, D-1000 Berlin 61. — Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., D-2240 Heide/Holst. Comparative karyology of three species of Elephant-shrew (Insectivora: Macroscelididae) By B. A. WEnHoLD and T. J. RoBINsoN Mammal Research Institute, University of Pretoria, Pretoria Receipt of Ms. 24. 3. 1986 Abstract Compared the chromosomes of three southern African species of elephant-shrew (subfamily Macros- celidinae) (Elephantulus rupestris, Macroscelides proboscideus and Petrodromus tetradactylus) by means of G-banding, C-banding and sılver-nitrate staining for the detection of nucleolar organizer regions. Extensive G-band homologies were evident in the chromosomes constituting the genomes of the three species. The difference in diploid number between P. tetradactylus (2n = 28) and E. rupestris and M. proboscideus (2n = 26) is attrıbutable to a tandem fusion of two chromosomes in the lineage common to the 2n = 26 species. The distribution of heterochromatin differed between the three species studied; although telomeric C-bands occurred in both E. rupestris and P. tetradactylus, the staining patterns were not identical. The distribution of NOR’s in the species’ genomes was, except for the common occurrence of one pair, species specific. Introduction Elephant-shrews, members of the family Macroscelididae, are endemic to the Afrıcan continent. The family, which comprises two subfamilies, the Macroscelidinae and Rhyn- chocyoninae, stands as a distinctive and relatively isolated group whose ordinal classifica- tion is the subject of some uncertainty. PATTERSON (1968) and RATHBUN (1979) are of the opinion that the Macroscelididae should be elevated to a separate order, the Macroscelidea, as members of the taxon are considered to have a more specialised morphology and life history than other Insectivora, a view supported by blood protein studies (GOODMAN 1974). These data suggest amonophyletic origin for the Macroscelididae which shares only primitive traits wıth the other insectivores. However, most taxonomic treatments of the group persist in the retention of the elephant-shrews within the family Macroscelididae, order Insectivora (CoRBET 1974; BUTLER 1978; SMITHERS 1983). Of the two subfamilies the Macroscelidinae is the more widely represented in the southern African subregion. It comprises the monotypic genera Macroscelides (A. Smith, 1829) and Petrodromus (Peters, 1846), and the relatively speciose Elephantulus (Thomas and Schwann, 1906) which has five species occurring in the subregion. The latter genus has a wide distribution in other parts of Africa and includes four extralimital species in East and North Africa and the Somali (SMITHERS 1983). Little cytogenetic information is available for elephant-shrews in general and, in particular, the southern African taxa. What information exists is restricted to unbanded karyotypes or the reports of diploid numbers of E. myurus (2n = 30; FORD and HAMERTON 1956), E. (nasılio) brachyrhynchus (2n = 26; Stimson and GOODMAN 1966) and the extralimital E. rufescens (2n = 34; CHu and BENDER 1962) and E. rozetı (2n = 28; MATTHEy 1954). Unfortunately, these data are of limited value in phylogenetic assessments since the absence of differential staining precludes unequivocable identification of shared derived chromosomal traits (synapomorphies) and, as has been pointed out previously U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5201-0001 $ 02.50/0 Z. Säugetierkunde 52 (1987) 1-9 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 2 B. A. Wenhold and T. J. Robinson (BAkKER and BıcKkHAM 1980; Haıpuk et al. 1981; Rosınson et al. 1986), grossly underes- timates the degree of chromosomal divergence among species. As part of an extensive investigation into the chromosomal relationships of the Macroscelididae we present chromosome banding data on three previously unreported taxa: M. proboscideus, E. rupestris and P. tetradactylus. By determining the extent of chromosomal concordance between these three genera we hope to provide preliminary information on their phylogenetic relationships. Material and methods Fibroblast cultures were established from skin biopsies or disaggregated kidney tissue. Air dried slides were routinely prepared from the fibroblast cultures and, occasionally, from bone marrow prepara- tions following a yeast dextrose pretreatment (LEE and ELDER 1980). The methods of Wang and FEDOROFF (1972) and SUMNER (1972) were used respectively for G-banding and C-banding. Nucleolar organizer regions (NOR’s) were visualized using a modified silver-staining technique (BLooM and GOODPASTURE 1976). The numbers of specimens analyzed and the collection localities with their corresponding grid references are presented: M. proboscideus: 1 8, van Wyksvlei (3021 BD), 1 d, Sutherland (3220 BC), 1 d, Vaalputs (3021 BA) E. rupestris: 1 2, Beaufort West (3222 BD) P. tetradactylus: 2 85, Sordwana Bay (2732 DA) Results Round-eared elephant-shrew, M. proboscideus (2n = 26) The G-banded chromosomes of this species are shown in Fig. 1A. The autosomal male karyotype comprises seven pairs of metacentric and submetacentric chromosomes and five pairs of subtelocentric and acrocentric chromosomes. The X ıs a small metacentric chromosome while the minute Y is acrocentric. A typical C-banded cell is presented ın Fig. 1B and shows the small amount of heterochromatin present in the genome of this species. Even in the best of preparations only approximately 50 % of the chromosomes regularly showed positive C-banding, this generally being concentrated in the proximal portion of the long arms immediately adjacent to the primary constrictions (for example pair 9-arrows). While the X chromo- some routinely appeared devoid of heterochromatin, the Y chromosome, ın contrast, is almost entirely heterochromatıc. Two nucleolar organizer regions (NOR’s), which are located at the terminal ends of autosomal pair 6, were visible following silver staining (Fig. 1C — arrows). Non-specific centromeric staining unrelated to NOR activity was observed infrequently. Smith’s rock elephant shrew, E. rupestris (2n = 26) The G-banded chromosomes of a female E. rupestris are presented in Fig. 2A. The karyotype comprises eight pairs of metacentric and submetacentric chromosomes and four pairs of subtelocentric and acrocentric chromosomes. The X chromosome is metacentric ın morphology; due to the absence of males in the test material the morphology of the Y remains unknown. A C-banded metaphase cell is presented in Fig. 2B. A striking feature of the hetero- chromatic distribution of this species is the presence of telomeric C-bands in many of the larger autosomes (Fig. 2B- arrowheads). In most preparations the telomeric heterochroma- tın stains less intensely than the centromeric C-bands. Karyology of elephant-shrews 3 B 12 Fıg. 1. A: G-banded karyotype of a male Macroscelides proboscideus (2n = 26). Asterisks indicate the positions of NOR’s. B: C-banded cell showing the a blocks of heterochromatin present on chromosome 9 (arrows). The heterochromatic Y is indicated. C: Silver-stained cell showing one pair of NOR’s ıdentified as being located on autosomal pair 6 (arrows) Evidence of two NOR bearing autosomal pairs was found in E. rupestris following silver staining (Fig. 2C). These were tentatively identified as being situated at the terminal ends of pairs 5 and 7 although invariably, silver deposit was found only on one homolog of each pair. Four toed elephant shrew, P. tetradactylus (2n = 28) The G-banded chromosomes of this species are illustrated in Fig. 3A. The male karyotype consists of eight pairs of metacentric and submetacentric chromosomes and five pairs of subtelocentric and acrocentric chromosomes. The sex chromosomes comprise a small 4 B. A. Wenhold and T. J. Robinson Fıg. 2. A: G-banded karyotype of a female Elephantulus rupestris (2n = 26). Asterisks indicate the Po es ot NOR’s. B: C-banded cell showing terminal (arrowheads) and centromeric (arrows) eterochromatin. C: Silver-stained cell showing two of the four NOR bearing chromosomes (arrowheads), the homologs in each case being free of silver deposit submetacentric X and an acrocentric Y which is the smallest chromosome of the comple- ment. A typical C-banded metaphase cell is presented in Fig. 3B. As with the preceding species the distribution of heterochromatin is both telomeric (arrowheads) and centromeric (arrows). A representative silver stained metaphase cell (Fig. 3C) shows clearly the presence of two NOR bearing chromosomes corresponding to pair 7. The NOR’s are situated at the terminal ends of the short arms of this autosomal pair. Karyology of elephant-shrews 5 X Y [2 12 Fıg. 3. A: G-banded karyotype of a male Petrodromus tetradactylus (2n = 28). Asterisks indicate the positions of NOR’s. B: C-banded cell showing differential staining of terminal (arrowheads) and centromeric (arrows) heterochromatin. The heterochromatic Y abutting on chromosome 1 is indica- ted. C: Sılver-stained cell showing a pair of NOR bearing chromosomes identified as being autosomal paır 7 (arrows) Interspecific comparison A comparison of the G-banded chromosomes of M. proboscideus, E. rupestris and P. tetradactylus is presented in Fig. 4. The euchromatic portions of all the chromosomes have banding equivalents in their specific genomes. The difference in diploid number between the two former species, each with 2n = 26, and P. tetradactylus (2n = 28), ıs due to the presence of two unfused chromosomal elements in the P. tetradactylus genome (chromosomes 4 and 13), which show banding homologies with the submetacentric chromosomes constituting pairs 3 and 4 in the respective 2n = 26 karyotypes (Fig. 4). 6 B. A. Wenhold and T. J. Robinson 455 566 6 7 7 788 8 93 10 10 10 1111 11 12 12 12 Fıg. 4. Comparison of the half karyotypes of three species. M = M. proboscideus, R = E. rupestris and P=P. tetradactylus. Dot indicates the absence of the Y of E. rupestris in the study material. The box contains the two chromosomes of P. tetradactylus involved in the tandem fusion, as well as the corresponding submetacentric fusion chromosome responsible for the difference ın diploid numbers observed between P. tetradactylus (2n = 28) and the 2n = 26 species M. proboscideus and E. rupestris. The corresponding position of the two unfused elements in the half karyotype comparison illustrated here ıs shown by an asterısk Subtle interspecific differences in chromosome morphology due to the presence or absence of heterochromatin were noted for several pairs, the most dramatic difference being evident in pair 9 of M. proboscideus. This chromosome corresponds in banding pattern to the autosomes constituting pair 2 of the other species. However, the addition of juxtacentromeric heterochromatin in the long arm of this chromosome (Fig. 4- arrow) has resulted in a change in its gross morphology from submetacentric in E. rupestris and P. tetradactylus to subtelocentric in M. proboscideus. This is responsible for the differential placement of this chromosome in the M. proboscideus karyotype. Likewise, the slight Karyology of elephant-shrews 7 differences in the morphology of the X chromosomes of, on one hand, M. proboscideus and E. rupestris (both metacentric) and P. tetradactylus (submetacentric) on the other, are thought to be due to minor heterochromatic differences. Discussion In this investigation use was made of G-banding, C-banding and silver-nitrate staining of the somatic chromosomes of M. proboscideus, E. rupestris and P. tetradactylus in order to determine their chromosomal relationships. The comparative G-band analysis showed conservation of homologous band patterns in each pair of chromosomes in all three species studied, while the difference in diploid number between P. tetradactylus (2n = 28) and the 2n = 26 species, E. rupestris and M. proboscideus, ıs attributable to the presence of the unfused elements, 4 and 13, in the P. tetradactylus genome. As is evident in Fig. 4 a tandem fusion, be it a centromere-telomere or a telomere- telomere translocation, would account for the reduction in diploid number from 2n = 28 in P. tetradactylus to the observed 2n = 26 in both E. rupestris and M. proboscideus. A fusion product resulting from the tandem translocation of two chromosomes could, depending on the position of the initial breakpoints, be monocentric or alternatively dicentric with one functional centromere (Hsu et al. 1975). However, interstitial hetero- chromatin may be expected following this form of rearrangement. Its apparent absence in the fusion chromosomes of M. proboscideus (number 3) and E. rupestris (number 4) probably indicates that the centromeric heterochromatin either did not participate, or persist, in the structuring of the fusion product. It is obviously difficult to determine the direction of karyotypic change from the limited data available. A definitive statement on whether 2n = 28 possibly represents the ancestral condition, with the fusion chromosome present in E. rupestris and M. proboscideus representing a shared derived characteristic, which arose in their last common ancestor, must await analysıs of the remaining species. Nevertheless, while evidence for tandem fusion and a concomitant reduction in diploid number has been documented for cultured cells (Hsu et al. 1975) and a variety of mammalian species (Liming etal. 1980; ELDER 1980; Rosınson and SKINNER 1983) the reverse process, fission or fragmentation would, in this case, represent a reversion of a fusion which had previously occurred. Consequently, the most parsimonious explanation argues for a reduction in 2n through tandem fusion. The unusual distribution of heterochromatin in the genomes of E. rupestris and P. tetradactylus deserves special mention (Figs. 2B and 3B). While criteria such as the shared presence of terminal heterochromatin have previously been used in phylogenetic assess- ments (VAN TUINEN and LEDBETTER 1983), in this study distinct differences in both the staining intensity and the number of chromosomes showing the material in E. rupestris and P. tetradactylus negate its usefulness as a synapomorphic trait. Likewise, the locations of NOR’s are, with a single exception, species specific and of no cladistic value. The common occurrence of a pair of NOR’s on homologous chromosomes ın all three species (chromo- some 6 of M. proboscideus and 7 of the other species) may however prove useful in future investigations incorporating the remaining taxa. The relative conservativeness of the karyotypic change exhibited by the three species investigated in this study is to some extent unexpected, in view of the social system of elephant-shrews. RATHBUN (1979) is of the opinion that elephant-shrews form monoga- mous pairs occupying small territories, a situation which may predispose to the fixation of structural rearrangements through inbreeding; a situation not dissimilar to foxes which show extensive karyotypic varıability (BusH 1975; MAKInEN and GusTavsson 1982). When interspecific variation in diploid number is not great, this often indicates that the euchromatic segments of chromosomes are conservative ın their structural arrangements, 8 B. A. Wenhold and T. J. Robinson which is borne out by the present investigation. Should the conservative nature of karyotypic change evident from the three genera analysed in this study hold for the other species in the order, comparative cytogenetic data may prove particularly useful in the assessment of the groups’ phylogenetic relationships. Acknowledgements We thank Dr. P. F. WooDatt for his help with the capture and identification of specimens, the Natal Parks Board for permission to trap P. tetradactylus and the University of Pretoria for financial support. Zusammenfassung Vergleichende Untersuchungen an Chromosomen von drei Arten der Elefantenspitzmäuse (Insectivora: Macroscelididae) Die Chromosomen der drei ın Südafrika lebenden Elefantenspitzmausarten (Macroscelididae) Zle- phantulus ae Macroscelides proboscideus und Petrodromus tetradactylus wurden in ihren G- und C-Banden sowie nach Silbernitratfärbung (Lage von NOR = nucleolar organizer region) verglichen. In den G-Banden stimmen die drei Arten weitgehend überein. Die gegenüber Petrodromus um ein Paar verminderte Chromosomenzahl bei den beiden anderen Arten ist das Ergebnis einer Verschmelzung der Chromosomen 4 und 13 von Petrodromus. Petrodromus und Elephantulus rupestris haben im Gegensatz zu Macroscelides endständiges Heterochromatin, unterscheiden sich aber ın der Stärke der C-Banden und der Anzahl betroffener Chromosomen. Ein NOR fand sich bei allen drei Arten auf homologen Chromosomen, bei Elephantulus rupestris zusätzlich auf einem weiteren. Literature BAKER, R. ]J.; BicKHam, J. W. (1980): Karyotypic evolution in bats: evidence of extensive and conservative chromosomal evolution in closely related bat taxa. Syst. Zool. 29, 239-253. Broom. $. E.; GOODPASTURE, C. (1976): An improved technique for selective sıilver-staining of nucleolar organızer regions in human chromosomes. Hum. Genet. 34, 199-206. BusH, G. L. (1975): Modes of anımal speciation. Ann. Rev. Ecol. Syst. 6, 336-364. BUTLER, P. M. (1978): Insectivora and Chiroptera. In: Evolution of African mammals. Ed. by V. S. Mario, H. B. $. CookE. Cambridge, Mass.: Harvard Univ. Press. Chu, E. H. Y.; BENDER, M. A. (1962): Cytogenetics and evolution of Primates. Science 133, 1399-1405. CoRBET, G. B. (1974): Family Macroscelididae, part 1.5 pp 1-6. In: The mammals of Africa: an identification manual. Ed. by ]. MEESTER, H. W. SETZER. Washington DC: Smithson. Inst. Press. ELDER, F. F. B. (1980): Tandem fusion, centric fusion, and chromosomal evolution in the cotton rats, genus Sigmodon. Cytogenet. Cell Genet. 26, 199-210. ForD, C. E.; HAMERTON, ]. L. (1956): Chromosomes of five rodent species. Nature 177, 140-141. GoopMman, M. (1974): Protein sequence and immunological specificity: their role in phylogenetic studies of primates. In: Phylogeny of the Primates. Ed. by W. P. LuckETT, F. S. SzaLay. New York: Plenum. Ha1DuR, M. N.; BAKER, R. ]J.; RoBBIns, L. W.; SCHLITTER, D. A. (1981): Chromosomal evolution in African megachiroptera: G-band and C-banding assessment of the magnitude of change in similar standard karyotypes. Cytogenet. Cell Genet. 29, 221-232. Hsu, T. C.; PATHAR, $.; CHEn, T. R. (1975): The possibility of latent centromeres and a proposed nomenclature system for total chromosomal and whole arm translocations. Cytogenet. Cell Genet. 15, 41-49. LEE, M. R.; ELDer, F. F. B. (1980): Yeast stimulation of bone marrow mitosis for cytogenetic investigations. Cytogenet. Cell Genet. 26, 36-40. Liming, $.; YInGYInG, Y.; XINGSHENG, D. (1980): Comparative cytogenetic studies on the red muntjac, Chinese muntjac, and their F1 hybrids. Cytogenet. Cell Genet. 26, 22-27. MAKINEN, A.; GUSTAVsSOoN, I. (1982): A comparative chromosome banding study in the silver fox, the blue fox and their hybrids. Hereditas 97, 289-297. MATTHEY, R. (1954): Les chromosomes de Macroscelides rozeti (Duverroy) (Mammalia - Insectivora). Existe-t-ıl une serie polyploide chez les Macroscelididae. Rev. Suisse Zool. 61, 669-677. PATTERSON, B. (1965): The fossil elephant-shrews (Family Macroscelididae). Bull. Mus. Comp. Zool. 133, 295-335. RATHBUN, G. B. (1979): The social structure and ecology of Elephant-shrews. Hamburg, Berlin: Paul Parey. Karyology of elephant-shrews 9 ROBINSON, T. J.; SKINNER, J. D. (1983): Karyology of the riverine rabbit, Bunolagus monticularis, and its taxonomic implications. J. Mammalogy 64, 678-681. RoBInsoN, T. J.; SKINNER, ]J. D.; Haım, A. S. (1986): Close chromosomal congruence in two species u squirrel: Xerus inaurıs and X. princeps (Rodentia: Sciuridae). S. Afr. J. Zool. 21, M. ne R. H. N. (1983): The mammals of the southern African Subregion. Pretoria: Univ. retoria. STIMSON, C.; GooDMan, M. (1966): Chromosomes of the Elephant-shrew, Nasilio brachyrhynchus brachyrhynchus. Mamm. Chrom. Newsl. 22, 188. SUMNER, A. T. (1972): A simple technique for demonstrating centromeric heterochromatin. Expl. Cell Res. 75, 304-306. Tuınen, P. van; LEDBETTER, D. H. (1983): Cytogenetic comparison and phylogeny of three species of Hylobatidae. Amer. J. Physical Anthropology 61, 453-466. Wang, H. C.; FEDOROFF, $. (1972): Banding in human chromosomes treated with trypsin. Nat. New Biol. 235, 52-54. Authors’ address: B. A. WENHoLD and Dr. T. J. Rogınson, Mammal Research Institute, University of Pretoria, Pretoria 0002, South Africa Aktionsräume und Verteilung erwachsener Luchse, Lynx Iynx (L.), im Hinteren Bayerischen Wald Von G. ZACHARIAE, W. ELSTRODT und INGRID HUCHT-CIORGA Institut für Zoologie der Universität Düsseldorf Eingang des Ms. 8. 4. 1986 Abstract Home ranges and dispersion of adult lynxes, Lynx lynx (L.), in the Bayerischer Wald, West Germany Studied movements and habitat use of three adult lynxes in the eastern Bayerischer Wald. Information on activities was obtained from tracking in snow, occasional traces during summer, kills and prey residues, scats, urine marks, and dens frequently used for rest; direct sightings occurred rarely. Individuals were identified by measuring footprints and toothmarks on prey bones and by traces of sex-related behaviour. Freshly found prey en were ınoculated with polystyrene granulate to obtain marked scats at distant localities. Home ranges consisted of a core area of ca. 30 km’ and a more extended peripheral zone. Ranges of two resıdent females overlapped with that of one resident male. Each core area included vast rocky sites with dens providing shelter under various weather conditions. These prerequisites had mainly determined the choice of habitat by the lynxes. They lived there for at least 7 years. Some studies from different countries support the concept of a core area as part of the home range. Other literature data, however, diverge extremely on the spatial requirements of the lynx. The relation between methods and results is discussed. Einleitung Zu den Ansprüchen des Luchses an Raum und Landschaftsstrukturen liegen wenige und widersprüchliche Angaben vor. Ursachen dafür sind die Schwierigkeiten der Freilandfor- schung und erhebliche Unterschiede in den eingesetzten Methoden. Der Luchs hält sich vor dem Beobachter verborgen, bewegt sich vorwiegend bei Nacht und über weite Entfernungen. Darum müssen seine Aktivitäten aus Spuren erschlossen und die Untersu- chungen auf überschaubare Gebiete und damit auf einzelne Individuen konzentriert U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5201-0009 $ 02.50/0 Z. Säugetierkunde 52 (1987) 9-20 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 10 G. Zachariae, W. Elstrodt und Ingrid Hucht-Ciorga werden. Aussagen von allgemeiner Gültigkeit sind nur zu erreichen, wenn Studien aus verschiedenen Landschaften verglichen werden. Vom Winter 1970/71 an gab es freilebende Luchse ım Hinteren Bayerischen Wald; sie waren dort illegal ausgesetzt worden. Bis zum Winter 1983/84 wurden wie regelmäßig festgestellt; danach und bis 1985 gab es nur noch vereinzelte Hinweise. Über unsere Untersuchungen zu Raumansprüchen und Jungenaufzucht wurde schon kurz berichtet (ELSTRODT et al. 1983). Der vorliegende Beitrag behandelt Größe und Lage der von den erwachsenen Luchsen gewählten Wohngebiete, die dabei wirksamen Faktoren und den Einfluß der Methoden auf solche Ergebnisse. Untersuchungsgebiet und Herkunft des Luchsvorkommens Eine langjährige Ansiedlung erfolgte nur in dem Gebiet, welches von der Grenze zur Tschechoslowa- kei und den Bergen Arber (1456 m), Falkenstein (1315 m) und Rachel (1453 m) umrahmt wird (Abb. 3). Die Hänge sind zumeist nach Süden oder Westen exponiert; zwischen 800 m und 1150 m Höhe befinden sich vielerorts freistehende Felswände und Blockströme. Schneezeit ist von November bis April; es werden Schneehöhen um 2 m erreicht. Nur in Tallagen um die wenigen Ortschaften sind Wiesen und Ackerflächen. Etwa 260 km? sind zusammenhängender Wald, davon drei Viertel Staatsforst, das übrige Privatbesitze. Wichtigste Baumarten sind Fichte (Picea abies), Rotbuche (Fagus sylvatica), Weißtanne (Abies alba) und Bergahorn (Acer pseudoplatanus). Rehe (Capreolus capreolus), en (Cervus elaphus) und Hasen (Lepus europaeus) waren die häufigsten Beutetiere für die Luchse. Die Aussetzungen erfolgten vermutlich ım südöstlich vom genannten Gebiet gelegenen National- park Bayerischer Wald; dort haben sich aber niemals freilebende Luchse angesiedelt. Nach STEHLIK (1979) waren die ersten Tiere Wildfänge aus der Slowakei. Zu weiteren Aussetzung gibt es keine nachprüfbaren Angaben. Wissenschaftliche Aussagen über die Anfangsentwicklung sınd nicht mög- lich, weil die Verantwortlichen alle nötigen Daten verschweigen; nur in Populärschriften erschienen Artikel mit unklaren und widersprüchlichen Angaben. Das Staatliche Forstamt Zwiesel bemühte sich in den Jahren 1972-1979, zufällig eingehende Hinweise und Beobachtungen zu sammeln. Wir führten selbst Geländeuntersuchungen ın allen Jahreszeiten durch. Nach unserer Kenntnis wurde das Vor- kommen in den Jahren 1977-85 nicht mehr durch weitere Aussetzungen verändert. Aber die festgestellten drei erwachsenen Luchse waren dort wahrscheinlich schon vor 1977 fest angesiedelt; das ist aus vorliegenden Meldungen zu schließen. Auch die letzten Nachweise aus den Jahren 1983-1985 bestätigen nur diese Luchse und ihre Jungen. Beobachtungen zu deren Aufzucht werden an anderer Stelle veröffentlicht. Material und Methoden Beweise für Anwesenheit und räumliche Bewegung der Luchse Während der drei Winter 1978/79 bis 1980/81 wurden insgesamt 113 km Fährtenstrecke in Höhen zwischen 600 m und 1250 m aufgenommen; die Zugangs- und Suchwege für uns selbst betrugen das Fünfzehnfache. Wegen der Kürze der Wintertage und anderer Aufgaben im Arbeitsprogramm konnten die oft langen Wege der Tiere nur in ee verfolgt werden; auch änderten sıch der Schneezustand und damit die Erhaltung von Spuren kurzfristig und abhängig von Höhe und Exposition. Die längsten erfaßten Strecken waren 5-9 km lang, die meisten aber kürzer. Auch in den Wintern 1981/82 bis 1983/84 wurden noch weitere Fährtenstücke aufgenommen. Außerdem suchten wir ın den Jahren 1979 bis 1985 während der schneefreien Zeit nach Aktivitätsspuren der Luchse. Wir fanden achtmal einzelne Trittsiegel oder kurze Schrittfolgen an Stellen, wo ein Luchs auf weichen Bodenmulm oder lehmigen Sand treten mußte. Die Luchsherkunft war zusätzlich gesichert durch den Zusammenhang mit Lagerplätzen, Haarfunden, Rißresten oder Losungen. Gerissene Rehe und Rotwildkälber wurden meistens wiederholt befressen. Das war bei Schnee gut kontrollierbar und dann Beleg für längere Anwesenheit eines Luchses. Im Sommer haben wir Beutereste einem Luchs nur zugeschrieben, wenn Tötungsbiß, Anschnitt oder Reißzahnspuren eindeutig waren. Wir konnten 1979-81 zehn frisch gerissene Beutetiere selbst untersuchen und die Nutzung überwachen. Weiter fanden wir 1979-83 siebenmal Knochenreste von Beutetieren mit eindeutigen Zahnspuren von einem Luchs. Für die Zeit September 1976 bis März 1979 nennen die Forstamtsberichte schon 23 zweifelsfreie Rißfunde. Dabei konnte in 11 Fällen das Luchsindividuum nachträglich ermittelt werden, wenn gleichzeitig zu seinen Besuchen am Riß auch die anderen Luchse in deren bevorzugten Aufenthaltsräumen festgestellt wurden. Aktionsräume und Verteilung erwachsener Luchse Lil Jeder Luchs hatte bestimmte Lagerplätze unter überhängenden Felsen oder Wanderblöcken, die er zu allen Jahreszeiten aufsuchte. Diese boten auch in der schneefreien Zeit die Möglichkeit, das Vorhandensein von Luchsen und sogar eines Muttertieres mit Jungen festzustellen. Beweise dafür waren Trittsiegel, Lagerspuren im Laub, Haar- und Knochenfunde, Harnmarken und Kotstellen (Hucht und Eıstropr 1983). Die Umgebung, vermutete Zugänge und Verbindungswege wurden abgesucht. Jedesmal wurden Tritte und Lagerspuren eingeebnet und Losungen fortgenommen; so ließ sich die Entstehungszeit späterer Funde eingrenzen. Auch heftige Winde ordnen die Streu neu, so daß Lagerspuren danach entstanden sein müssen. Im Frühsommer schlägt sich der Pollen der Waldbäume auf Losungen und anderen Spuren nieder, die vorher erzeugt wurden; auch dies ermöglicht eine ungefähre Datierung. Abb. 1. Konservierung und Vermessung von Trittsiegeln des Luchses. a: Abdruck der linken Vorderpfote auf weicher Erde eines Lagerplatzes im Sommer, Gipsabguß. Wenige Streuteile oder kleine Steine stören die Auswertung nicht. b: An Trittsiegeln erhobene Maße. Die innere der beiden mittleren Zehen steht vor. A,, A;= Gesamtlänge; B,, B;= Teillänge; C, D= Gesamt- und Teilbreite; E = Mittelballenbreite; & = Spreizungswinkel Markierung von gerissenen Beutestücken und Losungen In den Wintern 1979/80 und 1980/81 wurden frisch gerissene Beutetiere mit unverdaulichem Polystyrol-Granulat in verschiedenen Farben (Hersteller Hoechst AG, Frankfurt; Vertrieb Uhlig GmbH, Emmelshausen) markiert. Es entspricht den Forderungen für physiologische Unbedenklich- keit (FRANcK und MÜHLSCHLEGEL 1966/68). Die Körner sind 2,4-3,0 mm groß und haben stumpfe Kanten. Das durch tiefe Schnitte ins Fleisch gesteckte Granulat störte die Luchse nicht, zumal sie auch Knochen zerbeißen und fressen. Wir fanden acht markierte Losungen, davon drei erst im Sommer, in maximal 4 km Entfernung vom ursprünglichen Rißplatz. So war die Zuordnung von diesen Losungen, einigen Lagerplätzen und bestimmten Geländepartien zu einem im Winter identifizierten Luchs zusätzlich gesichert. Identifizierung der erwachsenen Luchse Meßdaten an erlegten Luchsen oder an Museumsmaterial wurden von DanıLov et al. (1979), HEPTNER und SrLuDsk1J (1980) aus der Sowjetunion, SUMINSKI (1973) aus Polen, HELL und SLADEK (1980) und StoLLmann (1963) aus der Tschechoslowakei und VasıLıu und DEcEI (1964) aus Rumänien mitgeteilt. Danach sind erwachsene männliche Luchse größer und schwerer als weibliche, aber in der Regel überschneiden sich die Variationsbreiten der Geschlechter. Im selben Areal variieren äußere Körpermaße und Schädelmaße in Verhältnissen zwischen 1:1,2 und 1:1,4. Wenn nur wenige Individuen vorhanden sind und diese derartige Unterschiede aufweisen, dann kann man sie identifizie- ren an den in den Spuren faßbaren Körpermaßen, nämlich Pfotenmaßen aus Trittsiegeln und Zahnabständen an Beuteresten. So sind auch Jungtiere im ersten Lebensjahr eindeutig von den Erwachsenen zu unterscheiden. Weil speziell diese Mafßzahlen in der Literatur fehlen, wurden sie 1982-85 an jungen und erwachsenen Luchsen im Zoo Wuppertal und in einigen anderen Zoologischen Gärten zum Vergleich überprüft. Die dort erhobenen Meßdaten werden später mitgeteilt. Zur Unterscheidung einzelner Individuen haben auch ZELTUCHIN (1984) beim Luchs (Lynx Iynx) und WINEGARNER (1985) beim Rotluchs (Lynx rufus) die Maße von Trittsiegeln herangezogen. Am negativen Fußabdruck ist genaues Vermessen schwierig; auch ergeben sich oft später noch Fragen nach Einzelheiten. Darum übernahmen wir Anregungen von der Kriminalıstik: Trittsiegel 12 G. Zachariae, W. Elstrodt und Ingrid Hucht-Ciorga wurden zusammen mit einem Maßstab genau senkrecht fotografiert und mit Modellgips ausgegossen. Bei richtiger Ausführung lassen sich feinste Einzelheiten konservieren (Abb. 1). Ein zunächst dünnflüssiges Gemisch aus gleichen Volumina Wasser und Gipspulver wird bis zum Beginn des Eindickens gerührt und dann schnell mit dem Löffel, vom en Punkt ausgehend, in die Spur gefüllt. Nun legt man zum Festigen Stäbchen aus frisch entrindeten Zweigen ein und bildet mit einer weiteren Gipsschicht eine allseits überstehende Platte; in diese werden die Funddaten eingeritzt. Damit Spuren im Schnee nicht zerschmelzen, müssen Wasser und Gips sehr kalt sein. Pulverschnee, Sand und Bodenmulm kann man zuvor mit Haarspray fixieren. Bei sehr lockerem Material und an schrägen Flächen wird zunächst eine dünne Schicht trockener Gips aufgestreut, mit Wasser besprüht und dann ausgegossen. An Fotos und Gipsausgüssen wurden die Mafe nach Abb. 1 bestimmt; dazu wurden bei den Zehenballen der Krümmungsmittelpunkt des Vorderrandes und beim Mittelballen der Flächen- schwerpunkt eingezeichnet. Vergleiche sind nur zulässig, wenn die Füße ohne Spreizung aufgesetzt wurden; bei abweichendem Spreizungswinkel « ist in gewissen Grenzen eine Korrektur möglich. Ein unveränderliches, gut fafSbares Maß ist die Breite des Mittelballens (E). Bisher wurden ın Veröffent- lichungen über freilebende Luchse nur die Maße des Abdruckumrisses angegeben. Aber diese verändern sich mit der Festigkeit und Tiefe vom Schnee und der Stärke des Haarkranzes am Pfotenrand; das erklärt die großen Widersprüche in den Angaben verschiedener Autoren. Wir konnten in jedem Fall den Kuder (A;= 78 mm, E = 50 mm) von den beiden Luchsinnen (A; = 69 mm, E = 45 mm) und diese von letztjährigen Jungen (im Januar/Februar A<62 mm, E<37 mm) unterscheiden. Schließlich war die Südost-Luchsin durch eine Anomalıe gekennzeichnet. Beim Aufsetzen der Vorderläufe drückte sie die Fußgelenke so tief durch, daß in weicher Schneedecke der Fußwurzelbal- len sichtbar wurde; dieser kommt bei normalen Tieren niemals zum Abdruck. Solche Anomalien wurden bisher nur bei Zoo- und Haustieren untersucht. Bei Katzenartigen können sie durch Ernährungsstörungen in der Jugend (Gass 1976; SLUSHER et al. 1965) verursacht werden. An den Knochen der Beutestücke fanden wir die Zahnspuren der Luchse, sofern die Reste nicht nachträglich von Fuchs, Wildschwein oder Mäusen benagt worden waren. Wenn der Luchs auf große Knochen beißt, hinterlassen die scharfen Spitzen der oberen Zähne P° undP* Scharten oder Abdrücke (Abb. 2); nach wiederholtem Zubeifßßen sind mehrere Paare solcher Spuren zu erkennen. Der Gegen- biß der unteren Zahnreihe erzeugt keine verwertbaren Spuren; das wurde im Zoo Wuppertal bei Fütterungsversuchen festgestellt und am Gebiß von drei toten Tieren nachgeprüft. Wir ermittelten den Reißzahnabstand Z an allen frischen Beutestücken, wo der daran fressende Luchs bekannt war. Die gleichen Maße stellten wir auch an sieben älteren, z. T. erst im Sommer gefundenen Rißresten fest. So ließ sıch in jedem Falle entscheiden, ob es der Luchskuder (Z = 15,0 mm) oder eine der beiden Luchsinnen (Z = 11,5 mm) war. Letztere konnten an den Zahnspuren leider nicht unterschieden werden. Abb. 2. Zuordnung von Zahnspuren. a: Schädel des Luchses. Zum Zerbeißen von großen Knochen werden die Prämolaren eingesetzt; b: Rechter Unterkiefer eines Rehbocks mit Zahnspuren vom Luchs; c: Schädelrest eines Rotwildkalbs mit Bißspuren vom Luchs am Stirnrand. Z = Abstand der Zahnspitzen von P? und P* Unterscheidung der Geschlechter Im Sommer und in der ersten Winterhälfte müssen Spuren von einem Erwachsenen mit Jungen immer von einer führenden Luchsin stammen. Der Kuder wurde nie zusammen mit Jungtieren festgestellt; nach seinen Fuß- und Zahnspuren zu urteilen, war er besonders groß. Im Freigehege beobachtete Jonsson (briefl. Mittlg. 1981), daß der Luchskuder in der Ranzzeit Aktionsräume und Verteilung erwachsener Luchse 13 manchmal grüne Triebe von Fichten zerbeißt; wir sahen das gleiche in Zoologischen Gärten. Im März 1979 sah Forstamtsrat H. SCHWARZ auf einer Fährte des Kuders, daß dieser Tannenspitzen zerbissen hatte, und im März 1981 fanden wir in einer Losung von ihm abgebissene grüne Tannennadeln. Ganz frische Losungen haben einen Überzug vom Schleimsekret aus dem Analbeutel; es kann abgeschabt werden und nach Trocknung wieder aufquellen. Dieses Sekret ist beim Männchen dunkelbraun, beim Weibchen gelb. Der Unterschied wurde zuerst von Baıtey (1974) für den Rotluchs (Lynx rufus) in Idaho/USA beschrieben; wir konnten ihn an den Luchsen im Bayerischen Wald und auch ım Zoo Wuppertal bestätigen. Im Bayerischen Wald spritzten erwachsene Luchse beiderlei Geschlechts ihren Harn gegen Baumstümpfe, Wurzelteller oder kleine Fichten (HuchHr et al. 1984). Aber nach Zoostudien von MATJuscHkIn (1978) und SCHROETER (1981) reibt nur der männliche Luchs gelegentlich seinen Kopf an der frischen Harnmarke. Wir konnten ım März 1980 solches Verhalten des Kuders vor einem Baumstumpf aus Spuren rekonstruieren. Einen weiteren Unterschied beobachtete ZELTUCHIN (1984) auf den Valdaj-Höhen/Nordwestrußland. Neben Liegeplätzen wurde auch in den Schnee geharnt; meistens war die Spritzspur der Luchsin kürzer und breiter als beim Kuder, weil dieser den Harn kräftiger und weiter spritzt. Darauf sollte künftig bei Freilandstudien geachtet werden. Daten von den Forstverwaltungen Für die Zeit vor 1979 und nach 1981 wurden die eigenen Beobachtungen durch Daten von den örtlichen Forstverwaltungen ergänzt. Das Staatliche Forstamt Zwiesel sammelte von 1972 bis zum Frühjahr 1979 alle Hinweise auf Luchsfährten, Risse und Sichtungen in jährlichen Dienstberichten. Die Angaben vor dem Winter 1976/77 lassen sich jedoch nicht bestimmten, auf Dauer angesiedelten Individuen zuordnen. Das Staatliche Forstamt Bodenmais und die Privatverwaltungen teilten uns einzelne Feststellungen noch bis 1985 mit. Viele Daten haben wir durch persönliche Rückfrage I pol. Nur Meldungen von zuverlässigen Beobachtern, meistens Forstbeamten, wurden berück- sichtigt. Ergebnisse Nur drei erwachsene Luchse hatten sich nach den Aussetzungen im Hinteren Bayerischen Wald auf Dauer angesiedelt. Ein Kuder, also ein männlicher Luchs, lebte ständig am Falkenstein und in dessen Umgebung. Südöstlich davon bis zum Rachel hielt sich ein Weibchen auf, im folgenden als Südost-Luchsin bezeichnet. Das dritte Individuum zwi- schen Falkenstein und Arber, ebenfalls ein Weibchen, nennen wir Nordwest-Luchsin (Abb. 3). Kerngebiet und Aktionsraum des erwachsenen Luchskuders Die Aktivitäten des Kuders konnten wir besonders gründlich beobachten. Wir ermittelten 34 Lagerplätze, die er regelmäßig oder gelegentlich benutzte. Alle befanden sich unter Felswänden oder großen Felsblöcken im Hangbereich zwischen 800 und 1150 m Höhe; je nach Form und Exposition waren sie bei verschiedenen Wetterlagen oder zu bestimmten Jahreszeiten besonders geschützt (HucHT und ELsTRODT 1983). Weiter war sein Aktions- raum gekennzeichnet durch 13 Beutestücke, die eindeutig von diesem Luchs gerissen wurden, und durch drei Stellen mit Spuren von mißlungenen Angriffen. Die tagsüber aufgesuchten Lagerplätze lagen also im oberen Hangbereich. Während des Winters blieben die wichtigsten Beutetiere Reh und Rothirsch aber am Unterhang und in den Tälern; dort wurden in jenen Jahren Wildfütterungen unterhalten, und dort setzte die Behinderung durch tiefen Schnee später ein als in den Hochlagen. Dorthin ging der Luchs bei Dunkelheit zur Jagd und zum wiederholten Fressen an Beutestücken. Diese räumliche Trennung von Ruhebereich und Jagdgebiet war sicher nicht im Verhalten des Tieres begründet, sondern von der Landschaft und durch die Wildhege vorgegeben. Sie bot während der Wintermonate günstige Voraussetzungen, um verschiedene Aktivitäten des Luchses zu unterscheiden. Zwischen seinen Lagerplätzen, dem Jagdgebiet und wiederholt besuchten Beutestücken ging dieser Luchs viele verschiedene Wege, doch wählte er bestimmte Strecken öfter als 14 G. Zachariae, W. Elstrodt und Ingrid Hucht-Ciorga andere. Außerdem wurden lange Fährten festgestellt, welche in ganz andere Richtungen führten und offenbar das Hinüberwechseln von einem Lagerplatz zu einem anderen, das Aufsuchen von anderen Wildfütterungen und Jagdchancen oder die Rückkehr von dort waren. Um die so regelmäßig durchstreifte Fläche zu erfassen, benutzten wir das 1 km weite Gitternetz der Topographischen Karte 1:25000. Jedes Quadrat wird gezählt, in welches Belege für dieses Individuum fielen (Abb. 3). Alle Funde, auch wenn es Rißreste, Losun- gen, einzelne Trittsiegel oder nur kurze Fährtenstücke waren, bedeuteten ja Aufenthalte, Hin- und Rückwege des Tieres und damit Aktivitäten nicht an einem Punkt, sondern in einer Fläche. »._./® . Ü Ne le zur 4 = BR —Q v. = . v N © ERROR ES v Bayer. Due GER CAR YA ( Eisenstein < BER N NL) / Jal EN SG Rs 1453 v u e4 Zwiesel Bodenmais 2se.. Lüchskuder ee NW-Luchsin aa SO-Luchsin === NW- oder SO-Luchsin Abb. 3. Kerngebiete der drei erwachsenen Luchse 1976/77-1983/84. Inset: Bundesrepublik Deutsch- land. Weitab gelegene Planquadrate mit einzelnen Nachweisen wurden in die Flächenberechnung nicht einbezogen. (Weitere Erläuterung im Text) Das so erfaßte Gebiet ist eine zusammenhängende Fläche von 33 km? (Abb. 3). Der gesamte von diesem Luchs bestrichene Raum muß jedoch größer sein. In der Winterzeit, als genaue Kontrolle möglich war, war er mehrmals eine Woche oder länger nicht feststellbar und hatte sich vermutlich weiter vom Kerngebiet entfernt. Viele Beobachtungen belegen, daß er Strukturen und Lagebeziehungen ın seinem Aktionsraum genau kannte. So überschritt er am 21.2. 1981 in der Nähe des Dortes Spiegelhütte zunächst einen schmalen Bachlauf, indem er eine Stelle mit flachem Ufer wählte und auf ein Schneebrett und Steine im Wasser trat. Um aber einen zweiten, breiteren und tief eingeschnittenen Bach zu überqueren, machte er einen weiten Umweg zur Straßenbrücke. Auf dem Weg dorthin mußte er die letzten 100 m in tieferem Gelände gehen und konnte wegen hoher Schneewehen Straße und Brücke gar nicht sehen; dennoch ging er gezielt auf die richtige Stelle zu. Aktionsräume und Verteilung erwachsener Luchse 5 Unsere letzten Nachweise für diesen Kuder sind drei Losungen vom Winter 1983/84, die wir bei von ihm bevorzugten Lagerplätzen am Falkenstein fanden. Seine Pfoten- und Gebißmaße, die Einsinktiefe im Schnee und die Höhe seiner Harnmarken über dem Boden bewiesen, daß er ungewöhnlich groß und schwer gewesen ist. Kerngebiet und Aktionsraum der Südost-Luchsin Ein weiblicher erwachsener Luchs bewegte sich in dem langgestreckten Waldgebiet zwischen Falkenstein und Rachel (Abb. 3). Die höchsten Bergkämme, auf welchen die Grenze zur Tschechoslowakei verläuft, bilden nach Nordosten einen natürlichen Abschluß; von dort fallen die Hänge vorwiegend nach Westen und Südwesten ab. Auch hier gibt es im klimatisch günstigen Hangbereich zwischen 800 m und 1150 m überhän- gende Felsen und Blockströme. Dieser Luchsin konnten wir nicht so oft nachgehen wie dem Kuder, weil wir unsere Arbeit auf dessen Kerngebiet konzentrierten. Aber seit dem Winter 1967/77 belegten die Riß- und Fährtenmeldungen vom Staatsforst und aus den angrenzenden Privatbezirken stets nur einen erwachsenen Luchs im genannten Gebiet. Schon im Februar 1979 machte Forstamtmann G. REINWALD Fotos von Luchsfährten, die am abgedrückten Karpalballen sicher identifiziert werden konnten. Wir selbst fanden Fährten, Spuren von zwei mißlunge- nen Angriffen, drei Risse bzw. Beutereste, mehrere Lagerplätze und Losungen, die eindeutig dieser Luchsin zuzuordnen waren. Gelegentlich jagte sie auch am Falkenstein; dort konnte ıhr ein erfolgloser Angriff auf ein junges Stück Rotwild nachgewiesen werden. Zahlreiche eindeutige Nachweise sind über das ganze Gebiet verteilt, ebenso die Beobachtungen eines erwachsenen Luchses ohne Identifizierung; sie dürfen mit großer Wahrscheinlichkeit diesem Individuum zugerechnet werden. Dennoch bleiben Lücken zwischen den belegten Quadraten. In Abb. 3 ist die engste Umrandung gezeichnet; das ergibt eine Fläche von 32 km?. Zwei weit abgelegene Nachweise wurden nicht einbezogen; sie bedeuten, daß sich auch diese Luchsin manchmal weit von ihrem bevorzugten Kernge- biet entfernte. Die letzten Hinweise auf sie sind Reste eines Rehs mit Reißzahnspuren vom Winter 1982/83 und frische Fährten im Januar 1983. Im Falkensteingebiet überschnitten sich also die Aktionsräume des Kuders und dieser Luchsin. Offenbar kannte auch sie dort Strukturen und Lagebeziehungen genau. Wahr- scheinlich fand im Überlappungsgebiet die Paarung statt; Beweise dafür fanden wir nicht. Aber jeweils im Februar/März 1980 und 1981, also zur Paarungszeit, erschien diese Luchsin am Falkenstein. Am 4./5. 3. 1981 besuchte sie hier auf einem Rundweg fünf geschützte Lagerplätze und hinterließ drei Harnmarken an diesen und acht weitere an Jungfichten, Baumstümpfen und Zaunpfählen. Solches Verhalten könnte den Kontakt zum Geschlechtspartner herstellen. Die Auswahl betraf nämlich solche geschützten Stellen, die nach Lage und Mikroklima besonders günstig waren und auch vom Kuder oft benutzt wurden. BaıL£y (1974) berichtet aus Idaho/USA, daß weibliche Rotluchse (Lynx rufus) dıe Höhlen mit Urin markieren, wo sie ihre Jungen ablegen. Solcher Zusammenhang könnte auch in unserem Fall bestehen. Wirklich fanden wir an den Lagerplätzen in diesem Gebiet während der Sommer 1980 und 1982 Fußabdrücke von Jungluchsen neben solchen von der Größe der Luchsin. Kerngebiet und Aktionsraum der Nordwest-Luchsin Eine weitere erwachsene Luchsin lebte im Raum nordwestlich des Falkenstein. In ihrem Kerngebiet befanden sich die Süd- und Westhänge des Arber; auch dort gibt es ın 800-1150 m Höhe freistehende Felsen und Blockströme mit einem vielfältigen Angebot an geschützten Plätzen. Diese Luchsin ist durch zahlreiche Fährten und Risse nachgewiesen. In den Jahren 1977-1979 wurde sie sogar achtmal allein und viermal zusammen mit Jungen 16 G. Zachariae, W. Elstrodt und Ingrid Hucht-Ciorga bei Dämmerlicht gesehen. Fünfmal erfolgten solche Beobachtungen zur gleichen Zeit, zu welcher auch die Südost-Luchsin auf der entgegengesetzten Seite des Falkenstein festge- stellt wurde. Damit ergaben sich zusätzliche Beweise für die Existenz beider Individuen. Auch diese Luchsin bewegte sich meistens außerhalb unseres Arbeitsgebietes; darum konnte der von ihr bevorzugte Raum nur lückenhaft erfaßt werden. Aber alle Meldungen der staatlichen und privaten Forstverwaltungen und unsere eigenen Daten ließen dort auf nur einen erwachsenen Luchs schließen. Wird wie oben die engste Umrandung um die Nachweise gezeichnet, so ergeben sich 33 km? (Abb. 3). Dabei wurden fünf weit abgele- gene Stellen - zwei nach Nordosten, eine nach Südwesten und zwei außerhalb der Karte im Westen — nicht einbezogen; hier hatte sie sich aus ihrem bevorzugten Kerngebiet entfernt. Im Februar 1982 war ihr ein gerissener Frischling zuzurechnen; das ist bemerkenswert, weil Wildschweine (Sus scrofa) selten Beute der Luchse waren. Noch in den Wintern 1984/ 85 und 1985/86 wurden vom Arber einzelne Luchsfährten gemeldet; wir konnten jedoch nicht dort sein und darum nicht feststellen, ob das diese Luchsin oder ihr Nachwuchs war. Diese Luchsin bewegte sich zeitweise auf der Westseite vom Falkenstein; sie ging also ın das Kerngebiet des Kuders hinein. Das Entsprechende wurde oben schon von der Südost- Luchsin berichtet. Leider ließ das Fundmaterial keinen Unterschied in Pfoten- und Gebißmaßen zwischen den beiden weiblichen Tieren erkennen. Auch die in zwei Sommern in Lagerplätzen am Falkensteinsüdhang gefundenen Spuren einer Luchsin mit Jungen waren nie so tief eingedrückt, daß die Fußanomalie der Südost-Luchsin hätte erkannt werden müssen; sie ließen sich darum nicht identifizieren. Die Lokalitäten (drei gestri- chelte Quadrate in Abb. 3) liegen zwischen den Gebieten mit eindeutigen Nachweisen für die beiden Luchsinnen. Unklar bleibt also, ob es dazwischen eine scharfe Grenze gab oder ob die eine wenigstens zeitweise auch in das Gebiet der anderen vordrang. Sicher ist jedoch, daß ihre eigentlichen Kerngebiete durch das Falkensteinmassiv getrennt waren. Diskussion Raumansprüche und Verteilung der Luchse im Hinteren Bayerischen Wald Die drei erwachsenen Luchse haben mindestens sieben Jahre in den angegebenen Räumen gelebt. Dieses Ergebnis beweist erstmalig die langjährige Bindung an ein geeignetes Wohngebiet; vermutlich wird es zeitlebens beibehalten. Nur ZeLtuchın (1984) stellte auf den Valdaj;-Höhen/Nordwestrußland bestimmte Luchse über drei Winter in denselben Gebieten fest. In Idaho/USA konnte BAıLEy (1980) individuell gekennzeichnete Rotluchse (Lynx rufus) drei Jahre und in einigen Fällen 7-10 Jahre in derselben Gegend bestätigen; auch dieser Autor nimmt an, daß sıe dort bis zum Lebensende bleiben. Die Auswertung unserer Beobachtungen und Daten ergibt gewiß nicht den maximal bestrichenen Raum. Wahrscheinlich machten die Luchse gelegentlich noch weitere Wege. Es gibt aber keine Beobachtungen außerhalb des geschlossenen Waldes. Zu überlegen ist, welche Bedeutung solche durch häufige Nachweise belegten Mindestflächen haben. Unter ‚Revier‘ oder ‚Territorium‘ versteht man ein gegen Artgenossen verteidigtes Wohngebiet; es kann durch Markierungen gekennzeichnet oder begrenzt sein. Wir fanden bei unseren drei benachbarten Luchsen jedoch keinen Hinweis auf eine aktive Reviervertei- digung. Zur gleichen Feststellung kommt BaıLEy (1974, 1980) beim Rotluchs. Der von uns gründlich beobachtete Kuder hinterließ Losungen und Harnmarken dort, wo er sich am häufigsten bewegte (HucHT et al. 1984). Seine gelegentlichen Fernwege führten vom bevorzugten Kerngebiet weg oder dahin zurück. Solche Fährten verliefen niemals so, daß sie Grenzen einer Fläche hätten sein können. Wir verwenden darum die neutrale Bezeich- nung „Aktionsraum“ (engl. home range) für das ganze vom Luchs durchstreifte Gebiet ım Sinne von IMMELMANN (1975) und HEyMER (1977). Unsere Suche nach Fährten, Beutestücken, Losungen, Harnmarken, Lagerstellen und Aktionsräume und Verteilung erwachsener Luchse 17 anderen Spuren mußte besonders dort Erfolg haben, wo sich das betreffende Tier am häufigsten aufhielt; das gleiche gilt für die Meldungen der Forstbeamten. Mit dieser Arbeitsweise werden vom Luchs seltener besuchte Gebiete des Aktionsraumes nur zufällig erfaßt. Andererseits wird hierbei ein bevorzugtes Kerngebiet erkennbar. Im Fall des Kuders war das eine geschlossene Fläche von 33 km?; in jedem Teil davon wurde er wiederholt festgestellt und identifiziert (Abb. 3). Die Auswertung für die Luchsinnen ergibt ebenfalls Gebiete, wo sie sich bevorzugt aufhielten, und wo darum die Wahrscheinlichkeit für Sichtungen und Spurenfunde am größten war. Sie hatten die gleiche Größe wie das Kerngebiet des Kuders. Aber einige Nachweise mit sicherer Zuordnung liegen weit entfernt von den Flächen, aus welchen die Mehrzahl der Meldungen kam (Abb. 3). Wahrscheinlich wurden dort Aufenthalte im Randbereich des Aktionsraumes erfaßt. Den Südhang des Falkenstein hat vielleicht nicht nur die Südost-Luchsin, sondern auch die Nordwest-Luchsin zeitweise besucht; das ließ sich leider nicht klären. Die Annahme eines Kerngebietes und das Überlappen der Aktionsräume beim Luchs werden durch Veröffentlichungen aus anderen Ländern gestützt. Auf den Valdaj-Höhen ermittelte ZELtucHın (1984) die Aktionsräume von Luchsen nach Fährten im Winter, Lagerplätzen und Angriffen auf Schneehasen, welche dort die häufigste Beute sind. Die Luchse wurden an Trittsiegeln identifiziert. Zwei Männchen und ein Weibchen wurden drei Jahre lang beobachtet. Die Aktionsräume überlappten sich; die maximal durchstreif- ten Gebiete waren 250, 130 und 70 km? groß, aber in einem Kerngebiet (russ. korennoj rajon) von 30 km? bewegten sie sich regelmäßig, in der peripheren Zone nur selten. DanıLov et al. (1979) unterscheiden bei Luchsen in Nordwestrußland ein Kerngebiet von 25-45 km? und eine seltener besuchte periphere Zone mit Überlappung bei benachbarten Tieren. Hacıunn (1966) beschreibt sogar weitgehendes Überlagern der Streifgebiete von einem Kuder und einer Luchsin während eines Winters ın Südschweden. Entsprechende Ergebnisse liegen vom Kanadaluchs (Lynx canadensis) vor. SAUNDERS (1963) ermittelte in Neufundland für zwei Männchen 18,1 und 20,7 km?, für eine Junge führende Luchsin 15,5 km? nach Fährten im Schnee und Fallenrückfang der markierten Tiere im Sommer. Fernwege zur Paarungszeit wurden nicht mitgerechnet, also nur bevorzugte Gebiete erfaßt. Parker et al. (1983) bestimmten auf Cape Breton Island/ Neuschottland mit Radiotelemetrie die Aktionsräume von zwei Erwachsenen, Männchen und Weibchen; es waren im Winter 12,3 und 18,6 km?, im Sommer 25,6 und 32,3 km?. Nur im Winter erkannten die Autoren ein Kerngebiet (engl. core area) von 5,3 bzw. 6,6 km’, aus welchem 75 % aller Peilungen kamen. Die niedrigen Zahlen mögen durch Artunter- schiede und Landschaft bedingt sein; interessant ist, daß im Sommer kein Kerngebiet bevorzugt wurde und daß die Aktionsräume bei gleicher Lage im Sommer größer als im Winter waren. Vielleicht gilt ähnliches auch in Europa; entsprechende Daten fehlen bisher. Von den drei im Bayerischen Wald angesiedelten Luchsen ist nach unseren Spurenfunden nur sicher, daß sie sich zu allen Jahreszeiten in den Gebieten aufhielten, die in Abb. 3 dargestellt sind. Für den Luchskuder war die Felsenzone am Süd- und Westhang des Falkenstein jahrelang das Zentrum, von wo aus er seinen Aktionsraum durchstreifte (HucHT und ELsTRopT 1983). Auch die beiden Luchsinnen hielten sich, mit und ohne Junge, zeitweise dort auf. Aber sie kehrten immer wieder zurück nach Nordwesten zum Arber bzw. nach Südosten an die Hänge oberhalb der Ortschaften Spiegelhütte und Buchenau bis zum Nordwesthang des Rachel. Auch dort gibt es, wie erwähnt, unter Felswänden und Wanderblöcken geschützte Stellen für jede Wetterlage und Windrichtung. Jeder der drei Luchse hatte also in seinem Kerngebiet eine große Auswahl solcher Plätze, die als Lager geeignet waren. Gewiß war das Vorhandensein von Beutetieren in der Landschaft eine Voraussetzung. Aber nicht das Beuteangebot, sondern die Komfortansprüche der drei erwachsenen Luchse 18 G. Zachariae, W. Elstrodt und Ingrid Hucht-Ciorga hatten offenbar bestimmt, wo sie sich auf Dauer ansiedelten. Ihre Kerngebiete und daran anschließende Zonen enthielten sowohl Bereiche, die von den Beutearten Reh und Rothirsch kaum aufgesucht wurden, als auch deren bevorzugte Tageseinstände und Äsungsplätze. Besonders beim Rehwild gab es in jenen Jahren auch kleinräumige Ände- rungen, bedingt durch forstwirtschaftliche Maßnahmen, Einrichten oder Aufgeben von Winterfütterungen. Ein Einfluß auf Lage und Ausdehnung der von den Luchsen gewählten Kerngebiete war nicht zu erwarten, weil diese sich über größere Entfernungen erstreckten. Raumbedarf und Komfortansprüche des Luchses können gegenwärtig nur noch in dem geschlossenen Waldgebiet erfüllt werden, welches sich 7-9 km breit längs der Landes- grenze erstreckt (Abb. 3). Westlich davon sind ungünstigere Bedingungen. Der Wald ist durch Landwirtschaft, Ortschaften, Straßßen und Eisenbahn stark zergliedert. Felsen und einzelne Blöcke gibt es dort nur ım Bereich unter 800 m Höhe, wo im Winter oft Tauwetter mit Regen und Naßschnee herrscht. Andere Schutzplätze, etwa unter gestürzten Bäumen, gibt es in den forstwirtschaftlich genutzten Wäldern jedoch nicht. Aber auch im Osten, auf tschechischer Seite, sind Süd- und Westhänge selten, die Talsohlen schon 1000-1100 m hoch und damit oberhalb der optimalen Zone. Unverständlich ist, warum FEstErTics (1980, 1981) dieses Luchsvorkommen bis 8 km in die Tschechoslowakei hinein darstellt, obwohl von dort keine Meldungen vorliegen. Für Luchse günstige Bedingungen wiederholen sich nur nordwestlich des Arber, wo 1978/79 am Hohen Bogen (1079 m) ein Luchs aufgetreten sein soll, und südöstlich vom Rachel am Lusen (1370 m) im National- park; dort erfolgte aber keine Ansiedlung. Die drei von uns beobachteten Luchse waren ım Winter 1976/77 bereits erwachsen. Sie wurden demnach mindestens 8-9 Jahre alt und haben vermutlich ein natürliches Lebens- ende erreicht. Vielleicht waren ım Bayerischen Wald noch mehr Luchse ausgesetzt worden; Hinweise darauf oder auf die in einigen Jahren aufgezogenen Jungtiere gab es auch außerhalb jenes optimalen Gebietes. Ihr Verbleib ist unbekannt, aber für die ın Populärschriften behaupteten Abschüsse durch Jäger fehlen ebenfalls Beweise. Vor Plänen zur Wiedereinbürgerung dieser Raubtierart wäre dringend nötig, die biologischen Voraus- setzungen für ihre Erhaltung zu klären. Ergebnisse aus anderen Ländern und Einfluß unterschiedlicher Methoden Als Kerngebiet bezeichnen wir den Landschaftsausschnitt, wo sich ein Luchs über mehrere Jahre regelmäßig aufhielt. Dort häuften sich die Nachweise durch Fährten, Beutereste, Losungen, Harnmarken, Lagerplätze und sogar manchmal Sichtungen. Nur zu diesem Gebiet hatte das Tier offenbar eine engere Bindung; darum sollte nur hier diskutiert werden, welche Gründe für seine Ansiedlung ausschlaggebend waren. Wie oben gezeigt, war das Angebot an Lagerplätzen unter Felsen ein entscheidendes Kriterium. Mit gleichen Methoden kommt ZELTucHIn (1984) in der ebenen Landschaft der Valdaj-Höhen/Nord- westrußland zu einem ähnlichen Ergebnis. Die Kerngebiete der Luchse sind dort etwa 30 km? groß und enthalten stets Windbruchflächen, in welchen sie geschützte Lagerplätze finden. Diese Arbeitsweise läßt also Zusammenhänge zwischen dem Verhalten der Tiere und Landschaftsstrukturen erkennen (HucHT und ELsTRODT 1983; HucHT et al. 1984); es ist im Bergland jedoch schwierig, die gelegentlichen Fernwege zu verfolgen. Einige Autoren versuchen, die Fläche des gesamten Aktionsraumes zu erfassen. Hacıun (1966) gibt aufgrund mehrtägiger Abfährtungen während eines Winters für einen Luchskuder in Südschweden ein „Revier“ von mindestens 300 km? an. Nicht geprüft wurde, ob jenes Gebiet wiederholt durchstreift wurde; auch mag die Auswertung mit einem Netz von 25 km? großen Quadraten ungenutzte Bereiche erfaßt haben. Jonsson (1980) berechnete für zwei Luchsinnen in Mittelschweden sogar Aktionsräume von 625 und 2000 km? Größe. Hier wurden vielleicht einmalige Erkundungswege einbezogen. Die Radiotelemetrie erlaubt, langfristig die Bewegungen von Luchsen und auch ihre Fernwege Aktionsräume und Verteilung erwachsener Luchse 19 zu kontrollieren. Aus einem solchen Arbeitsprogramm im Berner Oberland/Schweiz rechnen HALLER und BREITENMOSER (1984) mit 200-300 km? Aktionsraum. Zur Beurteilung solcher Zahlen muß man die wirkliche Form eines Aktionsraumes beim Luchs berücksichtigen. BaıLEy (1974) und ZeLrtucHın (1984) betonen, daß der periphere Teil oft aus einzelnen Lokalitäten besteht, die nur in größeren Zeitabständen besucht werden. Die Geländepartien dazwischen und längs der Wege dorthin können für das Tier bedeutungslos sein. Durch gerades Verbinden der äußersten Nachweispunkte erhält man aber eine kontinuierliche Fläche; sie wird größer sein als die, welche der Luchs wirklich durchstreift und deren Strukturen und Lagebeziehungen er im Gedächtnis speichern kann. Auch wird die vereinfachende Umrandung bei benachbarten Tieren mehr Überschneidung vortäuschen als der Wirklichkeit entspricht. Viele ältere Veröffentlichungen geben den Bestand von Luchsen in bestimmten Land- schaften an. Gewöhnlich wurde bei Schnee die Zahl der unabhängigen Fährtensysteme festgestellt und durch diese die untersuchte Gesamtfläche dividiert. Dabei wurden nicht nur die auf Dauer angesiedelten Luchse, sondern auch Durchzügler und Jungtiere erfaßt. Solche Erhebungen gibt es für viele Teile der Sowjetunion (HEPTNER und SrupskıJ 1980; MATJuScHKIN 1978) und auch aus der Slowakei (NovAkovA und Hanzı 1968). Es werden Zahlen zwischen 0,1 und 1,0 Exemplaren auf 10 km? angegeben. Sie sind produktionsbio- logisch von Interesse und haben praktische Bedeutung für die Pelztierjagd. Falsch ist jedoch, für alle diese Tiere eigene, exklusive ‚Reviere‘ anzunehmen und deren Größe aus solchen Zahlen zu berechnen. Behauptungen, daß Luchse sich auf 10 km? und weniger beschränken könnten, sind vermutlich so zustande gekommen. Auch die Schlußfolgerung, „daß sich die Reviergröße umgekehrt proportional zur Bestandsdichte der Art verhält“ (MATJuschkIn 1978, S. 91f.), ist sicher unrichtig. Alle diese Methoden liefern also sinnvolle Ergebnisse, aber zu ganz verschiedenen biologischen Aspekten. Sie dürfen nicht unkritisch miteinander verglichen werden. Darum sollte künftig bei der Diskussion um Raumansprüche des Luchses der Zusammenhang zwischen ermittelten Daten, eingesetzten Methoden und verfolgter Fragestellung beachtet werden. Danksagungen Für die Unterstützung unserer Freilandarbeit danken wir dem damaligen Leiter des Staatlichen Forstamtes Zwiesel, Herrn Forstdirektor E. ENGELSTÄDTER, und den Revierbeamten Forstamtmann R. GAEBEL, Forstamtmann H. THoMA, Forstoberinspektor L. TISCHLER, Forstamtmann G. ReEın- WALD, Forstamtsrat H. SCHWARZ und Forstoberinspektor E. ZITZELSBERGER. Zusätzliche Hinweise verdanken wir dem Staatlichen Forstamt Bodenmais, dem Fürstlich Hohenzollern’schen Forstamt Bayer. Eisenstein, der Gutsverwaltung Freiherr von Poschinger Oberfrauenau und der Gräflich Mellin-Wolffersdorff’schen Gutsverwaltung Oberzwieselau. Herr E. MEININGHAUS, Technischer Sachverständiger beim Landeskriminalamt Düsseldorf, gab uns Anleitung zur Spurensuche. Für die Arbeitsmöglichkeit im Zoologischen Garten Wuppertal danken wir Herrn Direktor Dr. G. Haas und Herrn Dr. U. SCHÜRER, für praktische Hilfe dem Tierpfleger Herrn U. Schmipr. Der Technischen Assıstentin Frau E. PoGGEL danken wir für Mitarbeit bei den Laboruntersuchungen. Zusammenfassung Bei drei erwachsenen Luchsen im Hinteren Bayerischen Wald wurden räumliche Bewegungen und Ansprüche an Landschaftsstrukturen untersucht. Ihre Aktivitäten wurden aus Fährten im Schnee, einzelnen Spuren im Sommer, gerissenen Beutetieren, Losungen, Harnmarken und Be erschlossen; Sichtungen waren selten. Die Tiere wurden nach den Maßen von Trittsiegeln und Zahnspuren an Beuteresten und nach Spuren geschlechtsbezogener Verhaltensweisen identifiziert. Frische Beutestücke wurden mit Kunststoffgranulat präpariert, was markierte Losungsfunde an entfernten Orten ermöglichte. Die Aktionsräume von zwei Luchsinnen überlappten sich mit dem eines Kuders. Sie bestanden aus einem bevorzugten Kerngebiet von etwa 30 km? und einer peripheren Zone. Die Lage der Kerngebiete war vom sehr an geschützten Lagerplätzen bestimmt; die Luchse lebten dort mindestens 7 Jahre lang. Studien aus anderen Ländern stützen die Annahme eines Kerngebiets solcher Größe ım Aktionsraum. In der Literatur werden die Raumansprüche des Luchses jedoch auch viel geringer oder größer angegeben. Die Beziehung zwischen Methoden und Ergebnissen wird diskutiert. 20 G. Zachariae, W. Elstrodt und Ingrid Hucht-Ciorga Literatur Baıtey, T. N. (1974): Social organızation in a bobcat population. J. Wildl. Manage. 38, 435-446. BaıLey, T. N. (1980): Factors of bobcat social organisation and some management implications. Worldwide Furbearer Conf. Proc. Ed. by J. A. CHapman and D. Pursrey. Frostburg: Univ. Maryland. Vol. 2, 984-1000. DanıLov, P. I.; Rusakov, O. $.; Tumanov, I. L. (1979): Chiscnye zveri severo-zapada SSSR. (Die Raubtiere der nordwestlichen USSR.) Leningrad: Izd. Nauka. ELSTRODT, W.; HUCHT, I.; ZACHARIAE, G. (1983): Aktionsräume erwachsener Luchse, Lynx Iynx (Linnaeus, 1758), und Belege für Jungenaufzucht ım Hinteren Bayerischen Wald. Dtsch. Ges. Säugetierkunde, 57. Hauptvers. Bonn, 6. FEsTETICS, A. (1980): Die Wiedereinbürgerung des Luchses in Europa. In: Der Luchs in Europa. Hrsg. von A. FEsTETIcs. Greven: Kilda. 224-254. FEstETics, A. (1981): Das ehemalige und gegenwärtige Vorkommen des Luchses, Lynx Iynx (Linne, 1758), ın Europa und seine Wiederansiedlung in einigen europäischen Ländern. Säugetierkdl. Mitt. 29 (1), 21-77. FRANCK, R.; MÜHLSCHLEGEL, H. (1966/68): Kunststoffe im Lebensmittelverkehr. Empfehlungen der Kunststoff-Kommission des Bundesgesundheitsamtes. Hrsg. von R. FRANck. Lief. 5: Polystyrol. Lief. 9: Farbmittel zum Einfärben von Kunststoffen. Köln: Heymann. Gass, H. (1976): Katzen, Schleichkatzen, Marder. In: Zootierkrankheiten. Hrsg. von H.-G. Krös und E. M. Lanc. Berlin und Hamburg: Paul Parey, 92-112. HaAGLunD, B. (1966): De stora rovdjurens vintervanor. I. Viltrevy 4, 81-299. HALLER, H.; BREITENMOSER, U. (1984): Der Luchs. Schweiz. Naturschutz, Basel, Jg. 1984 (2), 2-30, und Panda/WWF Schweiz, Zürich, 17 (1), 2-31. " Heıı, P.; SLADER, J. (1980): Ergebnisse der Luchsforschung in der CSSR, II. Teil. Über die Variabilität der Schädel- und Körpermaße des Luchses (Felix Iynx L., 1758) ın den Westkarpaten. Beitr. Jagd Wildforsch. 11, 260-275. HEPTNER, V. G.; SLupdsk1J, A. A. (1980): Raubtiere (Feloidea). In: Die Säugetiere der Sowjetunion. Hrsg. von V. G. HEPTNER und N. P. Naumov. Jena: VEB Gustav Fischer. Vol. 3. HEYMER, A. (1977): Ethologisches Wörterbuch. Berlin und Hamburg: Paul Parey. HuchHr, 1.; ELstRoDT, W. (1983): Nutzung von Lagerplätzen beı einem freilebenden Luchs, Lynx bynx (Linnaeus, 1758), im Hinteren Bayerischen Wald. Dtsch. Ges. Säugetierkunde, 57. Hauptvers. Bonn, 14. HuchHr, I.; ELSTRODT, W.; ZACHARIAE, G. (1984): Beobachtungen zum Harnmarkieren beim Luchs, Lynx bynx (Linnaeus, 1758), in Freiland und Gehege. Dtsch. Ges. Säugetierkunde, 58. Hauptvers. Göttingen, 23-24. IMMELMANN, K. (1975): Wörterbuch der Verhaltensforschung. München: Kindler. Jonsson, S. (1980): Erforschung und Erhaltung des Luchses ın Schweden. In: Der Luchs in Europa. Hrsg. von A. FEsTETIcs. Greven: Kilda. 170-180. MATJUSCHKIN, E. N. (1978): Der Luchs. Neue Brehm-Büch. 517. Wittenberg Lutherstadt: Ziemsen. NovAkrovä, E.; Hanzı, R. (1968): Contribution ä la connaissance du röle joue par le lynx dans les communautes sylvicoles. Schweiz. Z. Forstwes. 119, 114-126. PARKER, G. R.; MAxwELL, J. W.; MorTon, L. D.; SMITH, G.E. J. (1983): The ecology of the lynx (Lynx canadensis) in Cape Breton Island. Can. ]J. Zool. 61, 770-786. SAUNDERS, J. K. (1963): Movements and activities of the lynx in Newfoundland. J. Wildl. Manage. 27, 390-400. SCHROETER, W. (1981): Beobachtungen an den Nordluchsen, Lynx lynx (Linne, 1758), der Tiergrot- ten und des Nordseeaquariums Bremerhaven. Säugetierkdl. Mitt. 29 (4), 8-9. SLUSHER, R.; BISTNER, $. I.; KIRCHER, C. (1965): Nutritional secondary hyperparathyroidism in a tiger. J. amer. vet. med. Ass. 147, 1109-1115. STEHLIK, J. (1979): Znovuvysazeni rysa ostrovida (Lynx lynx L.) v nekterych evropskych zemich v letech 1970-1976. (Wiedereinbürgerung des Luchses (Lynx bnx L.) in einigen europäischen Ländern ın den Jahren 1970-1976.) Folıa venat. (Praha) 9, 255-265. STOLLMANN, A. (1963): Beitrag zur Kenntnis des Luchses (Lynx lynx L.) in den tschechoslowakischen Karpaten. Folıa zool. (Praha) 12, 301-316. SuMINsKI, P. (1973): Rys. (Der Luchs.) Warzawa: Pan. Wydaw. Roln. Lesne. VasıLıu, G. D.; Deceı, P. (1964): Über den Luchs (Lynx bynx) der rumänischen Karpaten. Säuge- tierkdl. Mitt. 12, 155-183. WINEGARNER, M. S. (1985): Pugmarks and the biology of the bobcat, Lynx rufus. Z. Säugetierkunde 50, 166-174. ZELTUCHIN, A. $. (1984): O suto&noj aktivnosti i razmerach ulastkov obitanija rysı v juznoj tajge verchnevol2’ja. (Über die tägliche Aktivität und die Reviergrößen des Luchses in der südlichen Taiga um den Oberlauf der Wolga.) Bjul. mosk. O-va Ispyt. Prirody, Otd. biol. 89, 54-62. Anschrift der Verfasser: Prof. Dr. GERHARD ZACHARIAE, Stud.Rt. WERNER ELSTRODT, Dipl.-Biol. InGrıD HUCHT-CioRGA, Institut für Zoologie der Universität Düsseldorf, Universitätsstraße 1, D-4000 Düsseldorf Dynamics of reproduction and development of weight in the Wild boar (Sus scrofa) in South-west France By Dominique P£rın, F. Spitz, G. JANEAU and G. VALET Institut National de la Recherche Agronomique, Castanet-Tolosan, France Receipt of Ms. 25. 3. 1986 Abstract Studied was the weight and reproductive stage of wild boar in relation to age and food conditions. 133 data were obtained between 1978 and 1981, among which 65 (49 %) relevant to tagged animals. Age of most anımals was determined by reference to their lens weight at death. Results indicate that the body weight evolution is similar ın both sexes until they reach about 45 kg at 12-15 months. Males maintain a fast weight gain while a tendency to a step by step change is observed in females, as a consequence of their annual investment in breeding. After a big oak mast, adults are heavier and the Dh occur earlier: this last fact is the only responsible of heavier piglets observed at the next hunting season. Introduction In the domestic sow, puberty occurs at an age which is characteristic of each bred, provided weight growth has not been too much slower than the average (ETIENNE and LEGAULT 1974). If young sows are subject to important food deficiency just before puberty or even earlier, their body development is slowed down and theır first oestrus delayed (Duncan and LODGE 1960: in SADLEIR 1969; ZIMMERMAN et al. 1960; Du£E and ETIENNE 1974; ETIENNE et al. 1983). A more energetic diet given sows just before mating results ın a significant increase of ovulation rate (the so-called “flushing’”) as told by SORENSEN et al. (1961). In the wild sow, OLorr (1951) reports that many young females may breed at 8-10 months old when good oak or beech mast occurs but not when mast is poor. He also precises that the rate of reproduction increases in years of plenty. MATSCHkE (1964) made equivalent observations, speaking of the so-called “flushing effect”’ of food. HEnRY (1966), BRIEDERMANN (1971) and AUMATTRE et al. (1984) observe that an increase in food supply results in higher litter size. Food excess or deficiency can also be involved in the timing of anoestrus-oestrus transition in autumn (MAUGET 1980; AUMAITRE et al. 1982). Actually, while the domestic sow is continuously fertile with some fluctuations (MARTINAT-BOTTE et al. 1981, 1984), the wild sow is sexually seasonal, whatever the quantity of food offered. On the other hand, weight development in wild boar (male or female) is affected to a certain extent by food conditions in the wild (OrLorr 1951; Kozro 1975; KrEın 1984). Relationship of food, weight and reproduction in the wild boar has never been considered in detail. The present work is an attempt to analyze this complex dynamics, on the basıs of observations collected in the wild in South West France. U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5201-0021 $ 02.50/0 Z. Säugetierkunde 52 (1987) 21-30 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 22 Dominique Pepin, F. Spitz, G. Janean and G. Valet Material and methods Study area Gresigne country, located in adjacent parts of Tarn and Tarn-et-Garonne departements, extends over about 350 square km. It presents a hilly landscape varying ın altitude from 100 to more than 500 m above sea level. On an economical viewpoint, ıt ıs considered as ““marginal’”’ and qualified “piedmont” or “mountain”. The entire area is characterized by an important forest cover (more than 50 %), scattered farmland patches dominated by forage cultivation, and a large proportion of more or less recently abandoned rangeland. “Gaillac” Band anende extend over the extreme south of Gresigne. Center of Gresigne is occupied by a State Forest (35 square km). All woodlots in the country present a diversified understory, with several important evergreen species. A small number of plots are planted with conifers, mixed with dense scrub vegetation. Dense understory and scrubland are highl favourable to the establishment of wild boar resting places. Wild boars are also provided en abundant food resource: acorn, chestnuts, grapes, several species of wild fruit, herbs and legumes. During the study, an exceptional oak mast occurs in autumn 1978. Wild boar hunting ıs performed by specialized teams (hound hunting with posted shooters). Hunting effort ıs high, rather constant year after year and uniformly distributed. Hunting bag is estimated 200-250 for the 350 square km of Gresigne country. Wild boar population density has been estimated about 2 to 3 per square km in summertime. (SPITZ et al. 1984). Trapping and handling live animals Small (8 square m) and corral-type (500 square m) traps have been located in the State Forest. Anımals less than 30 kg are immobilized with a small net. Others are forced into a handling box, or chemically immobilized. For each anımal date, place, weight, sex, apparent reproductive state are recorded. In addition blood samples and ectoparasıtes are collected. An ear tag (numbered plastic tag) ıs placed. 87 ee have been ear Eee among them 15 received a radıo-collar. All are released at trappıng place. Ear-tag recovery and examination of dead animals At the onset of trapping work in Gresigne, a large information campaign was organized ın neighbouring departements of Tarn, Tarn-et-Garonne and Haute-Garonne, in newspapers, by direct contact with hunting teams and through hunting organızations and game keepers. Hunters were requested to record ıdentification number of the tag, sex and weight of tagged anımals they killed, place and date of the shot, and to collect eyes in formaline (containers wıth formalıne had been distributed prior to the hunting season). In addition to this large-scale survey, our team conducted regular observations on nk shot in Gresigne State Forest during the 78-80 seasons: for each anımal some additional informations were recorded (macroscopic examination of reproductive apparatus, stomach content, dentition, ectoparasites) and female reproductive tracts were collected and examined later. Ovarıes were weighed and observed with binocular microscope. Uterus was dissected in search of embryos and placental scars. Age of embryos was deducted from their length according to HENRY’s (1968) as Age determination Age classification in live anımals is usually based upon body size or general morphology. However DARDAILLON (1984) considers as known age animals only those piglets distinctly striped, and similarly Kreın (1984) places in this category only piglets less than 10 kg. For dead or live anımals, an usual technic is based on milk tooth eruption a replacement, according to a key published by MATSCHKE (1967). Age determination following this method is approximative and relies upon a subjective decision (SWEENEY et al. 1970). Following indications from QUERE (pers. comm.) we selected eye lens weight as an age criterion similarly to EL MAsTour et al. (1982) and DarvaıLLon (1984). All these authors propose approximately the same figures as those published by Sprrz (1984). This author’s results are summarized by the following formula: u me eG where y is age in days and x weight of both lenses in mg. When thıs technic is applied to an individual, uncertainty of this estimation is #1 month at 6 months, #2 at 12 months and + 4 at 2 years (Spitz 1984). When applied to large samples it gives a good approximation of the average birth date. This practically pe to the general aspect of age Weight and reproductive stage of wild boar 23 distribution (one or several peaks?) and allows to evaluate the average age (average birth date) corresponding to each peak. In the present a we consider as known age anımals those ones the lenses of which were weighed. Age at trapping is deducted from age at death. This basic set is completed by some individuals trapped as striped piglets among a famıly group (very small anımals of same weight and general appearance), but the lens of which were not collected at death. Treatment of informations about reproduction According to AUMAITRE etal. (1982), females are classified as “cyclic”’ or “inactive”’ following weight of ovary and empty uterus. Rythmicity of birth is deducted from the trapping rate of young piglets and from estimated age of dead anımals. Fertility (litter size) and embryonic losses are evaluated. Treatment of informations about weight gain General age distribution of anımals examined in Gresigne is presented in Table 1. Due to incertainty in individual age determination, anımals were grouped in successive age classes of increasing width. We first notice ın Table 1 that data concerning hunted animals are the most numerous (71,4 % of the total). Secondly, we observe that trapped animals are mostly piglets less than 5 months old (81,6 %), while hunted anımals are very few in this age class (6,3 %). Finally, the whole set of data can be Table 1 Distribution of age, sex and types of wild boars examined in the Gresigne State Forest between 1979 and 1981 ee Caught in traps ns untagged type Ag) type Bo) (month) 3 0 3 3 0 jo" +0 1-2 25 3-4 176-219 4-5 220-254 5216 255-283 6- 8 284-328 8-10 329-362 10-12 363-390 12-15 391-425 15-18 426-453 18-24 454-497 24-36 498-560 >36 >560 Total N er) are le 10 11 6 68 65 m ea | mm eNF | NWWON | Won we — [nv NW DNNDOUSH+I ON NS NWVunNDN | PowuwununNn-» (1) age determined by lens weight, (2) age at capture determined a posteriori by lens weight, (3) caught at the same time as another animal of the latter ee (4) age at GEBR + time S between capture and shooting, (*) anımal caught as a juvenile and shot 760 days after divided in 38 (28,6 %) individuals weighed at trapping, 27 (20,3 %) tagged individuals recaptured by shooting and 68 (51,1 %) untagged animals weighed at death. From 86 wild boars tagged in the 78-81 period, 39 (45 %) were recaptured (Spırz 1984). In addition to 21 animals the age of which was estimated by eye lens weight, the age of 6 males was evaluated. Finally, among 39 recaptured animals 27 have been successfully aged (69 %). For each age class male and female weights are compared by utilizing Student’s test when varıances are equal (F of FisHER-SNEDECOR) or the test of SUKHATME when variances are unequal. When unsignificant difference between sexes is observed, results from both sexes are grouped. 24 Dominique Pepin, F. Spitz, G. Janeau and G. Valet Results Reproduction Characters of genital tracts At infant stage, for a body mass of about 30 kg, average ovary weight is about 1 g. The same value is observed in females 1 to 2 years old when sexually inactive. In females older than 2 years ovarıes of sexually inactive individuals ıs a little bit heavier (1,75 g; significant difference at 0.01 level). As soon as cyclic, or at the beginning of pregnancy, anımals present ovaries weighing 3 g (significant difference at 0.01 level; Fig. 1). 4 400 (9) (9) S z £ [2) nn 2 so 200 3 3 “a © 2 o > I oo ı 2100 (0) (0) Stage Infant Inactive Cyclic Infant Inactive Cyclic Age (year) 2 >] <] 1-2 >?2 >] Fıg. 1. Development of ovary weight and uterus weight in relation to age and stage of females Evolution of uterine weight At infant stage, uterine weight averages 40 g. In sexually inactive famales 1 to 2 years old, it increases up to 80 g (P <0.01), and up to 120g (P <0.05) in older females. When cyclic females present significantly heavier uterus, with an average 300 g (Fig. 1). During preg- nancy, development of uterus and foetal annexes goes on: ın a female with 6 embryos 45 days old, total weight of uterus, embryos and annexes was 1436 8. Onset of oestrus and pregnancy The following observations concern only females older than 1 year examined during 78-79 and 79-80 hunting seasons. During the first campaign, among 5 females (15-18 months old) dissected in october, 2 show ovulations and another (2-3 years old) is pregnant. Three individuals examined in November-December are pregnant, the younger in the 12-15 month age class. During the 79-80 season, all 4 females observed in October-November (among which 2 older than 3 years) are sexually inactive. Among 7 females examined ın December, only 3 are cyclic but without any sign of pregnancy. Birth date In 1978, all 8 piglets observed in the game bag were born in May-June. In 1979 births are more widely distributed, from January-February to July-August. In 1980 again, birth dates are restricted to a limited number of months (Fig. 2). Weight and reproductive stage of wild boar 25 oak-mast (N=8) (N=36) (N = 26) IF M)J so JF MJ so JF MJ so 1978 1979 1980 Fig. 2. Chronology of reproduction in wild boar from 1978 to 1980. Histogram of frequency of birth dates of piglets and percentage of observed ovulating or pregnant females (circles). Arrow points to oak-mast in autumn 1978 Body weight development Relation to sex Generally speaking, weight of male and female piglets of the first age class (1-2 months) is close to 5 kg. Weight gain ıs fast up to 1 year, when both sexes reach about 45 kg. During the 2nd year, weight dynamics change in relation to sex (Fig. 3). Males 1 to 2 years old maintain a fast weight gain, while females stay approximately at the same level. It results in 100 (kg) 75 a © Body Weight D a 1 2 3 >3 Age in Years Fig. 3. Development of body weight in relation with age and sex of anımals. —-A: males + females; @—-@: males; O--0: females an average difference in weight between sexes climbing up to 20 kgat the end of the second year. Afterwards males and females go on increasing in weight at similar rates so that animals older than 3 years present an average 100 kg in males and 75 kg in females. Difference between sexes stays highly significant (P <0.01). Fig. 3 shows that body growth in piglets, as well as in older males, ıs regular. If hypothesizing that weight varies at a constant rate, daily weight gain would be about 110 g. Fig. 4 presents the relative weight gain curve (monthly weight gain divided by actual 26 Dominique Pepin, F. Spitz, G. Janeau and G. Valet weight at the beginning of the month). Piglets show an increase rate 0f 26,3% . 2° per month in the interval 2-3 months, ® * decreasing down to 6,3 % in the inter- val 14-15 months. zu In older animals a further decrease ? is observed, with atendency toastep- => 15 by-step change in females. E 10 Influence of food resource An exceptional oak mast in 1978 re- sulted in an abundant food resource up to summer 1979. Individuals able a —— to take advantage of it are, first, piglets ? ; 5 = in their rapıd growth stage and, sec- ond, older animals. In order to test a Fig. 4. Relative body weight gain between 2 and 15 possible effect of this oak mast on months (in g/100 g of initial weight) both categories, we compare in Table 2 the average body weights of animals killed in 1979 to those in 1978 (prior to having taken advantage of acorn fall). Data available for piglets do not allow to demonstrate a significant difference between 1978 and 1979 for age classes 5-6 months and 6-8 months. For adults comparison concerns females 1 to 2 years old, and male 18 months-2 years. Individuals killed in 1979 are significantly heavier, with a difference greater for males (about 20 kg, P <0.01) than for females (about 10 kg, P <0.05). Age in Months Table 2 Comparison of average body weigth in various categories of wild boars in the Gresigne State Forest, in relation to food conditions Age class Oak-mast in previous autumn scarce abundant M 5-6 months 25.9 B 23.6 { rt EsENSs 6-8 months 26.0 h 27.9 4 rDENS 1-2 years 42.6 3 Sl \ r—28 20:05 1.5-2 years 61.1 - 81.0 { t— 7.9; 20.01 Discussion Methodological aspects Results of this study, realized from data collected in live and/or dead anımals are not exempt of criticizism. Let us first notice that most trapping take place in the March- September interval, while hunting season is limited from October to early January. A bias can thus be introduced by a seasonal cycle of the weight variation as observed in captive males fed ad libitum (Mauer et al. 1984), or in wıld anımals of East France (Kreın 1984). Let us also remind that age distribution of our two samples are dramatically different: it should be verified whether the adult seasonal cycle of weight gain (or loss) is similarly observed in piglets up to 1 year, or not. Moreover, when working with a completely wild population of boars, we must accept some unaccuracies, for instance when ageing Weight and reproductive stage of wıld boar 27: individuals more than 1 year. Grouping in classes decreases this bias, provided a sufficient number of animals in each class: it is not always the case in our study. Reproduction As in the domestic sow (Josse et al. 1979), observing and weighing ovaries and uterus allow to quickly determine the physiological state of the wild sow. Evolution with age, stage and season, observed in the present study is quite comparable to Aumaftre’s (1982) results. In Reden cock: author, we observed, by separing females less than 2 years and older, that ovary and uterus of inactive ones are heavier in the second category. The onset of the seasonal breeding activity is strongly related to food availability. MATSCHKE (1964) particularly mentions that an important food deficiency results in maintaining females in anoestrus. Conversely, an exceptional food resource, like oak or beech-mast, coincides with an earlier onset of oestrus. However, even in this case, a seasonal anoestrus phase occurs (in the northern hemisphere) in the July-September interval as revealed by hormonal profiles in MAuUGET (1982). Present observations indicate that the exceptional 1978 oak mast in Gresigne resulted in earlier births (likely as soon as January 1979). If we hypothesize a pregnancy length of 115-120 days (HEnrY 1968; Maucer 1972; VERICAD 1983), we estimate that the first fertile matings occured in the second half of September 1978, i.e. as soon as oestrus phase was able to begin. Due to a too small number of observations, our work cannot provide new elements concerning fertility or litter size increase in the case of food abundance. However this phenomenon has been frequently demonstrated, by directly recording ovulation rate and embryo number (HEnrY 1966; BRIEDERMANN 1971; STUBBE and STUBBE 1977; AUMAITRE et al. 1982) or pregnancy rate (SINGER and ACKERMANN 1981) or by observing sows with youngs (OLoFF 1951; MATSCHKkE 1964; DE Vos and Sassanı 1977). Weight growth in piglets and weight of adults In consequence of the difficulty of repeatedly measuring wild anımals, data concerning growth rates are rare. In wild boar, MAuGET (1980) observes that piglets from the captive population of the Chize forest exhibit a linear ponderal development, averaging 2.9 kg per month up to 6 months. In a more detailed study encompassing 4 different forests ın East France, KLeın (1984) shows that actually growth rate of piglets is rather variable from one place to another, seemingly due to the “potential value’ of each ecosystem. However, in a given place, he does not find any significant difference between successive years, up to the age of 6 months. This author also stresses that growth rate is similar in males and females as long as they do not reach 45 kg, i.e. approximately at the age of 12-15 months. Our own observations, although in good accordance with the above cited, lead to think that weight increase in piglets would tend to be linear up to the end of this “unisex” growth phase. Furthermore, our study shows that existence or absence of big oak mast does not affect the weight growth of piglets up to the age of 6-8 months. However in captivity, a severe or very severe food restriction can result in a very weak growth rate (MAUGET and P£rın 1985) or even a complete absence of growth and an increase of mortality (MATSCHkE 1963). We could conclude that in Gresigne area the trophic level is generally high enough to ensure a correct growth rate in piglets, even in absence of a good oak mast. Let us also stress that, in opposition to adults, juveniles do not tend to store fat (DE Vos and SassanI 1977). After a good oak mast, birth occur earlier and this fact is the only responsible of the heavier piglets observed at the next hunting season. In absence of age determination, we must be suspicious about all results concerning piglet weight growth (UECKERMANN 1972; Er MasTour et al. 1982). 28 Dominique Pepin, F. Spitz, G. Janeau and G. Valet In adult wild boars, KrEın (1984) stresses that a regular weight development in males is maintained up to 2 years of age, resulting in some individuals at 100 kg in “rich” areas, as compared to 75 kg only in “poor” ones. Oldest females, in comparison, reach only 70-80 kg. Our own work confirms this dynamic in relation to sex. The fact that females tend to a step-by-step weight increase must be put in parallel with their physiology. While adult males make a very small investment in reproduction, pregnant and lactating females invest in reproduction a great part of their energy input. This investment is so great that it prevents every weight gain (except during pregnancy or after lactation when food is overabundant). The difference observed between sexes, which is established after 1 year of age (Romıc 1975), could be assimilated to the part of the female energy budget which is invested annually in breeding. Wild boars appear able to adapt themselves with a great efficiency to a sudden and unpredictable food resource, essentially by modifying the rhythm and intensity of their investment in reproduction and body development. The fact that such dynamic adjust- ments occur without great latency confirm the great adaptability of the species. While ıt seems difficult to demonstrate the realization of internal fat storage in males, we can observe in females an ability to store abdominal fat. This ability, and their capacity of using the fat reserve during adverse periods is another remarkable adaptative character highly favourable in anımals establishing population strategy in free ranging conditions (YOUNG 1976). Acknowledgements Many hunting Saunen: and game keepers have contributed at this work. We are very grateful for their efficacious help. Thoughtful comments on the manuscript were provided by A. AuMmAITRE, F. Krein and R. MAucer. Special thanks to H. Roustan for typing. Zusammenfassung Fortpflanzung und Gewichtsentwicklung bei Wildschweinen (Sus scrofa) in Südwestfrankreich Gewicht, Alter und Fortpflanzungszustand wurden bei 133 südwestfranzösischen Wildschweinen in den Jahren 1978-1980 ermittelt und zur Bestimmung der Gewichtsentwicklung in Abhängigkeit von Alter und Geschlechts sowie der Fortpflanzungszeiten verwendet. Dabei war das Alter von als Frischlinge markierten Tieren genauer bekannt, bei der Mehrzahl wurde es nach dem Linsengewicht geschätzt. Bis zu einem Körpergewicht von 45 kg oder einem Alter von 12-15 Monaten verläuft die Gewichtszunahme bei beiden Geschlechtern gleich. Dann nehmen zunächst nur noch die Männchen gleichmäßig zu, wogegen die Weibchen ıhr Gewicht erst gegen Ende der Fortpflanzungsperiode im 2. Lebensjahr erneut steigern. Nach einem Mastjahr wurden die Jungtiere durchschnittlich früher ım Jahr gebornen als nach anderen Jahren. Gleichalte erwachsene Schweine waren nach Mastjahren deutlich schwerer als nach Normaljahren, wogegen sich bei gleichalten Jungtieren kein Gewichtsunterschied nachweisen ließ. Die höheren Gewichte von Jungtieren in der Jagdzeit nach einem Mastjahr beruhen offenbar allein auf ihrem höheren Durchschnittsalter. References AUMATTRE, A.; MORVan, C.; QUERE, J. P.; PEinIAU, J.; VALET, G. (1982): Productivite potentielle et reproduction hivernale chez la laie. (Sus scrofa scrofa) en milieu sauvage. Journ&es Rech. Porcine en France 14, 109-124. AUMAITRE, A.; QUERE, J. P.; PEıntau, J. (1984): Influence du milieu sur la reproduction hivernale et la prolificite de la laie. In: Symposium International sur le Sanglier. Ed. by F. Spitz, D. P£pın. Les Colloques de ’INRA, n° 22, 71-78. BRIEDERMAN, L. (1971): Zur Reproduktion des Schwarzwildes in der Deutschen Demokratischen Republik. Tag. Ber. Dt. Ad Landwirtsch. Wiss., Berlin, 113, 169-186. DARrDAILLoN, M. (1984): Le Sanglier et le milieu camargais: dynamique coadaptatıve. These Doct. 3eme cycle Univ. Paul Sabatier Toulouse, 1984. De Vos, A.; Sassanı, A. (1977): Eine Studie der Population des Schwarzwildes (Sus scrofa) in dem Mohammad Reza Shah National Park. Z. Jagdwiss. 23, 113-126. Weight and reproductive stage of wild boar 29 Du£r, P. H.; ErIEnNE, M. (1974): Influence de l’alimentation pendant la croissance de la truie sur la maturite sexuelle et les performances de reproduction: resultats preliminaires. Journees Rech. Porcine en France, 43-47. Er MASTOUR, A.; PERTHUIS, R.; Popescu, C. P. (1982): Recherches pr&liminaires sur la biologie et l’ecoethologie du Sanglier marocain (Sus scrofa barbarus). In: Symposium international sur la gestion et la conservation de la faune sauvage mediterraneenne. C.I.C. Ed. by Bureau technique au service de la chasse, Rabat, Maroc. 36 p. ETIENNE, M.; Camous, S.; CUVILLIER, A. (1983): Effets de restrictions alimentaires pendant la croissance des truies sur leur maturite sexuelle et leur reproduction ulterieure. Reprod. Nutr. Develop. 23 (2 A), 309-319. ETIENNE, M.; LEGAULT, C. (1974): Influence de la race et du regime alimentaire sur la precocite sexuelle de la truie. Journees Rech. Porcine en France, 57-62. HEnRY, V. G. (1966): European wild hog hunting season recommandations based on reproductive data. Proc. Ann. Conf. S.E. Assoc. Game and Fish Comm. 20, 139-145. — (1968): Length of oestrous cycle and gestation in european wild hogs. J. Wildl. Manage. 32, 406-408. JOSSE, J.; LE DENMAT, M.; MARTINAT-BOTTE, F.; SAULNIER, J.; VANNIER, P.; VAUDELET, J. C. (1979): A propos d’une enque£te sur les causes de reforme des truies. Journ&es Rech. Porcine en France, 365-370. Krein, F. (1984): Contribution a l’etude de la croissance du Sanglıer (Sus scrofa) par capture et recapture. In: Symposium International sur le Sanglier. Ed. by F. Spırz, D. P£rın. Les Colloques de ’INRA, n? 22, 55-67. Kozıo, P. G. (1975): The wıld Boar. Minsk. 1-223. (In Russ.). MARTINAT-BOTTE, F.; DAGORN, J.; TERQUI, M.; DAnDo, P. (1984): Effect of confinement, climatic conditions and litter parity on the seasonal varıations of the fertility rate and prolificacy. Ann. Rech. Vet. 15, 165-172. MARTINAT-BOTTE, F.; DANDO, P.; GAUTIER, J.; TERQUI, M. (1981): Variation saisonniere de la taılle de la portee en relation avec celles du taux d’ovulation et de la mortalit€ embryonnaire chez la truie. Journ&es Rech. Porcine en France, 269-276. MATSCHKE, B. H. (1963): An eyelens-nutrition study of penned European wild hogs. Trans. Ann. Conf. SE Ass. Fish and Game Comm. 17, 20-27. — (1964): The influence of oak mast on European wild hog population. Trans. Ann. Conf. SE Ass. Fish and Game Comm. 18, 35-39. — (1967): Aging European wild hogs by dentition. J. Wildl. Manage. 31, 109-113. MAUuGET, R. (1972): Observations sur la reproduction du Sanglier (Sus scrofa) a l’&tat sauvage. Ann. Biol. Anım. Bioch. Biophys. 122, 195-202. — (1980): Regulations ecologiques, comportementales et physiologiques (fonction de reproduction) de l’adaptation du Sanglier, Sus scrofa L., au milieu. These de Doctorat d’Etat, Universite de Tours, 1980. — (1982): Seasonality of reproduction in the wild boar. In: Control of pig reproduction. Ed. by D. J. A. Cox, G. R. Foxcrorr. London: Butterworths. 509-526. MAUGET, R.; Campan, R.; SpıTz, F.; DARDAILLON, M.; JANEAU, G.; P£pın, D. (1984): Synthese des connaissances actuelles sur la biologie du Sanglier, perspectives de recherche. In: Symposium International sur le Sanglıier. Ed. by F. Sprrz, D. P£rın. Les Colloques de ’INRA, n? 22, 15-50. MAUGET, R.; P£pin, D. (1985): La nubenıe chez le Sanglier: &tude preliminaire du röle de l’alımenta- tion. XVIIth Congress of Int. Union Game Biol. Ed. by $. DE CROMBRUGGHE. Brussels: Ministery of Agriculture. 191-197. E OrLorr, H. B. (1951): Zur Biologie und Ökologie des Wildschweines. Beitr. Tierkd. u. Tierzucht 2, 1-95. Romıc, $. (1975): Tjelesne dimenzije divlje svinje (Body dimensions of wild boar). Poljopr. znan. smotra 34, 13-24 (in Czech.). SADLEIR, R. M. F. S. (1969): The ecology of reproduction in wild and domestic mammals. Lond: Methuen. SINGER, F. J.; ACKERMANN, B. B. (1981): Food availability, reproduction, and condition of European wild boar in great Smoky national park. Res. Mgt. Report 43. Ed. by U.S. Dep. Inter. Nat. Park. Tenn. SOLENSEN, A. M. Jr.; THomas, W. B.; GossErT, J. W. (1961): A further study of the influence of level of energy intake and season on reproductive performance of gilts. J. Anım. Sci. 20, 17498 Spıtz, F. (1984): Demographie du Sanglier en Gresigne (Sud-Ouest de la France). In: Symposium International sur le Sanglier. Ed. by F. Sprrz, D. P£pın. Les Colloques de ’INRA, n? 22, 151-157. SPITZ, F.; JANEAU, G.; VALET, G. (1984): Elements de demographie du Sanglier (Sus scrofa) dans la region de Grösigne. Acta Oecol. Oecol. Applic. 5, 43-59. le STUBBE, W.; STUBBE, M. (1977): Vergleichende Beiträge zur Reproduktions- und Geburtsbiologie von Wild- und Hausschwein Sus scrofa L.). Beitr. Jagd- und Wildforsch. 10, 153-179. SWEENEY, J. M.; ProvoT, E. E.; SwEENEY, J. R. (1970): A comparison of eyelens weight and tooth 30 Dominique Pepin, F. Spitz, G. Janeau and G. Valet irruption pattern in age determination of feral hogs ($us scrofa)). Trans. Ann. Conf. SE Ass. Fish and Game Comm., 285-291. UECKERMANN, E. (1972): Zur jagdlichen Nutzungsfähigkeit von Rot-, Dam- und Schwarzwildbestän- den nach Beobachtung in einem Jagdgatt. Lev. Z. Jagdwiss. 18, 24-31. VERICAD, J. R. (1983): Estimacion de la edad fetal y periodos de conception y parto del jabalı (Sxs scrofa L.) en los Pirineos Occidentales. XVth Cong. Int. de Fauna Cinegetica y Sylvestre. Ed. by J. CAsTROVIEJO, J. M. Branc. Estacion, Biologica de Donana, Federacion Espanola de caza, 811-820. Young, R. A. (1976): Fat, energy and mammalıan survival. Amer. Zool. 16, 699-710. ZIMMERMANN, D. R.; Spies, H. G.; Rıcor, E. M.; SELF, H. L.; Casıpa, L. E. (1960): Effect or restricted feeding, cross breeding and season of birth on age at puberty in swine. J. Anım. Scı. 19, 687-694. Authors’ address: DOMINIQUE P£rPINn, FRANCcOIS SPITZ, GEORGES JANEAU, GILBERT VALET, Laboratoire de la Faune Sauvage, et I.R.G.M., Institut National de la Recherche Agronomique, BP 27, F-31326 Castanet-Tolosan Cedex, France Hinweise zur Kondition des Steinwildes Capra 1. ibex im Berner Oberland (Schweizer Alpen) Von SYLvIA ZUMBACH und P. Lürs Aus dem Naturhistorischen Museum der Burgergemeinde Bern und dem Jagdinspektorat des Kantons Bern’ Abstract Information about the condition of Alpine Ibex Capra ı. ibex in the Swiss Alps Increasing colonies of Alpine Ibex required that the number of this protected species be regulated. In the canton of Berne 242 Ibex were shot ın October of the years 1980 to 1984 ın a strictly organızed cull. Body weight, metatarsus marrow fat and kidney fat index showed a high correlation. The results showed that the Ibex were generally in good condition, though some deterioration in condition was observed in males older than 13 years. Einleitung Der Aufbau von Fettreserven in Zeiten günstiger Äsungsverhältnisse gestattet es einer Tierart, auch Perioden mit unzureichender Energiezufuhr ohne Schaden zu überstehen. Für das Steinwild ist die Einlagerung von Körperfett im Laufe des Sommers/Herbstes bekannt, wobei allerdings umfassende Zusammenstellungen fehlen. Andere Hinweise auf den konditionellen Zustand bei diesem einst nahezu ausgerotteten Vertreter der Paarhufer ım Alpenraum liegen nur in sehr fragmentarischer Form vor. Hegeabschüsse in den Jahren 1980 bis 1984 boten Gelegenheit, Daten und Material zu sammeln, mit dem Ziel, das Wissen über den Alpensteinbock zu erweitern. Zusammen mit ethologischen Aspekten sollen die Kenntnisse des konditionellen Zustandes der Tiere in den verschiedenen Kolonien eine Grundlage zur detaillierten Planung des Vorgehens bei zukünftigen Abschüssen liefern. Diese Abschüsse sind deshalb notwendig geworden, weil das Steinwild in den ständig wachsenden Kolonien Schäden am Wald, namentlich an 1 . .. . .. Mit Unterstützung der Brunette-Stiftung für Naturschutz. U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5201-0030 $ 02.50/0 Z. Säugetierkunde 52 (1987) 30-38 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Hinweise zur Kondition des Steinwildes Capra ıi. ibex im Berner Oberland 31 Aufforstungen in lawinengefährdeten Gebieten, aber auch an landwirtschaftlich genutzten Flächen verursacht hat - und dies noch tut. Material und Methode Alle 242 Tiere (133 &, 109 2) stammen von Hegeabschüssen ım Berner Oberland (BRÜLLHARDT und Lürs 1984; Lürs und ZUBER 1986; ZumBAcH 1986). Steinwild wurde nur dort zum Abschuß freigegeben, wo regelmäßig und in größerem Umfang Schäden festgestellt wurden. Die Abschüsse hatten jeweils zwischen dem 1. und 31. Oktober der Jahre 1980 bis 1984 zu erfolgen. Als durchschnitt- liches Abschußdatum für die Böcke konnte der 7., für die Geißen der 10. Oktober ermittelt werden (Lürs et al. 1986). Die Abschußorte lagen in der Regel zwischen 1600 und 2500m ü.M. Die Jäger erhielten den Auftrag, den Tieren beim Ausweiden dıe Nieren mit dem umgebenden Fett sowie ca. 20cm Enddarm zu entnehmen und diese Proben dem Wildhüter bei der obligatorischen Kontrolle abzuliefern. Dieser bestimmte das Alter des Tieres, wog und vermaß es im Fell und löste einen Metatarsus heraus. An der Trophäenschau im darauffolgenden Frühling, bei der die montierten Schädel mit Gehörn oder die präparierten Köpfe ausgestellt werden mußten, ließ sich das Alter nochmals überprüfen, gleichzeitig konnten le Messungen am Gehörn vorgenommen werden. In bezug auf den Abschußtermin ist das Material als recht einheitlich zu bezeichnen, wobei klimatische Unterschiede von Jahr zu Jahr hier nicht zum Ausdruck kommen können. Diese Einheitlichkeit bringt Vorteile, läßt aber andererseits keine allfälligen jahreszeitlichen Unterschiede im konditionellen Zustand erkennen. Sämtliche Angaben beziehen sich also auf Tiere, die gegen Ende der Vegetationsperiode erlegt wurden, wenn die Ernährungsbedingungen bereits wieder schlechter sein dürften als ım Hochsommer. Zu diesem Zeitpunkt hat die energieverbrauchende Brunft jedoch noch nicht eingesetzt. Die Tiere dürften, theoretisch betrachtet, konditionell nahe am Optimum stehen. Die Wildhüter bestimmten das Körpergewicht (KG) des ausgeweideten und ausgebluteten, aber nn en Tieres mit Haupt und Gehörn am Tag des Abschusses oder am darauffolgenden auf 1 kg genau. Die anderen Untersuchungen wurden nach Abschluß der Hegeabschüsse und nach vorübergehen- dem Tiefgefrieren der Organe im Labor durchgeführt. Es sind des — Messung der Länge des Metatarsus (Kanonenbein; Größte Länge auf 0,1 mm genau) nach von DEN DrizscH 1976 als Hinweis auf die Körpergröße. Der Metatarsus erreicht seine endgültige Länge bei S mit 4%, bei ? mit 3% Jahren (ZumBAcH 1986). Zunahmen des Körpergewichtes können somit weitgehend größenunabhängig festgestellt werden. — Ermittlung des Nieren (kidney fat index KFI). Bereits Harrıs (1945) hat auf die Aussagekraft des Gehaltes von Nierenfett für die Bestimmung des konditionellen Zustandes hingewiesen. Die Methode wurde durch Rıney (1955) an neuseeländischen Cervus elaphus metrisch überprüft und standardisiert. Ihre Aussagekraft ist seither bei verschiedenen Arten belegt worden (z.B. Hormann 1978, bei Rupicapra rupicapra; BucHı 1979, bei Cervus elaphus; JoHns et al. 1984, bei Odocoileus virginianus; SPINAGE 1984, bei Cobus ellipsiprymnus und Gazella granti). Bei den meisten Individuen ließ sich der Mittelwert aus linker or rechter Niere errechnen. Bei einzelnen Abschüssen war das Wiegen unmöglich oder auf eine Körperseite beschränkt (z.B. wegen starker Schußverletzungen). Diese Einzeldaten wurden analog den Durchschnittswerten behandelt. Da DaupHine (1975) bei Rangifer tarandus weder für Nierengewicht, noch für den Index KFI links-/rechts-Unterschiede feststellen konnte, wurde hier, ohne eine entsprechende Berechnung vorzunehmen, ein solches Vorgehen als statthaft betrachtet. — Bestimmung des Fettgehaltes im Knochenmark des Metatarsus (metatarsus marrow fat MMF). Im Mark vermag der Knochen beträchtliche Mengen an Reservefett zu speichern, das zu gegebener Zeit mobilisiert werden kann. In verschiedenen Untersuchungen konnte gezeigt werden, daß das Trockengewicht des Knochenmarks (Anteil fetthaltiger/nicht fetthaltiger Substanzen nach Entzug des Wassers und anderer flüchtiger Komponenten) den Wert des Marks als Fettreservoir recht gut wiederzugeben vermag (z. B. NEILAnD 1970, für Rangıfer tarandus). In der Regel wurde bei diesen Untersuchungen das Mark des Femurs untersucht (z.B. Hormann 1978; RansoM 1965). Aus praktischen Gründen (Fleischverwertung durch den Jäger) mußte hier das Mark des Metatarsus untersucht werden, wie dies auch bei den Untersuchungen Buchtis (1979) und AnDERSENS (1985) der Fall ist. Zwar weichen der Fettgehalt im Femur und Metatarsus bei einigen Arten leicht voneinander ab (SINCLAIR und Duncan 1972; FULLER et al. 1986), die Aussage bleibt aber dieselbe. Das in der Menge von rund 1 Gramm pro Tier gewonnene Mark aus dem Mittelbereich des Knochens wurde auf das Milligramm genau gewogen, im Wärmeschrank während 24 Stunden bei 100°C getrocknet, dann ım Exsıkkator während 24 Stunden abgekühlt und zum Schluß nochmals gewogen. Da für einzelne Altersjahrgänge zu wenig Daten vorliegen, um eine statistische Bearbeitung vorzuneh- men, wurden für verschiedene Berechnungen mehrjährige Altersklassen gebildet (s. BRÜLLHARDT und Lürs 1984): A bis 4% Jahre, B 5% bis 8%, C 9% bis 12" Jahre, D 13% Jahre und älter. 32 Sylvia Zumbach und P. Lüps Die Zahlen zweier Böcke und einer Geiß wurden bei der Bearbeitung ausgeschlossen: - Nr. 115, d, im 9. Jahr, mit Moderhinke (Tab. 2) - Nr. 132, ?, im 13. Jahr, mit durchschnittlichem KG, aber extrem tiefem MMF-Wert - Nr. 157, d, im 12. Jahr, mit durchschnittlichem KG, aber extrem tiefem MMF- und KFI-Werten, ohne äußerlich erkennbare Krankheitszeichen. Resultate Eine tabellarische Zusammenstellung von 10 Körpermaßen (u.a. Körpergewicht, Metatar- suslänge, Nierengewicht) der hier bearbeiteten Tiere findet sich bei ZumBAcH (1986). Körpergewicht (KG) Absolute Werte Böcke nehmen bis ins Alter von rund 10 Jahren an Gewicht zu (jahreszeitliche Schwan- kungen können hier nicht berücksichtigt werden, RArTı 1981; BRÜLLHARDT und Lüprs 1984). Ab diesem Alter tritt eine Stabilisierung ein. Böcke über 12% Jahren erreichen diese höchsten Herbst-Werte nicht mehr. Liegt das Körpergewicht für die 9%- bis 12%jährigen Böcke bei 66,8 kg, mit einer beträchtlichen Anzahl von Tieren über 70 kg, so überschreitet keines der 7 im 14. oder einem höheren Altersjahr stehenden Männchen die 70-kg-Limite x = 66,3 kg). Geißsen erreichen eine Stabilisierung des herbstlichen Körpergewichts bereits mit rund 5% Jahren, halten diese Werte aber über eine deutlich längere Zeit als die Böcke, ungefähr bis zum 15. Altersjahr (RAaTTı 1981; BRÜLLHARDT und Lürs 1984). Relatives Gewicht Um auszuschließen, daß es sich bei den etwas leichteren sehr alten Böcken nur um kleine Individuen handelt, wurden die Körpergewichte mit der Länge des Metatarsus in Bezie- hung gesetzt. Es zeigt sich dabei sowohl bei der Betrachtung der Gesamtheit (r = 0.5/ &@<1%), wie auch innerhalb der gebildeten Altersklassen (Abb. 1) eine statistisch gesi- cherte Zunahme des Körpergewichts mit zunehmender Länge des Metatarsus. Keine Korrelation ist bei den „greisen‘““ Böcken ab dem 14. Altersjahr zu erkennen. Diese Beobachtung weist auf eine hohe Streuung des Körpergewichts bei Alttieren hin, was mit einem Konditionsverlust im hohen Alter erklärt werden kann. Für die Gesamtheit der Geißen liegt ebenfalls eine gesicherte Korrelation zwischen der Länge des Metatarsus und dem Körpergewicht vor (r = 0,5/«<1%). Auch beı den weiblichen Tieren lassen sich innerhalb der Altersgruppen gesicherte Korrelationen nach- weisen (Abb. 1), wobei wie bei den Böcken die ältesten Tiere so stark streuen, daß keine gesicherte gegenseitige Abhängigkeit der beiden Maße vorliegt. Nierenfett (KFI) Das Nierengewicht nimmt mit steigendem relativem Körpergewicht zu (Abb. 2). Mit dieser Gewichtszunahme muß auch eine Erweiterung des Nierenvolumens erfolgen. Es stellt sich somit die Frage, ob das Nierenfett, und damit der Nierenfettindex, proportional zur Niere zunehmen oder ob mit zunehmender Nierengröße eine mit der relativen Abnahme der Nierenoberfläche zu erwartende relative oder absolute Abnahme der Fettre- serven einhergeht. Bei den Böcken nimmt der KFI bei steigendem Nierengewicht von Klasse A zu Klasse B zu. Trotz weiterhin ansteigendem durchschnittlichem Nierengewicht (und zunehmendem Hinweise zur Kondition des Steinwildes Capra i. ibex im Berner Oberland 80 Körpergewicht 70 60 so 40 30 20 10 11 33 D" C*’r-0,59 r=0.31 B’r-0.35 A'* r=083 er B*r=0.24 12 13 14 15 cm Metatarsus Abb. 1. Die Beziehung zwischen dem Körpergewicht (ausgeweidet) und der Länge des Metatarsus NG 100 90 80 707 60 3 SZ (Kanonenbein) ın den verschiedenen Altersklassen oCc oD oB eo »oQ, “D ec oA °B T T T T T T 2.5 4.0 4.5 50 rel.KG Abb. 2. Mit steigendem relativem Körpergewicht (rel. KG) und zunehmendem Alter (A bis D) erfolgt eine Zunahme des Nierengewichtes (NG) 34 Sylvia Zumbach und P. Lüps relativem KG) erfolgt keine weitere Zunahme des KFI zu Klasse C. Die tiefsten Werte für den KFI liegen für Klasse D (d = 13%) vor, dies bei hohem Nierengewicht (Tab. 1). Die Geißen zeigen einen kontinuierlichen Abfall des KFI mit zunehmendem Alter und ansteigendem Nierengewicht von Klasse A zu Klasse D. Tabelle 1 Nierengewicht; Nierenfettindex (KFI); relatives Körpergewicht (Körpergewicht KG in kg/ Metatarsuslänge in cm); Korrelationskoeffizient KFV/rel. KG Nierengewicht 8 56,7 76,6 3252 90,2 41,0 32,2 55,0 60,3 Diese Angaben weisen auf eine Abnahme des KFI bei steigendem relativem KG und hohen Nierengewichten hin. Bei alten Tieren wirkt sich zudem eine offensichtliche Abnahme der Kondition zusätzlich aus. Fettgehalt des Knochenmarks (MMF) Im Gegensatz zum Nierenfettindex weist dieses Konditionsmaß eine sehr geringe Variabi- lität auf. Lediglich 14 von 107 gemessenen Werten bei den Böcken liegen außerhalb der 92-98 %-Werte. Dabeı handelt es sich in 10 Fällen um Tiere im Alter von mindestens 11% Jahren. Bei den möglichen Mefsfehlern erstaunt dieses homogene Bild. Es weist auf eine hohe Tauglichkeit dieses Konditionsmaßes hin. Von 76 Geißen liegen 16 unter der 92 %-Grenze. Von diesen stammen 9 aus Alters- klasse D. Bei keiner alten Geiß (im 14. Jahr oder älter) lag der Fettanteil im Knochenmark über 94 %. Vergleich von Nierenfettindex und Fettgehalt des Knochenmarks Aus Abbildung 3 geht die schon erwähnte hohe Homogenität der MMF-Werte hervor. Bei den Böcken liegen lediglich 14 Werte deutlich unterhalb der 92 %-Grenze. Von diesen weisen 10 einen stark unterdurchschnittlichen KFI-Wert auf. Eine deutliche Abnahme des MME erfolgt erst bei KFI-Werten unter 110. Ein Wert bei den Böcken und zwei bei den Geifsen stehen isoliert da (relativ hohe KFI, tiefe MMF-Werte, ein Bock im 7. eine Geiß im 13. und eine im 16. Jahr). Inwieweit bei diesen drei normalgewichtigen Tieren Artefakte beı der Laborarbeit entstanden sind, ist nicht eruierbar. Wie alte Böcke haben auch alte Geißen in den beiden untersuchten Konditionsmaßen niedrigere Werte als der Durch- schnitt. Bei KFI-Werten unter 90 beginnt sich eine Abnahme des MMF-Gehaltes abzu- zeichnen. Die Daten weisen auf eine Mobilisierung der Reservefette im Knochenmark hın, wenn die Nieren-Fettreserven bereits stark abgebaut sind. Diese Erscheinung steht ın Übereinstimmung mit den bei andern Ungulaten ermittelten Resultaten (Harrıs 1945; SINCLAIR und Duncan 1972; ANDERSSON 1985; u. a.). Hinweise zur Kondition des Steinwildes Capra i. ibex im Berner Oberland 35 80 - T Tr za T 1 50 100 150 200 250 300 s 20 no E e 00 £ . s % e ° Br ____ _-__ 2020 a a an nn ne My nn nal un 90 .x [) 80 “eo 60 Be. T —— mel T SIE TE nme re en m De, 50 100 150 200 250 300 KFI Abb. 3. Der Fettgehalt im Mark des Metatarsus (MMF) in Prozent in Relation zum Nierenfettindex (KFI). a (oben) = 8, b (unten) = 2, Tiere aus Altersklasse D (> 13'jährig) sınd mit einem Stern speziell hervorgehoben Totfunde und ermattete Tiere Die durchwegs aus dem Oktober stammenden Abschüsse lassen keine Hinweise auf Jahreszeitliche Schwankungen des Körpergewichts zu. Ebensowenig gestatten sie Angaben über eventuelle jahreszeitliche Veränderungen im Reservefett von Knochenmark und Nieren. Damit fehlt ein notwendiger Bezug der vorgelegten Resultate zu andern weitge- hend. Die Möglichkeit zum Vergleich bietet das analog aufgearbeitete Material von Sonder- und Hegeabschüssen aus dem gleichen Zeitraum und den selben Kolonien (Tab. 2): Die fünf Böcke, die als altersschwach, als entkräftet oder wegen Moderhinke erlegt oder frischtot gefunden worden sind, stehen alle im 9. bis 16. Jahr, weisen deutlich unterdurch- schnittliche Gewichte auf und haben niedere Markfettwerte, wie dies bei keinem Bock aus den Kontingenten 1980-1984 der Fall ist. Die vier anderen Tiere liegen sowohl was das 36 Sylvia Zumbach und P. Lüps Tabelle 2 Erlegte d außerhalb der Kontingente 1980-1984 Altersjahr Datum Grund des Abschusses Gewicht Markfett % Nierenfett ausgeweidet 228 sah Beer un, 56 92,7 120,5 115 .10. ordentl. Hegeabschuß! 42 29,1 0,0 71 ! RAS! zusätzlicher Abschuß 94,6 97,7 229 i 10: zusätzlicher Abschuß 2220 159 : mid: zusätzlicher Abschuß 161 ; 2, Moderhinke ? 343 sat altersschwach kein Fett 432 > Ale entkräftet 7,8 320 Ä DE entkräftet ? | 'Bei Nr. 115 handelt es sich um einen Abschuß im Kontingent 1981, der vom Wildhüter nachträglich zur Ermittlung der Krankheitsursache eingeliefert worden ist. Er ist bei der statisti- schen Auswertung nicht berücksichtigt worden. Das Tier hatte Moderhinke. Körpergewicht als auch was die Fettreserven im Knochenmark und um die Nieren anbelangt im Bereich der als Hegeabschüsse zwischen 1980 und 1984 erlegten gleichaltri- gen Tiere. Ein Zusammenhang zwischen Körpergewicht und Fettreserven liegt hier zweifellos vor. Diskussion Das Ziel der Sammlung von Organen und der Abnahme von Körpermaßen war die Überprüfung des konditionellen Zustandes der erlegten Tiere im Herbst (Oktober). Die Kurve für das Verhältnis Nierenfett (KFI) zu Markfett (MMF) zeigt einen ähnli- chen Verlauf wie die für Odocoileus virginianus durch Ransom (1965), Connochaetes taurinus durch SincLAır und Duncan (1972) und Tragelaphus angası durch ANDERSON (1985) ermittelten. Nierenfettindices unter 30, welche z.B. für Odocoileus einen kritischen Wert darstellen (Ransom 1965; Kıe et al. 1983), wurden bei den Steinwild-Hegeabschüs- sen nicht festgestellt. Auch in bezug auf den Markfettgehalt darf die konditionelle Verfassung der untersuchten Tiere als gut betrachtet werden. Diese generelle Angabe wird eingeschränkt durch einige Tiere, vor allem solche aus Altersklasse D (> 13%jährig), die durch geringere Fettreserven auffallen. Die beiden nicht berücksichtigten Böcke (s. Methode), von denen einer als krank diagnostiziert werden konnte, und eine Geiß liegen völlig außerhalb der Punkteschar. Klar ersichtlich ist auch der Zusammenhang zwischen dem relativen Körpergewicht und den beiden die Fettreserven direkt widerspiegelnden Werten. Daß das Körpergewicht um einige Kilogramm unter demjenigen liegt, das ım Kanton Graubünden erreicht worden ist (RATTI 1981), mag sowohl methodische Gründe haben wie auch genetisch und/oder standortbedingt sein. Jahreszeitliche Schwankungen des Körpergewichts, wie sie für den Steinbock bekannt sind (COUTURIER 1964), des Nierengewichts und des Nierenfettindexes (vgl. dazu DaupHiınE 1975; SPINAGE 1984; VAN VUREN und CoBLENTZ 1985) können anhand der vorliegenden Daten aus dem Oktober nicht diskutiert werden. Ebenfalls können Fragen des Konditionsverlustes im Verlauf der Brunft und des Winters nicht beantwortet werden. Auffallend ist, daß MMF-Werte unter 90 % in erster Linie bei alten Tieren ab dem 14. Altersjahr und bei den Spezialabschüssen kranker oder ermatteter Tiere sowie bei Totfunden anzutreffen sind. Für eine Diskussion der Alterungsprozesse ist das Material noch zu wenig umfassend. Hinweise auf einen konditionellen Abbau ab dem 13. Altersjahr liegen aber vor. Hinweise zur Kondition des Steinwildes Capra ıi. ibex im Berner Oberland 37 Danksagung Allen bei der Aufnahme von Daten im Feld beteiligten Wildhütern und den bei der Ausmessung der Trophäen und der Überprüfung der Altersangaben zusätzlich beteiligten Herren H. AEGERTER und R. Fuchs danken wir für die zuverlässige Hilfe, den Vertretern des kantonalen Jagdinspektorates, den Herren Dr. H. BRÜLLHARDT UND M. ZUBER, zudem für die immerfort gute Zusammenarbeit. Ein spezieller Dank geht an Frau ROSEMARIE ALTHAUS für die stets lee Überarbeitung und die Reinschrift des Manuskripts. Zusammenfassung In den Jahren 1980-1984 wurden ım Berner Oberland (Schweizer Alpen) jeweils im Oktober insgesamt 242 Stück Steinwild Capra ı. ibex erlegt (133 8, 109 2). Für Ar meisten Tiere konnten durch Auswertung der folgenden im Feld und ım Labor genommenen Daten Hinweise auf den konditionellen Zustand gewonnen werden: Körpergewicht (ausgeweidet) absolut und relatıv zur Körpergröße (mit dem Metatarsus als Bezugsmaß), Nierenfettindex (KFI) und Fettgehalt des Kno- chenmarks im Metatarsus (MMF). Um dem starken Geschlechtsdimorphismus und dem langsamen Wachstum Rechnung zu tragen, wurden 4 Altersklassen geschaffen: 04%, 51-84, 9%-12'%, 13% Jahre und älter. ö nehmen bis rund zum 10., ? bis zum 6. Altersjahr an Gewicht zu. Hinweise auf einen (nicht gesicherten) Gewichtsverlust zeigen sich bei d ab dem 13. Altersjahr. Der Nierenfettindex nımmt mit steigendem Nierengewicht ab, bei alten Tieren liegt zudem eine Abnahme der Fettreserven vor. Die Werte für den Markfettgehalt des Metatarsus zeigen eine gute Übereinstimmung mit den Nierenfett- daten. Unterhalb einer 92 %-Grenze liegen in erster Linie Tiere aus Altersklasse D (> 13% Jahre), was auf konditionelle Verluste alter Tiere hinweist. Von wenigen erwähnten Ausnahmen abgesehen befanden sich die als Hegeabschüsse erlegten Tiere in guter Verfassung. Dies ist bei zusätzlich getätigten Spezialabschüssen kranker oder ermatteter Tiere und bei frischtot gefundenen Böcken nicht der Fall, was als zusätzliche Bestätigung der Resultate dieser Untersuchung gewertet werden darf. Literatur ANDERSON, J. L. (1985): Condition and related mortality of nyala Tragelaphus angası ın Zululand, South Africa. J. Zool., Lond. (A) 207, 371-380. BRÜLLHARDT, H.; Lürs, P. (1984): Entwicklung und Regulierung von Steinwild-Beständen (Capra ıbex L.) ım Berner Oberland. Mitt. Naturforsch. Ges. Bern N.F. 41, 153-169. Bucht, Ch. (1979): Zur Populationsdynamik, Kondition und Konstitution des Rothirsches (Cervus elaphus L.) im und um dee Schweizerischen Nationalpark. Diss., Zürich. COUTURIER, M. A. J. (1962): Le Bouquetin des Alpes. Grenoble Selbstverlag. Daupnint, T. Ch., jr. (1975): Kidney weight fluctuations affecting the kidney fat index in carıbou. ]. Wildl. Manage. 39, 379-386. DrisscH, A. von DEN (1982): Das Vermessen von Tierknochen aus vor- und frühgeschichtlichen Siedlungen. München Selbstverlag. FULLER, T. K.; Coy, P. L.; PETERson, W. J. (1986): Marrow fat relationships among leg bones of white-tailed deer. Wildl. Soc. Bull. 14, 73-75. Harrıs, D. (1945): Symptoms of malnutrition in deer. J. Wildl. Manage. 9, 319-322. Hormann, A. (1978): Unterschiedliche Entwicklungsleistungen von Appenzeller und Schwyzer Gemsen (Rupicapra rupicapra L.) und ihre möglichen Ursachen. Diss., Zürich. Jouns, P. E.; Smith, M. H.; Chesser, R. K. (1984): Annual cycles of the kidney fat index ın a southeastern white-tailed deer herd. J. Wildl. Manage. 48, 969-973. Kir, J. G.; Wuıte, M.; Draw, D. L. (1983): Condition parameters of white-tailed deer in Texas. ]. Wildl. Manage. 47, 583-594. Lüps, P.; Brüllhardt, H.; Zuger, M.; ZuMBachH, S$. (1986): Sonderabschüsse von Steinwild Capra 1. ibex im Berner Oberland: erste Erfahrungen und Resultate. Z. Jagdwiss. 32 (im Druck). Lürs, P.; Zuger, M. (1986): Das Steinwild im Berner Oberland. Jahrb. Thuner-Brienzersee 1986 (im Druck). N K. A. (1970): Weight of dried marrow as indicator of fat in carıbou femurs. J. Wildl. Manage. 34, 904-907. A: Ransom, A. B. (1965): Kidney and marrow fat as indicators of white-tailed deer condition. J. Wildl. Manage. 29, 397-398. Rattı, P. (1975): Bericht über die Untersuchung der Bündner Gemsstrecken der Jahre 1971-1974. Bündner Jägerz. 62, 89-95. Rartı, P. (1981): Zur Hege des Steinwildes im Kanton Graubünden. Z. Jagdwiss. 27, 2157 Rıney, T. (1955): Evaluating condition of free-ranging red deer (Cervus elaphus), with special reference to New Zealand. Part I: Description of techniques for determination of condition of red deer (Cervus elaphus). New Zealand J. Sci. Technol., Sec. B, 36, 429-463. 38 Sylvia Zumbach und P. Lüps SINCLAIR, A. R. E.; Duncan, P. (1972): Indices of condition in tropical ruminants. E. Afr. Wildl. J. 10, 143-149. SPINAGE, C. A. (1984): Seasonal influences and the kidney fat index in two equatorial African ungulates. Afr. J. Ecol. 22, 217-221. VUREN, D. van COBLENTZ, B. E. (1985): Kidney weight varıation and the kıdney fat index: an evaluation. J. Wildl. Manage. 49, 177-179. ZUMBACH, S$. (1986): Körpermaße von Steinwild Capra ibex L. aus dem Berner Oberland (Schweiz, Abschüsse 1980-1984). NMBE, Kl. Mitt. 12. Anschrift der Verfasser: SyLvıa ZUMBACH und Dr. PETER Lürs, Naturhistorisches Museum, Abt. Wirbeltiere, Bernastraße 15, CH - 3005 Bern Dental and skull anomalies in the Spanish wild goat, Capra pyrenaica Schinz, 1838 By C.R. Vıcar and AnnIE MACHORDOM Unidad de Zoologia Aplicada, Comunidad Autonoma de Madrid Receipt of Ms. 1. 4. 1986 Abstract One hundred eighty-six specimens of Spanish mountain goat (Capra pyrenaica Schinz, 1838) from two populations ot its distribution area were studied for skull and dentition abnormalıties. The anomalies encountered were grouped according to possible intrinsic or extrinsic origin. The incidence of some of the congenital abnormalities (intrinsic origin) varied significantly between populations, being noteworthy the elevated percentage of maxillary canınes in the Gredos population (20 %), as well as that of PM, hypodontia in Cazorla (45 % of the specimens with the complete molarıform series). In no case were significant differences found between sexes or situation on one side or another of the jaw. As regards abnormalities of extrinsic origin, the specimens from Gredos presented a significantly greater incıdence of periodontal affections (p < 0.001) than those from Cazorla. The effect of the abnormalities on individual survival was considered minimal. Introduction In an earlier study dedicated to establishing the sequence of dental eruption and replace- ment ın the Spanish mountain goat (VIGAL and MACHORDOM, in press), several variations from the general skull and dentition model of this species were observed. Anomalies of the skull, position and eruption of teeth and jawbone diseases have been described in various species of wild ungulates, including bovines (references in RUDGE 1970). Determination of the factors that contribute to presentation of these anomalıes is interesting for regional population studies (Runge 1970; NIETHAMMER 1971; BUCHALCZYK etal. 1981; HENRICHSEN 1982, among others). The objective of this study was to describe and compare the abnormalities encountered in two geographically isolated populations of Spanısh mountain goat. Materials and methods One hundred eighty-six complete skulls were studied (103 males and 83 females). Of these, 84 (44 males and 40 females) were from the National Game Reserve of the Gredos Mountains (central Iberian U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5201-0038 $ 02.50/0 Z. Säugetierkunde 52 (1987) 38-50 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Dental and skull anomalıes in the Spanish wild goat 9) Peninsula) and 102 (59 males and 43 females) from the National Game Park of the Cazorla and Segura Mountains (southeastern Spain). The specimens were collected for scientific purposes and none was obtained specifically because it was el Skulls were cleaned by ler: in water. Age was calculated by assuming May 15th to be the birth date (unpublished personal observations), the date of death being known, and counting the number of growth rings in the horn sheaths of anımals over one year (COUTURIER 1962). In the Spanish mountain goat, coincıding with Bovidae in general, the dental formulae are the following: 01 0C 3PM 3M/31 1C 3PM 3M x 2 = 32: definitive dentition 01 Oc 3pm/3ı Ic 3pm X 2 = 20: deciduous dentition The specimens were examined for the following abnormalities: plagiocephaly (assymetric head shape caused by abnormal development), heterotopic bones (small accessory Bones). supernumerary teeth (teeth exceeding the normal number), congenital agenesis (reduced number of teeth due to the fact that certain teeth do not develop), irregular placement (teeth in abnormal positions), alveolar thinning (buccal exposition of the tooth roots), dental disease (carıes) and periodontal disease (at the periodontitis, alveolar osteitis and curative or terminal stage). To calculate the frequency of maxillary canines, canine presence and the alveolar furrows or marks that testify to its having been present were considered. Incısor or premolar agenesis was diagnosed by counting the number of teeth or alveoli and was confirmed by roentgenogram. Teeth rotated more than 30° were considered irregularly placed. The characteristics of the different stages of periodontal disease were identified following the indications of PEKELHARING (1974). The frequency of abnormalıties in each population was compared, combining the results if no significant difference was found. Likewise, frequencies were compared by sex and side of the jaw on which the abnormality appeared (right or left). The statistical method used was the X? test ın 2x2 tables, applyıng the Yates’ correction. Comparison was made only when the frequency encountered was greater than 5. When a determined anomaly affected definitive dentition alone, only specimens of age to have the definitive teeth involved were considered: 34 months or more for PM2 agenesis, 35 months for dental and periodontal affections, 46 months for incisor agenesis and 63 months for canıne rotation (VIGAL ) MACHORDOM, in press). To study the frequency of the interfrontonasal bone, males with ossification of the interfrontal, frontonasal and internasal sutures, which impedes detection of this bone, were excluded. Results The skull and dental anomalıes observed in the sample were classified according to possible intrinsic or extrinsic orıgin (FELDHAMER 1982). Intrinsic origin Plagiocephaly This was encountered in 2 specimens from the Gredos Mountains. One was a 1 month-old male and the other, an 8 year-old female. The assymetry of the skull was similar ın both cases and caused by premature ossification of the suture between the right exooccipital and supraoccipital bones. In the 1 month-old specimen, this suture was already partially ossified. The normal period of ossification is the 4th year of life for males and the 3rd for females (personal, unpublished observation). These anımals presented several very deformed bones in the neurocranıum: supraoccipi- tal, exooccipital, basioccipital, basiısphenoid and temporal. Malformation of the bones resulted in distorsion of the skull on the sagital plane. In the female specimen, this distorsion was much more marked, with the facial area twisted to the right in relation to the skull case (fig. 1a). Heterotopic bones Of 155 skulls studied, including all the females and males under 6 years, in which the interfrontal, frontonasal and internasal sutures are still unossified, a small ovoid or rhomboid bone in the confluence of these sutures was detected in 21 instances. Among these were a 7 year-old male from Gredos and an 8 year-old male from Cazorla with the 40 C. R. Vigal and Annie Machordom heterotopic bone still evident in spite of the fact that the sutures involved had already begun to ossify. We call this bone the interfrontonasal (fig. 1b). Table 1 shows the frequency of this bone in our sample. As is apparent, this frequency is greater in the Gredos group, a significant difference existing between the two populations (X? = 3.89, p<0.05). There was no sexual dimorphism for this anomaly. Table 1 Frequency of appearance of the interfrontonasal bone in two Spanish mountain goat populations Without interfrontonasal With interfrontonasal total Gredos Cazorla Total 15 6 21 Supernumerary teeth Maxillary canınes: These are small teeth situated in the maxilla immediately posterior to its union with the premaxilla. Given their position and orientation forward, they do not emerge from the gum (figs. Ic and d) and are therefore undetectable until the skull is cleaned. Table 2 lists the frequencies of these canınes (or evidence of their presence) ın both populations. The difference in frequency was highly significant (X? = 17.40, p< 0.001). The only case from Cazorla was a 4 year-old male with a mark on the left maxilla. In the Gredos population there was a grea- ter incidence of upper maxillary canınes in males (12 of 44) than in females (5 of 40), although the difference was not significant. The anomaly appeared not only in young indi- viduals, but in adults as well (in the latter, ıt was generally detected by the alveolus or marks). Six percent of all cases (84) had an upper canıne in each hemimaxilla, while 14 % had only one canıne (7 % on the right and 7 % on the left) (Table 3). Presence of Pm!: Vestigial first premolars were found in the maxillae of two specimens with definitive premolars, both from the Cazorla Mountains. One was a 3 year-old male with left PM'; in the right hemimaxilla there was a small alveolus in the corresponding space. This small tooth was 6 mm long with a diameter of 3mm and was sıtuated immediately anterior to PM? (fig. 1e). The other specimen was a 2 year-old female with a right PM! of similar characteristics and location, although slightly broken, with no remains of this tooth or its alveolus on the left side. Supernumerary molars: No supernumerary molar was encountered, but there was an additional anterior cusp in an M; of a 6 year-old male from Cazorla (fig. 1f). Table 2 Frequency of appearance of maxillary canines in two Spanish mountain goat population Without With Total canınes canınes Gredos 67 17 84 Cazorla 101 1 102 Total 168 18 186 Congenital agenesıs Agenesis of PM2: Of 141 specimens with definitive premolars studied for this anomaly, 44 did not have the definitive mandibular PM; (figs. 2a and a’). In Table 4 can be seen that the difference in the frequency of hypodontia involving this tooth between the two Dental and skull anomalıies in the Spanish wild goat 41 Table 3 Distribution of the maxillary canines, alveoli or marks encountered in the specimens of the Gredos population by age, sex and side of the jaw where located (R: right, L: left) Canine Alveolus E. R. Tee > [oje] ® z jo" ano De: SS Io) Ion Wear eoamNm N m PerkeanNnNDDe \y SIen u c> I ID IST [e,0) JERRN 185) g populations was highly significant (X? = Table 4 17.38, p<0.001). For this reason, the two samples were examined separately. The 6 cases of PM, agenesis from Gredos corresponded to females (5 unilateral and Frequency of PM, hypodontia in two Spanish mountain goat populations With Without Total 1 bilateral). In Cazorla, 23 males and 15 fema- ° PM} PM} les presented this defect, with no significant en differences between the sexes. In 30 % of these De a 6 a Cazorla 46 38 84 cases, agenesis was bilateral and in 15 %, unila- Toal 97 44 141 teral. In some instances, the milk pm,, or its alveolus, clearly smaller than that of the defini- tive tooth, were found to persist (Table 5). Maxillary PM? hypodontia was much less frequent, appearing in only 4 cases (2 bilateral and 2 unilateral) from Cazorla; these individuals also presented mandibular PM, agenesis (Table 5). Incisor agenesis: Of 119 specimens studied, 3 had incisor hypodontia (Table 6). In some cases, the tooth affected could not be specified because of the similarities in appearance of the incisors, particularly when worn, although they were never confused with canines, which are easily distinguishable (fig. 2b). Irregular placement — rotatıon Rotation of incisors or canines: In 98 individuals studied for this anomaly, 7.8 % had rotatıion of incisors or canınes (fig. 2c). There were no significant differences between populations, sexes or location on one side or another of the jaw. Only 1 case of incisor rotation was observed, the right I; of a 3 year-old female from Gredos. The rest of the rotations involved canines (Table 7). Premolar rotation: Of 142 skulls examined, premolar rotation appeared in 13 (9 from Gredos and 4 from Cazorla), no significant differences being encountered (Table 8). The rotations involved the 3 maxillary premolars (fig. 1d) and the mandıbular PM.. Molar rotation: No molar rotation was observed, although in a 6 year-old male from Gredos, the first M; cusp was rotated in both mandibles. 42 C. R. Vigal and Annie Machordom Fıg. 1. Specimens that present some of the anomalıes described in this study. a: A 7 year-old female from Sierra de Gredos with plagiocephaly; b: A 1 month-old male from Sierra de Gredos with an interfrontonasal bone; c: The same specimen as in b) with maxillary canines; d: A 4 year-old male from Sierra de Gredos with upper canınes, left PM* rotation and alveolar crater around the right PM?; e: A 3 year-old male from Sierra de Gredos wıth PM'; f: A 6 year-old male from Sierra de Cazorla with supernumerary anterior cups in M; Dental and skull anomalıes in the Spanish wild goat 43 Table 5 List of the specimens from two Spanish mountain goat populations with PM2 hypodontia or persistence of the corresponding milk teeth, according to age, sex and side of jaw where located. (F: female, M: male) Hypodontia Persistence of milk PM2 Age RPM,; LPM, R.PM? L.PM? Rpm; L.pm; R.pm? L.pm? (Years) Gredos 12 15 20 Cazorla P) ag) Gel ag] ag] lSglzg) Ko u Pose ia u Eu u 2, 3 3 3 A 5 5 5 6 6 6 6 6 6 7 7 7 7 7 7 8 8 8 10 10 10 10 10 11 11 — — Do SS ” Eeiienenikeer ee ee ee ee ne nee mRmrearmepjf kb OO IWUWUN » Extrinsic origin Alveolar thinning This defect was infrequent - only 3 cases were found, 2 females (5 and 16 years) and a 12 year-old male, all from the Gredos population. Only maxillary premolars were affected (fig. 2d). 44 C. R. Vigal and Annie Machordom Table 6 List of specimens with incisor hypodontia in two Spanish mountain goat populations (I?: incisor affected with hypodontia not identified) Locality Age Sex (Years) Incisor hypodontia | Gredos Cazorla Gredos Table 7 List of specimens of Spanish mountain goat from the two populations with rotated canines Age Canine rotated (Years) R. 1E, Cazorla Dental disease - caries Out of 139 individuals with the complete mo- lar series, 12 specimens from 5 to 13 years of age with caries (11 from Cazorla and 1 from Gredos) were observed. Only molars were affected (Table 9). Periodontal disease Only periodontitis, alveolar osteitis and their curatıve or terminal stages could be recognized because the material was studied when clean. The first stage of periodontal disease, marginal gingivitis, was undetectable. Again, only speci- mens with complete molar series were consi- dered (139: 57 from Gredos and 82 from Cazorla). The difference in the frequency of affected anımals in each group proved to be very significant (X? = 14.86, p<0.001). There was no significant difference between sexes. As can be seen in Table 10, in the Cazorla population up to 6 years there were no dis- eased individuals. The frequency of affection increased with age. In contrast, in Gredos there was an elevated percentage of affected animals, even among the young, and no clear relation with age. Periodontal disease most often involved the maxillary teeth, especially the molars (fig. 2e). Table 8 List of specimens from the two populations of Spanish mountain goat with rotated premolars, according to age, sex and side of jaw involved Age (Years) Gredos ONNUWURAWW SB —_ Cazorla Seası Rotation Dental and skull anomalıes in the Spanish wild goat 45 Table 9 List of specimens with caries Age Sex Tooth affected (Years) M, M! M? Gredos 5 Cazorla Z ae Table 10 Frequency of periodontal disease, by age Gredos Cazorla Diseased Healthy Diseased Healthy 0) ) ) 3 2. 10 17. 92 However, the large number of PM? affected in the Gredos specimens is noteworthy (fig. 1d). In only 6 cases was the curative stage of the infection observed, with alveolı either healed, or in the process of healing, after loss of the tooth (1 M,,2M!' and 3 PM,) (fig. 2a). This generally implied malocclusion (overgrowth of the tooth facing the space) (fig. 2e). Discussion Different types of cranial asymmetry have been described in Caprini: COUTURIER (1962) communicates torsion of the craniofacial axis of an Alpine ıbex (Capra ibex) and CoLoMBO (1955, in CoUTURIER 1962) describes cranioscoliosis in a 15 year-old male from Gran Paraiso National Park (Italy). Plagiocephaly may be the result of genetic factors (FELD- HAMER 1982) or extrinsic agents (BEAVER et al. 1981). The plagiocephaly we encountered in two specimens was probably intrinsic, since both were from the same population, had a similar asymetry and presented no signs of previous trauma. We found no mention in the bibliography consulted of the interfrontonasal bone. For instance, COUTURIER (1962) does not list it with the other skull anomalıes in his extensive review of the Alpine ıbex. 46 C. R. Vıgal and Annie Machordom Fig. 2. Specimens that present some anomalıes described in this study. a: A 5 year-old female from Sierra de Cazorla with PM, agenesis and sıgns of the curative phase of a paradontal affection of PM,;; a’: A 6 year-old male from Sierra de Cazorla with PM, agenesis; b: A 4 year-old male from Sierra de Gredos with Ba of an incisor; c: A 7 year-old from Sierra de Cazorla wigh right canıne rotation; d: A 4 year-old female from Sierra de Gredos with alveolar thinning of the maxillary premolars; e: The same specimen as in fig. la with M? osteitis and abnormal M' growth due to the loss of the opposing mandibular tooth The appearance of maxillary canines is the most common example of recurrence of teeth that have disappeared phylogenetically (STEELE and ParamA 1979). In Cervidae, it is the most frequent dental abnormality in species that generally lack them: Norpquist (1941, in CHAPMAN and CHAPMAN 1973) found 2 mule deer (Odocosleus hemionus) with upper canınes in a sample of 4000; Van GELDER and HoFFMEISTER (1953) encountered the same anomaly in 11 (3.38 %) of 325 white-tailed deer (Odocoileus virginianus); CHAPLIN and ATKInsoN (1968, in CHAPMAN and CHAPMAN 1973) found 5 of 133 British roe deer (Capreolus capreolus) with two maxillary canınes or alveoli, while CHapman and CHAP- MAN (1973) communicate upper canine frequencies, in the same fallow deer (Dama dama) population, of 25 % of 68 newborns and only 0.8 % of 389 follow deer over 6 months. We found no data on the presence of upper canines in Bovidae. In an earlier communication (VıcaL and MACHORDONM, in press), based on the same sample studied here, we found that 3.03 % of the goats under 6 years (97) had maxillary canınes. This frequency rose to 20 % when the Gredos population was evaluated alone and animals over 6 years were included. Dental and skull anomalıies in the Spanish wild goat 47 In contrast wıth CHAPMAN and CHAPpMan’s (1973) observations in fallow deer, superior canınes, or indications of their presence were observed with relative frequency, not only in juveniles, but in adults and old goats (Table 3). As to PM!, no information on its presence in goats was found, although it is cited in other artiodactyls, such as the American elk (Cervus canadiensis) (MORAN and FAIRBANKS 1966, in MEcH et al. 1970) or the white-tailed deer (MeEc#H et al. 1970). As for the different explanations of the appearance of additional teeth (PavLınov 1975), we consider our cases of PM! to be typical of genetic atavism, in view of the situation of these teeth immediately anterior to PM’. PM2 agenesis has often been described among the dental anomalies of different ungulate species. In Cervidae, it was observed in Odocoileus hemionus (SHORT and SHORT 1964) and O. virginianus (MEcH et al. 1970; FELDHAMER and CHAPMAN 1980). Among the Bovidae, COUTURIER (1962) communicates 11 cases of bilateral PM, agenesis and 6 of unilateral agenesis in the Alpine ibex (sample size unknown). He also cites an instance of bilateral PM; hypodontia and another of upper and lower premolars in a 16 year-old female from the Gredos Mountains. Runge (1970) published a case of PM; agenesis and a case of vestigial PM; in a total of 269 wild goats (Capra hircus) from New Zealand. As regards the incisors, hypodontia has been described in milk teeth, as well as definitive teeth, in various species, such as fallow deer (CHapman and CHAPMAN 1969) and muskoxen (Ovibos moschatus) (HENRICHSEN 1981). Agenesis can be the result of retarded tooth formation and eruption or genetically induced loss determined by a phylogenetic reduction of the dental series (LA VELLE and Moore 1932, ın BEAVER et al. 1981). Other authors propose a traumatic origin to PM; disappearance (SHORT and SHORT 1964). The origin of the anomaly, in premolars as well as incisors, was assumed to be genetic because: 1. There was no evidence of earlier trauma. 2. No alveoli were observed where the teeth should have been, except for alveolı of milk tooth size. 3. Roentgenographs revealed neither vestigial nor impacted teeth (FELDHAMER and CHAPMAN 1980). In the radiographic study of our specimens with possible PM2 agenesis, we found 3 wıth PM; impacted in a mandible, while in the other mandible there was no sign of PM3;; these specimens were classıfied as unilateral agenesis. On the other part, we detected a case of retarded formation-eruption in a 3 year-old male from the Sierra de Gredos. In another 3 PM,, traumatism would be responsible for the hypodontia observed because we found signs of their having been present. The last 4 cases were not considered agenesis. COUTURIER (1962), and VıGaL and MACHORDOM (in press) have emphasized the physiological ıinutility of PM.. Its disappearance suggests the beginning of a numerical reduction in the dental series, linked to the progressive ““molarızation” (enlargement of the molars, tending to conform an expanded chewing surface in detriment of the premolars) described by Young (1971) in all the artiodactyls. COoLYER (1936, in FELDHAMER 1982) describes tooth rotation involving only premolars in eight deer species, supposedly resulting from the reduction in premolar space occasioned by the forward displacement produced by molar eruption, which is earlier than that of the premolars. The author also indicates that this positional variation may result from genetic or environmental factors. HENRICHSEN (1981, 1982) considers that genetic factors trigger premolar rotation, as well as three other dental anomalies (hypoplasia, canıne hypodontia and loss of maxillary molar entostyles) in the muskox and therefore would be useful as genetic markers. By thıs means he differentiated local populations within the natural distribution area. BUCHALCZYK et al. (1981) found that some of the dental anomalies of wolves can be treated as characteristic of the morphotypes, which are important in regional studies of different populations. In animal populations, the incidence of abnormalıties of intrinsic origin may be associated, to a certain degree, with the homozygosis present. FELDHAMER (1982) attributed the low incidence of these anomalies in the sika deer (Cervus nippon) population 48 C. R. Vigal and Annie Machordom of Maryland to the small number of anımals originally introduced and consequently, the “founder effect’”. As can be seen ın Table 11, analysis of the frequency of different intrinsic anomalies evaluated in this study permitted us to differentiate our two Spanish mountain goat populations, given the highly significant (PM, agenesis and presence of upper canines) or simply, significant (interfrontonasal bone) differences between both groups. It may be interesting to consider these differences from an evolutionary aspect. The appearance of upper canınes (an atavistic character) was observed in the Gredos population, while a process of disappearance of PM, was discerned in the Cazorla population. As has been mentioned, the loss of a tooth that apparently has no function may indicate an evolutionary direction. The cause of the possible “delay”’ in the Gredos population may be due to the small number of individuals that founded the actual population (CABRERA 1914), which would suppose a greater homozygosis with manifestations of archaic recessive characters. Table 11 Frequency of appearance of the different intrinsic anomalies considered in this study and statistical comparison Affection Gredos Cazorla Significance Plagiocephaly Aftected Unaffected Interfrontonasal bone Affected Unaffected Maxillary canınes Affected Unaffected Presence of PM! Affected Unaffected PM, agenesis Affected Unaffected Incisor agenesis Affected Unaffected Incisor and canıne Rotation Affected not sig. Unaffected Premolar Rotation Affected not. sig. Unaffected --- anticiped frequency <5; not. sig. P>0.05; * p<0.05; *** p<0.001. Alveolar thinning is the result of internal masticatory pressures and is not associated with any pathology (SMITH et al. 1977, in FELDHAMER 1982). The incidence of this defect ın the Spanish mountain goat was low (1.6 %). Caries (destruction of tooth enamel, dentin or cementum, particularly by microorgan- ism action) is more frequent in captive than in wild animals (Rosınson 1979). Of 2687 wild ungulates examined for this disease, only 4 specimens were affected (CoLyEr 1936, ın Rosınson 1979). COUTURIER (1962) considers caries infrequent in the ibex, most often Dental and skull anomalıes in the Spanish wild goat 49 affecting M’. In view of these findings, the high percentage (13.4 %) encountered in our sample of the Cazorla population is notable. In wıld anımals, periodontal disease ıs probably due to extrinsic factors, such as penetration of food in the soft tissues during chewing or the effects of chemical products of the impacted food that destroy the epithelial surface and produce deep infection (COLYER 1936, ın FELDHAMER 1982). COUTURIER (1962) communicates the penetration of cereal stalks or cyperacea stems (sometimes 2-2.5 cm long) in the alveolus of cheek teeth in the ibex. Moreover, the extreme abrasıon to which the gums are exposed results in injuries that become infected and eventually involve the alveolar bone (Runge 1970). NIETHAMMER (1971) accounts for the relatively high frequency of periodontal disease and greater dental wear in New Zealand chamois (Rupicapra rupicapra), as compared with anımals from Styria, by the more fibrous vegetation of the alpine pastures. Nonetheless, we did not observe more wear in the specimens from Gredos, although they presented significantly more periodontal disease in comparison with the Cazorla population. Therefore, the cause of this disparity in the incıdence of dental disease must be sought in the morphology (sharp leaves, etc.) or chemical composition (irritants, mineral deficiency, high fiber content, etc.) of the plants in the diet, the differences in soil (calcareous in Cazorla and siliceous, poorer in minerales, in Gredos), and the genetic maleup of the Gredos population. The age attained by some specimens with cranıal or dental anomalies suggest that they are little more than irrelevant to individual survival. Body weights and measurements of the affected goats were comparable to those of unaffected individuals, probably indicating that the general condition of the anımals is not negatively affected by these abnormalities, a conclusion already reached by other authors (Runge 1970; PEKELHARING 1974; BEAVER et al. 1981; FELDHAMER 1982; HENRICHSEN 1982). Acknowledgements We would like to thank Dr. FERNANDO Paracıos for his advice and help during the study. The authors wish to express their gratitude to Dr. RICARDO GARCiA-GONZÄLEZ for suggestions and reading the manuscript. Thanks also to BARBARA THoMmas for translating the manuscript ınto English. This study was made possible by a grant from I.N.I.A., and it is a part of the Spanish Mountain Goat Project financed by the Ministry of Agriculture, Fisheries and Nourishment. Zusammenfassung Zahn- und Schädelanomalien beim Iberiensteinbock, Capra pyrenaica Schinz, 1838 Zahn- und Schädelanomalien beim spanischen Steinbock (Capra pyrenaica Schinz, 1838) wurden an 186 Individuen aus zwei verschiedenen Populationen der Gesamtverbreitung untersucht. Die gefunde- nen Anomalien wurden in zwei Gruppen nach ihrer inneren oder lee Entstehungsursache geordnet. Das Vorkommen einiger en Anomalien (innere Entstehung) variierte sıgnifıkant zwischen den Populationen. Besonders bemerkenswert ist der hohe Prozentsatz von auftretenden Eckzähnen im Oberkiefer bei Individuen der Gredos-Population (20 %), wie auch eine u von PM, in der Population von Cazorla (45 % der Individuen mit vollständiger Backenzahnreihe). Signifikante Unterschiede zwischen den Geschlechtern sowie zwischen den Kieferhälften wurden nicht gefunden. In bezug auf Anomalien äußerer Entstehung ergab sich eine signifikant höhere Anzahl von Paradontoseschäden bei den Gredos-Exemplaren gegenüber denen von Cazorla. Auswirkungen der Anomalien auf individuelle Überlebenschancen werden als gering bewertet. References BEAVER, T. D.; FELDHAMER, G. A.; CHapMan, J. A. (1981): Dental and cranial anomalies in the river otter (Carnıvora: Mustelidae). Brimleyana 7, 101-109. BUCHALCZYK, T.; Dynnowski1, J.; SZTEYN, S. (1981): Variation in number of teeth and asymmetry of the skull in the wolf. Acta Theriol. 26, 23-30. CABRERA, A. (1914): Fauna Iberica. Mamiferos. Madrid. Mus. Nac. Cienc. Nat. de Madrid. CHAPMAN, D. I.; CHapman, N. G. (1969): The incidence of congenital abnormalities in the mandibular dentition of fallow deer (Dama dama L.). Res. Vet. Scı. 10, 485-487. 50 C. R. Vigal and Annie Machordom CHAPMAN, D. I.; CHAPMAn, N. G. (1973): Maxillary canine teeth ın fallow deer, Dama dama. ]J. Zool. 170, 143-162. COUTURIER, M. (1962): Le bouquetin des Alpes. Grenoble: Private Press. FELDHAMER, G. A. (1982): Cranıal and ee in sıka deer. Acta Theriol. 27, 353-357. FELDHAMER, G. A.; CHAPMAN, J. A. (1980): Mandibular dental anomaly ın white-tailed deer. Brimleyana 4, 161-163. HENRICHSEN, P. (1981): Dental anomalies in muskoxen from Greenland. Vidensk. Meddr. dansk Naturh. Foren. 143, 113-124. HENRICHSEN, P. (1982): Population analysıs of muskoxen, Ovibos moschatus (Zimmermann, 1780), based on the occurrence of dental anomalies. Säugetierkld. Mitt. 30, 260-280. MechH, L. D.; FRENZEL ]r., L. D.; Karns, P. D.; Kuehn, D. W. (1970): Mandibular dental anomalies in white-tailed deer from Minnesota. J. Mammalogy 51, 804-806. NIETHAMMER, G. (1971): Die Gemsen Neuseelands. Z. Säugetierkunde 36, 228-238. PAavLinov, 1. Y. (1975): Tooth anomalıes in some Canıdae. Acta Theriol. 20, 507-520. PEKELHARING, C. ]J. (1974): Paradontal disease as a cause of tooth loss in a population of chamoıis (Rupicapra rupicapra L.) in New Zealand. Z. Säugetierkunde 39, 250-255. Rosinson, P. T. (1979): A literature review of dental pathology and aging by dental means of nondomestic anımals. Parts I and II. J. Zool. An. Med. 10, 57-65; 81-91. Runge, M.R. (1970): Dental and periodontal abnormalities in two populations of feral goats (Capra hircus L.) ın New Zealand. N. Z. J. Sc. 13, 260-267. SHORT, H. L.; SHORT, C. P. (1964): Abnormal dentition in a Colorado mule deer. J. Mammalogy 45, 315: STEELE, D. G.; PARAMA, W. D. (1979): Supernumerary teeth in moose and variations in tooth number in North American Cervidae. J. Mammalogy 60, 852-853. VAN GELDER, R. G.; HOFFMEISTER, D. F. (1953): Canine teeth in white-tailed deer. J. Wildl. Manage. 17, 100. VıcaL, C. R.; MACHORDOM, A.: Tooth eruption and replacement in the Spanısh wild goat. Acta Theriol. (in press). Young, J. Z. (1971): La vida de los vertebrados. Barcelona: Ed. Omega. Authors’ address: CarLos R. VıcaL and AnnıE MACHORDOM, Unidad de Zoologia Aplıcada, Consejeria de Agrıcultura y Ganaderia, Comunidad Autönoma de Madrid, Apdo. 127, El Encin, Alcalä de Henares, Madrid, Espana WISSENSCHAFTLICHE KURZMITTEILUNGEN Zum Morgenflug von Nyctalus noctula (Schreber, 1774) (Mammalia, Chiroptera) VonL.P. A. PERRIN Naturhistorisches Museum Basel Eingang des Ms. 25. 4. 1986 Von Nyctalus noctula liegen aus der Literatur (Übersicht bei GaisLer et al. 1979 und Roßer 1982) und aus eigener Beobachtung zahlreiche Tagflugmeldungen vor. In der Regel wurden die Tiere an sonnigen Nachmittagen im Herbst beobachtet. Im September 1985 wurden an einem künstlichen Weiher an der nordöstlichen Peripherie der Agglomeration Basel am Morgen nach Sonnenaufgang jagende Nyctalus noctula beobachtet. Detaillierte Beobachtungen konnten am 14., 15., 16. und am 21. Sep- tember 1985 vorgenommen werden. Sie ergaben alle das gleiche Bild. Als Beobachtungshil- fen standen ein Fernglas TRINOVID 8x40 und ein Fledermausdetektor QMC S-200 U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5201-0050 $ 02.50/0 Z. Säugetierkunde 52 (1987) 50-51 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Zum Morgenflug von Nyctalus noctula (Schreber, 1774) 51 zur Verfügung. Die Artbestimmung wurde im Feld mit dem QMC S-200, eingestellt auf 25 kHz (Anr£n 1981), und durch Sichtbeobachtung (KLAwITTER und VIERHAUS 1975) vorgenommen. Im folgenden sei die Beobachtung vom 14. September 1985 als Beispiel kurz geschildert (Zeitangaben in MEZ): Beobachtungsbeginn: 05.00 Uhr; Sonnenaufgang: 06.05 Uhr. Um 05.15 Uhr konnte der erste N. n. mit dem QMC S-200 wahrgenommen werden. Der schwache Empfang ließß auf eine große Entfernung (über 50 m) schließen. Ab 05.20 Uhr war ein einzelnes Individuum regelmäßig hörbar. Um 05.27 Uhr kamen 3 bis 4 N. n. hinzu. Um 05.42 Uhr konnten 6 bis 7, 3 min später 10 N. n. gezählt werden. Das Maximum an gleichzeitig beobachtbaren N. rn. wurde mit ca. 30 Individuen um 06.00 Uhr erreicht. Die Tiere konnten mit dem Fernglas beim Jagdflug über der Wasseroberfläche beobachtet werden. Nach 06.00 Uhr änderte sich das Verhalten einzelner N. n. Sie begannen höher zu fliegen, ihre Jagdaktivität ließ nach, und sie verließen in kleinen Gruppen den Weiher. In 300-400 m Entfernung zum Weiher konnte man dann — zum Teil über offenem Feld und bei vollem Sonnenschein — bis um 06.25 Uhr kleine Gruppen von N.n. in 30 bis 50 m Höhe fliegen sehen. Diese Gruppen verließen nach und nach das Gebiet, ohne daß eine Abflugrichtung festgestellt werden konnte. Um 06.12 Uhr hatte der letzte N. n. den Weiher, 13 min später auch das umliegende Gebiet verlassen. Im September 1986 konnten erneut jagende Nyctalus noctula am selben Ort beobachtet werden. Ihre Flugaktivität endete aber immer kurz vor Sonnenaufgang, wie dies auch von CRANBROOK und BARRETT (1965) und STUTZ und HAFFNER (1985-1986) für Nyctalus noctula beschrieben wurde. Beim beobachteten Morgenflug handelt es sich um eine Ausdehnung der Jagdaktivität bis in die frühen Morgenstunden. Ob die beobachtete Verlängerung der Flugaktivität um ca. 30 min in Zusammenhang mit dem äußerst trockenen Spätsommer 1985 und der damit verbundenen Nahrungsknappheit zu sehen ist, muß offen bleiben. Es kann jedoch ein analoges Verhalten, wie es NyHoLm (1965) für Myotis mystacinus beschrieben hat, vermutet werden: Als Folge der geringen Nahrungsdichte (Frühling und Herbst in Finnland) verschiebt sich der Jagdbeginn von Myotis mystacinus in die frühen Nach- mittagsstunden. Dadurch können die Tiere ıhr Nahrungsdefizit kompensieren. Literatur AHLEn, I. (1981): Identification of Scandinavian Bats by their sounds. Sw. Univ. Agr., Dep. of Wildlife Ecology, Report 6, Uppsala, Sweden. CRANBROOK, EARL OF; BARRETT, H. G. (1965): Observations on noctule bats (Nyctalus noctula) captured while feeding. Proc. Zool. Soc. London 144, 1-24. GAISLER, J.; HANAK, V.; DungeL, J. (1979): A contribution to the population ecology of Nyctalus noctula (Mammalia: Chiroptera). Acta $c. Rat. Brno 13 (1), 1-33. KLAWITTER, ]J.; VIERHAUs, H. (1975): Feldkennzeichen fliegender Abendsegler, Nyctalus noctula (Schreber, 1774) und Breitflügelfledermäuse, Eptesicus serotinus (Schreber, 1774). Säugetierkdl. Mitt. 23, 212-222. f NyYHoLM, E. $. (1965): Zur Okologie von Myotis mystacinus (Leisl.) und M. daubentoni (Leisl.) (Chiroptera). Annal. Zool. Fenn. 2, 77-123. Roger, D. (1982): Tagflugbeobachtungen vom Abendsegler (Nyctalus noctula). Nyctalus (N.F.) 4/5, 445-446. Sturz, H. P.; HArrner, M. (1985-1986): Activity patterns of non-breeding populations of Nyctalus noctula (Mammalia, Chiroptera) in Switzerland. Myotis 23-24, 149-155. Anschrift des Verfassers: Laurent P. A. PERRINn, Naturhistorisches Museum Basel, Augustiner- gasse 2, CH-4051 Basel Evidence of long distance dispersal in the Common shrew (Sorex araneus) By H. TEGELSTRÖM and L. Hansson Department of Genetics, Uppsala University, and Department of Wildhfe Ecology, Swedish University of Agricultural Sciences, Uppsala Receipt of Ms. 27. 7. 1986 Dispersal is an important component in the dynamics of small mammal populations (Kress 1979; GAINnEs and MCCLENAGHAN 1980) and plays an important role in gene flow (GAInES 1981; SLATKIN 1985; BARTON 1987). Most of our knowledge concerning dispersal derives from continuous or pulsed removal of individuals from sampling areas contiguous with undisturbed areas of similar habitats. Other studies on animal dispersal rely on drift fences with continuous trapping of individuals moving away from the areas under study. Such studies often have indicated a limited number of dispersers (VERNER and GETZ 1985) and inflow of nearby residents to removal areas (BAırD and BırnEy 1982). Radiotracking of voles (McSHEA and Mapıson 1987) has disclosed only rare instances of dispersal. In general, there is a paucity of concrete observations of long distance dispersers. Here we report evidence of long distance dispersal in common shrews (Sorex aranenus). Numerous tracks of small mammals were observed on the snow-covered ice of the lake Ekoln (see figure) in south-central Sweden during one week in February, 1986. The snow was 2-5 cm deep giving a clear imprint of the tracks. During the study period night time temperatures were between —5°C and —-15°C and the day time temperature between —5°C and +2°C. The tracks were made by woodmice (Apodemus spp.) and shrews (Sorex araneus). Several tracks on the snow were followed and plotted on maps. Along a 5 km segment of the eastern shore, we found tracks of eight common shrews (Sorex araneus), which had set out from the shore and then turned back again. All these individuals had moved 0.3-0.4 km out on the ice before turning back. In two instances common shrews succeeded in crossing a 1 km wide bay (tracks A in the figure). Their tracks led directly away from the shore with no detours, and the anımals burrowed under the snow when reaching the opposite shore. Another shrew moved 3 km (track B in the figure), making several detours after about 1.5 km. This shrew was found dead 0.5 km from the opposite shore. It was an immature female, weighed 5.1 9, and exhibited no abnormalities. The mean January-February weight (x # SD) of snap-trapped female common shrews in this area was 6.1 + 0.6g (n = 24). Another common shrew (C in the figure) travelled in the opposite direction to the previous one. This animal also made several detours roughly halfway between the shores but succeeded in reaching the opposite shore after 4 km of running on the snow-covered ice. It made some 50-60000 footprints during this trip. Calculated distances for tracks B and C are minimum distances. If all small detours and deviations from a straight line are included the figures should be increased by 10 to 20 %, giving distances of 3.3-3.6 km and 4.4-4.8 km for tracks B and C, respectively. These observations of dispersal distances of 3-5 km in the common shrew demonstrate movements more than one order of magnitude larger than the normal home ranges of these U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5201-0052 $ 02.50/0 Z. Säugetierkunde 52 (1987) 52-54 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Evidence of long distance dispersal in the Common shrew 53 animals: SuıLLıTo (1963) found im- matures generally to remain within a small area (average length 37 m) but some moved more widely (119 m re- corded). Breeding anımals had larger ranges (at least 144m recorded). MiCHIELSEN (1966) observed some anımals to move at least 250-350 m. By successive trappıng she once found a shrew that covered a dis- tance of 800 m in a few weeks. How- ever, movements in the range of one to two km in exposed environments may be a typical behaviour even at subzero temperatures. Considerably longer trıps on lakes may be acciden- tal and non-adaptive as illustrated by the dead shrew found after a disper- sal distance of 3-4 km. Still, as de- monstrated by track C, even mam- mals as small as acommon shrew are capable of crossing an exposed strip 4-5 km wide. One Apodemus sp. (probably A. sylvaticus) moved almost one km on the ice of lake Ekoln. Other authors (e.g. LoMoLIno 1984 and references Map of the northern part of lake Ekoln with tracks of therein) have observed movement Sorex araneus (A-C) observed during a week in Februa- ry 1986 distances of small rodents extending from a few m to some 2 km on ice. However, the dispersal by $. araneus reported here ıs of at least the same order as the longest distance that seems recorded when radio-tracking a small mammal, viz. 3.3 km representing a ten day return journey of a rat (Rattus norvegicus) (TAYLOR and Quy 1978). Studies of island populations of the vole Microtus agrestis (Pokkı 1981) have demons- trated regular dispersal between ıislands by swimming. A similar summer dispersal by swimming seems to occur in S. araneus (Hanskı 1986). Such studies of dispersal between islands and across lakes represent movements in environments that are extreme for small terrestrial mammals. In terrestrial landscapes dispersal between various habitats should be easier. Extensive movements of voles in natural habitats have been demonstrated by MyrıyMmäÄkı (1970) and Hansson (1977). Taken together these observations ımply that long distance dispersal, and thus the possibility of gene flow, may affect the population dynamics of small mammals in even seemingly ıisolated habitats and that a frozen body of water is no barrier for gene flow. Acknowledgements We would like to thank RıcKARD FREDRIKSSON, GORDON J. KIRKLAND Jr. and Honor PRENTICE for valuable comments on the manuscript and BIRGITTA STINCHCOMBE for correcting the language. Financial support was provided by the Swedish Natural Science Research Council. Literature BAIRD, D. D.; Bırney, E. C. (1982): Characteristics of dispersing meadow voles Microtus pennsyl- vanicus. Amer. Midl. Nat. 107, 262-283. 54 H. Tegelström and L. Hansson BARToN, N. H. (1987): The genetic consequences of dispersal. In: Dispersal in Small Mammals. Ed. by N. C. STENSETH and LiDIckEr, W. Z. London: Chapman and Hall (in press). GAaInEs, M. $. (1981): Importance of genetics to population dynamics. In: Mammalian Population Genetics. Ed. by SMITH, M.H. nom I: Ahle: University of Georgia Press. GAINES, M. $.; MCCLENAGHAN, L. R. (1980): Dispersal in small mammals. Ann. Rev. Ecol. Syst. 11, 163-196. Hansk1, 1. (1986): Population dynamics of shrews on small islands accord with the equilibrium model. Biol. J. Linn. Soc. 28, 23-36. Hansson, L. (1977): Spatial dynamics of field voles Microtus agrestis in heterogeneous landscapes. Oikos 29, 539-544. Kress, C. J. (1979): Dispersal, spacing behaviour and genetics ın relation to population fluctuations ın the vole Microtus townsenduü. Fortschr. Zool. 25, 61-77. LoMoLino, M. V. (1984): Immigrant selection, predation and the distribution of Microtus pennsylva- nicus and Blarina brevicanda on islands. Am. Nat. 123, 468-483. McSHEA, W.]J.; Mapıson, D. (1987): Alternative approaches to the study of small mammal dispersal: insights from radio velemetry. In: Dispersal in Small Mammals. Ed. by STENSETH, N. C. and W. Z. LiDickeEr. London: Chapman and Hall (in press). MIcHIELSEN, N. C. (1966): Intraspecific and interspecific competiton in the shrews Sorex araneus L. and $. minutus L. Arch. Nedoda Zool. 17, 73-174. MyrryMmäkı, A. (1970): Population ecology and its application to the control of the field vole, Microtus agrestis (L.). EPPO Publ., Ser A 58, 27-48. Pokkr1, J. (1981): Distribution, demography and dispersal of the field vole, Microtus agrestis (L.) ın the Tvärminne archipelago, Finland. Acta Zool. Fennica 164, 1-48. SHILLITO, J. F. (1963): Observations on the range and movements of a woodland population of the common shrew Sorex araneus L. Proc. Zool. Soc. London 140, 533-546. SLATKIN, M. (1985): Gene flow in natural populations. Ann. Rev. Ecol. Syst. 16, 393-430. NAYLOR, KDD OU RS] (1978) Long dieee movements by a common rat (Rattus norvegicus) revealed by radio-tracking. Mammalıa 42, 63-71. VERNER, L.; GETZ, L. L. (1985): Significance of dispersal in fluctuating populations of Microtus ochrogaster and M. pennsylvaniıcus. J. Mammalogy 66, 338-347. Authors’ addresses: HÄKAN TEGELSTRÖM, Department of Genetics, Uppsala University, Box 7003, S-750 07 Uppsala, and LENNART Hansson, Department of Wildlife Ecology, Swedish University of Agricultural Sciences, Box 7002, S-75007 Uppsala, Sweden A rare nipple anomaly in Ringtails, Bassariscus astutus (Procyonidae) By INGEBORG POGLAYEN-NEUWALL and I. POGLAYEN-NEUWALL Receipt of Ms. 27. 7. 1986 Ringtails typically have 4 mammae and 1 to 4 young per litter (POGLAYEN-NEUWALL and ToweEıLL, in press). BURT and GROSSENHEIDER (1960) reported 6 mammae, CAHALANE (1947) and LECHLEITNER (1969) litters of up to 5 young. None of these authors indicated the source of his information. TOWEILL (pers. comm.) mentioned his pet ringtail (Bassaris- cus astutus flavus) having had 5 nipples, the supernumerary probably being non-functional. On 25 May 1986 we trapped a female Bassariscus astutus arizonensis in a canyon in the Santa Rita Mountains of southern Arizona. This animal had 5 prominent mammae with well developed nipples, which appeared barely suckled; the vulva was conspicuously inflated indicating very recent parturition. We had neither measuring tools nor camera with us, therefore released the animal immediately. On 7 June we returned to carry out another trapping operation. We recaptured this U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5201-0054 $ 02.50/0 Z. Säugetierkunde 52 (1987) 54-55 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 A rare nipple anomaly in Ringtails 55 Ventral view of ringtail with 5 mammary glands. (Photo by Dr. Ivo POGLAYEN-NEUWALL) female at the same location. The animal was anesthesized with Parke-Davis Vetular (Ketamine HC]) ı.m., and the mammary region examined. Her aureolas and nipples were even more enlarged and obviously suckled (see figure). All 5 nipples were functional as evidenced be oozing of milk when manipulated by the authors. It seems probable that the anterior and central nipples of the right sıde derived from the same “anlage’”’. Both are at the same distance from, but neither is in line with the anterior nipple of the left side. In the case reported by TowEILL (pers. comm.) the fifth nipple was also on the right side, but located posteriorly to the normal paired nipples. There is no correlation between the number of functional nipples and the number of young; in 9 females with 14 litters of less than 4 young all 4 nipples were functional (POGLAYEn-NEUWALL and POGLAYEN-NEUWALL, pers. observ.). Suckling kittens show no nıipple preference and freely change nipples from one suckling bout to another (POGLAYEN- NEUWALL and TOWEILL, in press). Acknowledgement A trapping permit was generously granted by the Arızona Game and Fish Department. References BuRT, W. H.; GROSSENHEIDER, R. P. (1960): A field guide to the mammals. The Peterson Field Guide Series. Boston: Houghton Mifflin Co. CAHALANE, V. H. (1947): Mammals of North America. New York: MacMillan Co. LECHLEITNER, R. R. (1969): Wild mammals of Colorado. Boulder, Colorado: Pruett Publ. Co. PoOGLAYEN-NEUWALL, 1.; TOWEILL, D. E.: Bassariscus astutus. Mammalian Species. Publ. Amer. Soc. Mammalogists (in press). Authors’ addresses: Dr. INGEBORG POGLAYEN-NEUWALL, 1765 N. Indigo Dr., Tucson, AZ 85745, USA and Dr. Ivo PogLAavyEn-NEUWALL, Dept. of Ecology and Evolutionary Biology, University of Arızona, Tucson 85721 USA Survival of wild gerbils (Mammalia: Rodentia) parasitized by larvae of the blowfly Cordylobia anthropophaga (Insecta: Diptera) By H. Korn and L. E. ©. BrAAcK 10.5. 30.5. 20.6. Date Infestation of Tatera brantsü by the blowfly Cordylobia anthropo- phaga. Lines indicate presence of different individuals on the grid. Days of obvious infestation are marked by white boxes. Crosses indicate trap mortality. Trapping periods are marked by dotted boxes on top of the graph Receipt of Ms. 27. 7. 1986 Because of their medical and veterinary importance as vectors and reservoirs of disease, intensive studies have been undertaken to describe the parasites associated with most of the rodent species occur- ring in southern Africa (for review see DE GRAAF 1981). Yet despite of the fact that the blowfly Cor- dylobia _anthropophaga (Calliphoridae: Diptera) is widespread in Africa and their larvae develop in many species of mammals, including humans (ZumPpT 1965; DE GrAAF 1981), little is known concerning the effect of the larvae on the survival of their small mammal hosts in the field. The study was carried out during a small mam- mal survey of the Trans- vaal Provincial Nature Reserve Nylsvley (Korn 1986). Live trappıng was performed on 23 nights betwen May2 and June 22, 1984, at the be- ginning of the cooler dry season. The grid extended over 2.63 ha and was lo- cated along the ecotone between grassland and Burkea type savanna in an U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5201-0056 $ 02.50/0 Z. Säugetierkunde 52 (1987) 56-57 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Survival of wild gerbils parasıtised by blowfly larvae 57 old field. A total of 45 Tatera brantsü (excluding two males caught on the last day of trappıng), 26 Dendromus melanotis, 15 Praomys natalensis, 13 Mus minutoides and 5 Tatera leucogaster (females only) were trapped. From the five species only Tatera brantsii and Tatera leucogaster were parasitized by larval Cordylobia anthropophaga. From the latter rodent species two of the five females hosted one and sıx larvae, respectively. Both survived for at least two more weeks, then moved from the study area. Wounds caused by the fly larvae had already healed when they disappeared. Of 30 Tatera brantsii caught until May 7, eleven individuals (= 36.7 %) were para- sıtized (Fig.). Larvae occurred on most parts of the body (limbs, feet, tail, back, ears, face or genitals) with sometimes two or three larvae in one foot. By May 10th all previously recorded larvae were out. Some days later a fresh infestation occurred but by mid-May heavy infestation of Tatera brantsii had ceased. All infested anımals survived (Fig.). The loss of one parasitized male was ascribed to migration or predation. It carried only one larva and was caught only on the very edge of the grid. Only one more larva was recorded in an anımal in mid-June (Fig.). Of 24 male Tatera brantsi only 4 (16.7 %) were parasitized by Cordylobia anthropophaga, while 10 (47.6 %) of 21 females hosted one or more larvae. This difference between sexes is statistically marginally insignificant (x? = 3.667, p = 0.0525). Average infestation intensity for parasitized males was 1.25 (range 1-2) and females 1.8 (range 1-3). Although no parasite-induced mortality was detected in Tatera brantsü or other rodents, the possibility of long-term effects on life-expectancy and fecundity cannot be excluded. Sexual differences in infestation rates and average number of fly larvae per parasitized Microtus pennsylvanicus (1.35 per male; 1.49 per female: Iverson and TURNER 1968) agree well with the observations on Tatera brantsii. Most studies have recorded increased parasıtism in males (Microtus: IvERsON and TURNER 1960; JACOBSEN 1966), Peromyscus (GOERTZ 1966; SELANDER 1961) or no difference between the sexes (Microtus townsendü: BooNsTRAa et al. 1980). Acknowledgements Prof. Dr. H. REMMERT from the Philipps-University in Marburg, and Brıan J. HUNTLEy from the CSIR in Pretoria made it possible for Horst Korn to work at the Nylsvley Nature Reserve. The CSIR is thanked for providing financial and logistic support. Literature BOONSTRA, R.; KrEBs, C. J.; BEACHAM, T. D. (1980): Impact of botfly parasitism on Microtus townsendi populations. Can. J. Zool. 58, 1683-1692. DE Graar, G. (1981): The Rodents of Southern Africa. Durban and Pretoria: Butterworth. GOERTZ, J. W. (1968): Incidence of warbles in some Oklahoma rodents. Am. Midl. Nat. 75, 242-245. HUNTER, D. M.; SADLEIR, R. M. F. S.; WEBSTER, J. M. (1972): Studies on the ecology of cuterebrid parasıtism in deermice. Can. J. Zool. 50, 25-29. Iverson, S. L.; TURNER, B. N. (1968): The effect of Cuterebra spp. on weight, survival and reproduction in Microtus pennsylvanicus. Manitoba Entomologist 2, 70-75. JacoBsen, B. (1966): Unidentified Cuterebra in mice and voles. Proc. Entomol. Soc. Manıtoba 22, 29. Korn, H. (1986): Populations- und Verhaltensstudien an freilebenden europäischen, afrikanischen und nordamerikanischen Nagetieren während der ungünstigen Jahreszeit. Diss. Universität Mar- burg. SELANDER, J. A. (1961): Hematological values in deermice in relation to botfly infection. J. Mammal- ogy 42, 47-60. ZuMmPT, F. (1965): Myiasis in man and animals in the old world. London: Butterworth. Authors’ addresses: Dr. Horst Korn, Fachbereich Biologie-Zoologie, Philipps-Universität Marburg, D-3550 Marburg, FRG, and Dr. L. E. ©. Braack, Krüger National Park, Private Bag X402, Skukuza 1350, Rep. South Africa Social position reflected by contact call emission in Gelada baboons (Theropithecus gelada) By HELGA AıcH, ELKE ZIMMERMANN and H. RAHMANN Zoologischer Garten “Wilhelma” and Institut für Zoologie der Universität Stuttgart-Hohenheim Receipt of Ms. 27. 8. 1986 Gelada baboons are terrestrial primates with complex social organızation (one-male- groups, all-male-groups, bands) inhabiting the highland savanna areas of Ethiopia. Although vocal behavior of this species has been discussed in various papers about its relatively well known ecology and social behavior (e.g. KumMErR 1975; Kawaı 1979; Dungar 1984), no detailed analysis of its acoustic behavior has ever been presented. Within a comparative phylogenetic and ontogenetic study of primate vocalizations, we therefore have investigated the acoustic communication of two captive one-male groups of gelada baboons (group 1: 1 harem male with 4 adult females and two juvenile males; group 2: 1 harem male with 7 adult females and one juvenile female), maintained at a large outdoor enclosure and a smaller indoor one at the Zoological Garden “Wilhelma” at -6 years Fıg. 1. Sociogram of social grooming relations of a one male-unit of gelada baboons. The width of the lines connecting different individuals represents the relative strength of their social grooming activity. (Values were calculated by dividing the observed time an individual spend on this behavior through the total observation time [= 100 %].) A subadult male of the neighbouring group is often found ın close association to the youngest female of the shown unit and was therefore included in the illustration by an interrupted line. Barbary sheeps living together with the geladas are sometimes groomed by members of the group, too U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5201-0058 $ 02.50/0 Z. Säugetierkunde 52 (1987) 58-60 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 (%) frequency nm —_ Social position and contact calls in Theropithecus gelada 3) ch © DR Maars 2 .ad® 7 years ı EEE BIBI EEE RE TEE WETTE DENE Dean yrealns = € 2 e. ad © -5 years [oe] je] \e}} [e,} = [eo] [= [= 3 [7] = _ -_ ren MS iau=as!aie > (77) [0] [©] £. u m [o;} [4] [@% © > © . ® [a3 oo > [= > [4] . © o o [) m _— [72] » oo Pe} [7] _ [)] oo © — Fe} Oo [77] a © [e% 5 [o} © = — m 1 2 3 4 5 6 7 8 910 behaviour & a0 2 2. 0EVAeraunzs Fig. 2. Relation of contact call emission with the behavior of the sender. Values are calculated as ın Fig. 1 1 2 3 4 5 6 7 8 9 10 Stuttgart. Techniques of behavioral data collection as well as sound recording and analysis correspond to that already described in prosimians species (ZIMMERMANN 1985). We present here a part of this study demonstrating the influence of social position to the frequency of contact call emission. A prerequisite for any qualitative and quantitative analysıs of acoustic communication is a precise knowledge of the socıal relations of each individual within a social group. To estimate this, we have established sociograms of social grooming activities between the members of one-male units, based on three months data collection by direct observation sessions. As illustrated by Fig. 1, social grooming activity is not equally distributed between the individuals within the group. The harem male a. concentrates most of his social grooming activity to one particular female b., who is the highest ranking female within the group. She receives grooming acts from all members of the unit, but, by herselt, prefers grooming the harem male and her male infant. The other individuals form grooming dyads preferring one particular grooming partner. Female d. spends more time in grooming female c. than reverse, they both spend an equal amount of time grooming the male infant of female c. Female e. gives most of her grooming acts to the infant male (f.) of female b. as well as to a male from the neighbouring one-male group (h.), who is often sitting close to her. She, by herself, receives the least grooming acts from all the other individuals. Comparing this data with the literature (DunBAr 1984), we can conclude that the rank order of the females is as the following: female b. > female c. > female d. > female e. Due to their age the two infant males have to be excluded from social hierarchy of adults. Contact between members of the own unit is also maintained vocally by nearly constant 60 Helga Aich, Elke Zimmermann and H. Rahmann ee Free ren ee eer ns OL IN@EFOPIKENG@ECNS 41 gienlrald’a | M h | 250 ms Fıg. 3. Sonagrams demonstrating representative acoustic patterns of gelada contact calls (below) and associated context (above) frequency calls with harmonic structure often superimposed by noise (see Fig. 2 for sonagrams, fundamental frequency 130-160 Hz, duration 200-280 ms) given as single call or call sequence with irregular intercall intervals before or during socio-positive social interactions (Fig. 3). Call emission is mainly associated with approaching one another, huddling together or grooming one another. It couldn’t be recognized in the two infants. Comparison of the frequency of contact call emission of the different individuals with the information gained from the sociogram revealed that there exists a close correlation between social position and calling activity. Thus, the harem male as well as the highest ranking female show the highest calling rate, which declines rapidly from 4.5 % in female b. to 0.3 % in female e. A similar trend could also be found in the neighbouring one-male group. The effects, that may induce these differences in contact call emission as well as the generality of these findings for the species by itself have to be prooved by further investigations. Acknowledgements We wish to thank Prof. Dr. W. NEUGEBAUER for the possibility of observing the monkeys and Dr. A. BROTZLER and the staff of the primate facilities at the “Wilhelma” for their support. We would also like to thank Prof. Dr. W. WICKLER, Max-Planck-Institut für Verhaltensphysiologie, Seewiesen, for giving us the opportunity to work with the spectrum analyser. References Kummer, H. (1975): Rules of dyad and group formation among captive gelada baboons (Theropithecus gelada). In: Proceedings of che ie congress of the international primatological society. Ed. by S. Konno, M. Kawar, A. EHARA, $. Kawamura. Tokyo: Japan Science Press. 129-159. Kawaı, M. (1979): Ecological and sociological studies of gelada baboons. Basel: S$. Karger. DunBaR, R. I. M. (1984): Reproductive decisions. New Jersey: Princeton Univ. Press. ZIMMERMANN, E. (1985): The vocal repertoire of the adult senegal bushbaby (Galago senegalensıs senegalensis). Behaviour 94, 212-233. Authors’ address: HELGA AıcH, Dr. ELKE ZIMMERMANN, Prof. Dr. H. RAHMAnNn, Institut für Zoo- logie, Universität Stuttgart-Hohenheim, Garbenstr. 30, D-7000 Stuttgart 70, FRG BEKANNTMACHUNG Einladung Die 61. Hauptversammlung der Deutschen Gesellschaft für Säugetierkunde e. V. findet von Sonntag, den 27. September, bis Donnerstag, den 1. Oktober 1937, in Berlin statt. Vorläufiges Programm Sonntag, 27. September: Anreise 19.00 Uhr: Begrüßungsabend im Zoo-Restaurant (Zoo- Eingang Budapester Straße/Elefantentor) Montag, 28. September: 9.00 Uhr: Begrüßung und Eröffnung der Tagung durch den 1. Vorsitzenden im Hörsaal des Institutes für Pflanzenphysiologie, Königin-Luise-Str. 12 9.30 Uhr: Hauptvortrag zu dem Schwerpunkt „Kommu- nıkation bei Säugetieren“, anschließend Kurz- vorträge 17.00 Uhr: Mitgliederversammlung 19.00 Uhr: Filmabend Dienstag, 29. September: 9.00 Uhr: Hauptvortrag zu dem Schwerpunkt „Funk- tionsmorphologie“, anschließend Kurzvorträge und Poster-Demonstration Abend zur freien Verfügung Mittwoch, 30. September: 9.00 Uhr: Vorträge zum Thema ‚Fortpflanzung‘ Poster-Demonstration 15.00 Uhr: Besuch des Zoologischen Gartens (Führung durch Prof. Dr. H.-G. Kıös, Dr. H. FRÄDRICH und Mitarbeiter) 19.00 Uhr: Gemeinsames Abendessen auf Einladung des Zoologischen Gartens im Zoo-Restaurant Donnerstag, 1. Oktober: Gelegenheit zu Exkursionen und zur Besichti- gung des Institutes für Anthropologie und Hu- manbiologie (Prof. Dr. C. NIEMITZ), Fabeck- straße 15, sowie der Zentralen Tierlaboratorien (Prof. Dr. W. WıLk), Krahmerstraße 6 Alle Interessenten sind zu der Tagung herzlich eingeladen. Neben den angeführten Schwerpunkten werden wir auch diesmal wieder der Vielfalt der säugetierkundlichen Arbeitsgebiete Rechnung tragen (Kurzreferate und Poster-Demonstrationen). Das Programm mit der Vortragsfolge wird allen Mitgliedern und auf Anfrage auch Nicht-Mitgliedern rechtzeitig vor der Tagung zugesandt. Falls besondere persönliche Einladungen gewünscht werden, wird gebeten, sich an den 1. Vorsitzenden, Prof. Dr. E. KULZER, Institut für Biologie III, Auf der Morgenstelle 28, D-7400 Tübingen, zu wenden. Bitte melden Sie die Vorträge, die nicht länger als 15 Minuten dauern sollen, und Poster- Demonstrationen für diese Tagung möglichst frühzeitig, spätestens aber bis zum 15. Mai 1987 beim Geschäftsführer der Gesellschaft, Prof. Dr. U. SCHMIDT, Zoologisches Institut, Poppelsdorfer Schloß, D-5300 Bonn 1, an. 62 Bekanntmachung Den Tagungsort betreffende und organisatorische Fragen beantwortet Prof. Dr. H.-G. Krös, Zoologischer Garten Berlin, Hardenbergplatz 8, D-1000 Berlin 30, Telefon (0 30) 261 11791. BUCHBESERECHUNGEN RÜLKER, J.; STÄLFELT, F.: Das Elchwild. Naturgeschichte, Hege und Jagd des europäi- schen Elches. Mit einem Kapitel über das Elchwild im östlichen Europa von Prof. Dr. RR. Dz1EcıoLowsk1. Hamburg und Berlin: Paul Parey 1986. 285 $., 188 Abb., davon 20 farbig u. 35 Tab. Leinen DM 94,-. ISBN 3-490-40212-X Der Elch ist unter den telemetacarpalen Hirschen eine besonders imposante Erscheinung. Schon für altsteinzeitliche Jäger in vielen Teilen Europas war er eine wichtige Jagdbeute. Heute ist sein Verbreitungsgebiet kleiner geworden; Elche blieben aber für viele Jäger, auch ın Ländern, in denen der Elch nicht mehr beheimatet ıst, ein begehrenswertes Jagdtier. An diese Gruppe wendet sich das vorgelegte Werk ın erster Linie; ın ihm sind aber auch viele Daten zu finden, die ein allgemeines Interesse an der Säugetierkunde beanspruchen. Nach knappen Angaben über Abstammung und Systematik sowie den geschichtlichen Wandel ın der Verbreitung wird die Variabilität des Erscheinungsbildes vielseitig geschildert. Die Körperent- wicklung wird durch Gewichtswerte aus verschiedenen Teilen des Verbreitungsgebietes belegt und die Besonderheiten des Geweihes sorgfältig dargetan. Cervine (= hirschartige Stangen-)Geweihe und palmare (= Schaufel-)Geweihe werden als erbliche Besonderheiten unterschieden. Im südlichen und mittleren Schweden kommen beide Typen, in unterschiedlicher Ausprägung, gleich häufig vor, im Norden Schwedens überwiegt der palmare Typ. Im ökologischen Teil werden nicht nur die Standorte und die Nahrung eingehend erörtert und die Probleme des Energieumsatzes angeschnitten, sondern auch die Problematik der Wanderungen, an denen sich nicht alle Individuen einer Population beteiligen. Vor allem ın den letzten Jahrzehnten haben sich Elche als höchst anpassungsfähig erwiesen. Galten sie früher vorwiegend als scheue Taigabewohner, findet man sie jetzt auch ım Kulturland, sogar ın Städten. In den meisten Gebieten des Nordens haben die Elchbestände in den letzten Jahren explosionsartig zugenommen. Im Zusammenhang damit finden die Fortpflanzungserscheinungen besonderes Interesse. Aber auch die Regelung der Bestandsdichten wird wichtig. Darüber werden eingehende Erörterungen vorgelegt. Der Elchjagd, ihrer Geschichte und ihren Vorschriften, wird eingehend Raum gegeben. Über Versuche, Elche zu domestizieren und die bisherigen (geringen) Erfolge, sind el Angaben zu finden. Insgesamt ein vielseitiges, anregendes Werk, weil in ihm viele Sachverhalte über den Elch in klarer Form zusammengefaßt sind. W. HERRE, Kiel DATHE, H.; ScHöPps, P. (Hrsg.): Pelztieratlas. Jena: VEB Gustav Fischer 1986. 323 S., 202-Abb,, 177 Ktn DM 75, - Im pelztierkundlichen Unterricht und im Rauchwarengewerbe sind viele Pelztiernamen gebräuchlich, deren eindeutige Zuordnung zur Tierart nicht immer ersichtlich ist. Das Anliegen der Herausgeber des vorliegenden Buches ist daher in erster Linie, diese taxonomische Zuordnung klarzustellen. Sie haben sich ferner zum Ziel gesetzt, alle diejenigen Säugetierarten zu erfassen, die als Pelztiere für den Menschen von Bedeutung sind und deren Biologie zu skizzieren. Zu diesem Vorhaben konnten sie 11 Fachwissenschaftler und eine Graphikerin des Tierparks Berlin gewinnen. Diese haben in unterschied- lichem Ausmaß verschiedene Kapitel erstellt. In einem vorangestellten allgemeinen Teil werden Anatomie und Morphologie des Haares dargelegt und auf artliche Vielfalt verwiesen. Weiterhin werden behandelt: Eintluß von Haltung und Ernährung auf den Pelz, Krankheiten und Schädigungen von Haut und Haaren, Parasiten. Im anschließenden speziellen Teil werden dann überraschenderweise zunächst mehrere Vogelarten (Lappentaucher, Seetaucher, Gänsevögel) behandelt, weil deren bedunte Häute zu Bekleidung verar- beitet werden können. Dann wird auf annähernd 190 Säugetierarten (Monotremen, Marsupialia, Placentalia) eingegangen. Einigen Haustieren sind gesonderte Abschnitte gewidmet. Die Beschreibung der Arten erfolgt nach dem einheitlichen Schema: Name, Körpermaße, Gewicht, Zahnformel, Verbreitung, Fortpflanzung, Lebensweise, Nahrung, Fellqualität, Fellfarbe, Fellstruktur, Haarwech- sel, Naturschutz. Verbreitungskarten von allen Arten und Abbildungen von der überwiegenden Mehrzahl sind willkommene zusätzliche Illustrationen. Buchbesprechungen 63 Das Buch bietet in straffer Form einen guten und detaillierten Überblick über die Biologie der behandelten Vögel und Pelztiere, unter denen eine große Anzahl bedrohter, geschützter und für heutige Pelznutzung unbedeutender Arten zu finden sınd. Gerade ım Hinblick auf Pelztiere wäre eine es ederuns der Farmpelztiere und mehr Information über deren Verbreitung, Haltung, Nutzung und Domestikationsgeschichte zusätzlich erwünscht. D. Kruska, Kiel SCHEBITZ, H.; WILKEns, H.: Atlas der Röntgenanatomie von Hund und Katze. (Atlas of Radiographic Anatomy of the Dog and Cat) 4. neubearb. u. erw. Aufl. Berlin, Hamburg: Paul Parey 1986. 244 $. mit 127 Röntgenbildern, 147 Röntgenskizzen, 79 Lagerungsskizzen und 4 Tab. DM 198,-. ISBN 3-489-56416-2 Der erstmals 1967 erschienene Röntgenatlas von Hund und Katze liegt nun bereits in 4. stark erweiterter Auflage vor. Die 127 Röntgenbilder, als Negative reproduziert, werden jeweils durch Röntgenskizzen und durch Lagerungsskizzen ergänzt (insgesamt 351 Abb.). Außer dem Skelet sind Larynx, Zähne, Thorakal- und Abdominalorgane berücksichtigt. Neu eingefügt sind unter anderem sehr brauchbare Röntgenskizzen zur postnatalen Entwicklung des Extremitätenskelets. Die Qualität der Röntgenbilder ist, wie ihre drucktechnische Wiedergabe, von höchstem Standard. Der knappe Legendentext in Deutsch und Englisch ist präzis und zuverlässig. Die Nützlichkeit dieses hervorra- genden und schönen Werkes ıst längst bewiesen. D. STARcK, Frankfurt/M. Duncker, H.-R.; FLEISCHER, G. (eds.): Functional Morphology in Vertebrates. Procee- dings of the International Symposium on Vertebrate Morphology, Giessen 1983. Fort- schritte der Zoologie Bd. 30. Stuttgart, New York: Gustav Fischer 1985. 752 pp., 156 pl., 486 figs. DM 298,-. ISBN 3-437-30442-9 In einer Zeit großer und bewundernswerter Erfolge der experimentellen, kausalanalytischen Diszipli- nen ın der Biologie sind vergleichende Methoden weitgehend verdrängt worden. Morphologie wird zwar vielfach noch als Basiswissenschaft akzeptiert, gılt aber oft als abgeschlossen und unmodern. Dabei wird meist nicht zur Kenntnis genommen, daß die vergleichende Morphologie in den letzten Jahrzehnten eine beachtliche Anzahl origineller Leistungen vorzuweisen hat, neue Fragestellungen erarbeitet und durch Erweiterung in Hinblick auf funktionelle und evolutionsbiologische Probleme überzeugende Beiträge zur allgemeinen Biologie geliefert hat. Dies wird eindrucksvoll durch den vorliegenden Sammelband über die Beiträge zu dem Internationalen Morphologie-Symposion in Giessen 1983 dokumentiert. Der Band enthält 141 Einzelbeiträge (über 200 Autoren), die hier nicht einzeln besprochen werden können. Dieses Referat muß daher begrenzt bleiben und berücksichtigt in erster Linie einige allgemein interessierende Übersichtsberichte und Beiträge, die säugetierkundliche Themen (55 Einzelbeiträge) behandeln. Die Vorträge befassen sich mit Themen aus allen Teilgebieten der Morphologie, da keine Diskussions-Schwerpunkte vorgesehen waren, doch hatten sich nach der Zahl der Berichtenden gewisse Schwerpunkte Bee scbilder So war dıe Anzahl der Vorträge aus den Gebieten Nervensy- stem, Sinnesorgane, Kauapparat und Lokomotion zahlreicher als die aus anderen Gebieten. Generell kann gesagt werden, daß Ihe meisten Autoren funktionell-morphologische Gesichtspunkte deutlich herausarbeiteten, während evolutionsbiologische Überlegungen dagegen etwas zurücktraten. Anato- mische Befundbeschreibungen traten ganz in den Hintergrund. Besondere Beachtung verdient das große Referat des Initiators der Tagung, H.-R. Duncker, über die gegenwärtige Situation der Morphologie und ihre aktuelle Bedeutung für Biologie und Medizin. Die Analyse des Verhältnisses zwischen Formen- und Funktionslehre aus den historischen Wurzeln und der jeweiligen Zeitsignatur ergibt, dafß der Antagonismus zwischen Morphologie und Physiologie nur ein Scheingegensatz ıst, der überwunden werden kann und wird. Das wırd nicht geschehen, indem die Morphologıe in Physiolo- gie und schließlich in Physik und Mathematik aufgeht, sondern dadurch, daß de Komplementarität beider erkannt wird und aus diesem dialektischen Verhältnis eine vernünftige Zusammenarbeit, Ergänzung und Integration erreicht wird. Überzeugende Beispiele aus dem eigenen Arbeitsgebiet des Autors begründen diese These. Der Vorwurf gegen die Morphologie, daß ihre vergleichende Methode intuitiv und daher unwissenschaftlich sei, wird mit klaren Argumenten widerlegt. ' Aus der Reihe umfassender Sammelreferate sei auf Folgende verwiesen: NORTHCUTT gibt eine Übersicht über die verschiedenen Hypothesen zur Phylogenie des Zentralnervensystems und kommt zu einer eindeutigen Ablehnung der ‚„Parcellation-Theorie“ von EBBEsson. GANS und NORTHCUTT bringen neue Deutungen zur Neuralleistenfrage (Homologie mit epidermalem Nervenplexus der Deuterostomier und Neuralleiste als Specificum der Vertebrata. Kopfbildung als Neomorphose). DunckeR legt neue Ergebnisse seiner morphometrischen Studien an der Vogellunge vor. NIEUWEN- Huys und MEER geben eine breite, vergleichende Übersicht über a g des Hirnstammes der Anamnia. ZıswILEr referiert gleichfalls auf breiter vergleichender Basıs über Bauelemente, Konstruktionsprinzipien, Ernährungsbiologie und Anpassungstypen bei Singvögeln, 64 Buchbesprechungen Papageien und Tauben. Besondere Aufmerksamkeit finden Adaptationen an gleiche ökologische Nischen bei Vertretern nıcht verwandter Gruppen. Arı widmet eine Gesamtübersicht über alle Wirbeltierstämme dem Feinbau der Retina. Die Mehrzahl der für den Säugetierkundler relevanten Vorträge, meist in sehr knapper Fassung publiziert, behandeln Spezialthemen aus allen Teilbereichen. Als Beispiele seien genannt: PREUSCHOFT und DEMES: Begrenzung der brachiatorischen Anpassung (Hylobatidae) durch die Körpermasse. Kwıecınsky: Ca-Stoffwechsel und Abbau bei Myotis in Winterschlaf, Gravidität und Laktationspe- riode. Kııma: Onto- und Phylogenese des Brustschultergürtels bei Säugetieren. DEMES untersucht mit neuer Methodik, welche Strukturen am Primatenschädel, speziell an der Schädelbasis, als Anpassung an spezielle mechanische Beanspruchung angesehen werden können. ZELLER kommt zu einer Revision der een en Vorstellungen über die Entwicklung der Fenestra rotunda in der Säugerreihe. Mehrere Beiträge über Kaumuskeln bringen neue Einsichten (SCHUMACHER: Funktionsvergleich Reptil - Säuger; HERRING: Ontogenese und Postnatalentwicklung der Kaumuskelfunktion De Schwein; FARINA: Kaumuskelfunktion bei G/yptodon; Druzınsky: Aplodontia). PERRIN und MADDOocK beschreiben bilokuläre Mägen bei einigen afrıkanischen Nagern (Mystryomys, Cricetomys). Der Vormagen enthält zwar Bakterien, do ist Cellulose-Verdauung nicht nachgewiesen. W. MAIER analysiert die Funktionsstruktur zalambdodonter Zähne. Busse bietet eine kurz gefaßte Übersicht über seine vergleichend-ontogenetischen Studien über Kopfarterien unter Berücksichtigung taxono- mischer Wertung. HELLwING untersucht Feinstruktur des Ovars und Ovulationsvorgang bei Croci- durinae und findet weitgehende Reduktion des Antrum folliculi kurz vor der Eiausstoßung; Befunde, die ın Hinblick auf das Vorkommen solider Reiffollikel bei Tenrecidae von großer Bedeutung sind. Mehrere Beiträge behandeln quantitative Befunde am Säugerhirn (STEPHAN: Relative Hirngröße bei Soricidae und Vespertilionidae im Vergleich. JunGERsS und Orson: Galagos und Loris. KAMITA et al.: Sirenia. RÖHRS: Cephalisation und Neencephalisation unter Domestikationseinfluß. PırLorT et al.: Makromorphologie des a: FLEISCHER resümiert seine wichtigen Ergebnisse über Funktionsmorphologie des schalleitenden Apparates. Bruns beschreibt, bei aller Ahnlichkeit im Bautyp des Innenohres der Säugetiere, eine Fülle neuer Spezialstrukturen verschiedener Anpassungs- typen und findet funktionelle Deutungen. In einem originellen Beitrag vergleicht NIEMITZ zwei Formen aus ganz verschiedenen Stämmen, Eule und Koboldmaki (Tarsıus), die im gleichen Lebens- raum vorkommen und ähnliche ökologische Ansprüche stellen und zeigt, wie auf verschiedenen Wegen mit differenten Mitteln in Struktur und Verhalten die Einpassung ın die gleiche ökologische Nische erreicht wird. Der vielfältige und überreiche Inhalt macht diesen Band zu einer wichtigen Informationsquelle und bietet zahlreiche Anregungen. Vor allem aber wird die heute wieder erstaunlich hohe Vitalität der Morphologie und der vergleichenden Methode eindrucksvoll dokumentiert. Druck und Ausstattung des Werkes sind von hervorragender Qualität. Herausgeber und Verlag verdienen unseren Dank. D. STARcK, Frankfurt/M. MEyER, H.; RöHrs, M. (Hrsg.): Studium generale. Vorträge zum Thema „Mensch und Tier“. Band IIVIV. W.S. 1984/85 und 1985/86. Hannover: Verlag M. und H. Schaper 1986. 152 S. DM 19,50. ISBN 3-7944.0144-1 Die Veröffentlichung der Vorträge des Studium generale der Tierärztlichen Hochschule Hannover, deren Doppelband IIV/IV nunmehr vorliegt, verdient weithin Beachtung bei allen, die am Gespräch zwischen Geistes- und Naturwissenschaften interessiert sind, wegen ihrer hohen Qualität und wegen des Bestrebens, komplexe Tatbestände und Probleme dem jeweils anderen durch sprachlich gediegene Darstellung verständlich und interessant zu machen. Aus dem Inhalt sei besonders verwiesen auf den Beitrag von E. SCHEUCH „Das Tier als Partner des Menschen in der Industriegesellschaft‘, der sowohl die Problematik der Hobby-Tiere wie die der Tierversuche klar und nüchtern analysiert und als wichtigen Faktor der heuten Kontroverse die Tendenz eines zunehmenden Anthropomorphismus, also einer Vermenschlichung des Tieres, erkennt. Der Aufsatz findet eine wichtige Ergänzung durch die Ausführungen von H. REMMERT über den „Menschen in den vorindustriellen Okosystemen der Erde“. Ethische Aspekte der Tierversuche behandelt G. Parzık. J. Hann behandelt experimentelle und ethische Fragen zum Thema „Embryo- nen-Transfer“. CH. VoGEL untersucht das Zusammenspiel von Natur und Kultur in der Menschwer- dung. Weitere Beiträge behandeln das Thema „Krankheit bei Tier und Mensch“ (L. C. ScHuLZ) und die „Rolle des Tieres in asiatischen Kulturen“ (H. FRANKE). Die Ausstattung des Bändchens ist einfach, dafür ist es aber erstaunlich preiswert. D. STARcK, Frankfurt/M. PAUL PARE Scientific Publishers Paul Parey Journals worldwide Scientific journals leading worldwide in the fields of Biology, Zoology, Ichthylogy, Entomology, Ethology, Botany, Veterinary Medicine, Agriculture, Agronomy and Crop Science, Plant Breeding, Animal Breeding, Animal Physiology, Forestry, Horticulture, Plant Protection, Phytopathology and Wa- ter Management are well-known as a link between science, and practice. Please make the current and valuable in- formation offered by Parey Journals also available to your staff and use one of the order forms. Anatomia, Histologia, Embryologia Journal of Veterinary Medicine, Series C Journal of the World Assn. of Veterinary Anatomists ISSN 0340-2096 Angewandte Botanik (A scientific journal of applied botany) ISSN 0066-1759 Anzeiger für Schädlingskunde, Pflanzenschutz, Umweltschutz (Reports on pest control, plant and environment protection) ISSN 0340-7330 Berichte über Landwirtschaft (Reports on agriculture) ISSN 0005-9080 Berliner und Münchener Tierärztliche Wochenschrift (A specialized journal for veterinarians) ISSN 0005-9366 Bibliography of Plant Protection ISSN 0006-1387 Erwerbsobstbau (A specialized journal for fruit production) ISSN 0014-0309 Ethology formerly Zeitschrift für Tierpsychologie ISSN 0179-1613 European Journal of Forest Pathology ISSN 0300-1237 Forstliche Umschau (Forestry abstracts) ISSN 0015-7988 REQUEST To: Acquisitions Librarian, Serials / The Library Commission From: Department: E] l/We have examined the Paul Parey Journal(s) = and recommend that the journal(s) will be purchased for the Main / Department Library ® I/We would like to see the Paul Parey Journal Sub- scription Price List with subscription information / back volumes 8] Please obtain sample copies on other related Paul Parey Journals and / or latest index / sample copy in- cluding volume-index SUBSCRIPTION ORDER FORM Send to your subscription agent or directly to Paul Parey Scientific Publishers Spitalerstr. 12 : P.O.Box 10 6304 D-2000 Hamburg 1 : West Germany Available in the USA, Canada, and Mexico from Paul Parey Scientific Publishers 35 West 38th St., Suite 3W - New York, NY 10018 E32] Please enter my / our subscription to (please give title or ISSN), starting with vol. Detailed subscription information is given inside the front page of each journal FE] This is anew order ® This is a renewal Name/Address lenclose my personal cheque bank draft UNESCO coupons Date Signature Forstwissenschaftliches Centralblatt (A scientific journal on forestry) ISSN 0015-8003 To: L] Acquisitions Librarian / Serials [] The Library Commission Gb+Gw Gärtnerbörse und Gartenwelt (A specialized journal on gardening and horticulture) - ISSN 0342-4731 Journal of Agronomy and Crop Science ISSN 0044-2151 Journal of Animal Breeding and Genetics ISSN 0044-3581 Journal of Animal Physiology and Animal Nutrition ISSN 0044-3565 Journal of Applied Entomology ISSN 0044-2240 Journal of Applied Ichthyology ISSN 0175-8659 Journal of Phytopathology ISSN 0031-9481 Journal of Veterinary Medicine, Series A (Animal Physiology, Pathology and ea re a ae N 1 Clinical Veterinary Medicine) ISSN 0721-0931 International Journal of Mammalian Biology ISSN 0044-3468 Zeitschrift für zoologische Systematik und Evolutionsforschung ISSN 0044-3808 Zuchthygiene Reproduction in Domestic Animals ISSN 0044-5371 | I I = e & = Journal of Veterinary Medicine, Series B (8 (8 us] fe) (Infectious Diseases, Immunology, Food Hygiene, Q 9 [) 2 | Veterinary Public Health) se z El ISSN 0721-1856 & ) < ) i > = Marine Ecology Kae | ISSN 0173-9565 Ben. en aan & | Meeresforschung (e) =, 50 3 Reports on Marine Research S oo, ,'= SI, ISSN 0341-6836 u: 2 = — x | Natur und Recht Q an o (A specialized journal on law ofthe environment Oo < ISSN 0172-1631 & ı 7 Plant Breeding + | ISSN 0179-9541 I | Wasser und Boden (Water management and supply, soil and environment protection) ISSN 0043-0951 ai 2 9 5 Zeitschrift für Jagdwissenschaft =); S | (A scientific journal on hunting) ®D & | ISSN 0044-2887 O & \ 3 Zeitschrift für Kulturtechnik A und Flurbereinigung ® (Journal of rural engineering and development 7 ISSN 0044-2984 za j 5 Zeitschrift für Säugetierkunde j I l I l I I I I I I I l I I Deutsche Gesellschaft für Säugetierkunde 60. Hauptversammlung in Stuttgart, 28.September bis 2. Oktober 1986 Kurzfassungen der Vorträge und Posterdemonstrationen. Herausgegeben von Fritz Dieterlen, Stuttgart. Zusammenstellung: Pia Wilhelm, Stuttgart. 1986. 99 Seiten. Kartoniert 24,- DM Ein Hauptziel der Deutschen Gesell- schaft für Säugetierkunde ist, auf ihren Jahrestagungen über Säugetiere arbeitende Wissenschaftler verschie- denster Fachrichtungen zusammen- zuführen, den Gedanken- und Erfah- rungsaustausch anzuregen, um so Erkenntnisse aus den einzelnen Forschungsgebieten zu integrieren. Da auf einer Jahresversammlung nicht alle Fachrichtungen gleichermaßen vertreten sein können, wurden - wie schon in den Vorjahren — wieder Schwerpunkte gesetzt: Erstens, die Paläontologie und, damit verbunden, Fragen der Systematik und Evolution der Säugetiere. Zweitens, ein Themenkreis, der vor- wiegend physiologische und etholo- gische Probleme bei Säugetieren um- faßt, und drittens, ökologische Themen, besonders zu Fragen des Artenschutzes und der Wildbiologie. Sehr erfreulich ist, daß eine größere An- zahl von Beiträgen den Fledermäusen gewidmet ist- der bei uns am stärksten bedrohten Säugetierördnung. Zu beziehen durch jede Buchhandlun Verlagsbuchhandlung PAJL Paul Parey Spitalerstraße 12 2000 Hamburg 1 P/REV Erscheinungsweise und Bezugspreis 1987: 6 Hefte bilden einen Band. Jahresabonnement Inland: 298,- DM zuzüglich 13,80 DM Versandkosten, Gesamtpreis 311,80 DM einschließlich 7% Mehrwert- steuer. Jahresabonnement Ausland: 298,- DM zuzüglich 24,- DM Versandkosten. Das Abonne- ment wird zum Jahresanfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens am 15. November eine Abbestellung ım Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, Spitalerstraße 12, D- 2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen il für Säugetier- kunde“ erhalten die Zeitschrift unberechnet im Rahmen des Mitgliedsbeitrages. Z. Säugetierkunde 52 (1987) 1, 1-64 Das klassısche Werk über dıe Anatomie der Wirbeltiere Dieses anschauliche Werk ist vor allem für Studenten der Biologie, der Medizin und verwandter Studienrichtungen gedacht; es gibt aber auch den Spezialisten auf engeren Forschungsgebieten der Biologie zuverlässige Grundlagen für ihre wıssenschaftliche Arbeit. Darüber hinaus wendet essich an jeden, der am Beispiel der Wirbeltieranatomie die wundervolle Mannigfaltigkeit der Organismen zu verstehen sucht. % Alfred Sherwood Romer / Thomas S$. Parsons: Vergleichende Anatomie der Wirbeltiere. Aus dem Amerikanischen übersetzt und bearbeitet von Hans Frick. Mit einem Geleitwort von Dietrich Starck. 5., neubearbeitete und erweiterte Auflage. 1983. 639 Seiten mit 1310 Einzeldarstellungen, davon 123 farbig, in 435 Abbildungen. Kunststoff 98,- DM Verlag Paul Parey Spitalerstraße 12 2000 Hamburg 1 PAUL PREV "Vol. 52 (2), 65--128, April 1987 ISSN 0044-3468 C2127AF a =.=ITSCHRIFT FÜR SAUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Organ der Deutschen Gesellschaft für Säugetierkunde Hamann, Ute: Zu Aktivität und Verhalten von drei Taxa der Zwerghamster der Gattung Phodopus Miller, 1910. — Studies on activity and ethology of three taxa of dwarf hamster, genus Phodopus Miller, 1910 65 Viro, P.: Size variation in cohorts ofthe Bank vole, Clethrionomys glareolus (Schreber, 1780), in Northern Finland. — Größenschwankungen bei der Rötelmaus Clethrionomys glareolus (Schreber, 1780) in verschiedenen Alters- klassen, untersucht in Nordfinnland 76 Stark, R.; Roper, T. J.; Mac Larnon, A. M.; Chivers, D. J.: Gastrointestinal anatomy of the European badger Meles meles L. A comparative study. — Der (Gastrointestinaltrakt des europäischen Dachses Meles meles L. Eine vergleichende Studie 88 Saldern, Alice von; Schliemann, H.; Kayanja, F. I. B.; Jacob, J.: Die Analbeutel von Civettictis civetta (Schreber, 1776) (Mammalia, Viverridae). Uber ihren Bau, die chemische Zusammensetzung ihrer Sekrete und ihre biologische Bedeutung. — The anal sacs of Civettictis civetta (Schreber, 1776) (Mammalia, Viverridae). Morpho- logy, chemical composition of the secretions and functional significance 96 Bauer, J. J.: Factors determining the onset of sexual maturity in New Zealand chamois (Rupicapra rupicapra L.). — Geschlechtsreifebestimmende Faktoren bei neuseeländischen Gemsen (ARupicapra rupicapra L.) 116 Wissenschaftliche Kurzmitteilung Rieger, I.; Walzthöny, Doris: Hair brush organs in Viscachas, Lagostomus maximus. — Haarbürstenorgan bei Viscachas, Lagostomus maximus 125 Bekanntmachung 128 Buchbesprechung 128 f in ee e (D = eb) (oO > eb) 3 "VO ®_ (D a4 = eb) 3 OÖ’ E (OO gs =) O, us) (D = =) a HERAUSGEBER/EDITORS P. J. H. van BREE, Amsterdam -— W. FIEDLER, Wien — H. Frick, München - W. HERRE, Kiel - H.-G. Krös, Berlin - H.-J. KuHnn, 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 SCHRIETLEITUNG/EDITORTIATZOHEIGE 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 ıst 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 ın 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 Siam und dem Verlag erhältlich. Sonderdrucke: Anstelle einer Unkostenvergütung erhalten die Verfasser von Originalbeiträgen und Wissenschaftlichen Kurzmitteilungen 50 unberechnete Sonderdrucke. Mehrbedarf steht gegen Berechnung zur Verfügung, jedoch muß die Bestellung spätestens mit der Rücksendung der Korrekturfahnen erfolgen. | Vorbehalt aller Rechte: Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei- ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ıst auch im Einzelfall grund- sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines Beitrages ist auch ım Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber- rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen. Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of the first page of an article in this journal indicates the copyright owner’s consent that copies of the article may be made for personal or internal use, or for the personal or internaluse of specific clients. This consent is given on the condition, however, that the copier pay the stated percopy fee through the Copyright Clearance Center, Inc., 21 Congress Street, Salem, MA 01970, USA, for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective, or for resale. a Syn: from back volumes of this journal see “Permissions to Photo-Copy: Publisher’s Fee List” of the Mit einer Beilage des Verlages Paul Parey Fortsetzung 3. Umschlagseite © 1987 Paul Parey. Verlag: Paul Parey, Hamburg und Berlin. Anschriften: Spitalerstr. 12, D-2000 Hamburg 1; Lindenstr. 44-47, D-1000 Berlin 61. - Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., D-2240 Heide/Holst. Zu Aktivität und Verhalten von drei Taxa der Zwerghamster der Gattung Phodopus Miller, 1910 Von UTE HAMANN Eingang des Ms. 23. 12. 1985 Abstract Studies on activity and ethology of three taxa of dwarf hamster, genus Phodopus Miller, 1910 Studied three taxa of hairy-footed dwarf hamster, genus Phodopus Miller, 1910, to find differences between closely related species wıth similar preference for dry and semi-arid biotopes. Due to their choice of different ecological niches it comes to a distinct separation within their mutual geographic distribution area. Furthermore there are conspicuous dissimilarities in ethology and activity pattern in the species. Results of a seven-months’ comparative study on the activity of the two dsungarian dwarf hamsters (Ph. sungorus sungorus Pallas, 1770; Ph. sungorus campbelli Thomas, 1905) and the roborowski dwarf hamster (Ph. roborowskii Satunin, 1903) are presented. An investigation covering a period of sixteen successive months revealed that Ph. roborowskı has a rapıd decrease in activity in February/March down to an average daily durance of less than 10 minutes. The sympatric taxon, namely Ph. sungorus campbelli, has its main actıvity peak particularly at that time of the year with a maxımum average activity of more than 6,5 hours daily. The two allopartric taxa Phodopus roborowskii and Ph. sungorus sungorus resemble each other in their activity pattern, the activity curves being strongly similar in their general shape and even more or less parallel. Einleitung Sympatrische Arten mit ähnlichen Lebensansprüchen haben oft bei genauerer Betrachtung in mindestens einem Bereich eine wirksame ökologische Sonderung. Vergleichende Studien sind hier besonders reizvoll, wenn es sich zusätzlich um eng miteinander verwandte Spezies handelt. Verglichen wurden zwei Unterarten des Dsungarischen Zwerghamsters Phodopus sungorus und zwar Ph. s. sungorus Pallas, 1770, und Ph. s. campbelli Thomas, 1905, mit dem Roborowski-Zwerghamster Phodopus roborowskii Satunin, 1903. Besondere Beach- tung fanden dabei Merkmale, die auf eine zwischenartliche Konkurrenzsituation hindeuten könnten. Während einer umfangreichen Untersuchung der Ethologie von Kurzschwänzigen Zwerghamstern der Gattung Phodopus Miller, 1910 (Hamann, in Druck) ergaben sich deutliche Hinweise darauf, daß sich die Unterschiede keineswegs nur auf das Verhalten beschränkten. Das Auftreten starker Aktivitätsschwankungen legte eine entsprechende eingehendere Untersuchung nahe. Material und Methode Die vorliegende Studie wurde ausschließlich an Hamstern aus der eigenen Zucht durchgeführt. Bei Phodopus sungorus campbelli und Phodopus roborowskü stammt die Nachzucht von je 2,2 Tieren aus dem Zoologischen Garten Berlin-Friedrichsfelde ab. Phodopus sungorus sungorus erhielt ich aus dem Zoologischen Institut Braunschweig und im Zoohandel. Die Tiere wurden mindestens paarweise gehalten, da Phodopus bei Einzelhaltung zunehmend träge wird, sich nur noch wenig bewegt und schließlich verfettet. U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5202-0065 $ 02.50/0 Z. Säugetierkunde 52 (1987) 65-76 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 66 Ute Hamann Tabelle 1 Gegenüberstellung von Ph. s. sungorus, Ph. s. campbelli und Ph. roborowskii Phodopus sungorus Phodopus roborowskiü Ph. s. sungorus Ph. s. campbellı Fell- Aalstrich Aalstrichnur kein Aalstrich färbung ö breit, oft un- : schmal, &#: scharf abge- a scharf abge- grenzt und bis grenzt auf die Kör- erseiten aus- end Sommerfell nal Sommerfell Sommerfell sand- grau ohne gr nn erausmit gelb, ins Ocker Gelbtöne, an Rue) gelblichem spielend stellenweise Se a Schimmer) schwärzlich use sslsıhellerals#2h! bis schwarz u rt 5. Sungorus (Flanke) Winterfell im ersten Lebens- Winterfell hellgrau ohne Winterfell sandgelb, winter oft fast reinweiß gelbbraune Tönung ins Graue spielend Bauchfell im Sommer weiß Bauchfell im Sommer weiß Bauchfell weiß mit vereinzelten schwarzen mit gelblichen Übergängen Haaren, ım Winter weiß zum Rückenfell hin, im Winter weıß Fort- bereits im ersten Lebenssommer erst ım Frühling nach 5 pflanzung dem ersten Lebens- winter Lebens- ca. 2-2% Jahre ca. 1%-2 Jahre, häufig dauer deutlich senil ım Alter von 17-19 Monaten Ver- nicht berührungsscheu sehr berührungsscheu halten geradlinige, direkte Annäherung Annäherung in Halb- kreisen seitlich lau- fend bei Störung am Nest: Flucht in Höhlen oder Ecken bei Störung am Nest: Flucht aufwärts, lan- ges, bewegungsloses Verharren auf erhöh- ter Stelle Markieren kreisförmiges Markieren fehlt kreisförmiges Markie- ren rund um Weib- chen im Paarungszu- sammenhang Urinmarkieren im Handstand fehlt Urinmarkieren im Handstand Paarung Paarungsmarsch ohne Kontaktbiß Paarungsmarsch mit Kontaktbiß Hoden stets von Fell bedeckt Hoden können bal- lonähnlich ausgestülpt werden und sind dort nicht von Fell bedeckt Karawanen Karawanenbildung fehlt Jungtierkarawanen Vergleich von Aktivität und Verhalten bei der Gattung Phodopus 67 Grundfutter war ALTROMIN-Hamster-Zuchtdiät. Während die Ernährung während der Win- terzeit eintönig gehalten wurde, bekamen die Hamster vom Frühjahr an ein ech abwechslungs- reiches Futter: eine reichhaltige Mischung aus kleinen Sämereien auf der Basıs von ‚Waldvogelfutter- Mischungen‘ für kleine Stubenvögel, ferner Obst, Gemüse, Kräuter und Quark. Hundekuchen und Wasser standen ad lıb. zur Verfügung. Die zwei Dsungarischen Zwerghamster Phodopus sungorus sungorus und Phodopus sungorus campbelli zeigen nur in bezug auf die Fellzeichnung und die bei ersterem auftretende Weißfärbung im Winter deutliche morphologische Verschiedenheiten. In ihrem Verhalten gibt es nur geringfügige Abweichungen. Deutliche Unterschiede findet man dagegen bei einem Vergleich der Dsungarischen mit den Roborowski-Zwerghamstern (Tab. 1). Schwer zu beschreiben ist dabei der für den Tierhalter auffälligste Gegensatz zwischen beiden Arten: die Agilität der Tiere könnte kaum unterschiedlicher sein. Phodopus roborowskil ist ein ausgesprochen flinkes, wendiges, lebhaftes Tier, immer sehr aktiv, mit ruckartigen, kurzen, fast nervös wirkenden Bewegungen. Phodopus sungorus dagegen ıst in der Bewegung behäbig, langsam und eher träge. Phodopus sungorus wird sehr leicht völlig zahm. Phodopus roborowsku bleibt immer recht berührungsscheu. Verbreitung Das Verbreitungsgebiet der Kurzschwänzigen Zwerghamster der Gattung Phodopus Miller, 1910 erstreckt sich vom südlichen West-Sibirien im Norden des Areals, in östlicher Richtung bis zur Mandschurei und nach Süden hın bis nach Nord-China (VEsELowsky und GRUNDOVvA 1965; JORDAN 1971). Ihr Vorkommen ist dabei weitgehend an trockene Landschaftszonen gebunden. Die Verbrei- tungsgebiete der beiden Arten überschneiden sich erheblich (FLınt 1966). Trotzdem dürften aufgrund unterschiedlicher Biotoppräferenzen (Tab. 2) zwischenartliche Begegnungen selten sein: Dsungarische Zwerghamster bevorzugen feste Böden, Roborowski-Zwerghamster dagegen lockere Sande (WoRoN- zow 1961). Das Biotop des Roborowski-Zwerghamsters beschränkt sich ausnahmslos auf reine Sandgebiete, deren Grenzen er niemals überschreitet. die Vegetation spielt für Ph. roborowsku keine wesentliche Rolle, nur darf eine gewisse Dichte nicht überschritten werden. Sobald die Pflanzen den Sand mit einem Rasen überziehen, fehlt der Roborowski-Zwerghamster (FLinT 1966). Tabelle 2 Biotoppräferenzen von Ph. sungorus und Ph. roborowskii Phodopus sungorus Phodopus roborowskii feste Böden bevorzugt lockere Sande bevorzugt vorwiegend Halbwüsten und Steppen ohne ausschließlich auf reinen Sandgebieten, Buschvegetation, ausnahmsweise auch Wie- die nur schwach bewachsen sein dürfen sen, Wälder und feuchtere Gebiete Kulturfolger (Gebäude) Kulturflüchter Roborowski-Zwerghamster sind ausgesprochene Kulturflüchter und kommen niemals in der Nähe von Siedlungen vor. Die Dsungarischen Zwerghamster dagegen dringen manchmal sogar in menschli- che Behausungen ein (FLinTt 1966). Im allgemeinen scheinen Dsungarische Zwerghamster anpassungsfähiger zu sein. Sie bevorzugen Halbwüsten und Steppengebiete ohne Buschvegetation. Aber an der Peripherie ihres Verbreitungsge- bietes kommen sie streckenweise auch ın Waldsteppen, Wäldern, ausnahmsweise sogar in feuchten Birken-Espen-Kolken, „hochgrasigen“ Wiesen und ın Flußtälern vor (VEsELOwsky und GRUNDOVA 1965, cıt. Bannıkow 1954). Manchmal findet man sie in Höhen bis 3000 m (VEsELowsky und GRUNDOVA 1965, cit. VINOGRAD und GROMOW 1952). In gebüschreichen und in steinigen Gebieten sind sie selten (VESELOwsKkY und GRUNDOVA 1965, cıt. BANNIKOW 1954). Fortpflanzung In meiner Zucht wird Phodopus sungorus vielfach bereits im ersten Lebenssommer fortpflanzungsfä- hig. Bei den vor Ende Mai En Jungtieren ist das die Regel. Die Weibchen sind häufig nicht nur eine Fortpflanzungsperiode lang fertil. Haben sie jedoch eine sehr hohe Vermehrungsrate erreicht, so züchten sıe im folgenden Jahr nur zu Beginn. 68 Ute Hamann Bei Phodopus roborowsküu hingegen werfen selbst im Februar geborene Tiere nicht vor dem Frühjahr des nächsten Kalenderjahres, sind in diesem Fall also schon 12 Monate alt. Die Sterblichkeit der Zuchttiere ist nach einer Fortpflanzungsphase meistens sehr hoch. Bei vielen zeigen sich schon während der letzten Sommermonate deutliche Vergreisungserscheinungen. Die Lebensdauer ist bei Phodopus roborowsku deutlich kürzer als bei Phodopus sungorus, was in Anbetracht der für so kurzlebige Nager ungewöhnlich spät einsetzenden Reproduktionsfähigkeit (oft erst ım letzten Drittel ihres Lebens) recht auffällig ıst. Da züchtende Roborowski-Zwerghamster so rasch vergreisen, dürfte es unwahrscheinlich sein, daß sie unter den strengen Winterbedingungen in ihrem Verbreitungsgebiet die kalte Jahreszeit überleben. Die Winterpopulation müßte daher bei Ph. roborowskii fast ausschließlich aus Jungtieren bestehen. Anders liegt die Situation bei Phodopus sungorus: hier überlebten in meiner Zucht auch bei extremen Witterungsbedingungen zumindest die im Spätsommer und Frühherbst geborenen Hamster auch ihren zweiten Lebenswinter, so daß ım Winter meist zwei Generationen zugleich vorkommen (HAMANN, ın Druck). Aktivitätsverhalten Auffallende Unterschiede sowohl in den tages- als auch jahresperiodischen Aktivitätsmustern legten eine en au an der drei Zwerghamstertaxa nahe. Dazu wurde zunächst geprüft, welche der beiden unten aufgeführten Mefßmethoden sich als sinnvoll erweist. Registrierung der Benutzung des Nestausganges Sämtliche zu Nesthöhlen führenden Gänge wurden durch Klapptüren verschlossen. Jedes Offnen derselben löste einen Mikroschalter aus, dessen Impuls über einen Camil Bauer Event Recorder registriert wurde. Diese Messungen waren lediglich Teil von Voruntersuchungen, die u.a. der Abklärung versuchstechnischer Fragen dienen sollten. Haltungsbedingungen hier: Koloniehaltung HÄUFIGKEIT IN % 100 HÄUFIGKEIT IN % 100 1 2 3 4 5 6 7 8 9 10 n 12 13 14 15 16 17 18 19 20 2] 22 Abb. 1a. Häufigkeit der Benutzung des Nesteingangs bei Ph. s. campbelli vom 1.-11. Febr. 1979, Langtagbeleuchtung (Kunstlicht von 1-9 Uhr, dann Naturlicht; Koloniehaltung: je 8 Tiere). b. hen. der Benutzung des Nesteingangs bei Ph. roborowskii (Haltungsbedingungen s.o.). Eın Vergleich beider Kurven, zeigt, daß die höchsten Werte etwa 3 Std. auseinanderliegen, bei Ph. s. campbelli (Abb. 1a) zwischen 18 und 19 Uhr, bei Ph. roborowskii zwischen 21 und 22 Uhr von je 8 Tieren, Zimmertemperatur, Langtag (Beleuchtungsbedingungen s. Angabe unter Abb. 1a und b). Zu Abb. 1-2: Für alle 24 Stunden des Tages wurde für jede Art stundenweise die Häufigkeit des Passierens errechnet. Der höchste dieser Durchschnittswerte (bei Ph. s. campbelli 26 Durchläufe zwischen 18 und 19 Uhr, bei Phodopus roborowsküi 25 zwischen 21 und 22 Uhr) wurde als 100 % angenommen und mit den übrigen Prozentwerten in ein Diagramm aufgetragen. Vergleich von Aktivität und Verhalten bei der Gattung Phodopus 69 aan IL FREENET TONER. GANG ee gg, 1 2 3 4 5 6 1 8 9 10 Il 12 13 14 15 16 17 18 19 20 a 22 23 zuuuR Abb. 2. Laufaktivität (Laufrad) und Benutzungshäufigkeit des Nestausgangs bei Ph. s. campbelli Messung der Laufaktivitat Bei Betätigung der Laufräder wurde ein Mikroschalter ausgelöst und dessen Impulse wie oben registriert. Länger andauerndes Laufen in den Rädern ergab durch die in geringen Abständen aufeinander folgenden Ausschläge einen durchgehenden Balken auf dem Mefßstreifen. Haltungsbedin- gungen: paarweise, gegen Niederschläge geschützt im Freien, Naturtag. Ergebnisse und Diskussion Die hier vorgelegten Ergebnisse geben erste Hinweise für eine weitergehende, systemati- sche Untersuchung der exogenen und endogenen Faktoren der Aktivitätsperiodik an einem umfangreicheren Tiermaterial. Da im Rahmen dieser Arbeit aus technischen Gründen nur eine geringe Individuenzahl gleichzeitig untersucht werden konnte, ist eine Beurteilung der beobachteten Phänomene hier nur in Ansätzen möglich. Registrierung der Benutzung des Nestausganges Verglichen wurden zwei gleichgroße Gruppen von Phodopus sungorus campbelli und Phodopus roborowskii. Die Tiere wurden für diese Voruntersuchung an Langtag-Bedin- gungen gewöhnt, da nicht-adaptierte im Spätwinter zumindest bei der Art Phodopus roborowsku zu wenig aktiv für diese Tests sind. Die Messungen erfolgten im Februar unter künstlicher Zusatzbeleuchtung und bei Zimmertemperatur. Bei der Interpretation der Ergebnisse dieser Voruntersuchung muß daher besonders vorsichtig vorgegangen werden, da eine Beeinflussung durch eine endogene jahresperiodische Rhythmik nicht auszuschlie- ßen ist. Einige Beobachtungen seien hier dennoch vorgetragen: Durch die Kontrolle des Nestausgangs lassen sich Aktivitäten erfassen, die bei einer Registrierung der Laufaktivität in den Laufrädern nicht aufgezeichnet werden. So ist bei den Abb. 1a und 1b der geringfügige Anstieg des Kurvenverlaufs während der Hellphase darauf zurückzuführen, daß die Hamster kurzzeitig das Nest verlassen um zu harnen. Roborowski-Zwerghamster (Abb. 1b) laufen anschließend meist ohne Zwischenaufenthalt zurück. Phodopus sungorus campbelli (Abb. 1a) hingegen verläßt das Nest in dieser Zeitspanne zwar ebenfalls zum Harnen, bleibt aber auch während der Hellphase längere Zeit draußen oder läuft kurz in das Nest zurück, um es anschließend wieder zu verlassen. Beide Arten benützen in dieser Zeit nie das Laufrad. Deshalb gibt eine Kontrolle des Nestausganges hier auch ein weit realıstischeres Bild als eine Registrierung der Aktivitäten mit Hilfe von Laufrädern. Die Kurven zeigen, daß die Aktivität von Phodopus roborowskii stärker an die Dunkel- periode gebunden ist als jene von Phodopus sungorus campbelli. Auch das kann bei einer Registrierung der Laufradbenutzung nicht erfaßt werden, weil Ph. s. campbelli zwar vielfach tagsüber recht aktiv ist, jedoch bei Helligkeit das Laufrad nicht berritt. Trotzdem ist unter den üblichen Haltungsbedingungen diese Untersuchungsmethode ungeeignet. Kurzschwänzige Zwerghamster wechseln sehr häufig das Nest. Sie transportie- 70 Ute Hamann ren dazu auch das Nistmaterial zu der neuen Stelle, durchlaufen also ungewöhnlich häufig den Gang. Außerdem suchen sıe während einer Aktivitätsperiode ıhr Nest immer wieder kurz auf, wodurch sich zweifellos ein ın vieler Hinsicht fehlerhaftes Bild ergibt. Die Kontrolle der Benutzung der Nestausgänge mag zusammen mit anderen Untersu- chungsmethoden vor allem bei solchen Säugern sinnvoll sein, die selten — möglichst nur bei Aktivitätsbeginn und Aktivitätsende — durch den Ausgang wechseln. Bei Phodopus ergab A Be [' on: UHR 22 u LAUFRAD 0. RECORDER 12345678910 1112 13 14 15 16 17 18 19 20 21 22 23 24 UHR 17 18 ı 20 2 2223 4 57er ı234ı56 788101 1218 14 IS 16 I7 ı8 19 20 21 Abb. 3. Verteilung und Dauer der Aktivitätsphasen (Laufrad) bei Ph. s. sungorus (links) Ph. s. campbelli (Mitte) und Ph. roborowskii (rechts). Jeweils 1,1 Tiere, Naturtag außen, Nov. 1979-Mai 1980. Ordinate = Tage, Abszisse = Uhrzeit, 7 = Defekt an Laufrad oder Recorder (an diesen Tagen registrierte Aktivität wird nicht gewertet) Vergleich von Aktivität und Verhalten bei der Gattung Phodopus 71 eine gleichzeitige Registrierung der Laufaktivität im Laufrad lediglich, daß meistens die höchsten Werte im Passieren der Nesteingänge weder am Beginn noch am Ende der Phase der Laufradbenutzung lag (Abb. 2). Phasen der täglichen Laufaktivität im Laufrad Untersucht wurden jeweils 1,1 adulte Tiere aller drei Taxa. Alle Paare waren als Jungtiere im Alter von drei Wochen zusammengewöhnt worden und gut miteinander verträglich. Die Laufaktivität von Phodopus roborowsküi konnte über 15 Monate von Februar 1979 bis Mai 1980 registriert werden, die von Ph. s. sungorus und Ph. s. campbelli aus technischen Gründen nur 6 Monate lang während der kühlen Jahreszeit (November 1979 bis Mai 1980). Die Laufaktivität aller Kurzschwänzigen Zwerghamster ist polyphasisch (Abb. 3), wobei die täglichen Aktivitätsschwankungen ziemlich unregelmäßig über die Dunkelpe- riode verteilt auftreten. Zu Beginn und gegen Ende der täglichen Dunkelzeit können jedoch zeitweise regelmäßige Aktivitätsphasen beobachtet werden. Bei Phodopus roborowskü liegt die hauptsächliche Laufaktivität im Sommer in den späten Abendstunden und fällt im Herbst und Winter mehr und mehr in die dunklen Stunden des Spätnachmittags. In den Monaten der geringsten Laufaktivität (Januar bis März) sind die Aktivitätsphasen sehr unregelmäßig, werden ın den folgenden Monaten länger und verschieben sich in die Abendstunden. Diese Feststellungen können bei den beiden Unterarten von Phodopus sungorus nıcht in dieser Deutlichkeit getroffen werden: die Aktivität von Ph. s. sungorus ist unregelmäßig in Bezug auf Zeitpunkt des Auftretens und tägliche Dauer; die Hauptaktivität von Ph. s. campbellı liegt, mit Ausnahme von drei Perioden im Januar und April, vorwiegend in den Abendstunden. Dauer der täglichen Laufaktivität im Laufrad Innerhalb des Vergleichszeitraums von November bis Mai gibt es bezüglich der Dauer der täglichen Laufradbenutzung zum Teil erhebliche tägliche Schwankungen (Abb. 4). Auf einige Tage mit großer Laufaktivität folgen zumeist einige mit weitaus geringerer. Diese Erscheinung ist bei allen drei Hamsterpaaren deutlich zu beobachten, jedoch nicht gleichzeitig. Besonders ausgeprägt sind solche Schwankungen bei Ph. s. sungorus in den Monaten März bis Mai, bei Ph. s. campbelli in den Monaten November bis Ende Januar. Bei Ph. roborowskii treten derartige Schwankungen im Vergleichszeitraum sehr unregel- mäßig und in kürzeren Abständen auf. Zusammenhänge mit Fortpflanzungszyklen waren nicht erkennbar. Da die Aktivitätsmaxima und -minima bei den dreı Paaren jeweils an verschiedenen Tagen auftraten, ist unwahrscheinlich, daß Witterungsbedingungen auslösend waren. Da aber verschiedene Arten möglicherweise verschieden reagieren, wurde versucht, eventuelle Zusammenhänge mit der Witterung aufzuzeigen. Luftdruck Die Kurzschwänzigen Zwerghamster sind nur sehr selten über mehrere Tage hın annä- hernd gleichlang aktiv. Während des Untersuchungszeitraumes gab es aber häufig langan- haltende Perioden mit gleichbleibendem Luftdruck. Weder hier noch bei abrupten Luft- druckänderungen ergaben sich erkennbare Zusammenhänge mit dem Aktivitätsverhalten der Hamster. 2 Ute Hamann HÄUFIGKEIT IN STDN. NOv.1979 HÄUFIGKEIT ul hl "JAN.1980 MÄRZ APR. MA! h Rn nm FEB. MÄRZ APR. MAI NOY.1979 DEZ. JAN.1980 HÄUFIGKEIT INSTDN. Abb. 4. Schwankungen der täglichen Aktivitätsdauer (Laufrad) bei Ph. s. sungorus (oben), Ph. s. campbelli (Mitte), Ph. roborowskiı (unten); Naturtag außen, Nov. 1979-Maı 1980 Temperatur Auch die Außentemperatur scheint keinen Einfluß auf die Aktivitätsdauer der Tiere zu haben. Dies wird besonders deutlich, wenn man die über 15 Monate registrierten Werte von Ph. roborowskü (Abb. 5) betrachtet. Hier gibt es in den Monaten Februar und März einen Tiefstand, bis Anfang Juni einen Anstieg und dann bis zum Oktober schubweise Phasen mit großer im Wechsel mit Phasen geringer Laufaktivität. Nach einem Maximum Ende Oktober nımmt die Aktivitätsdauer bis Ende Januar wieder ab. Die ausgeprägteste Aktivitätsperiode Ende Oktober liegt in einem Temperaturbereich, der zu keiner anderen Jahreszeit mit vergleichbaren Aktivitätsschüben gekoppelt ist. Vergleich von Aktivität und Verhalten bei der Gattung Phodopus Weder während Hitzeperioden noch während Kälteeinbrüchen können aus dem Rahmen fallen- de Aktivitätsschwankungen, die durch die Temperaturänderungen ausgelöst sein könnten, beobachtet werden. Sonnenscheindauer Auch die monatliche Sonnen- scheindauer kann das Aktivitäts- verhalten kaum wesentlich beein- flußt haben — was denkbar gewesen wäre, da die Fortpflanzungsperio- dik bei Phodopus stark photoperio- disch gesteuert ist. So fällt z. B. die Zeit der höchsten Aktivität bei Ph. roborowsku in einen Monat mit mittlerer Sonnenscheindauer (Ok- tober 1979: 137 Stunden). An- schließend, im Monat November, ist die Aktivität immer noch relatıv hoch, obwohl die Sonnenschein- dauer mit insgesamt 48 Stunden ın diesem Monat minimal ist. Die Aktivitätsdauer ist von der sich jahreszeitlich kontinuierlich ändernden Tagesdauer und den Mondphasen unabhängig. Weder das gehäufte Auftreten von trüben oder heiteren Tagen noch von Frost-/Eis-Tagen ergeben Anhalts- punkte für eine witterungsbedingte Abhängigkeit der Aktivität. Soweit sich Rückschlüsse aus Untersuchungen an einer so gerin- gen Individuenzahl ziehen lassen, ist nach den vorliegenden Ergeb- nissen zu vermuten, daß die Schwankungen in der Aktivitäts- dauer weitgehend endogen gesteu- ert sind. Selbstverständlich können auch individuelle Figenschaften er- hebliche Abweichungen von Nor- malwerten verursachen. Weitere Beobachtungen zum Aktivitätsverhalten Nach der Verbreitungskarte von WORONZoWw (1961) gibt es, anders als bei FLinT (1966), für die Dsun- Aktivität Temperatur | Bl Alm AHTER a m N ll a xl IV xl IX Vin VI IV 1980 1979 a KEITH HUN * Sonnenscheindauer: [_]-100h 73 1956 o © o - = = = --] oa n a - [7 <= > [3 [--1 ad [ Li rm nn & _ = = 03 ws = = ws - - 7 =) = —‘ us = = = re) < - [7 E43 kin u Abb. 5. Aktivitätsschwankungen (Laufrad) bei Ph. roborowsku von Febr. 1979-Mai 1980, Tempera- turkurve und Sonnenscheindauer 74 Ute Hamann garischen Zwerghamster, Phodopus sungorus, mehrere völlig voneinander isolierte Popula- tionen. Leider ordnet auch WORONZOW die Areale nıcht unmittelbar den beiden Unterar- ten von Phodopus sungorus zu. Es ist jedoch anzunehmen, daß das große Gebiet im Westen von Phodopus sungorus sungorus, stets als „nordwestliche‘“ (FrLınT) oder „westliche“ (VESELOwsKY und GRUNDOVA 1965) Form bezeichnet, besiedelt wird. Weder bei FLinT noch bei WoRONZow überschneidet sich dieser nordwestliche Teil des Verbreitungsgebietes mit dem von Phodopus roborowskü. Sehr interessant ist ein Vergleich der täglichen Aktivitätsdauern im Monatsmittel für die Laufaktivität bei den Arten mit allopatrischem Vorkommen wıe Phodopus sungorus sungorus und Phodopus roborowskü (Abb. 6). Die Kur- ven verlaufen weitgehend parallel, Aktivitäts- maxima treten jeweils im November auf, die Minimalwerte nur geringfügig verschoben bei Phodopus sungorus sungorus einen Monat spä- ter als bei Phodopus roborowskü. Trägt man jedoch die entsprechenden Meß- Ph.s.sungorus ergebnisse der Arten mit sympatrischem Vor- Mm Ph.roborowskii kommen auf (Abb. 7), so ergibt sich ein völlig gegensätzliches Bild. Im November, dem Mo- nat mit der höchsten Aktivität bei Phodopus roborowskü, hat Phodopus sungorus campbellı seinen Minimalwert. Umgekehrt erreicht Pho- dopus sungorus cambelli Spitzenwerte im Mo- nat Februar, der bei Phodopus roborowsku durch eine besonders geringe tägliche Aktıvi- tätsdauer hervorsticht. Einen entsprechenden Kurvenverlauf erhält man bei dem Vergleich von Phodopus sungorus sungorus mit Phodopus Abb. 6. Durchschnittliche tägliche Aktivi- TERSIER| campbellı (Abb. 23 tätsdauer (Monatsmittel) der allopatrischen Grundlage für die vorliegenden Kurven Arten Ph. s. sungorus und Ph. born sind Untersuchungen an jeweils nur 1,1 Tieren Nov. 1979-Mai 1980 der drei Zwerghamstertaxa. Bei der Bewertung der Meßergebnisse ist daher eine gewisse Vor- sicht angebracht. Jedoch sei erwähnt, daß die während der Wintermonate aufgezeichneten Minimal- und Maximalwerte bei Phodopus roborowskil und Phodopus sungorus mit den täglichen Beobachtungen auch der folgenden Jahre übereinstimmen. Stichprobenartige Messungen bei Phodopus roborowsku zeigten, daß bei dieser Art die besonders geringe Aktivität manchmal erst im März nachweisbar ist, und daß die Erscheinung auch bei in warmen Räumen gehaltenen Tieren zu beobachten ist. Beim Vergleich der Aktivitätskurven ist bemerkenswert, daß die Dauer der gemessenen Laufaktivität gerade bei den besonders aktiven Roborowski-Zwerghamstern stets weitaus geringer ist als bei Ph. s. sungorus und Ph. s. campbelli. Dieser Befund widerspricht den täglichen Beobachtungen, denn Ph. roborowskü bewegt sich insgesamt mit Sicherheit mehr als die beiden Unterarten der Dsungarischen Zwerghamster. Wahrscheinlich hängt dieser unerwartete Befund damit zusammen, daß die flinken Roborowski-Zwerghamster pro Zeiteinheit unverhältnismäßig größere Strecken laufen als die langsamen Dsungarischen Zwerghamster. Von einer Messung der Anzahl der Laufrad- umdrehungen wären hier allerdings kaum Erkenntnisse zu erwarten: Ph. roborowskii ist zwar während der Aktivitätsphase fast ständig in Bewegung, bleibt jedoch nur selten ausdauernd bei einer Tätigkeit. Die Tiere benutzen das Laufrad viel, springen aber immer wieder heraus, laufen — oft mehrmals hintereinander - rasch das Terrarium ab, steigen dann oa = X 1’80 I V Vergleich von Aktivität und Verhalten bei der Gattung Phodopus 73 (7 ER 7 N Ph.s.sungorus Einer sn.eampbreuialfi: IN Ph.roborowskii Abb. 7. Durchschnittliche monatliche Aktivitätsdauer der sympatrischen Arten. Links: Ph. s. camp- belli und Ph. roborowsküi; rechts: Gegenüberstellung der beiden Dsungarischen Zwerghamster Ph. 5. sungorus und Ph. s. campbelli wieder für einige Umdrehungen in das Laufrad, usw. Die Dsungarischen Zwerghamster bleiben sehr viel beständiger im Laufrad. Vor allem bei Phodopus roborowsk:i, in schwächerem Maße auch bei Ph. s. sungorus und Ph. s. campbellı, ist zu bemerken, daß mit zunehmender Gruppengröße immer seltener sämtliche Tiere gleichzeitig aktiv sind. Bei größeren Gruppen (bis 8 Hamster) können oft nur während einer Fütterung mit besonders begehrter Nahrung alle adulten Zwerghamster zugleich beobachtet werden. Bei Jungtieren findet man eine gleichzeitige Aktivität aller- dings noch häufig. Bei den beiden Dsungarischen Zwerghamstern ist die beschriebene Aufteilung der Aktivitätsphasen vor allem dann auffallend, wenn einzelne Weibchen in einer größeren Männchengruppe paarungsbereit sind. In einem mit 7,1 Tieren besetzten Terrarıum waren fast ausschließlich nur drei bis vier Böcke bei dem Weibchen. Die anderen ruhten, obwohl die Paarungsaktivitäten häufig über ihr Nest und sogar über ihre Rücken führten. Stieß ein Bock aus der schlafenden Gruppe zu den paarungsaktiven Tieren, so schied nach kurzer Zeit dafür ein anderer aus der „Paarungsgruppe“ aus. Es erfolgt also selbst in dieser Situation eine gestaffelte Benutzung des zur Verfügung stehenden Raumes. Danksagung Für Anregungen und Diskussionen danke ich herzlich allen, die meine Arbeit dadurch unterstützt haben: Herrn Dr. P. J. H. van BrEr, Herrn Dr. D. DiETERICH, Frau Dr. E. MoHR, Herrn Prof. Dr. M. Mikes, Herrn Dr. W. PopuscHka, Herrn Dr. I. Savı6, Herrn Prof. Dr. M. W. SCHÄFER, Herrn Prof. Dr. D. Starck und Herrn Dr. G. Storch. Sehr geholfen haben mir Herr Prof. H. DATHE und Herr Prof. H. KLıncer, von denen ich Zwerghamster erhielt. Zusammenfassung Alle Kurzschwänzigen Zwerghamster der Gattung Phodopus Miller, 1910, besiedeln bevorzugt semi- aride Biotope. Durch die Wahl unterschiedlicher ökologischer Nischen kommt es aber zu einer 76 Ute Hamann deutlichen Aufteilung innerhalb des gemeinsamen Verbreitungsgebietes. Auch in ihrer Ethologie und in der Aktivitätsperiodik gibt es einige weitere auffallende Unterschiede. Es werden Ergebnisse einer siebenmonatigen vergleichenden Untersuchung der zwei Dsungarischen Zwerghamster (Phodopus sungorus sungorus Pallas, 1770, und Phodopus sungorus campbellı Thomas, 1905) mit dem Robo- rowski-Zwerghamster (Phodopus roborowskıi Satunın, 1903) vorgestellt. Bei letzterem wurde in einer sechszehnmonatigen Aktivitätsstudie ein äußerst markantes Aktivitätsminimum im Februar/März mit einer durchschnittlichen täglichen Laufaktivität von weniger als 10 Minuten nachgewiesen. Der sympatrisch vorkommende Dsungarische Zwerghamster Phodopus sungorus campbellı hat gerade in dieser Zeit ein Aktivitätsmaximum mit einer täglichen Laufaktivität (im Monatsmittel) von mehr als sechseinhalb Stunden. Beim Vergleich der allopatrischen Arten Phodopus roborowsku und Phodopus sungorus sungorus ıst der Kurvenverlauf nahezu parallel. Literatur BANNIKOW, A. G. (1954): Mlekopitjustschie Mongolskoj narodnoj respubliky. Isdat. akad. nauk SSSR (russ.). FIGALA, J.; HOFFMANN, K.; GoLDAU, G. (1973): Zur Jahresperiodik beim Dsungarischen Zwergham- ster Phodopus sungorus Pallas. Oecologia 12, 89-118. FLint, W. (1966): Die Zwerghamster der paläarktischen Fauna. D. neue Brehmbücherei Bd. 366, Wittenberg. HAMANN, U.: Das Verhalten der Kurzschwänzigen Zwerghamster der Gattung Phodopus Miller, 1910. D. Zool. Garten. VESELOVSKY, Z.; GRUNDOVA, $. (1965): Beitrag zur Kenntnis des Dsungar-Hamsters Phodopus sungorus (Pallas, 1773). Z. Säugetierkunde 30, 305-311. VINOGRAD, V. C.; GROMov, J. M. (1952): Gryzuny fauny SSSR. Isdat. akad. nauk. SSSR (russ.). WORoONZow, N.N. (1961): Paläarktische Zwerghamsterarten (Cricetinae Rodentia) in statu nascendi. Sowjetwiss. naturwissenschaftl. Beiträge 2, 157-162. Anschrift der Verfasserin: Ute Hamann, Anne-Frank-Str. 40, D-6000 Frankfurt/M. 50 Size variation in cohorts of the Bank vole, Clethrionomys glareolus (Schreber, 1780), in Northern Finland By P. Vıro Department of Zoology, University of Oulu, Oulu, Finland Receipt of Ms. 14. 4. 1986 Abstract Studied size variation in the bank vole, Clethrionomys glareolus (Schreber, 1780) in three localıties during different phases of population fluctuation. The 1424 specimens were obtained from trappings carried out once a month ın 1966-72. The average size decreased in winter, but no clear trend was found in the size variation in the males and females, though there were some significant differences between them. The pregnant females from the autumn cohort were significantly longer in spring (III-V) than the non-pregnant ones, but this difference had disappeared by the summer (VI-VII). A 3-year population rhythm was found at the three trapping localities. The voles born during the increase phase were larger than those born during the other phases in both cohorts until the spring. The differences between the voles from the different phases were more significant in the autumn cohort than in the spring cohort. Size was thus found to be connected with sex, age class and the phase of population fluctuation. U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5202-0076 $ 02.50/0 Z. Säugetierkunde 52 (1987) 76-88 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Size varıation in cohorts of the Bank vole in Northern Finland ZU Introduction Studies concerning the size and growth of the bank vole, Clethrionomys glareolus (Schreber, 1780), have shown that birth time has a great influence on growth (WASILEwSKI 1952; Kusık 1965; BuJaLskA and GLıwiıcz 1968; CRAwWLEY 1970; ZEJDA 1971; KAIKUSALO 1972; WIGER 1979). The mean body size in a bank vole population diminishes ın winter (HAITLINGER 1965; GRUBER and KAHMANN 1968; HyvÄrINEN and HEIKURA 1971; KaAıkusaLo 1972). Sex-related sıze differences have been mentioned in some studies (STEIN 1956; HAITLINGER 1965), but no differences are recorded in the papers of BuJaLskA and Gwiwıcz (1968) and KaıkusaLo (1972). Bank vole populations fluctuate in the northern parts of the distribution area, with the phase of fluctuation affecting body size (Hansson 1969). The aim of this work is to explain the influence of factors such as season, sex, age class, trapping sıte and phase of population fluctuation on the size of the bank vole, and to study growth ın bank voles on the basıs of accurate determinations of age and regular monthly trapping results. The material permitted the above data to be examined over an extended period of time in the northern part of the distribution area of this species, where its population fluctuations are clear. Material and methods The material consists of bank voles trapped by the Department of Zoology at the University of Oulu from October 1966 to September 1972. The trapping sites were chosen from the same latitude, as follows: Tupos (Kempele) 64°55’ N: 25° 30’ E, Muhos 64°50’N: 26°05’E and Paljakka (Hyrynsalmı) 64° 40’ N: 28° 05’ E. Two biotopes were chosen at each locality: a spruce forest and abandoned field. At Tupos and Muhos the field was an open area, but at Paljakka ıt comprised the banks of ditches and partially overgrown forest clearıngs. Paljakka also differs from the other sites ın having a thicker snow cover (Vıro 1974a). The trappings were carried out once a month for one night at each locality, involving 100 traps both ın the forest and in the fields (for details, see HyvÄrınEn and HEIKURA 1971). 1424 specimens were obtained ın all. The numbers of specimens measured are given in Table 1. The bank voles were weighed and their length from the tip of the snout to the anus was measured (SIIVONEN 1977). Age determination was based on the length of the roots of the first mandibular molar (Vıro 1974b). The material was classified into two age classes or cohorts, those born before July being placed ın the spring cohort (C,) and those born during or after July in the autumn cohort (O;). Table 1 Numbers of bank vole specimens measured Tupos Muhos Paljakka Forest Fields Forest Fields Forest Fields Weight [6X0) 517 159 176 non-pregnant 22 129 132 126 pregnant ?? 16 16 22 > 307 324 Length 6%o) 133 164 non-pregnant ?? 106 113 pregnant ?? 16 20 > -kew-ysıe = Zurids “Arenigag—aqumss9q = I9IUIM “OqUDAON-ISqwaIdag = uumne Ysndny->unf = J9urung 110409 uunıne = 7 “1oyo9 Sunds = !7 100°0 > 6/ Erz 0728 8l ses 656 soo0> CHIEE3977 10388 II Ye S’06 uummnane IS] esgelfed 1000 > a i 198 60€ . J9JUIM 15] 100°0 > IS - 6/8 e9r ; 3 ; i ; uumnne 1S] soynW sono > 95 E c98 ce i uummıne 1S] sodn], ww 3y3ua]j Apogq JIWUWINS puz uwn)ne 35] eyyelfeg JOJUIM IS] uuınıne 1S] soynW uuınıne Ss] sodn, 3 1y31ı9m Apogq x sofeul9 (sopewoy aueudasd Zurpnf>x>) SJ10409 U} U99MJ>G S2[JoA YJueq ayJ ur yy3ua] pue Iydram Apoq ur s>9US153771P Juesipusis Afeaysiyess CA IZE, Size varıation in cohorts of the Bank vole in Northern Finland 1) Since the number and size of the embryos have a decisive influence on the total weight of a pregnant female, embryos longer than 5 mm have to be taken into account when measuring the body weights of females (MAzZAx 1962). Consequently all the visibly pregnant females (72 individuals) were excluded from the calculations of mean weights. The lengths of the visibly pregnant females were also treated separately because of the importance of maintaining homogeneity in the material. The lactating females were included with the other female voles. Each trapping site showed a 3-year rhythm of population fluctuation during the period studied. This was synchronized at Tupos and Muhos, but delayed by one year at Paljakka. At Muhos the peak was so unclear (VıRO 1974a) that the material was omitted from the present comparison. The material was analysed by taking the specimens born in the peak year as a sıngle group, which was then traced through to the following crash year. The specimens born during the crash and increasing years were then dealt with in the same way. Results Seasonal variation in body size The bank voles varied in weight and length according to season in a different manner in each age class (Figs. 1 and 2, Table 2). Those born in the first half of the summer (C,) had grown a little by the autumn, but their mean length seemed to diminish in winter, while the 108 28r 104 en L le; ae 100 1 NY 3 E 5 4r K E il ; c 96 a ; [o)} ‘D N „ ö 220 5 \ = 92 & Y = ae) ° 8 18 MD 88 16 84 14 (N Se —L_—_St I — len} SA WS S A W S 1Sseasons SI RT AWIEISTEST ERWIES” "Siseasons Mom78r 3a 21 290 8 nLC 169 8 40 49 33 11 NG, 438 139 120 78 30 9 3 3nC, SEE EI ER N Son Fig. 1. Seasonal varıation in mean body length in the bank voles (excluding pregnant females). Sym- Fig. 2. Seasonal variation in mean body weight in the bank voles (excluding pregnant temales). bols: = C,---=C. Vertical lines repre- Symbols: —@&0 222 1@, Nertiealhlınes sent one standard error on either side ofthe mean represent one standard error on one side of the mean Table 3 Body length in the bank voles in autumn and winter (excluding pregnant females) lst autumn 1st wınter lst autumn 1st winter 2nd autumn 2nd winter n A ——— 0.05 438 < 0.001 139 se 065 80 1067 C Body length mm S A WEES S A seasons ZU 270,8307 iS 142 37 7 11 16 ll Fig. 3. Seasonal varıation in mean body length in the bank voles of the spring cohort (C}), excluding pregnant females. Symbols: @-@ = younger age group, € —* = older age group . Vıro mean length of the autumn cohort (C,) did not diminish until the second winter (Fig. 1, Table 3), when about 2% of the specimens from the first autumn were still alive. The voles belonging to the autumn cohort were shorter than those belonging to the spring cohort in their first autumn (Table 3). In the C, there was a difference of five months between the youngest and oldest in- dividuals in every season, and consequently this cohort was divided further into two age groups and the length variations ın these studied separately (Fig. 3, Table 4). Since the younger age group accounted for nearly 53 % of this cohort in autumn and 79 % in winter, their small size served to reduce the mean length of C,. There was no correspond- ing difference in mean length in C, even though the age range inside the age class was more than five months. Table 4 Body length in the bank voles of C, in autumn and winter Autumn older age group younger age group Winter older age group younger age group 6.23 4.58 934 89.4 < 0.01 Older age group = individuals born in Aprıl-May, younger age group = individuals born ın June. Growth was vigorous in spring, and the voles reached a mean length of 103.4 + 0.83 mm in the summer (Fig. 1). The animals in the two cohorts grew differently in spring, however, the C, individuals at Paljakka diminishing in weight from March to May and ın length from April to May, the same also being seen separately in the males and females, whereas the C, specimens increased steadily in weight and length during the spring at every trapping site. Size variation among pregnant and lactating females The visibly pregnant females ın the spring cohort at Paljakka were significantly longer in summer than the non-pregnant ones (t = 2.179, p < 0.05), as also were the pregnant females of the autumn cohort in spring at Muhos (t = 4.173, p < 0.005) and at Paljakka (t = 2.883, p < 0.01). The trend was similar at Tupos, but the difference was not statistically significant (Fig. 4). The lactating females were heavier and longer than other non-pregnant females at the same period (summer and autumn) (Table 5). Size varatıon in cohorts of the Bank vole in Northern Finland 81 Table 5 Body weight and length in the female bank voles in summer and autumn (excluding pregnant females) Weight (g) lactating non-lactating Length (mm) lactating non-lactating Fig. 4. Seasonal varıation in mean body 110 length in the female bank voles. Symbols: O---0 = pregnant, ®—@ = non-pre- 102 gnant. Vertical lines represent one stan- dard error on one side of the mean. The 0 94 asterisks represent statistical significance =; 786 an“ : S AA W S Sseasons Size variation between Er Dre Me, 2 a trapping sites = .. . . . . CT, There was some variation in sıze be- 98 tween the voles caught at the three ®»% AR N trapping localities, with those from 05 Tupos in the first autumn smaller = 0 than the others (in Cı t = 2.704, d Z = 76,p < 0.005 and in C,t = 1.966, 2 oe df = 436, p < 0.05). Body size was similar in both Er biotopes in all cases except the C, < females at Paljakka in spring, where 190 the females were heavier in the fields = than in the forest (fields: x = 20.5 g, 82 Er een S.D. = 4.19, n = 9 and forest: x — „34 18 10.7 2a a7 23,23 12 15.9 g, S.D. = 2.04, n = 18; t = U rn ee 3.148, p < 0.025). This shows that the field area at Paljakka provides good living conditions, as reflected in the number of specimens captured (Table 1). Because of the great similarity in the body size between the forest and field areas, the catches from the two biotopes are combined in the following chapters. Size variation between cohorts and sexes The differences in size between the cohorts were apparent before wintering (Figs. 1 and 2). In the first autumn of their lifespan the C, specimens were significantly longer than those in C; (Table 2), a corresponding difference in weight being seen at Tupos and Muhos only among the females. In winter the differences disappeared, while at Paljakka the C, males remained heavier than those of C, even after wintering. Significant differences in size between the sexes were found irregularly at different 82 P. Vıro trapping sites and in different seasons, but the females were most often heavier, whereas the males were characteristically longer (Table 6). Since the differences between the sexes were small and irregular, the males and females were combined. Table 6 Statistically significant differences in body weight and body length between the male and non- pregnant female bank voles Females S.D. Body weight g Muhos C, 1st winter Paljakka C, 1st autumn C, 2nd summer 2nd autumn Body length mm Tupos C, 1st winter Muhos C, 1st autumn lst wınter Paljakka C, 1st summer lst autumn C, 2nd autumn Size variation during population fluctuation The voles born during the increasing year were heavier in winter only in C, at Paljakka compared with those born during the peak year (Fig. 5, Table 7). There were more differences between the fluctuation phases in the autumn cohort than in the spring cohort (Figs. 5 and 6, Table 7). At Tupos the voles born during the increasing year and still alive in the peak year, were the heaviest through to the spring. The size relationships at Paljakka were similar, but not as clear, those born during the increasing year being the largest and those from the low year the smallest. The weight of the wintered males was greater during the peak phase than during the low phase of the fluctuation (Fig. 7), the mean being 23.3 g ($.D. = 2.09, n = 37) at the peak and 20.7 g (S.D. = 2.07, n = 24) in the low phase. The difference was highly significant (t = 4.896, p < 0.001). The wintered males did not differ in length from one phase of fluctuation to another (peak: x = 102.3 mm, $.D. — 4.93, n = 34; low phase: x = 101.1 mm, S.D. 197m 22). Discussion Seasonal variation in size The mean length and weight of the bank voles diminished in winter. A smaller size would be more economical on energy in winter (KarELA 1957; IvErson and TURNER 1974; BEACHAM 1980), but because of the limitations in this material the individual reduction of Size varıation in cohorts of the Bank vole in Northern Finland 24 1067 c BOITC, hi =3> „20 (7) ° a S16 5 Fr (m N 12 E ee E (Ss =) = ® Do > fe >, 2 = 106 OD > m=20 °oo © Ox ns mx & 16 = a © a 12 SAWS S AW SS _A seasons SI aT 572g - : 39 10*(2 5 2 60-l8,,108 6,2 24 12 12 7 13 64 35) "5 24 Fig. 5. Seasonal varıation in mean body weight in the bank voles (excluding pregnant females) during the different phases of population fluctuation. Symbols: a Fig. 6. Seasonal variation in mean body length in the bank voles (excluding pregnant females) during the different phases of population fluctuation. Symbols as in @®@ = born during peak years, b, ©---@ = born Fig. 5 during decreasing years, c, O---O = born during increa- sing years. The asterisks represent statistical significance 8 al 7 phase 6 s) o = 4 ei ? 2 a 1 o 076 low S phase 3: 53 ee Fig. 7. Body weight in the wintered bank vole males 1 during the peak and low phases of population fluctua- la 1977720723725 727 tion. The voles were captured in June-August and repre- sent combined material from Tupos and Paljakka Body weight g weight cannot be proved. On the other hand, weight may diminish only in one age class (ZEIDA 1971) or only during certain phases of the population fluctuation (BEACHAM 1980). It is obvious that at least this has some influence on the winter reduction in body weight. The loss ın weight ı is clearer here in C,, but can be detected in both cohorts. Declines ın mean weight in the bank vole in winter are also mentioned by HAITLINGER (1965), ZEJDA (1971) and KaıkusaLo (1972), and the same phenomenon is also encountered in other microtines, e.g. Clethrionomys rufocanus (KALELA 1957), Microtus pennsylvanicus (IVER- son and TURNER 1974), M. oeconomus (WHITNEY 1976) and M. townsendi (BEACHAM 1980). Enzyme activity in the bones of the bank vole is low ın winter, (HyvÄrınen 1969a; HyväÄrınen and HEIKURA 1971), which should explain the cessation in growth. In the ‘9 ‘317 ur duo] Apoq 107 9soyl pure g “arg ul U9AlS 1er 1y39M Apoq 107 s9AInD OU], ‘oseyd 3ursea109ut 9yI 3ummp uog =D ‘oseyd MO] UI Zununp woq=g ‘oseyd sjead 9yI Zuump wog = V c0o> 05 6L'€ c88 95 r8’e c88 ccoo> G8 06€ L'98 uunne ıs] 9 espgelfeg 8 66 / 296 coo0> (e ce 7 WS 100> 6C ccH 076 3unds 15] 9 sodn], wu yn3uaf Apog so0n'0 > 176 Ze rc ; awums puz c0> GE os pl c0> 8l ce l LET 61 cal Lac) J9JUIM IS] sccoo > (ce) rg LSl 1% 18° 1 OReT J9JUIM IS] 'D espgelfed co0> 6 cye e’6l c0> T 870 6cl 0€ 36€ 091 Zuuids 15] 1000 > rl v2 NN co00> < Ir 9.5] LIC AT 6€el JOJUIM IS] “ı e/'0 Ir 00.0 > Aal Sa Isı uumane Is] sodn, 3 1y3Mm Apog d u ars x d u ‘Ts X d SURHEEV. SR EN U99MI9G U99MI9A UIMIOG sJ9uU919JJI ®) sP9U9A1JJLA q s99uU919JI V apA9 uonendod ayy Jo saseyd 39143 3yJ UaMIaq yy3us] Kpoq pue Iy3ıam Apoq ur sSSJU913JJ1P yuesipusis AesuseIs NL Size variation in cohorts of the Bank vole in Northern Finland 85 shrew, the skeleton is even reduced in size in winter through diminution of the interverte- bral discs and cartilage (HyvÄrınen 1969b), and the skull of the bank vole sımilarly shrinks in winter (HAITLINGER 1965). A cessatıon ın growth is also noted by ZEjpaA (1971) and KaıkusaLo (1972). The winter decline in the mean length of a bank vole population is explained by HAITLINGER (1965) as being due to those juveniles born in the second half of the breeding season, which grow slowly. The same phenomenon could also be seen in this material if cohorts were combined, but here the voles of C, “diminish” in length, too. In ZEejpa’s (1971) material the mean length decreases in the first of the three cohorts, because about half of the individuals in this cohort die in the autumn, and particularly the longest ones. The same has been noted ın M. oeconomus (MARKOWwskI 1980). Division of the present C} into two age groups shows that the youngest voles remain the smallest in winter and serve to reduce the mean length of their great number. Thus no real shrinking of the skeleton is needed to explain the decline. The onset of growth in spring occurs in March, as has also been observed to be case in Southern Sweden (BERGSTEDT 1965), Poland (HAITLINGER 1965) and England (CRAwLEY 1970). The voles belonging to C} at Paljakka form an exception to the normal spring growth model. Here the largest anımals, which cause the clear peak in March, belong to the material from 1969, which represents an increasing phase. The only winter pregnancıes in the whole material are from the same year and month. The fact that large individuals mature earlier than small ones (MARKOwSsKI 1980) is observed in this study at Paljakka in winter 1969 and in the females of C} in the spring. Size variation by cohort and sex The juvenile bank voles born in autumn are smaller than those born in spring or summer (MAZAK 1962), and their growth rate is slower than in those born in spring (WAsILEwsKI 1952; HAITLINGER 1965; Kugık 1965; BERGSTEDT 1965; CRAwWLEY 1970; ZEJDA 1971). Sıze differences between the cohorts are seen in the present material before wintering, when the voles in C, are bigger, but this normally diısappears after wintering, except in one case, at Paljakka in spring, where the males were heavier ın C; than in C,. This is connected with the exceptional size development of the voles at Paljakka. On the other hand, the identical growth rates reported for voles born in cold seasons, i.e. spring and autumn (SVIRIDENKO 1959; BuJaLska and GLiwicz 1968) cannot be seen in this material. One reason for this may be the short breeding period, so that only two cohorts exist. The differential growth rate is connected with differences in physiological maturation in rodents (SCHWARZ et al. 1964). In bank voles the individuals in the autumn cohort are physiologically younger than those of the same age born earlıer in the breeding season (FEDyYK 1974), and they also survive best in winter (WIGER 1979). Some individuals of C; even live over the second winter of their lifespan, having been born outside the normal breeding period (cf. pregnant females in March 1969) and thus having in a way lived through only one and a half winters. Differences in size occurred between the males and females at all seasons except in spring, at all the trapping sites and in both cohorts, the longer specimens being mostly males and the heavier ones females. There are some bank vole populations, however, in which no sex-related variations in body size have been noted (HAITLINGER 1965; BUJALSKA and GLiwıcz 1968; Zejpa 1971; KaıkusaLo 1972). In spring the C, pregnant females are longer than the non-pregnant ones, which are still growing. ZEJDA (1967) also notes that the pregnant females grow at the beginning of the breeding season, at a time when the males no longer increase in weight at all. 86 P. Viro Size variation during population fluctuations Various strategies have been put forward to explain the correlation between mean size and the fluctuation phases of a population. K or «-strategists (1.e. slow maturation, large size) are predominant during the peak phases and r-strategists (1.e. early maturation, small size) dominate during the increasing phases (see MACARTHUR and WıLson 1967; PıankA 1970; STENSETH 1978; WIGER 1979; BEACHAM and Kress 1980). The existence of genetically selected phenotypes is regarded as the basis for these two strategic models noted in Microtinae populations (see CHıtty 1960; BoonsTRA and Kress 1979). WIGER (1982), however, in his study concerning Clethrionomys species, concludes that there could be some response by the individuals to changes in their social environment. The existence of large individuals during peak population phases ıs known in many fluctuating species, both ın the genus Microtus (CHITTY 1952; BoonsTRA and Kress 1979; BEACHAM 1980) and in the genus Clethrionomys (STEIN 1956; Hansson 1969; WHITNEY 1976; WIGER 1979, 1982), although exceptional populations have also been described (C. rutilus and C. gapperi FULLER 1969). Large individuals are found at the peak phase in this material in some cases, the wintered males in June (i.e. those born during the increasing phase) being the heaviest during the peak phases. Similar results have been obtained from data on bank vole populations in Northern Sweden (GUSTAFSSON, pers. comm.). The weight of the males after wintering is lowest in the years with low numbers, as was also found by Hansson (1969). This is ın contradiction to the situation in the total material of this study, where the C; voles born during the low phases are the smallest and weakest. By winter these count as belonging to the increasing phase, in which small size has been noted in some other Microtinae populations (WIGER 1979; BEACHAM and Kress 1980). A rapıd growth rate (Kress et al. 1973, FuLLER 1977; BEACHAM 1980), good condition (HEIKURA 1977) and early maturation in the increasing phase (WIGER 1979; BEACHAM and Kress 1980) have also been observed in some Microtinae species. This material makes it possible to follow the trends in body size in successive seasons and phases. It is obvious that the selection of the material has an important influence on the results concerning the relation between body size and the phase of population fluctuation. Acknowledgements This work forms part of a project for which a research group led by Assoc. Prof. SEPPO SULKAVA has received a grant from the National Research Council for the Natural Sciences, Academy of Finland. Sincere thanks are due to Prof. ERKKI PULLIAINEN, Assoc. Prof. SEPPO SuLkAvaA and Dr. TORGNY GUSTAFSSON, who have kindly read the manuscript and offered criticism, which has been greatly appreciated. Mr. MarLcoLMm Hıcks M. A., kindly improved the English of the manuscript. Zusammenfassung Größenschwankungen bei der Rötelmaus Clethrionomys glareolus (Schreber, 1780) in verschiedenen Altersklassen, untersucht in Nordfinnland Die Größenschwankungen bei der Rötelmaus Clethrionomys glareolus wurden in drei Gebieten während verschiedener Stadien der Populationsdichte untersucht. Das Untersuchungsmaterial stammt aus Fängen der Jahre 1966-1972. Die mittlere Größe nahm während des Winters ab. Ein deutlicher Zusammenhang zwischen Schwankungen in der Größe einerseits und dem Geschlecht andererseits konnte nicht nachgewiesen werden, obwohl es in einigen Fällen erhebliche Unterschiede zwischen den Geschlechtern gab. Die im Herbst geborenen tragenden Weibchen waren ım Frühjahr (III-V) erheblich länger als die nichttragenden. Dieser Unterschied verschwand bis zum Beginn des Sommers (VI-VII). Eine Fluktuation der Bestandsdichte im 3-Jahres-Zyklus wurde an allen drei Fangplätzen festgestellt. Die Rötelmäuse, die in einer Anstiegsphase geboren wurden, waren größer als die, die nicht in einer solchen Phase geboren wurden. Die Unterschiede der in den verschiedenen Zyklen geborenen Rötelmäuse waren bei den im Herbst geborenen deutlicher als bei den ım Frühling geborenen. Es wurde festgestellt, daß die Größe der Rötelmaus mit dem Geschlecht, der Altersklasse und dem Zyklus der fluktuierenden Bestandsdichte in Zusammenhang steht. Size variation in cohorts of the Bank vole in Northern Finland 87 Literature BEACHAMm, T. D. (1980): Growth rates of the vole Microtus townsendii during a population cycle. Oikos 35, 99-106. BEACHAMm, T. D.; Kress, C. J. (1980): Growth rates of aggressive and docile voles, Microtus townsendiü. Amer. Midl. Nat. 104, 387-389. BERGSTEDT, B. (1965): Distribution, reproduction, growth and dynamics of the rodent species Clethrionomys glareolus (Schreber), Apodemus flavicollis (Melchior) and Apodemus sylvaticus (Linne) ın southern Sweden. Oikos 16, 132-160. BooNSTRA, R.; Kress, C. J. (1979): Vıability of large- and small-sized adults ın fluctuating vole populations. Ecology 60, 567-573. BUJALSKA, G.; GLıwIcz, J. (1968): Productivity ıinvestigation of an ısland population of Clethrionomys glareolus (Schreber, 1780). III. Individual growth curve. Acta theriol. 13, 427-433. Chitty, D. (1952): Mortality among voles (Microtus agrestis) at Lake Vyrnwy, Montgomeryshire, in 1936-1939. Phil. Trans. R. Soc. B 236, 505-552. — (1960): Population processes ın the vole and their relevance to general theory. Can. J. Zool. 38, 99-113. CRAWLEY, M. C. (1970): Some population dynamics of the bank vole, Clethrionomys glareolus and the wood mouse, Apodemus sylvaticus in mixed woodland. J. Zool., London, 160, 71-89. FEDYK, A. (1974): Gross body composition in postnatal development of the bank vole. II. Differentia- tion of seasonal generations. Acta theriol. 19, 403-427. FULLER, W. A. (1969): Changes in numbers of three species of small rodents near Great Slave Lake, N.W.T. Canada, 1964-1967, and their significance for general population theory. Ann. Zool. Fennicı 6, 113-144. — (1977): Demography of a subarctic population of Clethrionomys gapperi: sıze and growth. Can. ]. Zool. 55, 415-425. GRUBER, V.; Kaumann, H. (1968): Eine biometrische Untersuchung an alpinen Rötelmäusen (Clethrionomys glareolus Schreber, 1780). Säugetierkdl. Mitt. 13, 310-338. HAITLINGER, R. (1965): Morphological analysıs of the Wroclaw population of Clethrionomys glareolus (Schreber, 1780). Acta theoriol. 10, 243-272. Hansson, L. (1969): Spring populations of small mammals in central Swedish Lapland. Oikos 20, 431450. HEIKURA, K. (1977): Effects of climatic factors on the field vole Microtus agrestis. Oikos 29, 607-615. HYvÄRINEN, H. (1969a): Die jahreszeitliche Variabilität in der Aktivität der alkalischen Phosphatase des Knochens bei der Rötelmaus (Clethrionomys glareolus, Schreber) als Indikator des Wachstums. Z. Säugetierkunde 34, 109-113. — (1969b): On the seasonal changes in the skeleton of the common shrew (Sorex araneus L.) and their physiological background. Aquilo, ser. Zool. 7, 1-32. HyvÄRINEn, H.; HEIKURA, K. (1971): Effects of age and seasonal rhythm on the growth patterns of some small mammals ın Finland and in Kirkenes, Norway. J. Zool., London, 165, 545-556. IvERsoN, S$. L.; TURNER, B. N. (1974): Winter weight dynamıcs in Microtus pennsylvanicus. Ecology 66, 1030-1041. KaıkusALo, A. (1972): Population turnover and wintering of the bank vole, Clethrionomys glareolus (Schreb.), in southern and central Finland. Ann. Zool. Fennici 9, 219-224. KAreLa, ©. (1957): Regulation of reproduction rate in subarctic populations of the vole Cle- thrionomys rufocanus (Sund.). Ann. Acad. Sc. Fenn. s.A. Biologica 34, 1-60. KreBs, C. J.; GAINEs, M. $.; KELLER, B. L.; Myers, J. H.; Tamarın, R. H. (1973): Population cycles ın small rodents. Science 179, 35—41. Kusık, J. (1965): Biomorphological variability of the population of Clethrionomys glareolus (Schreber, 1780). Acta theriol. 10, 117-179. MACARTHUR, R. H.; Wırson, E. ©. (1967): The theory of ısland biogeography. Princeton: Princeton Univ. Press. MARKOWSKIT, J. (1980): Morphometric varıability in a population of the root vole. Acta theriol. 25, 155-211. MAZAR, V. (1962): Zur Kenntnis der postnatalen Entwicklung der Rötelmaus, Clethrionomys glareolus Schreber, 1780 (Mammalıa, Microtidae). Acta Soc. Zool. Bohemoslovenicae, Prague, 26, 77-104. Pıanka, E. R. (1970): On r- and K-selection. Am. Nat. 104, 592-597. SCHWARZ, $. S.; POKROVSKI, V. A.; ISTCHENKO, V. G.; OLENJEV, V. G.; OVTSCHINNIKOVA, N. A.; PjastoLova, O. A. (1964): Biological peculiarities of seasonal generations of rodents, with special reference to the problem of scenescens in mammals. Acta theriol. 8, 11-41. SIIVONEN, L. (1977): Pohjolan nısakkäät (Mammals of Northern Europe). Keuruu: Otava. STEIN, G. H. W. (1956): Natürliche Auslese bei der Rötelmaus. Z. Säugetierkunde 21, 84-100. STENSETH, N. C. (1978): Demographic strategies in fluctuating populations of small rodents. Oecologia (Berl.) 33, 149-172. 88 P. Vıro SVIRIDENKO, P. A. (1959): The growth and development of the vole Clethrionomys glareolus Schreb. Zool. Zhurnal 38, 756-766 (in Russian). VıRo, P. (1974a): Age structure in populations of the bank vole, Clethrionomys glareolus (Schreb. 1780), ın Northern Finland. Aquilo ser. Zool. 15, 18-24. — (1974b): Age determination in the bank vole, Clethrionomys glareolus Schreb. 1780, from the roots of the teeth. Aquilo ser. Zool. 15, 33-36. WASILEwsKI, W. (1952): Morphologische Untersuchungen über Clethrionomys glareolus glareolus Schreb. Ann. Univ. Marie Curie - Sklodowska C 7, 119-211 (in Polish). WHITNEY, P. (1976): Population ecology of two sympatric species of subarctic microtine rodents. Ecol. Monogr. 46, 85-104. WIGER, R. (1979): Demography of a cyclic population of the bank vole Clethrionomys glareolus. Oıikos 33, 373-385. — (1982): Roles of self regulatory mechanisms ın cyclic populations of Clethrionomys with special reference to C. glareolus: a hypothesis. Oikos 38, 60-71. ZEJDA, J. (1967): Mortality of a population of Clethrionomys glareolus Schreb. ın a bottomland forest in 1964. Zool. Listy 16, 221-238. — (1971): Differential growth of three cohorts of the bank vole, Clethrionomys glareolus Schreb. 1780. Zool. Listy 20, 229-245. Author’s address: Phil lic. PıRkKo VIRo, Department of Zoology, University of Oulu, SF-90570 Oulu, Finland Gastrointestinal anatomy of the European badger Meles meles L. A comparative study By R. STARK, T. J. ROPER, A. M. Mac Larnon and D. ]. CHIvERS School of Biological Sciences, University of Sussex Brighton Receipt of Ms. 18. 4. 1986 Abstract The gastrointestinal tract of the Furopean badger (Meles meles L.) consists of a simple elongated stomach, tortuous small intestine and simple smooth-walled colon. There is no caecum. Allometric comparison of the internal surface area of the main gut compartments with comparable data from a variety of mammal species shows that the badger has a marginally larger stomach, larger small intestine and smaller colon than expected for its body weight. Multivarıate analysıs based on the. absorptive areas of the three main gut compartments, taking into account body size, places the badger close to other mustelids and within a cluster of species most of whose diets are faunivorous. Introduction Kruuk (1978a, b; Kruuk et al. 1979; Kruuk and ParısH 1981) has characterised the European badger as a specialist predator on earthworms, especially Lumbricus terrestris. Stomach contents and faeces of badgers from Osefordshire and Scotland reveal a high incidence of earthworm remains, while behavioural observations suggest not only that badgers spend most of their foraging time looking for and consuming worms, but also that wormhunting effort increases or decreases to compensate for changes in prey availability. U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5202-0088 $ 02.50/0 Z. Säugetierkunde 52 (1987) 88-96 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Gut anatomy in badgers 89 In Kruux’s view worm availability is sufficiently crucıal to the badger to determine both territory size and group size (see also Mac DonaLD 1983 and von SCHANTZ 1984a). That worms are an important food for badgers is now beyond dispute (see NEAL 1977 and Kruuk 1978a, b, for a review of supporting evidence from a variety of geographical areas). However it is equally clear that badgers also eat many other foods, including plant as well as anımal matter. Constituents of the diet include other invertebrates, expecially beetles and insect larvae (e.g. ANDERSEN 1955; CıamPiLını and Lovarı 1985); small vertebrates such as amphibians, rodents and immature rabbits (e.g. ANDERSEN 1955; LIKHACHEV 1956; Kruuk and PArısH 1981); and a variety of vegetable matter including fruits, berries, nuts, tubers, garden vegetables and cereals (e.g. SkooG 1970; BRADBURY 1974; NEAL 1977; Kruuk and DE Kock 1981; Kruuk and ParısH 1981; Harrıs 1984; CIAMPILINI and Lovarı 1985). The present study attempts to relate the gross anatomy of the badger gut to the species’ taxonomic status and diet. It is suggested (CHIvERS and Hranık 1980) that gut anatomy in mammals reflects dietary habits. Species that predominantly forage for fruits, seeds, flowers (generally termed as frugivores), possess a gut that is of a relatively unspecialised type in which simple stomach, small intestine, colon and caecum all are present. Those species, that eat leaves, grasses, stems, barks and gums (folivores) show enlargement of either stomach (in the case of “foregut fermenters”’) or of caecum and sometimes colon (in the case of “midgut fermenters”’) to form one or more chambers for the processing of the longchain carbohydrates. Species devouring anımal matter including invertebrates (faunivores) exhibit a gut that tends overall to be short and to be dominated by small intestine rather than by stomach, colon or caecum. Given the badger’s taxonomic status as a carnıvore we expected it to exhibit a gut of the faunivore type. On the other hand since fruit, tubers and cereals do seem to be common constituents of the badger’s diet the gut may have evolved partially towards frugivory. The ıdea that gut anatomy reflects diet is not of course new: a pioneering attempt to relate the two was made by Cuvier (1805). In most cases, however, comparisons have been made on the basıs of gut length in relation to body length (e.g. SK0oG 1970). There are two important problems with this type of comparison. First, digestive capacity is more closely related to the inner surface area of a particular gut compartment than to its length (e.g. MancoıD 1950, cited in SKooG 1970). Second, simple ratios such as intestine length to body length are misleading in that they are not “size-free”’; they vary according to body size irrespective of any difference due to diet. In this paper we present for the fırst time detailed quantitative anatomıcal data on gut morphology in the badger, including measurements of the inner surface area of different gut compartments. We also use an allometric method of analysis, developed by MarrTIn et al. (1985) and Mac Larnon et al. (1986), to compare the badger’s gut morphology with that of other mammal species for which comparable data are available. Material and methods Specimens Anatomical data were obtained from 11 road-killed badgers; 8 males and 3 females (see Table 1). One male and one female were from East Sussex; the remainder were from East Anglıa. All were found between late February and early April 1984, a season of the year when road mortality ıs high (Davızs et al., ın press). Anatomical measurements The anatomical techniques have been described in detail by CHivers and Hrapık (1980), and so will only be summarised here. Animals were first weighed intact, and body length was measured from bregma to ischial tuberosity. The complete gastrointestinal tract was then removed; the gut wall was 90 R. Stark, T. J. Roper, A. M. Mac Larnon and D. J. Chivers opened out and flattened without stretching; and measurements were made of the length and breadth of stomach, small intestine and colon. The primary data of interest were the surface areas of stomach, small intestine and colon, and these were calculated from lengths and a series of breadths. In addition the three main gut compartments were weighed after removal of excess water. The nine East Anglian specimens were examined and measured in a fresh state as soon as possible after collection; the two East Sussex specimens were deep frozen for about a month prior to dissection. Allometric analysis of data For a cross-species comparison of quantitative gut anatomy it is necessary to take into account differences in Da size. MARTIN et al. (1985) have shown that when the internal surface area of each of the gut compartments is plotted against body weight on logarıthmic coordinates for a variety of mammals, a line of slope 0.75 can be fitted through the data points for each gut compartment. This ıs compatible with KLEIBER’s law, whereby basal metabolic rate scales to body weight with an exponent ot 0.75 (KLEIBER 1961). The line of fixed slope then gives the expected value of gut compartment surface area for a typical mammal of given body weight. If the data pont for a particular species falls above or below the line this indicates that the gut compartment ın that species has a larger or smaller surface area than expected. Deviation from expectation is given by a quotient for each gut compart- ment. In accordance with Mac LARnNoN etal. (1986) we have used log gut quotients (log observed surface area minus log expected surface area) rather than absolute quotients ee expected surface area) in our analysıs. However when discussing individual quotients ın the text we have cited the absolute quotient values, since these are more readily understood (the expected value of the absolute quotient for a typical mammal being unity). In order to compare the overall pattern of quotient values for the four gut compartments across a spectrum of species, a multivariate technique is necessary. In studies by Marrın et al. (1985) and Mac LARNON et al. (1986) the four gut quotient values from a variety of mammal species were used to generate a matrıx of Euclidean distances between all pairs of species, and a multidimensional scalıng technique was then applied to this matrix to provide a two-dımensional representation (a “mul- tidimensional scalıng plot””) of these distances. On the multidimensional scalıng plot species which have similar overall gut anatomy lie close together in space, so that the plot provides a pictorial representation of similarities and differences in gut morphology across a wide range of species. In the present study we present a multidimensional scalıng plot for 79 mammal species previously described by Marrın et al. (1985) and Mac LArnon et al. (1986), and we incorporate into this plot new data from the badger Meles meles. Since the badger was found to possess no caecum (see below), the present analysıs was based on three quotient values for stomach, small intestine and caecum plus colon respectively. Results Anatomy of the gut The gastrointestinal tract of Meles meles (see Fig. 1) consists of stomach, small intestine and colon: there is no caecum. A summary of anatomical data obtained from 11 road-killed animals is given ın Table. The stomach of Meles meles is simple but elongated. The small intestine is very long and tortuous (mean length 536cm) and provides about 80 % of the surface area of the gastrointestinal tract. There is a very large Peyer’s Patch (lymphoid), about 33 cm long, at the distal end. The colon is simple and fairly short (mean length 29cm), wıth numerous Iymphatic nodules in the last 10 cm. In the absence of a caecum the start of the colon can be identified by the abrupt transition from the villi of the ileum to the crypts of the colon. This corresponds with a colour change to the mucosa a few cm after the large Peyer’s Patch, and with the ileocolic artery from the cranıal mesenteric artery. Inspection of the data in Table 1 suggests no sex difference in the absolute or relative size of the three gut compartments, but a larger sample of females is needed to confirm this statistically. In the single juvenile specimen (number 11 in Table 1) the surface area of stomach and colon and the length of the colon were within the range of values provided by the 10 adult animals; but the small intestine of the juvenile was shorter and smaller ın Gut anatomy in badgers 91 surface area. In most speci- mens the stomach con- tained remains of earth- worms together with grass and earth. Allometric analysis Figure 2 shows the surface areas of stomach, small in- testine and colon plotted against body weight on logarıthmic coordinates, for 80 mammals species in- cluding Meles meles. The data points for Meles meles represent the mean values from seven of the eleven anımals included in Table 1: the remaining four anı- mals were starved or dam- aged, and so were excluded from the analysıs. Data for the 79 comparison species were provided by A. Mac- Larnon. The absolute gut quotient values for Meles meles are: stomach quo- tient, 1.1; small intestine quotient, 1.75; colon quo- tient, 0.40. Thus it can be Fig. 1. Gut of a mature female badger weighing 8.5kg, showing concluded that the surface stomach, small intestine and colon. An arrow marks the beginning area of the stomach of the of the colon (the scale is 20cm) Table 1 Anatomical data from 11 road killed animals Surface area (cm)? Weight (qm) Length (cm) Stom. ST. Colon Stom. SI. Colon SI. Colon 2280510095159 74 ASP 540278272312 139, 587 546 3007 190 184 575 497, 28511082117, 178 359 3167720217219 110 470 460 2910 208 170 394 30322 1176745=151 132 Ha, 460 2892 266 195 553 4355,2.16852194 103 542 9065. 72062655.180 150 574 441 2006 226 82 458 409 2663 197 138 536 !EA=E. Anglia; ES = E. Sussex. — *Immature anımal, data not included in mean. — Data not included ın allometric analysis. — *Measured from bregma to ischial tuberosity. 92. R. Stark, T. J. Roper, A. M. Mac Larnon and D. J. Chivers Fig. 2. Surface area of stomach, small intestine and 0 colon plotted against body weight on logarithmic 4 e coordinates for 80 mammal species including Meles meles. Each data point represents a single species. A line of fixed slope 0.75 has been fitted to each set of points. For a list of the species in question see Table 2 STOMACH badger is approximately that expected of a “typical” mammal of the same body weight; the small intestine is larger than expected; and the colon is smaller than expected. Figure 3 ıs a multidimensional scaling plot generated from the three log quotient values (stomach, small intestine, caecum plus colon) for the same spectrum of mammal species. The varıous mammal species represented on the plot are listed numerically in Table 2. Dotted lines have been drawn around groups of species whose diets are either primarily faunivorous (group A), primarily frugivorous (group B) or primarily folivorous (groups C and D). The positioning of these lines is necessarily approximate since accurate data concerning natural diet in some species is not 2 available (for references on which the dietary groups ın Fig.3 are based see MarTInN et al. 1985, and MAcLarnon et al. 1986). Neverthe- less there clearly is a gross correlation between 1.0 Zn 3 TREE the diet of a species and the position that it occupies, by virtue of its gut anatomy, in the multidimensional scaling plot. In general faunivorous mammals fall into the upper right-hand quadrant of the plot; frugivorous mammals form a cluster to the left of centre; and folivores form two clusters at the bottom and left-hand extremities of the plot, according to whether they possess enlarged fore-gut or enlarged mid-gut respectively. Within the faunivorous group, species whose diet is mainly insectivorous tend to fall to the left of cluster A. Meles meles (species numer 80) can be seen to fall within the faunivorous cluster. e PRIMATE 1 ° MAMMAL V MELES MELES SMALL INTESTINE Log), SURFACE AREA (cm2) COLON Log,, BODY WEIGHT (g) Discussion In ıts gross anatomy the badger’s gastrointestinal tract follows the typical faunivore pattern: there is a simple stomach, tortuous small intestine and simple smooth-walled colon, none of which is elaborated in any obvious way. In common with a number of other faunivores the badger lacks a caecum. This condition is shared, for example, by the mustelids Mustela nivalis and M. erminea, by the cetaceans Phocaena phocaena and Tursiops truncata, and by some tropical insectivorous mammals such as Manis tricuspis and Potomogale velox (see CHivErs and Hıanvık 1980, Table 8). However a few non- faunivorous mammals also lack a caecum, e.g. Nandinia binotata, which is primarıly frugivorous (CHARLES-DOMINIQUE 1978). Gut anatomy in badgers B) . 5. A a ® K Kan & & B a FILTER ei $ ., a | 1 ® 3 ı EH „8 19 nn WE I NE ei — leo — = 8 2 79 738 "3, Be 1% 21 3u° Mi : @.” BE al + &) Carnivora, Cetacea, Insectivora, 2,9529 B Pinnipedia, Pholidota 2303 3 E23 8 Artiodactyla, Edentata, Hydro- iy 8 coidea, Lagomorpha, Marsupalia D is F & Rodentia (Sciuridae) : GH 12 Primates ll Meles meles Fıg. 3. Multidimensional scaling plot based on three log gut quotient values (stomach, small intestine and colon plus caecum) for 80 mammal species, including Meles meles. For a key to the species represented in the plot see Table 2. Species have been roughly grouped according to diet as follows: cluster A: faunivores; cluster B: frugivores; cluster C: folıvores (mid-gut fermenters); cluster D: folıvores (fore-gut fermenters) The multidimensonal scaling plot shown in Fig. 3, which gives an overall picture of the degree of anatomical similarity between the gastrointestinal tracts of different species, places Meles meles within a cluster of primarily faunıvorous mammals including all other members of the Carnivora for which data are available (species 43 to 55 in Fig. 3). In particular, the gut anatomy of Meles meles clusters with that of other primarily faunivorous mustelids (Mustela nıvalıs, 45; Mustela sp., 51) A particularly interesting comparison is with the fox, Vulpes vulpes, which like the badger eats a wide range of foods including small vertebrates, worms and other invertebrates, fruits, berries, tubers, carrıon and scavenged items (e.g. ENGLUND 1965; MAcDonALD 1980; HARRIS 1981; CIAMPILINI and Lovarı 1985). Vulpes vulpes (46) lies further towards the inner end of the faunivorous cluster ın Fig. 3 than does Meles meles, suggesting a degree of morphological adaptation to frugivory. Of course the ability of an anımal to digest particular types of food depends on factors other than gross gastro-intestinal anatomy: for example enzyme secretion and dentition are obviously important. In addition, it should be noted that all our road-killed specimens were collected in the spring, and it ıs possible that the gut may undergo minor morphologi- cal adjustment in the autumn in response to a more frugivorous diet (see CHIVERS and Hrapık 1980). Nothing seems to be known of enzyme activity ın the badger gut, and detailed analysıs of badger faeces is required to determine what if anything is digested out of the various fruits and cereals that are known to be eaten. As regards dentition, NEAL (1977) comments that the last premolar and molars are enlarged and flattened for crushing and grinding, and hence are indicative of an omnivorous diet (see also SkooG 1970; BoRN- MUELLER 1974). However the unusual jaw articulation of the badger, in which the squamosal is folded almost completely over the condyle to form a transverse hinge, restricts sideways movement of the lower jaw relative to the upper and so permits very little grinding action (MAYNARD SMITH and SavaceE 1959; PoLyakovA 1974); and inspec- tion of badger faeces suggests that most seeds and berries are subjected only to repliminary crushing (STARK 1984). sepıpodossey “erferdnsse 9epıpodApeig ereauspg JepImaeIo1g “e9prosesÄH sepIueM “eIOPIJoUg Jepıfe3owolo] “eIOANISSUJ 9eprıoda | “eydıowose] JepLINIIS "eNUSPoy JepLInIg “enuspoy JepLINIIg “eNuUspoy JepILINIIS “eNUspoy Jeprurydjaq, ‘e9se139 Jeprumydfaq ‘e99e199 eprsoyg “eIpodiuurg 9epınb “er4ı9eponIy JepIAI9) eÄIDePonIYy JeprAaog ‘efÄReponIy 9epıaog “eAeponIy Jepıng “eI4I9ePONIYy JepIfpg "eIoATUIe) JEPLLIOALA “BIOATUIET JepIUOAIOIT "TIOATUIET JepIUoAI0I] TIOATUIET JepıfpIsn\ “eIOATUIEY JEPLLI9AL A “BIOATUIET JEPLLIIALA “EJOATUIET) JEPLLI9ALA “BIOATUIET JePLIIHAL “BIOATUIET JepIuen) “eIOATUIEY JepıfsisnN “eIOATUIET epıuen “eIOATUIET JepIJ2 ] "EIOATUIET JeprumwWof 9epı3uog 9epı3uog 9epı3uog sajaıu soo sn/nı sndossry snıdapvpıua sndAkpvig sıjvs10p xvAalgoApualz ' sıdsns141 stuv xoja% 9703014030.] d snynMund SNEVIODÄAO SISUAUNJ0AVI SHAMIIS wunıgauvagofnaı sndanDsoyafz ' ga snioxıd7 sıuvajna SNANDS vıv3unA sdoisın m vuav39ogd vu9090gJ ' sndKkı3 snaaogaypH " snıjpgqv2 sunb7 snqdvja sna4a) saııv SIAO sn34ug vAdvZ v[oAs snS s1uaı] DAogIuog * (ds v113u35 val1ApUu onNSYN suadyn[ snuny ' (ds vjossnw vgıjvaA4as v119U9H 1uospirgq9la vuviog ' pjvjoung vıupurN ' snsoupnjvd xrpuw sadjna sodyn A sıppanu vJoIsny suvmuo/ sıuv) vI1]s3Wop sa] s SILLVWIAd-NON suaıdvs owod ' vın4o3 vp4on) s214p078041 up] snavwmshd 03u0J ' JepnegofAH JepnegoJÄH Jeprmayııdo91a7 Jepmayı1ıdo9199 Jepmayı1do9a1a] Jeprmayı1do91a) Jepmayıırdo9199 Jeprmayııdo9197 9epmayı1do9199 Jepmayı1do9199 Jepmayı1do9197 Jepmayıdos1s7 Jepmay11d091957 9epmayı1dos129 9epmayı1dos199 Jepmayı1do9199 Jepmayı1dos1a) 9epmayı1dos1a) 9epmayı1do9139 Jepmay1ıdo9139 NE) SERID A) SED a) ano) HD) JepIsLIo] JepLınwa7 JepIsLIoT] Jeprınwa7 JepLınwa7 JepIsLIo] JepisuIo] JepIstIo] JepLınwaT SZPIEPNI JepIsLIo] sndpvpuds (snZuojvgduds) saavgojlky sniwand sorvgoykg snovwmau X1gp3h] ' sn1004v] sypsvN vpunngnA sı1lqsaAd sogdojrjaw sıahqsoig vandsgo sUKgsaAl vavIS119 SIIÄGSaAL sajarua sıılqsaug = we sng0]05 spjpd sngo9o4qıdkaz xuıgds o1do.] stuvgn19sv[ v3aVa0 vaıuıs v9vVIP snuvayds vovooW vuadıqjp SNQ2903425 surmaıu sn92qN1d09495 sn»aj3au sn99qNdos43) snqda3 sn99q11d03495 uodrjvi snaagndoıy vq9114108v7 X11q10807 vavııypd vaapnoyy ' snumndv3 sng2J 11P915490 1unuıvs N sojaV x SNIVSADLA] SNIOV 1Ko4ffoa3 snumndvs 01104 snappıpo4Aa] z snunmw SngaI0nW snpv4Bıpavı sı407 sndosnaj anwandeT snumassnw anwandaT naopıuap 03v]vH 1uayjp o8vjvH, Snnıuv3oja Sn9110n7 aolvrw snojv3onag) 4a81ur] 1gvaY SISU9ADGYJYI SNGAIOWAV SLLVWINd p>seq 318 € pue 7 'sdı] yaIyMm uo eyep ayy ur papnpur sarads jewureur O8 Jo 3517 CAT, NA NOoNon Gut anatomy in badgers 95 To conclude, the morphological evidence is consistent with a characterisation of the badger as a specialist faunivore preying primarily on worms and insects (Kruuk 1978a, b; Kruuk and PArısH 1981). But given the relatively long small intestine and the consequent large surface area for digestion and absorption, the badger is probably able to process with reasonable efficiency small fleshy fruits and tubers containing shortchain sugars. The alimentary system is not, however, morphologically well suited to mastication or digestion of cereals. Acknowledgements Road-killed badgers were made available through the Veterinary Investigation Centre, Cambridge, and we thank M. Lamont and A. GRESHAM for allowing us access to the Centre’s facilities. The F iatoeraph shown as Fig. 1 was taken by D. Crinton. In particular we thank Dr. P. H. Harvey for is expert advice for our study. The research was supported in part by financial support to R. STARK and a Royal Society grant to T. J. RoPER. Zusammenfassung Der Gastrointestinaltrakt des europäischen Dachses Meles meles L. Eine vergleichende Studie Der Gastrointestinaltrakt des europäischen Dachses (Meles meles L.) setzt sich zusammen aus einem einfachen, ungegliederten Magen, einem in Schlingen gelegten Dünndarm und einem einfachen, leicht gewellten Dickdarm. Der Dachs besitzt keinen Blinddarm. Ein allometrischer Vergleich der inneren Oberflächen von Magen, Dünn- und Dickdarm zwischen verschiedenen Säugetierarten ergibt, daß der Dachs einen nur wenig vergrößerten Magen hat. Die Innenfläche des Dünndarms ist größer, die des Dickdarms kleiner als man für das Körpergewicht des Dachses erwarten würde. Eine Varianzanalyse der Gastrointestinaltrakte, die die unterschiedlichen Körpergewichte der untersuchten Säugetierarten berücksichtigt, zeigt, daß der Dachs zusammen mit anderen Musteliden und weiteren Vertretern der Carnıvora eine Gruppe bildet, die gekennzeichnet ist durch faunivore Ernährung. References ANDERSEN, J. (1955): The food of the Danish badger (Meles meles danica) with special reference to the summer month. Danish Rev. Game Biology 3, 1-75. BoRN-MUELLER (1974): Zur Biologie des Dachses in der Region Basel mit besonderer Berücksichti- gung der Nahrung. Dipl.-Arbeıt Univ. Basel. BRADBURY, K. (1974): The badger’s diet. In: Badgers of Yorkshire and Humberside. Ed. by R. ]. PAGET, and A. L. V. MipperTon. pp. 119-141. York: Ebor Press. CHARLES-DOMINIQUE, P. (1978): Ecology et vie sociale de Nandinia binotata (Carnivores. Viverrides): Comparison avec les prosimiens sympatriques du Gabon. Terre et Vie 32, 477-528. CivErs, D. J.; Hrapık, C. M. (1980): Morphology of the gastro-intestinal tract in primates: comparisons with other mammals in relation to diet. J. Morphology 166, 337-386. CIAMPILINI, B.; Lovarrı, $. (1985): Food habits and trophic niche overlap of the badger (Meles meles L.) and the red fox (Vulpes vulpes L.) in a Mediterranean coastal area. Z. Säugetierkunde 50, 226-234. CUVIER, G. (1805): Lecons d’Anatomie Comparee. Crochard; Paris. Davıss, J. M., RoPpERr, T. J., SHEPHERDSON, D. ]. (in press): Seasonal distribution of road kills ın the European badger (Meles meles L.) J. Zool., London. EGLunD, J. (1965): Studies on food ecology of the red fox (Vulpes vulpes) in Sweden. Viltrevy 3, 375-485. Harrıs, $. (1981): The food of suburban foxes (Vulpes vulpes) with special reference to London. Mammal Review 11, 151-168. Harıs, $. (1984): Ecology of urban badgers Meles meles: distribution in Britain and habitat selection, persecution, food and damage in the city of Bristol. Biol. Conservation 28, 349-375. KLEIBER, M. (1961): The Fire of Life: an Introduction to Animal Energetics. New York: John Wiley. Kruuk, H. (1978): Spatial organisation and territorial behaviour of the European badger Meles meles. J. Zool., London, 184, 1-19. Kruukx, H. (1978): Foraging and spatial organisation of the European badger, Meles meles L. Behav., Ecol. Sociobiol. 4, 75-89. Kruuk, H.; Kock, L. de (1981): Food and habitat of badgers (Meles meles L.) on Monte Baldo, northern Italy. Z. Säugetierkunde 46, 295-301. Kruuk, H.; ParısH, T. (1981): Feeding specialization of the European badger Meles meles ın Scotland. J. Anımal Ecol. 50, 773-788. 96 R. Stark, T. J. Roper, A. M. Mac Larnon and D. J. Chivers Kruuk, H.; ParısH, T.; Brown, C. A. J.; CARRERA, J. (1979): The use of pasture by the European badger (Meles meles). J. Applied Ecol. 16, 453-459. LIKHACHEV, G.N. (1956): Some ecological traits of the badger of the Tula Abatis broadleaf forests. In: Studies of Mammals in Government Preserves. Ed. by G. YURGENsoN. pp. 72-94. Ministry of Agriculture of the USSR, Moscow (1961). MacDonaLD, D. W. (1980): The red fox, Vulpes vulpes, as a predator upon earthworms, Lumbricus terrestris. Z. Tierpsychol. 52, 171-200. MacDonALD, D. W. (1983): The ecology of carnıvore social behaviour. Nature 301, 379-384. Mac LArnon, A. M.; MARTIN, R. D.; CHIvERs, D. J.; HLADık, C. M. (1986): Some aspects of gastro- intestinal allometry ın primates and other mammals (in press). MARTIN, R. D.; CHIVERS, D. J.; Mac LAarnon, A. M.; Hrapık, C. M. (1985): Gastrointestinal allometry in primates and other mammals. In: Size and Scaling in Primate Biology. Ed. by W.L. JUNGERS. pp. 61-89. New York: Plenum Press. MAYNARD SMITH, ]J.; SAVAGE, R. J. G. (1959): The mechanics of mammalıan jaws. School Science Review 141, 289-301. NEAL, E. G. (1977): The Badger. Poole, Dorset: Blandford Press. PoLyakova, R. $. (1974): Comparative morphology of the masticatory apparatus of Mustelidae. Trudy Zoologischeskogo Inst. Acad. Nauk, Leningrad, 54, 99-122. SCHANTZ, T. von (1984a): Spacing strategies, kin selection, and population regulation in altrıcıal vertebrates. Oikos 42, 48-58. SCHANTZ, T. von (1984b): “Non-breeders” in the red fox Vulpes vulpes: a case of resource surplus. Oikos 42, 59-65. SKo0g, P. (1970): The food of the Swedish badger. Viltrevy 7, 1-120. STARK, R. (1984): The Badger (Meles meles L., 1758): a study of morphological behavioural and ecological correlates. Unpubl. MSC-Thesis, Univ. Sussex. Authors’ addresses: R. STARK, MSc, Albert-Ludwigs-Universität Freiburg, Zoologie, Albertstraße 21a, D-7800 Freiburg ı. Br., T. J. ROPER, School of Biological Sciences, Univer- sıty of Sussex, Brighton BN1 9QG, UK; A. M. Mac LARNoNn, Department of Anthropology, University College London, Gower Street, London WCI1E 6BT, UK; D. J. CHivers, Sub-Department of Veterinary Anatomy, Tennis Court Road, Cambridge CB2 1QS, UK Die Analbeutel von Civettictis civetta (Schreber, 1776) (Mammalia, Viverridae) Über ihren Bau, die chemische Zusammensetzung ihrer Sekrete und ihre biologische Bedeutung! Von ALICE VON SALDERN, H. SCHLIEMANN, F. I. B. KayanJa und J. JACOB Eingang des Ms. 14. 5. 1986 Abstract The anal sacs of Civettictis civetta (Schreber, 1776) (Mammalıa, Viverridae). Morphology, chemical composition of the secretions and functional significance The anal sacs of Civettictis civetta lying on both sides of the anal canal consist of an epithelium lining the central lumen of these organs, a lamina propria rich in free cells, connective tissue and smooth muscle fibers, and a thick layer of skeletal musculature being derived from the M. sphincter anı ext. The lamina propria contains about 10 complexes of sebaceous glands each of which is associated with apocrine glands. Histology and ultrastructure of the epithelium and the glands are analysed emphasiz- ing the conspicuous development of the smooth endoplasmatic reticulum of lipogenic cells. No highly ' Mit Unterstützung der Deutschen Forschungsgemeinschaft (Schl 98/7-1). U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5202-0096 $ 02.50/0 Z. Säugetierkunde 52 (1987) 96-115 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Die Analbeutel der Zibetkatze 97 volatile components have been found in the secretion, the most important constituents of which are: squalene, cholesterol, cholesterol esters, mono- and diester waxes, triglycerides, alkane-1,2-diols, free fatty acids, and free alcohols. The biological significance of these secretions as chemical signals in the dung heaps of Civettictis is discussed. Einleitung Analbeutel sind zusammengesetzte Drüsenorgane beiderseits des Analkanals. Sie besitzen ein zentrales Lumen, eine muskulöse Hülle und einen Ausführungsgang, der in Afternähe mündet. Sie sind bei Säugetieren und insbesondere bei Carnivoren weit verbreitet und sind häufiger mit anderen Drüsenorganen und -geweben dieser Körperregion verwechselt worden. SCHAFFER (1940) widmet in seinem grundlegenden Werk über die Hautdrüsen- organe der Säugetiere den „‚Anal- und Zirkumanaldrüsen“ ein umfangreiches Kapitel. Er stellt u.a. die Synonyme zusammen, unter denen in älteren Publikationen Studien über Analbeutel bekannt wurden, und geht detailliert auf die lange Geschichte der Bearbeitung dieser auffälligen Drüsenorgane ein. Unter den von ihm genannten Autoren befinden sich bekannte Persönlichkeiten, so BLasıus, DAUBENTON und Tyson und aus dem letzten Jahrhundert Cuvier, JOH. MÜLLER, STANNIUS und GEGENBAUR. Bei den drei zuerst genannten Autoren finden sich sogar Angaben über die Analbeutel der Zibetkatze. Auch Pocock (z.B. 1915, 1916) hat sich in mehreren Arbeiten u.a. mit den Analdrüsen von Viverriden auseinandergesetzt. SCHAFFER selbst hat kein eigenes Material von Civettictis untersucht, referiert jedoch Befunde der älteren Bearbeiter in kurzer Form. An einem anderen Viverriden, Herpestes ichneumon, konnte er feststellen, daß sich am Aufbau eines Drüsenorgans, das sich anhand seines Materials nicht eindeutig als Analbeutel oder Drüsenapparat der Analtasche identifizieren ließ, Talg- und apokrine Drüsen beteiligen. Neuere Arbeiten über den Bau der Analbeutel der Säugetiere im allgemeinen und der Viverriden im besonderen sind selten. ORTMAnN (1960) publizierte eine umfangreiche, vergleichend-anatomische Studie über die Analregion der Säugetiere. In dieser Arbeit finden sich zahlreiche Hinweise auf die Analbeutel der Säugetiere und damit auch auf diese Organe bei Viverriden. GoRMan et al. (1974) befaßten sich mit dem Bau der Analbeutel von Herpestes auropunctatus und teilten zugleich Befunde über die chemische Zusammen- setzung der Analbeuteldrüsen-Sekrete und die Bakterienflora der Analbeutel mit. Diese Arbeit ist im Zusammenhang mit dem Interesse der Autoren an dem Phänomen der chemischen Kommunikation zu sehen. Entsprechendes gilt für KayanJa und SCHLIEMANN (1981), die Befunde zur Histologie und Ultrastruktur der Talgdrüsen der Analbeutel von Genetta tigrina veröffentlichten. Während über die chemische Zusammensetzung der Sekrete der Analbeutel der Carnı- voren eine ganze Reihe von Veröffentlichungen vorliegen, sind die Viverriden hierbei bislang weitgehend vernachlässigt worden. Neben der erwähnten Arbeit von GORMAN et al. (1974) haben lediglich Jacoß und SCHLIEMANN Befunde über die Chemie der Sekrete der Analbeuteldrüsen von Viverriden publiziert, und zwar von Civettictis und Genetta (1983, 1986). Die meisten Viverriden besitzen neben den Analbeuteln weitere, sehr auffällige Drüsen- organe, die Perinealdrüsen, die im Bereich des Dammes liegen. Bei Civettictis enthält das Sekret dieser Drüsen neben vielen anderen Verbindungen (MAURER et al. 1979) das Zibeton, ein zyklisches Keton (RuzıckA 1926). Das Sekret der Perinealdrüsen von Civettictis hat für die Parfümherstellung eine bedeutende Rolle gespielt. Die Funktion dieser voluminösen Organe ist durch Beobachtung der Tiere ohne weiteres zu erkennen; sie dienen dem Markierverhalten. So markiert Civettictis auf der Höhe der Perinealdrüsen liegende Gegenstände, indem die Drüsen gegen diesen Gegenstand gepreßt und dabei die Sekrete abgesetzt werden; selten werden hierbei laterale Bewegungen des Beckens beob- achtet, wie sie von anderen Viverriden bekannt sind. 98 Alice von Saldern, H. Schliemann, F. I. B. Kayanja und ]. Jacob In zahlreichen verhaltensbiologischen Untersuchungen (BEARDER und RAnDALL 1978; Dücker 1965; EwER und WEMMER 1974; GORMAN 1976; GORMAN et al. 1974; KLEIMAN 1974; LEYHAUSEN 1969; RAnDALL 1979; Rasa 1973; ROEDER 1978, 1983; VERBERNE und LEYHAUSEN 1976; WEMMER 1977) ist im übrigen die Bedeutung des Markierverhaltens und damit die von Drüsensekreten für die chemische Kommunikation der Viverriden verdeut- licht worden; soweit diese Arbeiten nicht Vertreter der Herpestinae betreffen, sind es im wesentlichen die Sekrete der Perinealdrüsen, die zum Markieren verwendet werden. Mit dem Markierverhalten in Gefangenschaft gehaltener Zibetkatzen haben sich EwER und WEMMER (1974) im Rahmen einer umfänglicheren ethologischen Studie über diese Schleichkatze auseinandergesetzt, und RanDaLL (1979) konnte Beobachtungen über das Markierverhalten freilebender Zibetkatzen mitteilen. Die biologische Bedeutung der Anal- drüsen ist weder hierbei noch im allgemeinen annähernd erkannt worden. Das ist um so erstaunlicher, als Analbeutel ja bei Landraubtieren mit Ausnahme einiger Ursiden (EwER 1973) regelmäßig vorkommen. Hinweise auf ihre Funktion ergeben sich jedoch u.a. aus den Beobachtungen und Überlegungen von KıEıman (1972), BEARDER und RAnDALL (1978) und MAcDonALD (1980). Die vorliegende Untersuchung soll mit Hilfe morphologi- scher und biochemischer Befunde weitere Informationen über dieses sicher wichtige Drüsenorgan am Beispiel von Civettictis beisteuern und seine Funktion erhellen helfen. Civettictis civetta ıst mit einem Körpergewicht von bis zu 17 kg und einer Schulterhöhe von fast 40cm die größte afrıkanische Schleichkatze. Sie ist nahezu über den gesamten afrıkanischen Kontinent südlich der Sahara verbreitet und besiedelt die unterschiedlichsten Landschaften vom Regenwald bis zur Savanne, trockenere Bereiche ıhres Verbreitungsge- bietes eher meidend (CoETZEE 1971). Trotz ihrer Häufigkeit liegen kaum Freilandbeob- achtungen vor (RAnDALL 1979); bruchstückhafte Kenntnisse ihrer Biologie verdanken wir vorwiegend der Untersuchung von Gefangenschaftstieren (DÜckER 1965; EwER und WEMMER 1974). Civettictis hat eine ausschließlich nächtliche Lebensweise, ist ausschließ- lich bodenlebend und als nahezu omnivor zu bezeichnen (BOoTHMA 1965; ROSEVEAR 1974; VERHEYEN 1951). Soweit bekannt, leben diese Tiere solitär, und es gibt Anhaltspunkte dafür, daß sie territorial sind — hierauf wird im Zusammenhang mit den Erörterungen über die Funktion der Analbeutel zurückzukommen sein. Material und Methode Für diese Untersuchungen standen vier weibliche und zwei männliche Zibetkatzen zur Verfügung. Ein im Zoo Berlin verstorbenes Exemplar diente der makroskopischen Analyse, die übrigen stammten aus Liberia. Eines dieser Tiere lebte für eine längere Zeit in der Tierhaltung des Zoologischen Instituts und Zoologischen Museums Hamburg; ihm wurden die Analbeutelsekrete für die biochemischen Untersuchungen entnommen. Für die Darstellung der Lage der Analbeutel und ihrer topographischen Beziehungen zu den Strukturen des Beckenbodens wurde diese Körperregion makroskopisch präpariert. Das für die Histologie benutzte Material war in neutralem Formalın, Sublimat oder Bouinscher Flüssigkeit fixiert. Die Serienschnitte von 5-10 um Dicke wurden verschiedenen Färbungen unterzogen; im einzelnen gelangten die Azanfärbung nach HEIDENHaIn, Färbungen nach CRossMoN, GOLDNER und PAsınI sowie die Eisenhämatoxylinfärbung nach HEIDENHAIN zur Anwendung (RoMEIS 1948). Das Material für elektronenoptische Untersuchungen wurde in 5%igem Glutaraldehyd fixiert, mit Osmiumtetroxid nachfixiert und in Epon eingebettet. Semidünnschnitte wurden mit Toluidinblau gefärbt. Für die chemische Untersuchung wurden die frisch gewonnenen Sekrete in Chloroform/Metha- nol/Wasser (2:1:1; 60 ml) verteilt; hierbei verbleiben die Lipide im Chloroform. Die freien Fettsäuren wurden mit methanolischer NaOH als Na-Salze abgetrennt und durch Ansäuren mit HC] und Extraktion mit Chloroform wiedergewonnen. Die verbliebenen Lipide wurden durch SıO;- Säulenchromatographie (5g Kieselgel, 14,53 % Wassergehalt) unter Benutzung eines Gradienten von Cyclohexan/Benzol/Chloroform getrennt (Jacog 1975, 1977; JacoB und GRIMMER 1973). Die freien Fettsäuren wurden verestert, Cholesterinester, Monoesterwachse und Triglyceride mit 5%iger metha- nolischer HCl (20 ml; 45 min; Siedetemperatur) umgeestert. Nach Hinzufügen von Wasser (20 ml) erfolgten die Extraktion der umgeesterten Substanzen mit Chloroform und deren Trennung ın Die Analbeutel der Zibetkatze 99 Methylester und Alkohole durch SiO,-Säulenchromatographie. Anschließend wurden die Alkohole mit CrO,/Essigsäure in Cyclohexan oxidiert, um die entsprechenden Fettsäuren zu erhalten, die nachfolgend in der beschriebenen Weise verestert wurden. Die Bestimmung der Doppelbindungen ungesättigter Fettsäuren erfolgte nach Oxidation mit OsO, ın Pyridin/Dioxan (MuRAWSsKI et al. 1971) und anschließender Silylierung mit Trisil/N,O-bis-trimethyl-silyltrifluoracetamid (Jacog 1978) durch massenspektrometrische Analyse. Die Bestimmung aller Bestandteile wurde durch gaschromatogra- phischen und massenspektrometrischen erelecch mit authentischen Referenzsubstanzen vorge- nommen. Ergebnisse und Diskussion Zur Topographie der Analbeutel Die paarıgen Analbeutel, die sich in beiden Geschlechtern finden, liegen in der Tiefe der Muskulatur des Beckenbodens. Sie münden über kurze Ausführungsgänge mit stecknadel- kopfgroßen Mündungen beiderseits des Anus in die Zona cutanea. Neben den Analbeuteln besitzt Civettictis wie andere Viverriden auch Perinealorgane, die bei adulten Zibetkatzen eine besonders kräftige Ausbildung erlangen. Bei männlichen Tieren erstrecken sich diese Organe zwischen Scrotum und Vorhaut, während sie bei Weibchen den Raum zwischen Anus und Vulva einnehmen. Die Perinealregion ist bei diesen Tieren durch eine in der Medianlinie verlaufende, reichbehaarte Grube gekennzeichnet, in deren Tiefe, durch einen Längswulst getrennt, zwei mächtige Drüsentaschen mit je einer Öffnung münden (Abb. 1, Abb. 1. Lage der Analbeutel, der Perinealorgane, des Afters und der äußeren Geschlechtsorgane der Zibetkatze (Civettictis civetta); links Männchen, rechts Weibchen; Ansicht von ventral. A = Anus, Ab = Analbeutel (Lage durch gestrichelte Linien angegeben), Ag = Ausführungsgang der Analbeu- tel, C = Circumanalwulst, P = Perinealorgan, Pe = Penis, Pr = Praeputium, Ps = Perinealspalt, S = Scrotum, V = Vulva Die Lage der Analbeutel im Beckenboden macht die Analyse eines Teiles der Muskula- tur des Diaphragma pelvis erforderlich. Hierbei werden im folgenden das Diaphragma anale, dessen Muskulatur sowohl von der Schwanzmuskulatur wie auch vom M. sphincter cloacae abstammt, und das Perineum mit seiner vom Sphincter cloacae herzuleitenden Muskulatur berücksichtigt. 100 Alice von Saldern, H. Schliemann, F. I. B. Kayanja und ]J. Jacob ventral dorsal Rectalschleimhaut M.sphincter ani int. Urethra - An Perinealdrüsentasche | Sekretgranula “ Analbeutel Nager “ emasen x " x D ö 4 q S ö u ML, => - a A M. sphincter ani ext. Abb. 2. Der anogenitale Bereich einer juvenilen weiblichen Zibetkatze (Civettictis civetta) in einem Sagıttalschnitt (halbschematisch); die Lage des rechten Analbeutels ist durch eine gestrichelte Linie angegeben. (Vergr. ca. 2,5fach) = N M.ext.(cr) N X M.ext.(cr) M.ext.(ca) Abb. 3. Analbeutel, Perineal- organ und Muskulatur des Beckenbodens und Umge- bung einer weiblichen Zibet- katze (Civettictis civetta) ın der Ansicht von ventral, cau- dal und lateral; das Präparat wurde in der Mittellinie ge- spalten. A = Anus, Ab = Analbeutel, C.a. = Analka- nal, M.b.f. = M. biceps fe- moris, M.c. = M. coccygeus, M.cf. = M. caudofemoralis, M. ext. = M. sphincter anı ext. (ca = aha cr = cra- nial), X M. ext. = gekreuzte Muskelfasern des M. sphinc- ter ani ext., M.ic. = M. ılio- coccygeus, M.ıt. = M. inter- transversarıus caudae, M. lev. = M. levator ani, M.obt. M. obturator int., M.pc. M. pubococcygeus, M.p.1. = M. perinealis ons M.rc. = M. rectococcygeus, M.sc. = M. sacrococcygeus, M.sm. = M. semimembrano- sus, M.st. = M. semitendino- sus, P = Perinealorgan, Ps = Perinealspalt, R= Rectum, Sf = Schwanzfascie Die Analbeutel der Zibetkatze 101 Muskulatur des Diaphragma anale Hierher gehören der M. sphincter anı ext. und der M. levator anı sowie der M. coccygeus, der für diesen Zusammenhang ohne Bedeutung ist. Der vom Sphincter cloacae abzulei- tende M. sphincter anı ext. läßt eine caudale Pars superficialis und eine craniale Pars profunda unterscheiden. Die Pars superficialis liegt der Haut am nächsten, und ihre Muskelfasern bilden den kräftigen caudalen Anteil des Schließmuskels. Nach NITSCHKkE (1970) entspringen die Muskelfasern beim Haushund an der dorsomedian auf dem Anal- rohr liegenden bindegewebigen Raphe anı; die Fasern enden im subcutanen Bindegewebe der Haut. Die weiter cranial gelegene Pars profunda des M. sphincter ani ext. gliedert sich in zwei nicht immer eindeutig voneinander trennbare Portionen: Eine obertlächliche Schicht nımmt ihren Ursprung teils sehnig an der Schwanzfascie und teils fleischig an der Raphe anı; in ihrem caudalen Bereich umhüllt sie noch strangartig den After und trägt zum Schließmuskelcharakter dieser Portion bei. Die Muskelbündel ziehen seitlich vom Anal- kanal in craniale Richtung, umhüllen die paarigen Analbeutel (Abb. 3) und laufen jeweils in eine lange Sehne aus, die sich jederseits an der Spina ischiadica anheftet. Muskelfasern von der Ventralseite der Analbeutel sind auch am Schambein befestigt. Die tiefere Schicht des cranıalen Sphincteranteils wird größtenteils von der oberflächli- chen verdeckt. Sie entspringt beidseitig mit einer dünnen Aponeurose von den Schwanz- wirbeln zwischen M. levator ani und M. rectococcygeus, umgreift den Analkanal und erfährt unter zum Teil gekreuztem Faserverlauf rostral vom Anus eine kontinuierliche Fortsetzung in dem M. perinealis longitudinalis (Abb. 3) (s.a. NITSCHKE 1970). Der M. levator anı erscheint bei Civettictis als großer, dreieckiger, flacher Muskel, der aus zwei voneinander getrennten Teilen besteht: dem lateralen, schwächeren M. ıliococcy- geus, der an der Innenfläche des Darmbeines entspringt, sowie dem medial davon liegenden, wesentlich stärker ausgebildeten M. pubococcygeus, der seinen Ursprung an der Innenfläche des Schambeines und der Schambeinsymphyse nımmt. Beide Muskelteile verlaufen schräg dorsocaudad, verschmälern sich auf dem Wege durch das kleine Becken zum Schwanz und bilden, mit den Muskeln der anderen Seite konvergierend, ein trichter- förmiges Gebilde, dessen Ventralfläche das Rectum anliegt. Beide Muskelteile verjüngen sich zu flachen Sehnen, die sich an der Ventralfläche der Schwanzwirbel anheften (Abb. 3). Von cranialen und medialen Anteilen des M. pubococcygeus verlaufen Muskeltfasern in Richtung auf die Wand des Analkanals. ELLENBERGER und BAUM (1943) sowie NICKEL et al. (1979) beschreiben bei verschiedenen Carnivoren, daß mediale Teile des Levator anı die äußere Sphinktermuskulatur vervollständigen. Ebenso fand NITscHke (1970) muskulöse Ausstrahlungen des M. pubococcygeus, die teils zum Septum vaginale, teils zur Pars profunda des Sphincter ani ext. zogen. Da zahlreiche Lymphknoten bei dem verfügbaren Material die Präparation erschwerten, kann zu diesen Befunden nichts beigetragen werden. Muskulatur des Perineums Bei vielen Säugetieren ist der Damm nur kurz und u.a. durch transversal verlaufende Perinealmuskeln gekennzeichnet. Civettictis mit den stark entwickelten Perinealorganen besitzt ein ausgedehntes Perineum, dessen Muskulatur nach den vorliegenden Befunden nur aus längsverlaufenden Fasern besteht. Diese Mm. perineales longitudinales umhüllen die Perinealorgane und verbinden den Sphincter anı ext. mit dem Diaphragma urogenitale (M. bulbospongiosus) (Abb. 3). Das Fehlen transversaler Dammuskeln steht in Überein- stimmung mit den Befunden von NITSCHkE (1970). Zur Morphologie der Analbeutel Die Analbeutel stellen drüsenreiche Hohlorgane dar, die sich nach Kröruıng (1927) aus soliden epidermalen Zapfen beiderseits des Anus entwickeln. Bei juvenilen Zibetkatzen 102 Alice von Saldern, H. Schliemann, F. I. B. Kayanja und J. Jacob besitzen sie die Größe einer Haselnuß, bei adulten Tieren erreichen sie Walnußgröße. Der im frischen Zustand gelbliche und dickflüssige Beutelinhalt wird durch je einen nach caudo-medial verlaufenden, drüsenlosen Ausführungsgang, der ebenfalls von quergestreif- ter Muskulatur umhüllt ist, entleert. In das makroskopisch weiße, fibröse Gewebe der Beutelwand sind fünf bis zehn, möglicherweise aber auch mehr, runde bis ovale, kuppelförmig in das Lumen vorragende Talgdrüsenkomplexe von jeweils 3 bis 5 mm Durchmesser eingelassen. Sie sind mehr oder weniger regelmäßig über die innere Oberfläche der Analbeutel verteilt. Am Scheitel eines jeden Lagers befindet sich eine trichterförmige Sekretzisterne, in die alle dem Lager angehörenden Talgdrüsenalveoli sezernieren (Abb. 4, 5). dorsal ug SAT z E BR RER” Schnittfläche des M.sphincter ani ext. Talgdrüsenkomplexe ventral \ Schleimhaut der Beutelinnenwand Abb. 4. Der rechte Analbeutel einer juvenilen Zibetkatze (Civettictis civetta) in einem Längsschnitt von lateral eröffnet; Blick auf die mediale Wandung und Ausführungsgang des Analbeutels (halbsche- matisch). (Vergr. ca. 4fach) Abb. 5. Der rechte Analbeutel einer Zibetkatze (Civettictis civetta) und sein Ausfüh- rungsgang in einem Längs- schnitt ae (Vergr. ca. 6fach). A = apo- krine Drüsen, B = Beutel- lumen, Bg = Ausführungs- gang, C = Sekretzisterne, E = mehrschichtiges, unver- horntes Plattenepithel, glIM = glatte Muskelfasern, L = Lymphtollikel, Lp = Lamina propria, M = M. sphincter ani ext., MBg = Muskulatur des Ausführungsganges, T = Talgdrüsen, Tg = Talgdrü- senausführungsgang, V = Verhornung im Ausführungs- gang und in seiner Nähe, Z = zentrale Sekretzisterne Die Analbeutel der Zibetkatze 103 In der Umgebung jedes dieser Talgdrüsenkomplexe liegen apokrine Drüsen, die bereits bei makroskopischer Inspektion andeutungsweise zu sehen sind. Ein Querschnitt durch die Analbeutelwandung zeigt eine dreigliedrige Schichtung: Zum Lumen hin wird der Analbeutel von einer Schleimhaut ausgekleidet, deren Epithel auf einer bindegewebigen und von Drüsen reich durchsetzten Lamina propria ruht; in der Peripherie folgt eine kräftige muskulöse Umhüllung der Organe. Bei frischem Material zeigt die Schleimhaut keine nennenswerte Faltenbildung (Abb. 4, 5). Das Epithel der Analbeutelwandung und des Ausführungsganges Das Epithel der Beutelwandung einer juvenilen Zibetkatze ist bei kubischen und z. T. auch prismatischen Zellen dadurch gekennzeichnet, daß es mit zwei bis vier Zellschichten vergleichsweise niedrig ist. Bei adulten Tieren ist die Epidermis des Beutellumens ein mehrschichtiges, nicht ver- hornendes Plattenepithel, B das 15 und mehr Zell- schichten umfassen kann (Abb. 6). Die Kategorisie- rung dieses Epithels als Plattenepithel bedarf im weiteren noch der Ergän- zung, da die obersten Zell- schichten nicht an allen Stellen die für diesen Epi- theltyp charakteristische Gestalt aufweisen. Das Stratum basale grenzt mit einer mäßig ge- gliederten Unterseite an die Lamina propria. Eine Vielzahl von Wurzelfüß- chen, die mit Halbdesmo- somen ausgerüstet sind, verankert die basale Zell- schicht ın der Basallamina. In den untersten Schichten babensndıe Zellen "ihre größte Ausdehnung senk- recht zur Epitheloberflä- che, die ovalen Kerne sind verhältnismäßig kleinvolu- mig und stehen dicht ge- Abb. 6. Das mehrschichtige, unverhornte Epithel, das die Analbeu- drängt; sie besitzen hier tel der Zibetkatze (Civettictis civetta) auskleidet. Färbung Toluidin- kessai Es blau. (Vergr. ca. 450fach). B = Beutellumen, BZ = basale Zell- Den weiter Zur Ep schicht, CP = Coriumpapille mit Kapillarschlinge, EO = Epithel- theloberfläche einen deut- oberfläche, IC = Interzellularraum der mittleren Zellschichten lichen Nucleolus. Das Cy- (Stratum spinosum), LP = Lamina propria toplasma ist reich an To- nofilamenten und freien Ribosomen; auffällig ist die große Zahl an Mitochondrien; das rauhe endoplasmatische Retikulum ist wenig entwickelt. Benachbarte Zellen sind an ihren Grenzen stark verzahnt. Zur Epitheloberfläche hin nehmen Zellen und Zellkerne an Umfang zu. Die Zellen sind von polygonalem Umriß bzw. haben ihre größte Ausdehnung parallel zur Epitheloberflä- 104 Alice von Saldern, H. Schliemann, F. I. B. Kayanja und ]J. Jacob che (Abb. 6). Im Cytoplasma finden sich hier ebenfalls freie Ribosomen und zahlreiche Tonofilamente. Die Zellgrenzen zeigen die Charakteristika einer Stachelzellschicht mit einem ausgesprochen reich entwickelten Interzellularraum. In seinen obersten Schichten ist das Wandepithel an manchen Stellen durch stärker abgeflachte, aber kernhaltige Zellen gekennzeichnet, an anderen Orten jedoch ist die oberflächliche Zellage mit blasig aufge- quollen erscheinenden Zellelementen versehen, die stellenweise pilzförmige Protrusionen in das Analbeutellumen aufweisen. Keratohyalıngranula und Keratin sind nicht eindeutig nachzuweisen. Elektronenoptisch sind die oberflächlichsten Zellschichten von unter- schiedlicher Dichte, manche der Zellen enthalten kleine membranumgebene, vakuolenar- tige Gebilde, deren Natur bislang nicht bekannt ist. Daneben kommen hier, aber auch in tiefer liegenden Zellschichten vereinzelt Vakuolen mit homogenem Inhalt vor, die für kleine Lipidvakuolen gehalten werden. Die histologische Organisation des Beutelwandepithels steht in einem deutlichen Gegensatz zu der bei Genetta und Paradoxurus (SCHLIEMANN et al. 1985) sowie bei anderen Carnıvoren (KrÖLLING 1927; SCHAFFER 1940; KRÖLLING und GRAU 1960). Dort sind die Analbeutel von einem relativ niedrigen, verhornenden Plattenepithel ausgekleidet. Es kann unterstellt werden, daf3 diese Bauunterschiede der Epithelien nicht ohne funktio- nelle Bedeutung sind und daß dem Analdrüsensekret von Civettictis an der Epitheloberflä- che abgegebene, relativ intakte Zellen oder Zellteile, auf keinen Fall aber Keratinmaterial beigemischt werden. Es ist nicht bekannt, wie sıch dieses auf die stoffliche Zusammenset- zung des Sekrets auswirkt. Schon SCHAFFER war der Ansicht, daß die Epithelien der Analbeutelwandung für die Sekretbildung eine Rolle spielen, und beschreibt für die Analbeutel von Mephitis eine nicht verhornende Epithelauskleidung, in deren Zellen er Vakuolen findet; er meint, daß dieses Epithel sich an der Lipidbildung beteilige, und vergleicht es mit einem entsprechenden Epitheltyp aus der Kehldrüse von Molossus. ORTMANN (1960) ist ebenfalls überzeugt, daß die Wandepithelien der Analbeutel zu den Sekreten beisteuern. Bei Pelea capreolus fanden STARCK und SCHNEIDER (1971) auf der Glans penis ein Epithel, daß sie als Flächendrüse mit holokriner Sekretion deuteten und mit anderen als holokrine Flächendrüsen angesehenen Epithelien am Praeputialorgan des Bibers sowie des Wiesels und in den Armtaschen von Saccopteryx (STARCK 1958) vergli- chen. STARcK (1982) betont noch einmal die Wichtigkeit solcher Befunde für die Möglich- keit, die Entstehung polyptycher Drüsen zu verstehen. Die geschilderten Befunde an dem Analbeutelepithel von Civettictis würden in dieses Bild passen, wenn die elektronenopti- schen Befunde ausgeprägtere Anzeichen einer Lipidbildung erkennen ließen. Da jedoch nur vereinzelt für Lipidvakuolen gehaltene Strukturen gefunden wurden, für die Lipoge- nese charakteristische Organellen dagegen zu fehlen scheinen, kann lediglich davon ausgegangen werden, daß dieses Epithel den cytoplasmatischen Inhalt von Oberflächenzel- len ganz oder teilweise dem Sekret hinzufügt. Der Ausführungsgang weist auf den Querschnitten ein sternförmiges Lumen auf. Sein Epithel ist von geringerer Dicke als dasjenige, das die Beutelwandung bedeckt. Es ist darüber hinaus verhornt und besitzt nahe am Beutellumen an seiner Unterseite eine stärkere Verzahnung mit der Lamina propria. Verhornung und die vergleichsweise reich strukturierte Unterseite des Epithels sind Anpassungen an die durch die Austreibung des Sekretes bedingten mechanischen Belastungen des Epithels des Ausführungsganges und seiner Öffnung zum Beutellumen hin. Die Lamina propria und ıhre Drüsen Die Lamina propria ist eine kollagenfaserreiche Schicht von z. T. mehr als einem halben Millimeter Dicke, in der die Hautdrüsen liegen und die basal an die Muskelschicht der Analbeutel grenzt (Abb. 5). Es läßt sich eine oberflächliche, an das Epithel grenzende Zone von einer tiefer Die Analbeutel der Zibetkatze 105 liegenden unterscheiden. In der oberflächlichen Zone sind die gewellt verlaufenden Kolla- genfaserbündel weniger dicht gepackt, in ıhr fällt der Reichtum an Kapillaren auf, die vielfach bis nahe unter das Epithel reichen; dieses ist besonders deutlich in den bindegewe- bigen Papillen, in denen die Kapillaren haarnadelförmige Schlingen bilden. Das lockere Bindegewebe dieser oberen Schicht weist außerdem zahlreiche freie Zellelemente auf, unter denen die Makrophagen und die Mastzellen besonders auffallen. An der Grenze zur tiefer gelegenen Schicht der Propria finden sich einzelne oder zu Bündeln und auch Lagen zusammengefaßte glatte Muskelfasern, die sich überkreuzend durch das Bindegewebe ziehen. Diese tiefe Zone ist faserreicher, in ihr liegen stärkere Blutgefäße und vor allem die Talgdrüsenkomplexe und die sie begleitenden apokrinen Drüsen (Abb. 5). Die Lamina propria der Analbeutel von Civettictis ist sehr viel kräftiger entwickelt, als dieses bei den anderen von uns untersuchten Viverriden (Genetta, Paradox- urus) und etwa auch bei /chneumia (ORTMANN 1960) der Fall ist; soweit sich dieses nach einer Abbildung beurteilen läßt, gleicht die Propria der Zibetkatze in ihrer Mächtigkeit derjenigen des Haushundes (SCHAFFER). Die Lamina propria ist sehr reich an Lymphocyten. Hierauf soll gesondert hingewiesen werden, da dieser Befund bislang auch nur an der Zibetkatze erhoben werden konnte und ihm im Zusammenhang mit dem nicht verhornenden Fpithel eine besondere Bedeutung für die Infektabwehr zugeschrieben werden muß. Lymphfollikel finden sich bevorzugt in oberflächlichen, aber auch in tieferen Schichten der Propria. Stellenweise ist das Epithel der Analbeutelwandung durch Lymphocyteninfiltration zu einem Iymphoepithelialen Zellver- band aufgelöst. Die Talgdrüsen Die Talgdrüsenkomplexe werden durch von der Propria einstrahlende, zunächst kräftige und dann zum Zentrum hin zarter werdende Bindegewebssepten in Läppchen unterglie- dert (Abb. 7, 8). Diese Septen enthalten neben Kapillaren und freien Bindegewebszellen sehr häufig auch glatte Muskelfasern. Die peripheren Zellen der Drüsenalveoli ruhen auf einer kräftigen Basallamina, sie sind relativ flach und in auffälliger Weise untereinander und mit den über ihnen liegenden intermediären Zellen an den Zellgrenzen verzahnt (Abb. 9). Ihre ovoiden Zellkerne sind heterochromatinreich, ihr Cytoplasma enthält u.a. freie Ribosomen und Mitochondrien. Dieser undifferenzierte Zelltyp ist jedoch nicht überall in der Alveolusperipherie zu finden, vielfach begegnet man dort auch Zellen, die bereits Lipidvakuolen enthalten. Die zuvor beschriebenen Matrixzellen liegen offenbar bevorzugt in den proximalen Drüsenteilen. Auf die basalen folgen intermediäre Zellen, die bereits alle Anzeichen der Lipogenese zeigen. Ihr Cytoplasma enthält neben zunächst noch kleinen Lipidvakuolen meist homo- genen Inhalts ein sehr ausgedehntes glattes endoplasmatisches Retikulum (Abb. 10), dessen Zisternen blasıge Erweiterungen aufweisen, die u.a. in der für Lipide produzierende Zellen bekannten Weise den Vakuolen anliegen. Das glatte endoplasmatische Retikulum scheint nahezu den gesamten Zellraum auszufüllen, ein Befund, der von den bei Genetta beobach- teten Verhältnissen abweicht (KayanJa und SCHLIEMANN 1981). In Richtung auf die Zerfallszone zeigen die Zellen Vakuolen, die extrahiert sind und an Umfang und Zahl zunehmen. Die großvolumigen, runden und heterochromatinarmen Zellkerne werden häufig von den Vakuolen eingedellt. Schließlich werden die Zellkerne pyknotisch, und die Zellen zerfallen zu Sebum. Die Sekretzisternen distal der Zerfallszonen erscheinen auf den Präparaten leer; sie werden von einem niedrigen, schwach verhornten Plattenepithel gesäumt, das in Richtung auf den in das Beutellumen mündenden Hauptausführungsgang des Komplexes an Höhe gewinnt und dementsprechend ein kräftiges Stratum corneum besitzt. Nahe der Mündung dieses Ganges in das Beutellumen verliert sich die Keratinschicht, und das Gangepithel geht kontinuierlich in das Epithel der Beutelauskleidung über. Der Ausführungsgang wird o>AfeussnıpSfe] = L ‘snjo9AY sap JUOZXLNEM J9p uajppZ 3Jaydırad = Zg “ursserpoysnyy aue3 = MW ‘eldord eumwme]T = 7 “usjonyeaprdr] usysrsapyez ur usjfaZ >aeıpaunazur = ZI "gqamadapurg = Hg '(ypeJozz ‘ea ‘1319 \) nejquipinjo,] 3ung.eg ‘(9119019 517011100157)) 9z1espagıZ 1aura Sunpuemjsinsgfeuy 19p eridord eurure] I9p 3uU0Z uapon I9p sne aruysssny pun 1o9AfeuUssnIpSTe], 19191yoWw allaydulag alq '8 '4gYV SIXa[IwoyUasnIpZTE], sap I[OIAJY = L snıq usyaAıdouou 19p [oyadg sadnysıyssum = 7 ‘josafe- pun uayanejyssussniqg UP UIYISIMZ Jgamadapuıg = IgG “Osnıq usuLıyode 1ap Suesssuns -ynjsny = HY ‘ussnig Junyode = Y '(yseIcz] 'eI 'S1o\) "uezy 3ungseJ "Zuedsdunsynjsny pun uaysnejyssussniq) usuLıyode usıleqyseusg Mur (9119809 S119111217) SZIEMOgIZ J9ur Junpuem -oinsgjeuy I9p xafdwoyuasnıpae], wur sne Hruyassny '/ 'gqW Die Analbeutel der Zibetkatze 107 Abb. 9. Peripherie eines Talgdrüsenalveolus aus der Analbeutelwandung einer Zibetkatze (Civettictis civetta). Beachte die Verzahnung der peripheren Zellen. (Verg. ca. 3 500fach). BL = Basallamina des Alveolusepithels, IZ = intermediäre Zelle mit reich entwickeltem, glatten endoplasmatischen Reti- kulum, LP = Lamina propria, LV = Lipidvakuolen, LY = Lysosomen, M = Mitochondrien, PZ = periphere Zellen mit starker Verzahnung von einer kräftigen Manschette von Bindegewebe umgeben und von glatten Muskelfasern begleitet. Von diesen Muskelfasern und denjenigen in den Septen zwischen den Drüsen- alveoli und Läppchen wird angenommen, daf3 sie die Austreibung des Sekretes unterstüt- zen. Das Gewebe um den Ausführungsgang herum ist reich an Lymphocyten. Die apokrinen Drüsen Die apokrinen Drüsen liegen in der Peripherie der Talgdrüsenkomplexe, eingebettet in die gut kapillarisierte Lamina propria (Abb. 5, 7). Zuweilen erscheinen die apokrinen Drüsen zwischen den Talgdrüsenkomplexen und der muskulösen Umhüllung der Analbeutel nur als flachgedrückte Lager. Unsere Befunde bestätigen die Ansicht Kröruınss (1927), nach der die apokrinen Drüsen auf Grund ihrer Entstehung vom Haarbalg aus in derselben Zahl wie die holokrinen Drüsen vorhanden sind. Bei Genetta ıst eine Zuordnung der sehr ausgedehnten Lager apokriner Drüsen zu einzelnen Talgdrüsenkomplexen nicht möglich (SCHLIEMANN et al. 1985). Das Lumen dieser Drüsen zeigt häufige, ampullenartige Erweiterungen. Aussagen über das Vorkommen von Verzweigungen der Drüsenschläuche sind nicht möglich. Kennzeich- nend für monoptyche Drüsen ist ihr einschichtiges Epithel (Abb. 11), das hier lichtmikro- skopisch an vielen Stellen Anzeichen apokriner Sekretion erkennen läßt. Sezernierendes Epithel besitzt Zellen zylindrischer Gestalt, das Epithel ruhender Drüsenteile ist flacher. Die großvolumigen, mit einem Nucleolus versehenen Kerne liegen in der unteren Zell- hälfte; hier weist das Cytoplasma eine deutliche Basophilie auf. Auf den Semidünnschnit- ten zeigt das Cytoplasma oberhalb der Zellkerne eine Vielzahl kleiner, extrahierter Vakuolen (Abb. 12). Die Myoepithelzellen zwischen dem Epithel und der Basallamına 108 Alice von Saldern, H. Schliemann, F. I. B. Kayanja und J. Jacob Abb. 10. Ausschnitt aus intermediären Zellen eines Talgdrüsenalveolus aus der Analbeutelwandung der Zibetkatze (Civettictis civetta). Beachte das ausgedehnte, in den einzelnen Zellen aber unterschied- lich entwickelte, glatte endoplasmatische Retikulum. (Vergr. ca. 14000fach). ER = glattes endoplas- matisches Retikulum mit z. T. stark erweiterten Cisternen sowie in unmittelbarer Nähe der Lipidva- kuolen, LV = Lipidvakuolen, LY = Lysosom, M = Mitochondrien besitzen bei senkrecht angeschnittenen Tubuli die typische Gestalt kompakter, dreieckiger Cytoplasmakörper. Der weitlumige Ausführungsgang (Abb. 7) ist von einem einschichtigen, kubischen Epithel ausgekleidet. Er mündet in der Regel in den zentralen Ausführungsgang des zugehörigen Talgdrüsenkomplexes. Hierzu kann er entweder an der Peripherie der Talg- drüsen entlangziehen und Kontakt zum Ausführungsgang bekommen oder aber dazu das Talgdrüsengewebe vorher durchdringen. Gelegentlich kann beobachtet werden, daß ein Ausführungsgang direkt in das Analbeutellumen mündet. Ob diese Ausmündungen im Sinne von Kröruıng (1927) als solche primär freier Drüsen zu deuten sind oder ob es sich hierbei um eine sekundäre Verlagerung einer Drüse handelt, die — wıe gewöhnlich - zu einer Haarbalganlage gehört (GREER und CALHOUN 1966; STARCK 1982), kann anhand des Materials adulter Tiere nicht entschieden werden. Bei etwa einem Drittel der Drüsenkomplexe lassen sich Reste von Haaren bzw. Haaranlagen nachweisen (s. hierzu ORTMANN 1960), und gelegentlich ist der Zusammen- hang zwischen dem Haarbalg und seinem Drüsenapparat zu beobachten. Die Talgdrüsen syanejy9suasnig sap uaunT = 7 ‘pyndyg sapusı -91U19Z9S “sO3nyaryssum = T Pgqamsdapurg = Hg “nruyssgeffoz Any1> 391999 ru ofeyrde = IV '(yPeJ0g6 'e> 1319) "nequipimjo], 3ungse,] ‘(913009 1121179017) 9zre>2gIZ 19p Zunpuemjoinsgfeuy J9p sne yanejy9sussnig) uaunıyode uaur y9anp NIuyagS ‘zz 'gYV Sfppzppynd3oAW = JW ‘uswnT S911919M19 änseuajpndure = 7 Osnıq uay94Aıdouou 19p [ysdy sadnysryasurs = 7 PgIMaFapurg = Yg '(YPRJOGE 'e9 1319) "nejquıpıno], 3ung -18,] (0119029 5172117915) JzeI>2gIZ 19P Zunpuemjoinsgjeuy 19p sne asnıq) usunıyode 19urI xafdwoy] up sne nruysssny ‘77 'ggYV 110 Alice von Saldern, H. Schliemann, F. I. B. Kayanja und ]J. Jacob münden dabei in den Haarbalgtrichter, und weiter distal öffnet sich die apokrine Drüse in diesen Trichter, der zur gemeinsamen Sekretzisterne wird, deren Wandung sich nach Krörnıng (1927) aus der äußeren Wurzelscheide entwickelt. Zur Biochemie der Sekrete Tierische Hautdrüsensekrete stellen mehr oder weniger komplexe Lipidgemische dar, die zum überwiegenden Teil aus Squalen, Cholesterinestern, Monoesterwachsen, Diester- wachsen, Triglyceriden, Alkoholen, Sterolen, Diolen und freien Fettsäuren bestehen, jedoch sind nicht alle diese Stoffklassen bei jeder Spezies vertreten. Ein auffälliges Kriterium dieser Sekrete ist, daß die an ihrem Aufbau beteiligten Komponenten (z.B. Fettsäuren und Alkohole) Strukturen besitzen, die sich deutlich von denselben in anderen Organen vorkommenden (z.B. in der Leber) unterscheiden. So zeichnen sich die Fettsäu- ren der Hautlipide häufig durch ungewöhn- Tabelle] liche Doppelbindungslagen oder Methylsub- stitutionen aus. Die hier untersuchten Analdrüsensekrete von Civettictis civetta setzen sich aus Chole- sterinestern, Monoesterwachsen, Diesterwach- sen, Triglyceriden, Alkoholen, Cholesterin und freien Fettsäuren zusammen (Tab. 1), zei- Gholesternester ' 1 | gen aber eine durchaus hautspezifische Kom- Monoesterwachse position der Lipidklassen. Unter den am Auf- Diesterwachse bau dieser Lipide beteiligten Fettsäuren finden Triglyceride Ta sich allerdings nur sehr geringe Mengen von Choleaeen (w-1)- und (w-2)-substituierten Individuen. Freie Fettsäuren Die Doppelbindungen der ungesättigten Fett- säuren liegen in ungeraden Positionen (bei Mo- noenfettsäuren in 7-, 9-, 11- und 13-Position, bei Diensäuren in 8,11-; 9,12- bzw. ın 10,13-Position, also in aus ersteren hergeleiteten Positionen (Tab. 2). Es kann nicht sicher entschieden werden, ob für die Bildung der ungesättigten Fettsäuren unterschiedlich spezifische Desaturasen verantwortlich sind (z.B. eine 9- und eine 11-Desaturase im Falle der C,;-einfach ungesättigten Säuren) oder ob z.B. die ın 11-Position ungesättigte Cjs-Säure durch Kettenverlängerung einer primär in 9- Position desaturierten C},-Säure entsteht (Abb. 13). In jedem Fall zeigen die hier gefundenen Fettsäurestrukturen mehrheitlich keine hervorragenden Besonderheiten. Die chemische Zusammensetzung der Sekrete wie auch die Strukturen ihrer Bestandteile unterscheiden sich jedoch signifikant von den bei Genetta gefundenen Verhältnissen (JACOB und SCHLIEMANN 1986). Quantitative Zusammensetzung der Lipide der Sekrete der Analbeuteldrüsen der Zibetkatze (Civettictis civetta) Zur Funktion der Analbeutel Wie bereits dargelegt, beziehen sich die Informationen über das Markierverhalten von Civettictis nahezu ausschließlich auf die Benutzung der Perinealdrüsen. Eine detaillierte Beschreibung dieses Verhaltens findet sich bei EwER und WEMMER (1974), die zwei Paare dieser Tiere in Gefangenschaft beobachten konnten. WEMMER (1977) geht in seiner ethologischen Studie über Genetta ebenfalls auf die Zibetkatze ein und beschäftigt sich mit deren Markierverhalten. Nach den Beobachtungen dieser Autoren werden für den Vor- gang des Markierens das Organ evertiert und die den Perinealspalt begrenzenden, mit Sekret bedeckten Flächen bloßgelegt und an die zu markierende Stelle gedrückt. Horizon- tale Gegenstände werden in normaler Positur, im Stehen auf den vier Extremitäten also, oder im Niederkauern markiert, eine Haltung, die auch zum Markieren von Stellen auf Die Analbeutel der Zibetkatze blall Tabelle 2 Quantitative Zusammensetzung der freien Fettsäuren und Fettsäuren der Monoesterwachse, der Cholesterinester und Triglyceride (% des GC - Flächenintegrals) Monoesterwachse/ Triglyceride Freie Fettsäuren Cholesterinester Gesättigte, unverzweigte Fettsäuren (total) (56.8) (66.1) Oo0000000 (3.9) De, 1.0 BE, . 2.9 Ungesättigte Fettsäuren (total) (39.3) 16:1 (7)* 3.9 16:1 (9) 2.3 18:1 (9) 14.3 Bd) Ä 8.9 20:1 (13) a 1832 (&, HE) 20:2 (10, 13) 20:3 20:4 20: polyen * Lage der Doppelbindung dem Boden benutzt wird. Zum Markieren vertikaler Objekte richten dıe Tiere den Hinterkörper bei steil aufgestelltem Schwanz gegen den zu markierenden Gegenstand und pressen das Perinealorgan fest auf das Objekt. Handstandähnliche Stellungen konnten nicht beobachtet werden. Vertikale Gegenstände werden offenbar mehr von Männchen, horizontale mehr von Weibchen markiert. Interessant sind die Deutungen von EwWER und WEMMER hinsichtlich der Situationen, in denen die Tiere markierten. Markiert wurden unbekannte Gegenstände oder bekannte, auf denen sich eine fremde Marke befand. Die Tiere näherten sich solchen Objekten stets mit Vorsicht, und es schien beiden Autoren, daß die Tiere erst durch das Setzen der eigenen Duftmarke „Sicherheit“ empfanden. In Situationen des Alarmiertseins wurde nicht markiert (s. hierzu Rasa 1973, 1984), jedoch schien Ängstlichkeit und Unsicherheit eine Indikation für dieses Verhalten zu sein. Diese Beobachtungen stimmen mit denjenigen von RAnDALL (1979) überein, der feststellte, daß Zibetkatzen in selten besuchten Arealen ihres Wohngebiets mit deutlich erhöhter Frequenz markieren, wie dieses auch geschieht, wenn andere Individuen in der Nähe sind. Nach RANDALL markieren Zibetkatzen sehr häufig in der Nachbarschaft ihrer Dunghaufen. Die Funktion der Analdrüsen ist wahrscheinlich im Zusammenhang mit diesen Dung- haufen zu sehen. Civettictis und auch andere Viverriden sowie Carnivoren im allgemeinen haben die Angewohnheit, immer wieder benutzte Dunghaufen anzulegen (z. B. ROSEVEAR 1974; VERHEYEN 1951); dieses konnten EWER und WEMMER sogar an ihren Gefangen- 112 Alice von Saldern, H. Schliemann, F. I. B. Kayanja und ]. Jacob a) für gesättigte Fettsäuren Startersubstrat Elongationssubstrat Endprodukt 0 a FSS CH3C- 5-CoA = [HO-C-CHZC-SCoA] m CHZICH,) COOH Acetyl-Coenzym A Malonyl - Coenzym A (Fettsauresynthase) geradzahlige ‚unverzweigte Fettsäuren „0 N % ESS : n. Ins 3CH3 an 7 [ Malonyl-Coenzym A ] — Che, CH COOH Propionyl-Coenzym A ungeradzahlige ‚unverzweigte Fettsauren CH 20 N CH 3>CH-C. *7 [ Malonyl - Coenzym A ] en esse (CH3 )]= COOH CH, SCoA CH3 Isobutyryl -Coenzym A ungeradzahlige «-1) -methylsubstituierte Fettsauren (2-Methylpropionyl-CoA) CH „DO 3SCH-CHJ-C +42 [Malonyl - Coenzym A ] Au me Ci; >CH- (CH) - COOH CH3 22 2n+1 SCoA CH3 Isovaleroyl -Coenzym A geradzahlige, (o-1)-methylsubstituierte Fettsäuren (3-Methylbutyryl -CoA) ZURN CH CHZ CH-C_*2 [Malonyl=Coenzym A ] en CH3-CH2-CH- (CHy) COOH Anteisovaleroyl - CoenzymA geradzahlige, (w-2 Jmethylsubstituierte Fettsäuren ( 2-Methylbutyryl-CoA) b) für ungesättigte _ Fettsäuren (Beispiel: C1g- Säuren) 9 M A ?- Desaturase CH3= (CH)),g COOH — — = —e CH (CH CH= CH = (CH))7 - COOH Stearinsäure Ann- ? Octadec -9-ensäure - 1(Olsaure) &Sar;; Vras,, -Verlän CH3 (Holz CH=CH ICH,), on COOH mit Malonyl-CoA Hexadec-9-ensäure -1 Octadec - 11- ensäure-1 (Palmitölsäure) Abb. 13. Fettsäuren aus dem Analdrüsensekret von Civettictis civetta und deren mögliche Biosynthe- sewege schaftstieren beobachten. Nach Ranvarı legen die Zibetkatzen ihre sog. „Civetries“ bevorzugt auf Lichtungen und an Wegen an, also an Plätzen, die leicht zugänglich und an denen die Dunghaufen möglicherweise auch optisch auffällig sind. Interessanterweise konnte WEMMER (1977) feststellen, daß den Dunghaufen von Genetta ein langanhaltender Duft anhaftet. Mit RanDAaLıL und RosEVEAR vermuten wir, daß den Faeces die Sekrete der Analdrüsen nicht nur bei Genetta, sondern auch beı Civettictis und anderen Carnıvoren zugegeben werden und daß es diese Drüsensekrete sind, die für die langanhaltende Siıgnalwirkung der Dunghaufen verantwortlich sind. In diesem Zusammenhang weisen wir noch einmal darauf hin, daß in den Analdrüsense- kreten von Civettictis und im übrigen auch von Genetta (JacoB und SCHLIEMANN 1986) keine leichtflüchtigen Substanzen gefunden wurden. Dieser Befund steht im Gegensatz zu den Verhältnissen bei Herpestes, für dessen Analbeutelsekret GoRMan et al. (1974) eine ganze Reihe kurzkettiger Fettsäuren beschrieben haben. Leichtflüchtige Sekretbestandteile sind bei vielen anderen Carnivoren gefunden worden, so etwa bei Canis (ALBONE und Die Analbeutel der Zibetkatze 1418 PERRY 1976), bei Vulpes und Panthera (ALBONE und GRÖNNEBERG 1977) und bei verschie- denen Musteliden (z.B. ANDERSEN und BERNSTEIN 1975; Brınck et al. 1978; CRumP 1980a, 1980b; PrETI et al. 1976; SCHILDKNECHT et al. 1981; SoKoLov et al. 1980). Leichtflüchtige Substanzen können für den lang anhaltenden Duft der Dunghaufen nicht verantwortlich sein, sehr wohl jedoch ist vorstellbar, daß dieses für die bei Civettictis gefundenen Sekrete gilt, wenn man davon ausgeht, daß die Sekrete in den Dunghaufen eine bakterielle Zersetzung erfahren, während derer über einen längeren Zeitraum hinweg flüchtige Substanzen freigesetzt werden. Die Vielzahl der gefundenen Verbindungen läßt es denkbar erscheinen, daß mit ihrer Hilfe eine größere Zahl verschiedener Signale erzeugt werden kann. Durch Rasa (1973, 1984) und GorMAaNn (1976) sind wir darüber unterrichtet, wie weitgehend die Informationen sind, die durch das Markieren gewonnen werden. Über die biologische Bedeutung von Dunghaufen haben sich in letzter Zeit verschie- dene Autoren geäußert, u.a. KrEıman (1972), MACDONALD (1980) und GORMAN und Mırıs (1984). Es gibt unverkennbare Hinweise darauf, daß über die ständig wieder benutzten Dunghaufen wichtige Informationen weitergegeben werden, die sich nicht unbedingt nur auf die Abgrenzung eines Reviers beschränken müssen. GORMAN und Mırıs haben dargelegt, daß Tiere mit kleinen Territorien eine andere Strategie in der Anlage der Dunghaufen verfolgen als Tiere mit großen Territorien; erstere markieren bevorzugt an den Territoriumsgrenzen, während es für letztere ökonomischer ist, im Territoriumsinneren zu markieren. Wenn auch nicht sicher bekannt ist, ob Civettictis Territorien verteidigt, läßt sich für eine nächtlich und solitär lebende Form gut vorstellen, daß es eine Reihe von biologischen Vorteilen ergibt, Signale ausströmende Dunghaufen zu besitzen, die anzeigen, daß das betreffende Areal bewohnt und seine Ressourcen genutzt werden. Partnerfindung und das „Heimischfühlen“ des markierenden Tieres sind sicher auch mit in Betracht zu ziehen. Welche Bedeutung in diesen biologischen Zusammenhän- gen den Analbeuteln und welche den Perinealdrüsen ın einzelnen zukommt, ist nicht geklärt. Möglicherweise werden mit dem Markieren durch die Perinealdrüsen Kurzzeit- und mit dem Markieren durch die Sekrete der Analbeutel Langzeiteffekte erzeugt. Danksagung Wir danken Herrn Prof. Dr. R. SacHs und Herrn Dr. H. VoELKER vom Bernhard-Nocht-Institut für Schiffs- und Tropenkrankheiten sowie Herrn Prof. Dr. H.-G. Krös und Herrn Dr. H. FRÄDRICH vom Zoo Berlin für die Überlassung des Untersuchungsmaterials. Frau W. ScHMIDT und Herrn G. Raag sind wir für ihre ausgezeichnete technische Assistenz zu Dank verpflichtet. Die Deutsche Forschungsgemeinschaft unterstützte unsere Arbeiten dankenswerterweise mit einer Sachbeihilfe. Zusammenfassung Die beiderseits vom Analkanal liegenden Analbeutel bestehen bei Civettictis civetta aus einem das Lumen auskleidenden, mehrschichtigen, nicht verhornenden Plattenepithel, unter dem eine an Binde- gewebs- und glatten Muskelfasern sowie freien Zellen reiche Lamina propria liegt. In dieser finden sich ca. zehn Talgdrüsenkomplexe, denen jeweils apokrine Schlauchdrüsen zugeordnet sind. Die Histologie und Ultrastruktur der Drüsen und des Epithels werden analysiert und die auffallende Entwicklung des glatten endoplasmatischen Retikulums in den mit der Lipogenese befaßten Talgdrü- senzellen hervorgehoben. In der Peripherie werden die Analbeutel von einer kräftigen Lage querge- streifter Muskulatur umgeben, die vom M. sphincter anı ext. abzuleiten ist. In dem Sekret der Analbeutel von Civettictis werden keine leichtflüchtigen Substanzen gefunden. Die Lipide der Analdrüsensekrete setzen sich aus Cholesterin, Cholesterinestern, freien Fettsäuren und Monoester- wachsen in großen sowie geringeren Mengen an Diesterwachsen, Triglyceriden und freien Alkoholen zusammen. Die biologische Bedeutung dieser Sekrete als Signalgeber ın den Dunghaufen der Ziber- katze wird diskutiert. Literatur ALBONE, E. $.; GRÖNNEBERG, T. O. (1977): Lipids of the anal sac secretions of the red fox (Vulpes vulpes) and the lion (Panthera leo). J. Lipid Research 18, 474-479. ALBONE, E. $.; PERRY, G. C. (1976): Anal sac secretion of the red fox, Vulpes vulpes. Volatile fatty 114 Alice von Saldern, H. Schliemann, F. I. B. Kayanja und ]J. Jacob acıds and diamines. Implications for a fermentative hypothesis of chemical recognition. J. Chem. Ecol. 2, 101-111. ANDERSEN, K. K.; BERNSTEIN, D. T. (1975): Some chemical constituents of the scent of the striped skunk, Mephitis mephitis. J. Chem. Ecol. 1, 493-499. BEARDER, $. K.; RAnDALL, R. M. (1978): The use of fecal marking sites by spotted hyaenas and civets. Carnivore 1, 32-48. BOTHMA, ]J. DU P. (1965): Random observations on the food habıts of certain Carnıvora (Mammalia) in southern Africa. Fauna Flora Pretoria 16, 16-22. BRINCK, C.; GERELL, R.; ODHAM, G. (1978): Anal pouch secretion in mınk, Mustela vison. Oikos 30, 68-75. COETZEE, C. G. (1971): Order Carnıvora. In: The Mammals of Africa. An Identification Manual. Ed. by J. MEESTER, H. W. SETZER. Washington: Smithsonian Inst. Press. CRUMP, D.R. (1980): Thietanes and dithiolanes from the anal gland of the stoat, Mustela erminea. ]. Chem. Ecol. 6, 341-347. — (1980): Anal gland secretion of the ferret (Mustela putorius forma furo). J. Chem. Ecol. 6, 837-844. Dücker, G. (1965): Das Verhalten der Viverriden. In: Handbuch der Zoologie. Hrsg. J.-G. HELMCKE, H. v. LENGERKEN, D. STARCK. Bd. 8, 1-48. Berlin: de Gruyter. ELLENBERGER, W.; Baum, H. (1943): Handbuch der vergleichenden Anatomie der Haustiere. 18. Aufl. Nachdruck 1977. Berlin, Heidelberg, New York: Springer. EWER, R. F. (1973): The Carnivores. London: Weidenfeld and Nicolson. EWER, R. R.; WEMMER, CH. M. (1974): The behaviour in captivity of the African civet, Civettictis civetta. Z. Tierpsychologie 34, 359-394. GORMAN, M.L. (1976): A mechanısm for individual recognition by odour in Herpestes auropunctatus (Carnıvora: Viverridae). Anımal Behaviour 24, 141-145. GOoRMAn, M. L.; Mırıs, M. G. L. (1984): Scent marking strategies in hyaenas (Mammalıa). J. Zool. Lond. 202, 535-547. GORMAN, M. L.; NEDWELL, D. B.; SMITH, R.M. (1974): An analysıs of the contents of the anal scent pockets of Herpestes auropunctatus. J. Zool. 172, 389-394. GREER, M. B.; CALHoun, M.L. (1966): Anal Sacs of the Cat. Am. ]. Vet. Res. 27, 773-781. Jacos, J. (1975): TLC, GLC and MS of complex lipid mixtures from uropygial secretions. ]. Chromatogr. Sci. 13, 415-422. — (1977): Bürzeldrüsenlipide. Fortschr. Chem. org. Naturst. 34, 373-438. — (1978): Sex-dependent Composition of Cuticular Lipids from the Beetle Rhagonycha fulva. Hoppe-Seyler‘s Z. Physiol. Chem. 358, 653-656. JACOB, ].; GRIMMER, G. (1973): Lipidzusammensetzung gesunder und pathologischer Hautbezirke bei Psoriasis vulgaris. Z. klin. Chem. klin. Biochem. 11, 297-300. JACOB, J.; SCHLIEMANN, H. (1983): Chemical composition of the secretion from the anal sacs of Civettictis civetta (Schreber, 1776). Z. Naturforsch. 38c, 497-500. JACOB, J.; SCHLIEMANN, H. (1986): The Anal Sac Secretion of Viverrids from the Genus Genetta. Z. Naturforsch. 41c, 325-336. KayanjJa F. I. B.; SCHLIEMAnN, H. (1981): Sebaceous glands of the anal sacs of Genetta tıgrına (Schreber, 1778). Z. Säugetierkunde 46, 26-35. KLEIMAN, D. G. (1972): Social behavior of the Maned wolf (Chrysocyon brachyurus) and Bush dog (Speothos venaticus): a study in contrast. J. Mammalogy 53, 791-806. — (1974): Scent marking ın the Binturong, Arctictis binturong. J. Mammalogy 55, 224-227 (1974). Kröruıng, ©. (1927): Entwicklung, Bau und biologische Bedeutung der Analbeuteldrüsen beı der Hauskatze. Z. Anat. Entwickl.-Gesch. 82, 22-69. Kröuuıng, O.; GRAU, H. (1960): Lehrbuch der Histologie und vergleichenden mikroskopischen Anatomie der Haustiere. Berlin und Hamburg: Paul Parey. LEYHAusen, P. (1969): Über die Funktion der relativen Stimmungshierarchie, dargestellt am Beispiel der phylogenetischen und ontogenetischen Entwicklung des Beutefangs von Raubtieren. In: Antriebe tierischen und menschlichen Verhaltens. Hrsg. K. Lorenz, P. LEYHAUSEn. München: Piper. 169-271. MACDONALD, D. W. (1980): Patterns of Scent Marking with Urine and Faeces amongst Carnıvore Communities. In: Olfaction in Mammals. Ed. by D. M. SroppaArT. Symposia Zool. Soc. London 45, 107-139. MAURER, B.; GRIEDER, A.; THOMMEn, W. (1979): (cis-6-Methyltetrahydropyran-2-yl)acetic Acıd, a Novel Compound from Civet (Viverra civetta). Helv. Chim. Acta 62 (Nr. 6), 44-47. MurAwskt, U.; EGGE, H.; ZILLIKEN, F. (1971): Massenspektren synthetischer methylen- und nicht- methylen-unterbrochener cis, cis-Oktadecadiensäure-methylester als Tetra-O-trimethylsilyläther. Z. Naturforsch. 26b, 1241-1252. NiIckEL, R.; SCHUMMER, A.; SEIFERLE, E. (1979): Lehrbuch der Anatomie der Haustiere. Bd. 2. Berlin und Hamburg: Paul Parey. NITScHKkE, T. (1970): Diaphragma pelvis, Clitoris und Vestibulum vaginae der Hündin. Anat. Anz. 127, 76-125. Die Analbeutel der Zibetkatze 115 ORTMANN, R. (1960): Die Analregion der Säugetiere. In: Handbuch der Zoologie. Hrsg. J.-G. HELMCKE, H. v. LENGERKEN, D. STARCK. Bd. 8, 1-68. Berlin: de Gruyter. Pocock, R. J. (1915): On the feet and glands and other external characters of Viverrinae, with description of a new genus. Proc. Zool. Soc., London (1915), 131-149. — (1916): On the external characters of the Mongooses (Mungotidae). Proc. Zool. Soc., London (1916), 742-755. PRETI, G.; MUETTERTIES, E. L.; FURMANn, J. M.; KEnnELLY, ]J. J.; JoHns, B. E. (1976): Volatile constituents of dog (Canis familiaris) and coyote (Canis latrans) anal sacs. J. Chem. Ecol. 2, 177-186. RAnDALL, R. M. (1979): Perineal gland marking in free-ranging African civets, Civettictis civetta. ]. Mammalogy 60, 622-627. Rasa, O. A. E. (1973): Marking behaviour and its socıal significance in the African dwarf mongoose, Helogale undulata rufula. Z. Tierpsychologie 32, 293-318. — (1984): Die perfekte Familie. Stuttgart: Deutsche Verlagsanstalt. ROEDER, J. J. (1978): Marking behaviour in genets (G. genetta L.) seasonal varıations and relation to socıal status in males. Behaviour 67, 149-156. — (1983): Memorisation des marques olfactives chez la Genette (Genetta genetta L.): duree de reconnaissance par les femelles ie marques olfactives de males. Z. Tierpsychologie 61, 311-314. ROSEVEAR, D. R. (1974): The Carnivores of West Afrıca. London: Brit. Mus. (Nat. Hist.). Romeiıs, B. (1948): Mikroskopische Technik. München: Leibniz. Ruzıcka, L. (1926): Zur Kenntnis des Kohlenstoffringes. I. Über die Konstitution des Zibetons. Helv. Chim. Acta 9, 230-248. en J. (1940): Die Hautdrüsenorgane der Säugetiere. Berlin und Wien: Urban und Schwarzen- erg. SCHILDKNECHT, H.; BIRKNER, CH.; Krauss, D. (1981): Struktur und Wirkung der Musteliden- Okomone. II. Erweiterte Analyse des Analbeutelsekretes des Nerzes Mustela vison L. Chemiker- Ztg. 105, 273-286. SCHLIEMANN, H.; JACOB, J.; KayaAnjJa, F. I. B. (1985): Vergleichend-morphologische und -biochemi- sche Untersuchungen an Analbeuteln von Schleichkatzen (Mammalıa, Viverridae). Ein Beitrag zum Verständnis chemischer Kommunikation der Landraubtiere. Funkt. Biol. Med. 5, 319-329. SORKOLONYANV FE: ALBONE, E. S.; FLoonp, P. P.; HrAP, P. E.; Kacan, M. Z.; VASILIEVA, V. S.: Roznov, V. V.; ZINKEVICH, E. P. (1980): Secretion and secretory tissues of the anal sac of the Mink, Mustela vison. Chemical and histological studies. J. Chem. Ecol. 6, 805-825. STARCK, D. (1958): Beitrag zur Kenntnis der Armtaschen und anderer Hautdrüsenorgane von Saccopteryx bilineata Temminck 1838 (Chiroptera, Emballonuridae). Morph. Jb. 99, 3-25. — (1982): Vergleichende Anatomie der Wirbeltiere auf evolutionsbiologischer Grundlage. Bd. 3. Berlin, Eelbers New York: Springer. STARCK, D.; SCHNEIDER, R. (1971): Zur Kenntnis, insbesondere der Hautdrüsen von Pelea capreolus (Artiodactyla, Bovidae, Antilopinae, Peleini). Z. Säugetierkunde 36, 321-342. VERBERNE, G.; LEYHAUSEN, P. (1976): Marking behaviour of some Viverridae and Felidae. Time- interval analysıs of the marking pattern. Behaviour 58, 192-253. VERHEYEN, R. (1951): Contribution A l’etude ethologique des Mammiferes du Parc National de l’Upemba. Brüssel: Institut des Parcs Nationaux du Congo Belege. WEMMER, CH.M. (1977): Comparative ethology of the large-spotted genet (Genetta tıgrina) and some related viverrids. Contrib. Zool. 239, 1-93. Anschriften der Verfasser: Dipl.-Biol. ALıcE von SALDERN, Prof. Dr. HARALD SCHLIEMANN, Zoolo- gisches Institut und Zoologisches Museum der Universität Hamburg, Martin-Luther-King-Platz 3, D-2000 Hamburg 13; Prof. Dr. FREDERIK I. B. Kayanja, Dept. of Veterinary Anatomy and Histology, Makerere University, P.O. Box 7062, Kampala, Uganda; Prof. Dr. JÜRGEN JACOB, Biochemisches Institut für Umweltcarcınogene, Sieker Landstraße 19, D-2070 Ahrensburg Factors determining the onset of sexual maturity in New Zealand chamois (Rupicapra rupicapra L.) By J. J. BAUER Department of Zoology, University of Canterbury, New Zealand, and Zoologisches Institut der Albert-Ludwigs-Universität, Freiburg, FRG Receipt of Ms. 17. 4. 1986 Abstract During a study on the dynamics of New Zealand chamois populations, subadult anımals (less than two years) were examined for evidence of sexual maturity and reproduction. High incidence of sexual maturity ın yearlings was found. This phenomenon seems to be restricted in European chamoıis to introduced populations found in low altitudes. Furthermore sexual maturity (gonad development, follicle maturation, ovulation) and actual reproduction (presence of embryo) was found in 7 month old females. It could be shown, that sexual maturation within this age class in both sexes ıs being determined by crıtical body weight and daylight values, both acting as synergistic double stımulı. Daylength, being a fixed parameter acts hereby as primary stimulus, which depends on a critical body weight for activation. Should the anımal faıl to reach this critical weight before the sensitive phase passes, sexual maturity is delayed for at least one year. Comparison of two regions with different plane of nutrition and final adult body size showed, that this crıtical body weight does not appear to be an absolute value, but seems more lıkely to be a specific percentage of the final adult body weight. Implications of these results for the rate of increase of populations are discussed briefly. Introduction The age of sexual maturity in different species of artiodactyls appears to be rather uniform; varlability within one species being larger than between species. Under optimal conditions females of all Caprinae species can give birth at the age of two years. The beginning of reproduction at the age of one year, known from pronghorn antilopes (Antılocapra americana; McKean 1954; WRIGHT and Dow 1962; MITCHELL 1967), Sika deer (Cervus nippon; KıDDıE 1972; CHapMAN and HorwooD 1968), Mule deer (Odocosleus hemionius; Cowan 1956), White-tailed deer (Odocosleus virginianus; CHEATUM and MoRToN 1946; MORTON and CHEATUM 1946; CHEATUM 1942), Reindeer (Rangifer tarandus; KLEIN 1968) and Saiga (Saiga tatarıca,; HEPTMER et al. 1966), is less common amongst the Caprinae and seems to be confined to domesticated forms like Ovis ammon f. arıes (JEWELL and GRUBB 1974) or Feral goats (Capra aegagrus f. hircus; RUDGE 1969) or breeds living in captivity. Typically, both species show high incidence of twinning when adult. In the majority of wild living populations of Caprinae however, females do not give birth before their third or fourth year of age, Alpine ibex (Capra ibex) may fail to bear offspring until five to six years of age (NIEVERGELT 1966). Sexual maturity in chamois is reached between the ages of 3%%-4% years (COUTOURIER 1938; SCHRÖDER 1971). The start of reproduction at the age of 1% years, formerly believed to be confined to odd individuals (TRarz 1957; Knaus 1960; CoUTOURIER 1938; KRÄMER 1969) was shown in later investigations to be quite common in populations of good condition. CAUGHLEY (1970) showed for New Zealand chamoıis from the Rangitata/Copland area, that 39 % of the yearling females gave birth for U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5202-0116 $ 02.50/0 Z. Säugetierkunde 52 (1987) 116-125 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Sexual maturity in New Zealand chamois 117 the first time (LAMBERT and BARTHGATE 1977). In the Wanaka area the fecundity of yearlıng females (100%; N = 10) was even higher (BAUER 1982; BAUER 1983a, 1985). European, colonizing populations of chamoıs show similar fecundity patterns. SALZ- MANN (1977), analyzing corpora albicantia from six 2% year old females from the Swiss Jura, found that four of them had given birth previously. In the Black Forest (BAUER in prep.) and the Vosges (BAUER in prep.) it has also been observed, that a certain number of 1 year old females may already conceive and give birth the following spring. Consequences of the varıability of sexual maturity on the rate of growth of populations are considerable. Core (1954) and LEwoNTIN (1965) pointed out, that in faster growing populations the age of sexual maturity is being reduced to the physiological minımum of the species. This phenomena to increase the growth rate of the population should be of particular importance ın species which are colonizing readily under suitable conditions, however are not capable of varyıng their number of offspring, as is the case in chamois (BAUER in prep.). Populations living under colonizing conditions like the Wanaka chamois in New Zealand (BAUER 1982) are ıdeal objects for the investigation of factors determining the sexual maturity of this species and of local populations, respectively. Material and methods During a study, comparing the condition of colonizing (Wanaka region - high plane of nutrition, Fig. 1) and stable (Harper-Avoca region - low plane of nutrition, Fig. 1) chamoıs populations in New Zealand (BAUER 1977; BAUER and Gossow 1978; BAUER 1982, 1983a, 1983b, 1983c, 1985; BAUER and Gossow 1983), considerable differences were found in adult body weights, fecundity and sexual maturity. Differences in growth rates of fawns were equally high. Whereas in stable populations 10-12 months old fawns averaged 9.9 kg (N = 15), sıx to eight months old fawns of peripheral (colonızıng) populations had mean body weights of 15.2kg (N = 40). Equally dissimilar fecundity and maturity patterns between the populations suggested some cor- SSL KAHUTERA relation between growth rate and L sexual maturity as it has been % 1979 found in rats (WıDDowson 1968), HARPER-AVOCA hare (Frux 1967), cattle (JOUBERT stable population 1954, 1963; Hansson 1956), roe deer (ELLENBERG 1978), and red deer (MoorE 1982). As uterus, and testis development are strongly accelerated during 5 wATafiX ö KIRKLISTAN RANG berty (Amir et al. 1968), they an r ” 5 served as indicators of sexual N 1977 . ö colonizing maturity during the study. population Gonads were taken from anımals 20-120 min alter their death and EERRTON stored in 5% formalın. Testıs & were dried and weighed, after AT MT. COOK Ben of the tunica albuginea, EE1900 - 1930 with the epidydimis). Uteri were iR : ne a dried, RN1930 - 1950 weighed to the nearest 0.01 g, dis- E31950 - 1970 sected and searched for blastocy- stes or embryos. Mean testis Em] > 1975 weights were obtained by averag- je u, ing the weights of both testes. Fig. 1. Distribution and history of colonization of chamois in Mean weight differences in New Zealand (N.Z.) regions mentioned in the text are indicated. gonads of strong/weak kids were (A distribution map of the New Zealand Forest service was tested with Students t-test for sig- modified according to own observations) 118 J. J. Bauer nificance, assuming normal distribution. The number of examined males less than two years of age was 27, the number of females, less than two years, was 56. In both regions, high incidence of sexual maturity in yearlings was found (BAUER 1982, 1982a, 1983a, 1983b, 1985). Furthermore, sexual maturity at the age of 7 months was recorded in the Wanaka region as well as in the Rangitata/ Copland region (CAUGHLEY 1970). The location of these areas is shown in Fig. 1. As maturity at the age of 7-8 months is physiologically possible, the study on which factors determine sexual maturity in chamois, was concentrated on kıds. In the Wanaka region data were collected from May-June (rut), in the Harper-Avoca region from October-February (parturition). Results Establishment of the hypothesis According to the investigations of HammonD (1949), HAFEZ (1952) and WATSON and GAMBLE (1961), stimulation of gonad growth by diminishing daylıght is only effective in animals which have reached a critical body size. Uterus growth (which ıs mainly stimulated by estrogen and progesteron produced by the ovaries; HEAP 1972) ıs fastest during puberty. Relatively faster uterus growth (in relation to body growth) shows therefore maturation of sexual organs (Amir et al. 1968). If this also applies to chamois of the Wanaka region, New Zealand, the following testable hypotheses can be established. 1. Uterus weights of weak and strong kids should not differ significantly before a daylength stimulus becomes effective. 2. During a period of dıminishing daylıght, uterus weights of heavy kids should increase proportionally more to those of weak kids. These relationships should also hold for the testis growth of male kids. Female kids were considered as weak, if their body weights were less than 13 kg (eviscerated), they were considered as strong, if they exceeded 15 kg. The corresponding weight classification in male kids was 13.5 and 15.5 kg, respectively. Testing of the hypothesis Sexnal maturation in female kids from the Wanaka region The average weights of female kids between the two sampling dates (18.5. and 24. 7.) increased by 7 % and the uterus weights by 30 % (Tab. 1). Although both increases are not significant, the proportionally larger increase of the gonads during this period, indicates that already in their first year gonads mature during the rut. Plotting uterus weights against body weights of female kids ( N = 18), shows a rather high linear correlation which holds especially well within weight ranges of 12-16 kg. From 16 kg onwards, the uterus growth seems to be accelerated (Fig. 2). Table 1 Body weight and uterus weight increase of chamois kids (absolute and relative) in the Wanaka region, New Zealand, between the 18. 5. 1977 and the 24. 7. 1977 (N = 17) Increase (%) Body weight (kg) Uterus weight (8) 9920. .00 + * Differences in body weights not significant. ** Differences in uterus weights: FG = 11, t = 1.91, p <0.10 Sexual maturıty in New Zealand chamoıs 119 % 300 Fi Bodyweight Uterusweight Gay) 200 Uterusweight Deviatiation =) oO 2 15 ls 16, 17 1erıs Bodyweight (kg) May July Fig. 3. Mean body weight (kg) and mean uterus Fıg. 2. Uterus weights of 7 month old chamois weight (g) of strong chamois kids (Z 15 kg) in the kids (N = 18) which have not yet ovulated, in Wanaka region, N. Z. in May/July, expressed as the Wanaka region, (N.Z.), as a function of positive deviation (%) from the mean body weight body weight (y = -3.3 +0.4x; r = 0.70) (uterus weight) of weak kıds (<13 kg), (N = 13) Table 2 Mean uterus weights and mean body weights of female chamois kids from the Wanaka region, New Zealand (N = 13) Weak kids Strong kids Body weight Uterus weight N Body weight Uterus weight (kg) (8) (kg) mean + SD ale mean + SD 119.0) 22. (0,5 3 i 165 =::09 122522=2.055 3 ; 164 ==.055 In Tab. 2 average body weights and uterus weights in May and July are shown. Despite small sample sıze (N = 13) differences in weight proportions of uteri and body weights of weak and strong kids (Tab. 2) are evident ın July (FG = 4, t = 3.02, p < 0.05), whereas ın May, strong and weak kids do not differ (FG = 5, t = 1.49, not significant). Whereas the gonads of weak kids decrease between May and July significantly by about 40 % (FG = 4, t = 3.16, p <0.05), the gonads of strong kids increase during the rut by 40% (FG = 5, t= 1.23, not significant). The relative deviations (positive) of uterus weights (g) and body weights (kg) of strong kids and those of weak kids are comparable in May, whereas ın July the relative deviation of uterus weights exceeds by far the deviation in body weights (Fig. 3). Sexnal maturation in male kids from the Wanaka region The development of testis weights of male kids is consistent with the previous findings ın females. The growth acceleration between the first period of decreasing daylength between May and July is distinct (Fig. 4). Whereas body weight increases only slightly (12 %), the 120 J. J. Bauer jr Bodyweight RR BE Bodyweight [ EEE | > Testisweight 500 se Testisweight er) en! Du) > ce c 2 7m 1) > w = > ® = =) [®) 100 May July May July Fig. 5. Mean body weight (kg) and mean testis Fıg. 4. Relative increase (%) ın body weights and weight (g) of strong (Z 15.5 kg) chamois kids of testis weights of male kids of the Wanaka region, the Wanaka area N. Z. in May and July, expressed N. Z. between the 15.-20. May and 10.-25. July as percentual deviation from mean body weights (NEE) (testis weights) of weak kids (= 13.5 kg). (N = 14) increase in testis weight is very high (544 %). Here, a differentiation of gonad development in contrast to body weight growth is evident, too. Differences in growth acceleration of the testis between weak and strong kids indicate, that stimulation of testis development ıs dependent on a crıtical body weight. Fig. 5 shows the relative deviation of body and testis weights in May and July of strong kids from the respective weights in weak kids. Sexnal maturation in juvenile females from the Harper-Avoca region In female kids of the Harper-Avoca region there is no evidence of accelerated uterus development or follicle maturation at the age of 12-15 months. However 43 % (N = 7) of 24-27 months old females had given birth and must therefore have conceived as yearlıngs. The weights of nine 22-36 month old females averaged only 13.3 + 3.1 kg. Sexnal maturation in juvenile males from the Harper-Avoca region Males show also very low weights as kids and yearlings. Four 12-15 month old males weighed on the average 11.6 + 1.7 kg. The largest yearlings shot had testis weights of 2.3 + 0.43 g (N = 3). The mean testis weight of four male kids weighing 9 kg was 1.34 g. Accelerated testis development between November and February was found in all age classes but was highest in yearlings (Tab. 3). Whereas testis weights of yearlings increased significantly (+ 414 %) in this time period (FG = 5, t = 2.99, p < 0.05), increase in other anımals is less pronounced. Testis weight of yearlings in November differed highly Sexual maturıty in New Zealand chamoıs 121 Table 3 Mean average testes weights of different chamois age classes from the Harper-Avoca region in November and May (N = 17) November February Increase (%) Mean + SD testis weight Mean + SD testis weight (8) 10.8 4.10 16.8 4.63 20.3 SD significantly from respective weights of adult males (FG = 6, t = 8.14, p <0.001). In February however, the testis weights of yearlings did not differ significantly from the testis weights of older anımals. Discussion The differences in uterus weights between sexually immature kids (mean uterus weight 3.13 # 1.06 g, N = 18), one ovulating (uterus weight: 18 g) and one pregnant kid (uterus weight: 50 g, at gestation period of about 30 days) are due to the role of hormones activated during sexual maturation. Uterus growth, stimulated by estrogen and to a smaller extent progesterone, is due to the highly increased production of theca cells in the ovaries. This can lead to pronounced uterus growth during the course of a few days (HEaAP 1972). In the case of female chamoıis kids of the Wanaka region this suggests that sexual maturity and successive pregnancy is not only due to overproportionally developed uteri of few anımals. Other animals with comparable body weights might have reached sexual maturity as well, possibly only several days later. This possibility is further suggested by the incidence of mature graafıan follicles in the ovaries of most strong kids. Similarly, further evidence for weight and daylıght dependant sexual maturation is found in male kids. The heaviest male kid which was shot, weighed 19.7 kg and was shot in May. Although its mean testis weight of 2.7 gexceeded the mean testis weight of weaker animals (mean body weight: 12.15 + 1.2 kg; N = 4), shot at the same time (mean testis weight: 1.3 + 0.55 g; N = 4), the difference is not great, which might suggest, that body weight itself could not stimulate testis growth, but that another stimulus (daylength?!) was necessary. No evidence of sexual maturation of male and female kids (11-13 months) of the Harper-Avoca region was found. This is not surprising as kids gain weight during their first year of life continuously (SCHRÖDER 1971) and it is highly improbable that any of these anımals reached a body weight of more than 12 kg by the age of 6-8 months (rut). As expected, testis weights of 12 month old anımals from the Harper-Avoca region do not show any differences to testis weights of Wanaka kids before puberty. This critical weight seems to be equally important for sexual maturation in 1% year old males. As ın the Wanaka region all 18 months old males are well above this limit (21.4 kg, N = 11), weight as critical factor for the development of testis is not crucial. In Harper-Avoca animals, however, variation around this critical weight indicates further delay of sexual maturation in some animals. Whereas yearlings within the range of the critical weight show strong acceleration of testis growth (mean testis weight = 14.6 8), one weak animal, shot in February showed much less developed testes (6.4 g). Although chamois of the Harper-Avoca region do not reach 15 kg at their second period of sensitivity to diminishing daylight conditions, they do become sexually mature. This is less surprising, if adult body size is being considered. Whereas adult males and females of the Wanaka region weigh on the average 24-29 kg, Harper-Avoca anımals weigh 122 J. J: Bauer between 13-20 kg (BAuEr 1982a). If body size at the time when all (Wanaka) or two thirds (Harper-Avoca) of the anımals reach sexual maturity (18 months) is expressed as the fraction of body weight in comparison to adult anımals, values of 0.90 for Wanaka animals and 0.74 for Harper-Avoca anımals can be determined. SADLEIR (1969) assumes that four main factors influence or determine sexual maturation in mammals. These are nutrition, temperature, population density and light regime. Tempera- ture, nutrition and population density (the latter does not only increase intraspecific competition for food, but also causes stress induced endocrine changes which influence puberty; CHıiTTy 1962; CHRISTIAN 1956) havıng effect upon puberty mainly by changes ın growth rate (FLux 1967; AMıR et al. 1968; JoUBERT 1954; 1963; HamMmonD 1949; Hansson 1956; Davıs 1951; MoUSTGAARD 1959). Experimental research on domesticated species of ungulates shows, that although final adult size seems to be genetically fixed (Garz 1969, Hansson 1956), rate of growth (Hansson 1956) and sexual maturity (REıD 1960) are determined by the plane of nutrition. Whereas in cattle (Hansson 1956, JOUBERT 1954, 1963), as well as in hare (FLux 1967) or in rats (Davis 1951, 1953); WıDDowson 1968) a strong correlation between nutrition, growth rate and body sıze on the one hand, and sexual maturity on the other hand, was observed, similar experiments in sheep show somewhat inconclusive results (HamMonD 1949), although THomson and AITkEn (1959) found cues, that food supply during winter influenced fertility of female sheep. High plane of nutrition increases the relative number of breeding females at 1% years of age, as has been shown in thar (CAuGHLEY 1970), chamois (BAUER 1982, 1983, 1985), red deer (ALgon et al. 1983; MooRE 1982), white- tailed deer (CHEATUM and MoRTON 1946) or roe deer (ELLENBERG 1978). A protein as well as phosphorous deficient diet ıs also known to delay sexual maturity (MOUSTGAARD 1959; NEHRING 1972). The influence of the light regime on the time of ovulatıon in mammals is described by several authors (HamMonD ]J. 1944; CAUGHLEY 1971; HART 1950; Marshall 1937) and seems to be of major importance to the onset of puberty, at least in sheep (HAmMoND ]. 1944; Watson and GAMBLE 1961; SADLER 1969) and in Whitetailed deer (Buppe 1983). Analysıs of chamois, presumably living under optimal conditions in the Wanaka area, New Zealand, show, that the physiological minimum of sexual maturity of this species ıs six months. This seems to apply for both sexes. The onset of sexual maturity being determined by critical body weight and daylength values, both acting as synergistic double stımulı. Daylength, being a fixed parameter acts hereby as primary stimulus, which is dependent on.acritical body weight for actıvation. Should the animal fail to reach this critical weight, before this sensitive phase passes, sexual maturity is delayed for at least one year. Results from the Wanaka and Harper-Avoca regions further suggest, that this critical body weight, required to reach sexual maturity, differs between populations. Whereas it could be seen that sexual maturation is physiologically possible at a very early age, there are few studies which allow estimation of actual reproductive success of age classes, which have not reached their final body sıze and are socially immature. Reproduc- tive success of male immature anımals ıs mainly affected by dominance relationships to other males. In the absence of older males or within populations of imbalanced sex ratio however, even yearlıng males of 2% year old males can participate substantially at reproduction as has been shown in chamois (BAUER, in prep.) and roe deer (BAUER etal. in prep.). Reduced reproductive success in very young females will be due to physiological as well as behavioural and social factors. The size of the birth channel will not be optimal in not fully grown females, leading to birth complications as has been suggested in mountain sheep (GEısT 1971) and observed in chamois yearlıng females (BAUER in prep.). Appropri- ate behavioural responses to the newborn offspring will affect survival of the young. This Sexnal maturity ın New Zealand chamoıs 123 has been shown in roe deer (Kurt 1968; ELLENBERG 1978) and red deer (CLuTTON-BRock et al. 1982). Actual contribution of females reproducing for the first time, to the rate of increase of a population will therefore be substantially lower as expected. Acknowledgements This study benefited greatly from a range of institutions and people. It was made possible by a scholarship from the German Academic Exchange Service, as well as the University of Freiburg, Department of Forest Zoology, University of Otago, Dunedin, New Zealand, Department of Zoology, University of Canterbury, New Zealand, Forest Research Institute, Christchurch, New Zealand. I wish to express my special thanks to the following people; Prof. Dr. G. OscHe, Zoologisches Institut, Freiburg, Prof. Dr. H. Gossow, Vienna, Dr. K. WEsTERsKov, Dunedin, and Prof. Dr. W. CLArk, Christchurch, New Zealand. For the recovery of animals, I wish to express my thanks to Arran Duncan, Alpine Helicopters and his family. My special thanks go to my friend Ray HENDERSON, Forest Research Institute, for hıs invaluable field assıstance. Last but not least I wısh to thank GABRIELE SANDMANN, Bundesforschungsanstalt für Landschaftsökologie und Naturschutz, Abt. Wild- und Jagdökologie, for typing the manuscript and drafting the figures. Zusammenfassung Geschlechtsreifebestimmende Faktoren bei neuseeländischen Gemsen (Rupicapra rupicapra L.) Im Verlauf einer Studie zur Dynamik neuseeländischer Gemsenpopulationen wurden die Gonaden von 27 männlichen und 56 weiblichen subadulten (weniger als zwei Jahre) Tieren auf Geschlechtsreife bzw. Fortpflanzung untersucht. Innerhalb beider Vergleichspopulationen wurde ein hoher Anteil geschlechtsreifer Tiere innerhalb dieser Altersgruppe festgestellt. Dieses Phänomen scheint bei europäischen Gemsen nahezu ausschließlich bei Populationen in Mittelgebirgen nach Einbürgerungen aufzutreten. Weiterhin jedoch wurde in Neuseeland bei Gemsen, die ein neues Areal kolonisierten, Geschlechtsreife (Follikelreifung, Gelbkörperentwicklung) bei sieben Monate alten weiblichen Kitzen festgestellt. Nicht nur Geschlechtsreife, sondern auch tatsächliche Fortpflanzung (Embryo) konnte weiterhin bei einer Kitzgeiß nachgewiesen werden. Es konnte gezeigt werden, daß innerhalb der Kitzaltersklasse sexuelle Reifung in beiden Geschlechtern sowohl durch eın kritisches Körpergewicht als auch durch bestimmte Tageslängenwerte eingeleitet wird. Hierbei wirken beide als synergistische Stimulantien. Tageslänge, als unabhängiger Parameter, wirkt hierbei als Primärauslöser, der jedoch der Erreichung eines kritischen Körpergewichts zur Aktivierung bedarf. Erreicht ein Kitz dieses kritische Gewicht nicht, bevor die sensitive Lichtphase vorüber ist, wird der Eintritt der Geschlechtsreife wenigstens für ein Jahr verschoben. Der Vergleich zweier Regionen mit unterschiedlichen Ernäh- rungsgrundlagen, Wachstumsraten und Körpergröfßsen der adulten Tiere zeigte, daß dieses Schwellen- gewicht keinen absoluten artspezifischen Wert darstellt. Vielmehr scheint er mit den Wuchsverhältnis- sen jeweiliger Populationen ın Zusammenhang zu stehen. Schlußfolgerungen dieser Ergebnisse für die Sozıiobiologie der Gemsen werden kurz besprochen. Literature Aıson, $. D.; MITCHELL, B.; Staıngs, B. W. (1983): Fertility and body weight in female red deer: A density-dependent relationship. J. Anım. Ecol. 52, 969-980. AMIR, S.; Karı, J.; VoLcanı, R. (1968): Influence of growth rate on reproduction and lactation in dairy cattle. In: Growth and development of mammals. Ed. by. G. A. Lopge, and G. E. Lamminc. London: Butterworths. 234-263. BAUER, J. J., 1977: Population dynamics and annual growth rate of horns in three populations of chamois (Rupicapra rupicapra L.) at Lake Wanaka, New Zealand. Thesis, University of Otago, Dunedin, New Zealand. | — (1982): Untersuchungen zur Dynamik stabiler und kolonisierender Gemsenpopulationen (Rupicapra rupicapra L.) in Neeseeland. PhD Thesis, Fac., Albert-Ludwig-Univ. Freiburg. — (1983a): Selection for low age of maturity in New Zealand chamois (Rupicapra rupicapra |.) in New Zealand. 15. Pacific Science Congress, Dunedin, New Zealand. Absır. ar — (1983b): Sexual maturity and fecundity of stable and colonizing populations of chamois in New Zealand. 4. Int. Sym. on Chamois and other mountain Ungulates, Pescasseroli, Italy. — (1985): Fecundity patterns of European and New Zealand chamois populations. In: The Biology and Management of Mountain Ungulates. Ed. by S. Lovarı. Beckenham, Kent: Croom-Helm. — Populationsökologie, Verhalten und Bewirtschaftung der Schwarzwaldgemse. Forschungsergeb- nisse einer zweijährigen Studie mit Mitteln des LJV Bad.-Württbg. (in prep.). 124 J. J. Bauer BAUER, J. J.; Gossow, H. (1978): Zur Populationsdynamik der Gemse in Neuseeland. Proc. 3. Int. Gamswild Symp., Mayrhofen, Tirol, Österreich. 63-75. — — (1983): On diseases as a mayor regulatory factor in the population dynamics of New Zealand chamois (Rupicapra rupicapra L.). 15. Int. Congr., Game Biol., CSSR, Abst. BAUER, J. J.; PFLIEGER, R.: Populationsstatus der Vogesengemse — Modellsimulation überbejagter Gemsenbestände (in prep.). BAUER, ]. J.; RIEGER, F.; STROHHÄCKER, U.: Age structure and reproductive effort/success ın male roe deer (in prep.). Buppe, W. (1983): Effects of photoperiod on puberty attainment of female white-tailed deer. J. Wildl. Memt. 47, 595-604. CAUGHLEY, G. (1970a): Eruption of ungulate populations, with emphasis on Himalayan Thar in New Zealand. Ecology 51, 53-72. — (1970b): Population statistics of chamoıs. Mammalia 34, 194-199. — (1970e): Liberation, dispersal and distribution of Hımalayan Thar (Hemitragus jemlahicus) in New Zealand. N. Z. J. Scı. 13, 220-239. — (1971a): The season of births of northern hemisphere ungulates in New Zealand. Mammalıa 36, 204219. — (1971b): Demography, fat reserves and body size of a population of red deer (Cervus elaphus) in New Zealand. Mammalıa 35, 367-383. CHAPMAN, D. J.; HORwWooD, M. T. (1968): Pregnancy in a Sıka deer calf (Cervus nıppon). J. Zool., London 155, 227-228. CHEATUM, E. L. (1942): On the occurrence of pregnancy in white-tailed deer fawns. J. Mammalogy 23, 210-211. CHEATUM, E. L.; MORTOoN, G.H. (1946): Breeding season of white-tailed deer in New York. J. Wild. Memt. 10, 249-263. CHıtty, D. (1962): Body weight in relation to population phase in Microtus agrestis. Symp. Theriol. Brno, 1962 (Proc. Internat. Symp. Methods of Mammalogical Investigation) 77-86. CHRISTIAN, J. J. (1956): Adrenal and reproductive responses to population size in mice from freely growing populations. Ecology 37, 258-73. CLUTTON-BROcK, T. H.; GuInNneEss, F. E.; ALBon, $. D. (1982): Red Deer. Behavior and ecology of two sexes. Chicago: Chicago Press. Cots, L. C. (1954): The population consequences of life - history phenomena Quart. Rev. Biol. 29, 103-137. COUTOURIER, M. A. J. (1938): Le chamois. Grenoble: Arthond. Cowan, J. McT. (1956): What and where are the mule and black-tailed deer. Ed. by. W. P. TAyLor. In: The deer of North America. Harrisburg, Penns.: Stackpole. 335-359. Davıs, D. E. (1951): A comparison of reproductive potential of two rat populations. Ecology 32, 469-475. Davıs, D. E. (1953): The characteristics of rat populations. Quart. Rev. Biol. 28, 373401. ELLENBERG, H. (1978): Zur Populationsökologie des Rehes (Capreolus capreolus L., Cervidae) ın Mitteleuropa. Spixiana, Supal' 2, 1-210. Frux, J. E. L. (1967): Reproduction and body weights of the hare (Lepus europaeus Pallas) in New Zealand. N. Z. J. Sci. 10, 357401. GaLL, G. A. E. (1969): Gentics of growth. In: Animal growth and nutrition. Ed. by E. S. E. Harez and I. A. Dyer. Philadelphia: Lea and Fabiger, 63-82. GEIST, V. (1971): Mountain sheep —- a study in behaviour and evolution. Chicago, London: Univ. Chicago Press. Harez, E. S. E. (1952): Studies on the breeding season and reproduction of the ewe. Part III. The breeding season and artificial light. J. Agrıic. Sci. Cambridge 42, 232-265. Hammono, J. (1944): On the breeding season in the sheep. J. Agrıc. Sci. Cambridge 34, 97-105. HamMmono, J. (1949): Physiology of reproduction in relation to nutrition. Brit. J. Nutrit. 3, 79-83. Hansson, A. (1956): Influence of rearing intensity on body development and milk production. Proc. Br. Soc. Anım. Prod., 51-56. HART, D. $. (1950): Photoperiodicity in Suffolk sheep. J. Agric. Sci. Cambridge 40, 143-149. Heap, R. B. (1972): Role of hormones in pregnancy. In: Hormones in reproduction. Cambridge: Ed. by C. R. Austin, and R. V. SHORT. Cambridge Univ. Press. 73-106. HEPTNER, V.; Nasımovic, A.; Bannıcov, A. (1966): Die Säugetiere der Sowjetunion. Vol 1. Paarhufer und Unpaarhufer. Jena: G. Fischer. JEwELL, P. A.; Gruss, P. (1974): The breeding cycle, the onset of estrus and conception in Soay sheep. In: Island survivors, the Soay sheep of St. Kilda. Ed. by P. JEewELL, C. MILNER, and J. Bovp. London: Athlone Press. 224-241. JousErT, D. M. (1954): The influence of winter nutritional depressions on the growth, reproduction and production of cattle. J. Agrıc. Sci. Cambridge 44, 5-66. JoOUBERT, D. M. (1963): Puberty ın farm anımals. Anım. Breed. Abstr. 31, 295-306. Sexual maturity in New Zealand chamoıs 125 Kıppız, D. G. (1972): The Sika Deer (Cervus nıppon) ın New Zealand. New Zealand Forest Service, Wellington, N. Z. Kıeın, P. R. (1968): The introduction, increase and crash of reindeer on St. Matthew Island. J. Wildl. Mgmt. 32, 350-367. Knaus, W. (1960): Das Gamswild. Hamburg und Berlin: Paul Parey. KrÄMER, A. (1969): Soziale Organısation und Sozialverhalten einer Gamspopulation (Rupicapra rupicapra L.) der Alpen. Z. Tierpsych. 26, 889-964. Kurr, F. (1968): Das Sozialverhalten des Rehes. Mammalıa depicta, Beihefte z. Z. Säugetierkunde. Hamburg und Berlin: Paul Parey. LAMBERT, R. E.; BATHGATE, ]J. L. (1977): Determination of the plane of nutrition of chamoıs. Proc. N. Z. Ecol. Soc. 24, 48-56. Lewonrtin, R. C. (1965): Selection for colonizing ability. In: The genetics of colonızing species. Ed. by H. G. Baker and G. L. STresBins, New York and London: Academic Press. MARSHALL, F. H. A. (1937): On the change over in the estrus cycle ın anımals after transference across the equator with further observations on the incidence of the breeding seasons and the factors, controlling sexual periodicity. Proc. Roy. Soc. London (B) 122, 413—428. McKean, W. T. (1954): Preliminary observations of ovulation incidence in North Dakota deer and antelope. Project Rept. North Dakota State Game and Fish Dept., Federal Aid Project. W-7-R-9 to 13. MITCHELL, G. ]J. (1967): Minimum breeding age of female pronghorn antelope. J. Mammology 48, 4EIA4R. Me G.H. (1982): Deer, Red Deer, Calving and Weaning. Farm Prod. Pract. N.Z., Agric. and Fish. MoRrTONn, G. H.; CHEATUM, E.L. (1946): Regional differences in breeding potential of white-tailed deer in New York. J. Wildl. Mgmt. 10, 242-248. MOUSTGAARD, J. (1959): Nutrition and reproduction in domestic anımals. Ed. by H.H. Core and P. T. Cupps. Reproduction in domestic anımals. New York: Academic Press. NEHRInNG, K. (1972): Lehrbuch der Tierernährung und Futtermittelkunde. Radebeul: Neumann Verlag. NIEVERGELT, B. (1966): Der Alpensteinbock. Mammalıa depicta, Beihefte z. Z. Säugetierkunde. Hamburg und Berlin: Paul Parey. Runge, M. (1969): Reproduction ın feral goats near Wellington, N. Z. N. Z. J. Scı. 12, 817-827. SADLEIR, R. M. F. (1969): The ecology of reproduction in wild and domestic mammals. London: Methuen. SALZMANN, H. C. (1977): Gewicht, Schädelgröße und Gehörnwachstum von Gemsen aus dem Jura und ein Vergleich mit anderen Populationen. Z. Jagdwiss. 23, 69-80. Author’s address: Dr. J. J. BAUER, UNDP, P. ©. Box 107, Kathmandu, Nepal NZESSIEINISSOHPVEILETOFTEZKTURZMITTEILUNG Hair brush organs in Viscachas, Lagostomus maximus By I. RıEGER and Dorıs WALZTHÖNY Receipt of Ms. 27.7. 1986 Viscachas are rare zoo animals and not of great economic importance. This is probably one of the main reasons why the amount of scientific data of the species is rather small. Only four zoos in the world are breeding this species (OLNEY 1982). Recently, the Zurich Zoo viscacha females started to abandon their offspring. Therefore, the head keeper of the Small Mammals Section, Mr. HEınz KOHLER, was obliged to hand-raise them. Viscachas of all ages have a rather unexpected and striking morphology of their hind feet: the dorso-medial fleshy part of the longest, i.e. second toe is, compared to the U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5202-0125 $ 02.50/0 Z. Säugetierkunde 52 (1987) 125-127 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Fig. 1. Habitus of a young viscacha, three weeks of age. Note the remarkably enlarged distal part of the second hind toe. Fig. 2. View of a viscacha’s hind foot, three weeks of age. First and second toes are clearly visible. The upper part (more proximal region on the dorsal “o of the second toe shows normal hair cover, whereas the lower part (more disto-medial region on the dorsal side) is covered with thick bristle-like hair thus forming a brush. Fig. 3. Hind foot of a three weeks old viscacha with widely spread toes. Both main brush-like organ on the second toe and smaller brush-like organ on the first toe are visible Hair brush organs in Viscachas, Lagostomus maximus 12% corresponding parts of the neighbouring toes, bulb-like enlarged (fig. 1). Macroscopical investigation shows that the upper part of this enlarged area (1.e. more proximal region on the dorsal side of the toe) is covered with fur not differing from the one on the neighbouring toes (fig. 2), whereas the lower part of the area in question (i.e. more disto- medial region on the dorsal side of the toe) shows a completely different structure: the hairs are shorter, thicker, and stiffy. The distance between one haır to another is larger than between hairs in “normal” fur. Thus, the hairs of the lower part of the enlarged area (i.e. the distomedial part which is nearest to the claw) on the second hind toe of viscachas can be described as bristles, and the whole area in question is a brush-like organ. A similar, but smaller brush-like organ is found on the first toe of the hind foot (fig. 3). The scratching movement of a viscacha’s hind leg seems to be rather non-effective because the claws do not scratch the fur as it is the case in grooming behaviour of most mammals. On the contrary, the observer gets the impression that the claws touch neither fur nor skin, thus the claw movement is running idle. Most probably, during this scratching movement by the hind leg, the hair brush organ on the dorso-medial part of the tips of both second and first toes are touching and therefore grooming the fur. Thus the real grooming organ is not the claw itself but rather this brush- like organ described above. The evolution of such a specialized grooming organ indicates the importance of grooming behaviour in ground-dwelling mammals. Analogous organs are found in other species of the same ecological niche, e.g. hyraxes, Procavia spp., with a “double” claw on the second toes of the hind feet (Hann 1959), and Lemurs, Lemur spp. with comb like lower jaw incisors and canınes (AnkEL 1970). In the last century, Hupson (1872) mentioned this brush-like organ in a footnote. Later, the knowledge of this very specialized structure was forgotten. Neither SEELE (1968) in his monographic overview on skin and hair of Chinchillidae nor Dusost (1970) in his comparison of the grooming claws of several mammalian species did refer to this unique hair-brush organ in viscachas. Acknowledgements We like to express our sincere thanks to the staff of Zurich Zoo and especially to Mr. HEınz KOHLER, head keeper of the Small Mammals Section of Zurich Zoo, who gave us the opportunity to take care for these interesting youngsters. Mr. DIETER HARTMANN, St. Gallen, kindly helped us with viscacha references. References Anke, F. (1970): Einführung in die Primatenkunde. Stuttgart: G. Fischer. Dusost, G. (1970): Die Umwandlung von Hinterfußkrallen zu Putzorganen beı Säugetieren. Z. Säugetierkunde 35, 56-60. Harn, H. (1959): Von Baum-, Busch- und Klippschliefern. Wittenberg-Lutherstadt: A. Ziemsen Verlag. in H. (1872): On the Habits of the Vizcacha (Lagostomus trichodactylus). Proc. Zool. Soc. London 1872, 822-833. OLney, P. ]J. S. (ed.) (1982): International Zoo Yearbook. Vol. 22. Zool. Soc. London. SEELE, E. (1968): Haut und Haar der Chinchillidae. Zool. Anz. 181, 60-75. Authors’ address: Dr. Ingo RıEgEr and Dr. Dorıs WALZTHöny, Zoologischer Garten, Zürichberg- straße 221, CH-8044 Zürich, Switzerland BEKANNTMACHUNG Die Internationale Kommission für Zoologische Nomenklatur teilt mit: The following Opinions have been published by the International Commission on Zoological Nomenclature, vol. 43, part 2 on 9 July, 1986: Opiniıon No. 1393 (p. 142) Choeropsis Leidy, 1852 (Mammalıa, Artiodactyla): conserved vol. 43, part 3 on 6 October, 1986: Opıinion No. 1400 (p. 229) Simia fascicularıs Raffles, 1821 (Mammalıa, Primates): conserved 1409 (p. 247) Adianthus bucatus Ameghino, 1891 (Mammalia): neotype designated. 1413 (p. 256) Delphinus truncatus Montagu, 1821 (Mammalıia, Cetacea): conserved. BÜCTEBESBREETMING CHAanIn, P.: The natural history of otters. London, Sidney: Croom Helm 1985. 179 pp., 17 colored pl., numerous figs. and tables. £ 7.95. ISBN 0-7099-3460-2 Das vorliegende Büchlein über Otter von PauL CHanin ist der erste Band in einer geplanten Reihe über einige Säugetiere, die von ERNEST NEAL herausgegeben wird. In dieser Reihe sollen neuere Erkenntnisse über die Biologie der vorgesehenen Arten zusammengefaßt dargestellt werden. Dieser Band über die Lutrinae ist ın 7 Kapıtel gegliedert. Zunächst werden Systematik, Verbreitung, Artenkennzeichen und besondere morphologische und physiologische Merkmale beschrieben, dann Nahrungszusammensetzung und Ernährung. Es folgen Ausführungen über Beziehungen zu anderen Tieren (Konkurrenz) und soziale Organisation, Territoriengröße, Ortsveränderung, räumliche Vertei- lung und Formen der Kommunikation. Im Kapitel über Lebensgeschichte werden Reproduktion und Jugendentwicklung behandelt sowie Lebenserwartung und Mortalitätsursachen. Schließlich ıst ein letzter Abschnitt den Beziehungen zwischen Otter und Mensch gewidmet. Hier wird auf die Geschichte der Otterjagd eingegangen, u.a. ausführlicher auf die Situation in Großbritannien, aber auch auf Schutzbemühungen. Bei den Beschreibungen stehen die Erkenntnisse an Lutra lutra, L. canadensis und Enhydra lutrıs im Vordergrund, weil am besten untersucht. Neuere Studien an den übrigen Arten sind zusätzlich in den Text eingefügt. Einige Farbbilder und Strichzeichnungen illustrieren die Ausführungen, zahlrei- che Grafiken und Tabellen tragen zum besseren Verständnis bei. Der Text ist nicht immer frei von Wiederholungen, jedoch für Fachleute und Laien in gleicher Weise ansprechend. Am Ende jedes Kapitels ist eine sinnvolle kurze Zusammenfassung angefügt. Insgesamt liegt eine anregende Doku- mentation unserer Kenntnis über die Otter der Welt vor, die alle Säugetierfreunde interessieren wird. Es ist ein schöner und gelungener Anfang der geplanten Reihe. D. Kruska, Kiel Deutsche Gesellschaft für Säugetierkunde 60. Hauptversammlung in Stuttgart, 28.September bis 2. Oktober 1986 Kurzfassungen der Vorträge und Posterdemonstrationen. Herausgegeben von Fritz Dieterlen, Stuttgart. Zusammenstellung: Pia Wilhelm, Stuttgart. 1986. 99 Seiten. Kartoniert 24,— DM Ein Hauptziel der Deutschen Gesell- schaft für Säugetierkunde ist, auf ihren Jahrestagungen über Säugetiere arbeitende Wissenschaftler verschie- denster Fachrichtungen zusammen- zuführen, den Gedanken- und Erfah- rungsaustausch anzuregen, um so Erkenntnisse aus den einzelnen Forschungsgebieten zu integrieren. Da auf einer Jahresversammlung nicht alle Fachrichtungen gleichermaßen vertreten sein können, wurden - wie schon in den Vorjahren — wieder Schwerpunkte gesetzt: Erstens, die Paläontologie und, damit verbunden, Fragen der Systematik und Evolution der Säugetiere. Zweitens, ein Themenkreis, der vor- wiegend physiologische und etholo- gische Probleme bei Säugetieren um- faßt, und drittens, ökologische Themen, besonders zu Fragen des Artenschutzes und der Wildbiologie. Sehr erfreulich ist, daß eine größere An- zahl von Beiträgen den Fledermäusen gewidmet ist- der bei uns am stärksten bedrohten Säugetierordnung. Zu beziehen durch jede Buchhandlun Verlagsbuchhandlung BAJL Paul Parey Spitalerstraße 12 2000 Hamburg 1 P/REYV Erscheinungsweise und Bezugspreis 1987: 6 Hefte bilden einen Band. Jahresabonnement Inland: 298,— DM zuzüglich 13,80 DM Versandkosten, Gesamtpreis 311,80 DM einschließlich 7% Mehrwert- steuer. Jahresabonnement Ausland: 298,- DM zuzüglich 24,- DM Versandkosten. Das Abonne- ment wird zum Jahresanfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, Spitalerstraße 12, D- 2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesellschaft für Säugetier- kunde“ erhalten die Zeitschrift unberechnet im Rahmen des Mitgliedsbeitrages. | Z. Säugetierkunde 52 (1987) 2, 65-128 | Winfried Ahne Grundriß der Zoologie für Tiermediziner Von Prof. Dr. rer. nat., Dr. med. vet. habil. Winfried Ahne, Institut für Zoologie und Hydrobiologie, Tierärztliche Fakultät der Lud- wig-Maximilians-Universität München. 1986. 308 Seiten mit 133 Abbildungen und 11 Tabel- len Kartoniert DM 39,-. Dieser Studientext basiert auf der langjäh- rigen Lehrtätigkeit des Verfassers im Fach Zoologie für die Studierenden der Tiermedi- zin an der Ludwig-Maximilians-Universität in München. Die Stoffauswahl richtet sich weitgehend nach den Bestimmungen der Approbationsordnung für Tierärzte und ent- hält im wesentlichen den in der Studienord- nıng enthaltenen Lehrzielkatalog für das Fach Zoologie. Neben dem systematischen Überblick über die Tierstämme werden ins- bssondere Grundkenntnisse der Genetik, der Fortpflanzung und der Entwicklung, der Zytologie und der Histologie sowie der Evo- lution, der Abstammungslehre und der Ökologie vermittelt. Ausgehend von der rein zoologischen The- matik wird auch das vielseitige Aufgaben- gebiet des Tierarztes berücksichtigt. Dazu werden sowohl die Tiergruppen ausführlich im Kapitel „Spezielle Zoologie“ beschrie- ben, die in der tierärztlichen Tätigkeit eine Rolle spielen, wie etwa die klassischen Haus- tiere, als auch neue, tierartbezogene Betäti- gungsfelder angesprochen, wie z.B. die tier- ärztliche Betreuung von Aquakulturen (Weichtiere, Krebse, Fische) und Bienen- bestände oder die Behandlung von wechsel- warmen Wirbeltieren (Fischen, Amphibien, Reptilien), die spezielle zoologische Kennt- nisse auch über exotische Tierarten voraus- setzen. In erster Linie als Vorlesungsbegleittext und Nachschlagewerk für die Studierenden der Tiermedizin geschaffen, vermittelt das Buch aber auch fächerübergreifende Grundlagen und bietet dem praktizierenden Tierarzt eine schnelle und übersichtliche Orientierungs- hilfe. Aus dem Inhalt: Allgemeine Zoologie Die Zelle - Das Gewebe - Genetik - Fortpflan- zurg - Entwicklung - Evolution - Ökologie - Zoologische Systematik. Spezielle Zoologie - Protozoa (Urtiere, Einzeller) - St. Sarcomasti- gophora - St. Labyrinthomorpha - St. Api- complexa - St. Microspora - St. Ascetospora - St. Myxozoa - St. Ciliophora. Metazoa (Gewebetiere) - St. Mesozoa - St. Schwämme (Porifera) - St. Nesseltiere (Cnidaria) - St. Acnidaria. Bilateria - Protostomia - St. Platt- würmer (Plathelminthes) - St. Schnur- würme (Nemertini) - St. Kelchwürmer (Kampotozoa) - St. Schlauchwürmer (Ne- mathelminthes) - St. Priapswürmer (Priapu- lida)) - St. Weichtiere (Mollusca) - St. Spritzwürmer (Sipunculida) - St. Igelwür- mer (Echiurida). Gliedertiere (Articulata) - St. Ringelwürmer (Annelida) - St. Stummel- füßer (Onychophora) - St. Bärtierchen (Tar- digrada) - St. Zungenwürmer (Linguatulida) - St. Gliederfüßer (Arthropoda) - St. Kranz- fühler (Tentaculata). Deuterostomia - St. Pfeilwürmer (Chaetognata) - St. Stachel- häuter (Echinodermata) - St. Kragentiere (Branchiotremata) - St. Bartwürmer (Pogo- nophora) - St. Chordatiere (Chordata) - Anhang - Literaturverzeichnis - Bildquel- lennachweis - Sachverzeichnis. Berlin und Hamburg PAUL P/REV Vol. 92 (5), 129-200, Juni 1957 100N 0044-9498 % SITSCHRI PEeE aah n 2 SAÄUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Organ der Deutschen Gesellschaft für Säugetierkunde Maddalena, T.; Mehmeti, A.-M.; Bronner, G.; Vogel, P.: The karyotype of Crocidura flavescens (Mammalia, Insectivora) in South Africa. — Der Karyotyp von Crocidura flavescens (Mammalia, Insectivora) in Südafrika 129 Grimmberger, E.; Hackethal, H.; Urbanczyk, Z.: Beitrag zum Paarungsverhalten der Wasserfledermaus, Myotis daubentoni (Kuhl, 1819), im Winterquartier. - Contribution to the mating behaviour ofthe Daubenton’s bat, Myotis daubentoni (Kuhl, 1819) in a winter quarter 133 Stein, Barbara R.: Phylogenetic relationships among four arvicolid genera. — Phylogenetische Beziehungen zwischen vier Arvicoliden-Gattungen 140 Maier, W.; Schrenk, F.: The hystricomorphy of the Bathyergidae, as determined from ontogenetic evidence. — Die Hystricomorphie der Bathyergidae, nachgewiesen durch ontogenetische Befunde 156 Schröder, W.; Mensah, G. A.: Reproductive biology of Thryonomys swinderianus (Temminck). — Zur Fortpflan- zungsbiologie von Thryonomys swinderianus (Temminck) m 164 Breitenmoser, U.; Haller, H.: Zur Nahrungsökologie des Luchses Lynx Iynx in den schweizerischen Nordalpen. — Feeding ecology of the Iynx (Lynx Iynx) in the Swiss Alps 168 Wissenschaftliche Kurzmitteilungen Van Waerebeek, K.; Reyes, J. C.; Luscombe, B. A.: A second record ofthe Pygmy sperm whale Kogia breviceps (de Blainville, 1838) (Cetacea, Physeteridae) from the Peruvian coast. — Der zweite Fund eines Zwergpottwals, Kogia breviceps (de Blainville, 1838) (Cetacea, Physeteridae), an der peruanischen Küste 192 Kock, D.: Tadarida teniotis (Rafinesque, 1814): Zweiter Nachweis für Marokko, w-paläarktische Arealgrenzen und taxonomische Anmerkung (Chiroptera: Molossidae). - Tadarida teniotis (Rafinesque, 1814): Second record for Marocco, w-paläarctic distribution limits and taxonomic note (Chiroptera: Molossidae) 194 Kratochvil, Z.: Oligodontie bei Hasenartigen (Lagomorpha). — Ologodonty in Lagomorpha 196 Gewalt, W.: Pferdeantilopen (Hippotragus equinus) äsen unter Wasser. - Roan antilopes (Hippotragus equinus) browse under water 198 Bekanntmachung 199 Buchbesprechungen 200 Verlag Paul Parey Hamburg und Berlin HERAUSGEBER/EDITORS P. J. H. van BREE, Amsterdam — W. FIEDLER, Wien -— H. Frick, München - W. HERRE, Kiel - H.-G. Kıös, Berlin - H.-J. Kuhn, Göttingen - B. Lanza, Florenz - T. C. S. MORRISON-SCOTT, London - J. NIETHAMMER, Bonn — H. REICHSTEIn, Kiel - M. Röhss, Hannover — D. STARcK, Frankfurt a. M. — F. Strauss, Bern -— E. THEnıus, Wien — W.VERHEYEN, Antwerpen SCHRIETLEITUNG/EBTTORTAT OERTEE 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 ın englischer Sprache Bd. 43, H. 2 beigefügt; in ıhnen 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 ass und dem Verlag erhältlich. Sonderdrucke: Anstelle einer Unkostenvergütung erhalten die Verfasser von Originalbeiträgen und Wissenschaftlichen Kurzmitteilungen 50 unberechnete Sonderdrucke. Mehrbedarf steht gegen Berechnung zur Verfügung, jedoch muß die Bestellung spätestens mit der Rücksendung der Korrekturfahnen erfolgen. Vorbehalt aller Rechte: Die ın dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei- ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch im Einzelfall grund- sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines Beitrages ist auch ım Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber- rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sınd mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen. Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of the 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 Mit einer Beilage des Verlages Paul Parey Fortsetzung 3. Umschlagseite © 1987 Paul Parey. Verlag: Paul Parey, Hamburg und Berlin. Anschriften: Spitalerstr. 12, D-2000 Hamburg 1; Lindenstr. 44-47, D-1000 Berlin 61. — Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., D-2240 Heide/Holst. The karyotype of Crocidura flavescens (Mammalia, Insectivora) in South Africa By T. MADDALENA, A.-M. MEHMETI, G. BRONNER and P. VOGEL Institut de Zoologie et d’Ecologie Animale, Universite de Lausanne, Switzerland, and University of Natal, Durban, South Africa Receipt of Ms. 18. 6. 1986 Abstract Karyological results give evidence that the South African Crocidura flavescens (Geoffroy, 1827) is not conspecific with the West and East African giant shrews of the taxa spurrelli, manni, kivu and olivieri. Introduction The African giant shrews are characterized by a high variability in body size. The smallest form, Crocidura flavescens (Geoffroy), weighing about 20 g, is found in South Africa. The biggest, Crocidura manni Peters of 50 to 100 g, lives in the drainage-basin of the Niger. The giant shrews have been the subject of many taxonomic studies describing local morphotypes, sometimes as good species, and sometimes as subspecies associated with a species of a neighbouring geographic region. The difficulty of delimiting and defining these taxonomic units on a morphological basıs and the absence of evident sympatric forms lead HEIM DE Barsac and BARLoy (1966) to consider all these taxa as conspecific. They regroup them under the oldest available name C. flavescens (Geoffroy, 1827). This interpretation has been followed by the authors of the most recent taxonomic syntheses, e.g. HEIM DE BaLsac and MEESTER (1977), CoRBET (1978), CORBET and Hırı (1980) and HUTTERER et al. (1982). However, a verification of this interpretation by another method than skull biometry is absolutely necessary. As a first approach, classic cytotaxonomy seems to provide such a control. In fact, the list of REUMER and MEYLAN (in press) shows that ın the genus Crocidura the karyotype often permits good discrimination between species. The karyotypes of four forms of the African giant shrew have been studied: kıvu Osgood, 1910 of Zaire by MEyLan (1967); spurrelli Thomas, 1910 of Ivory Coast by MEyLANn (1971) and Mevran and VoceL (1982); olivieri (Lesson, 1827) of Egypt by DE Honor (1974); and manni Peters, 1878 of Mali, Cameroun and Nigeria by MEYLAN and Vocer (1982). These forms share acommon karyotype of2 n = 50, NF = 66 and NFa = 62. This suggests that these four forms belong to the same species. However, it is now crucial to investigate the karyotype of the nominate form from South Africa; i.e., flavescens (Geoffroy, 1827), in order to test the interpretation of HEım DE Barsac and BARLoOY (1966). Material and methods We analysed 7 specimens of flavescens. They were captured in the vicinity of Durban (Table 1). The chromosomes were prepared following a modified spread procedure in the laboratory of Prof. J. U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5203-0129 $ 02.50/0 Z. Säugetierkunde 52 (1987) 129-132 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 130 T. Maddalena, A.-M. Mehmeti, G. Bronner and P. Vogel Table 1 Origin of the seven C. flavescens analysed The material is conserved at the Institut de Zoologie et d’Ecologie Anımale, Lausanne, Switzerland No collection o (q) ” Locality in South Africa IZEAL 2567 IZEAL 2568 IZEAL 2569 IZEAL 2346 IZEAL 2347 IZEAL 2348 IZEAL 2349 Campus University, Durban Stainbank Nature Reserve Stainbank Nature Reserve Pigeon Valley, Durban Pigeon Valley, Durban Pigeon Valley, Durban Pigeon Valley, Durban E M M E M FE M MEESTER in Durban. Slides were stained with Giemsa and analysed in Lausanne, where they could be compared with chromosome preparations of spurrelli of a laboratory colony, from Ivory Coast. Results The karyotype of the South African C. flavescens analysed (Fig. 1a, b) is 2n = 50. It ıs characterized by 4 small pairs of metacentric chromosomes, 7 pairs of submetacentric/ subtelocentric and 13 pairs of acrocentric chromosomes. The X is submetacentric, the Y is acrocentric, therefore NF = 74 and NFa = 70. In the West African spurrelli (Fig. 1c) the diploid number is also 2n = 50, but the karyotype is distinguished by 17 acrocentric pairs as described by MEyLAn and VOGEL (1982) resulting in NF = 66 and aNFa = 62. Discussion The results reveal that C. flavescens from South Africa is characterized by a distinct karyotype which differs from that shared by the taxa kivn, spurrelli, olivieri, and manni. The important morphological differences, which cannot be explained by Robertsonian processes, seem to exclude the possibility of intraspecific polymorphism. We conclude, therefore, that the South African species C. flavescens (Geoffroy, 1827) is not identical with the West and East African forms which should be regrouped under another specific name. MeEyLan and VoGeL (1982) tentatively regrouped the forms kivu, spurrelli, olivieri and manni under the name C. occidentalis (Pucheran, 1855) partially on the basis of the old interpretation of ALLen (1939). If ever East and West African giant shrews are really conspecific, the name Crocidura olivieri (Lesson, 1827) would have priority, all the more since CoRBET (1978) has designated a neotype in order to end the discussion on the validity of this name arising from the work of HEım DE Barsac and BarLoY (1966). However, a common karyotype does not definitively prove conspecific unity of these forms. Let us remember that C. giffardi de Winton, 1898, of West Africa has a karyotype which is identical with that of our four forms (MEyLAnD and VoGeL 1982), while HEIM DE Barsac (1970) regarded this taxon as a separate species, but conspecific with C. odorata (Leconte, 1857). In conclusion, other techniques are necessary to unravel the taxonomic status of West, East and Central African forms of the giant shrews. An electrophoretic study of allozymes may help to clarify the phyletic relationship between these forms. This technique has already proved valuable in an investigation of European shrews of the genus Crocidura Karyotype of Crocidura flavescens 131 Eu 1. 7 u 7 m a 1) BLEI Fig. 1. Karyotypes of Crocidura flavescens from South Africa and the form spurrelli from Ivory Coast. a: female no. IZEAL 2346 of Durban; b: male no. IZEAL 2349 of Durban, both with 2n = 50 and 13 acrocentric pairs; c: female no. IZEAL 2407 from Ivory Coast with 2n = 50 and 17 acrocentric pairs 132 T. Maddalena, A.-M. Mehmeti, G. Bronner and P. Vogel (CATZEFLIS et al. 1985; VOGEL et al. 1986). A preliminary application of these methods to African shrews has also revealed promising results (MADDALENA in preparation). Acknowledgement We thank, in particular, Prof. J. MEESTER for his kind collaboration. Zusammenfassung Der Karyotyp von Crocidura flavescens (Mammalıa, Insectivora) in Südafrika Der Karyotyp der südafrikanischen Riesenspitzmaus Crocidura flavescens (Geoffroy, 1827) unter- scheidet sich von jenem der west- und ostafrikanischen Riesenspitzmäuse der Taxa spurrelli, manni, kivu und olivieri. Letztere müssen deshalb einer anderen Art zugeordnet werden. References ALLEN, G. M. (1939): A Checklist of Afrıcan Mammals. Bull. Mus. comp. Zool. Harv. 83, 1-736. CATZEFLIS, F.; MADDALENA, T.; HELLWING, $.; VOGEL, P. (1985): Unexpected findings on the taxonomic status of East Mediterranean Crocidura russula auct. (Mammalia, Insectivora). Z. Säugetierkunde 50, 185-201. CoRBET, G. B. (1978): The Mammals of the Palearctic Region: a taxonomic review. British Museum (Natural History). London/Ithaca: Cornell University Press. CoRBET, G. B.; Hırr. J. E. (1980): A World List of Mammalıan Species. Brit. Mus. Nat. Hist., London. DE Honor, H. A. (1974): Karyological studies on two Insectivores of Egypt. Proc. Egypt. Acad. Sci. 25, 171-174. HEIM DE Barsac, H. (1970): L’unite specifique entre Crocıdura giffardi de Winton et C. goliath Th. se trouve demontree gräce aux plus recentes captures effectu&es au Cameroun. Bonn. zool. Beitr. 21, 83-88. HEIM DE Barsac, H.; BARLoY, ].-J. (1966): Revision des Crocidures du groupe flavescens-occiden- talis-manni. Mammalia 30, 601-633. “ HEIM DE Bausac, H.; MEESTER, J. (1977): Order Insectivora. In: The Mammals of Africa, an identification manual. Ed. by J. MEESTER and H. W. SETZER. Washington: Smithson. Inst. Press. pp- 1-29. HUTTERER, R.; JONES, G. $.; RossoLIMO, O. L.; VAN GELDER, R. G.; Wang, $. (1982): Family Soricidae. In: Mammal species of the world. Ed. by J. H. Honackı, K. E. Kınman, and J. W. Koeppr. Lawrence, USA: Allen Press and Ass. Systematics Collections. Meyran, A. (1967): La formule chromosomique de Crocidura occidentalis kivu Osgood (Mammalıa, Insectivora). Rev. suisse Zool. 74, 685-691. Meyran, A. (1971): Chromosomes de Soricides de Cöte d’Ivoire (Mammalıia, Insectivora) Rev. suisse Zool. 78, 603-613. MEyLAn, A.; VoceL, P. (1982): Contribution ä la cytotaxonomie de Soricides (Mammalia, Insecti- vora) de l’Afrique occidentale. Cytogenet. Cell Genet. 34, 83-92. REUMER, J. W.; MEyLan, A. (1987): New developments in vertebrate cytotaxonomy. Chromosome numbers in the order Insectivora (Mammalıa). Genetica (in press). VOGEL, P.; MADDALENA, T.; CATZEFLIS, F. (1986): A contribution to the taxonomy and ecology of shrews from Crete and Turkey (Crocidura zimmermanni and C. snaveolens) (Soricidae, Mammalia). Acta theriol. (in press). Authors’ addresses: Prof. Dr. P. VoGEL, A.-M. MEHMETI, T. MADDALENA, Institut de Zoologie et d’Ecologie Anımale, Bätiment de Biologie, CH-1015 Lausanne Switzerland; G. BRONNER, University of Natal, King George V Av., Durban 4001, South Africa Beitrag zum Paarungsverhalten der Wasserfledermaus, Myotis daubentoni (Kuhl, 1819), im Winterquartier Von E. GRIMMBERGER, H. HACKETHAL und Z. URBANCZYK Eingang des Ms. 11. 6. 1986 Abstract Contribution to the mating behaviour of the Danbenton’s bat, Myotis daubentoni (Kuhl, 1819) in a winter quarter A description of the mating behaviour of Myotis daubentoni in a overwintering site is given. In Ba the paper considers the microclimatic conditions, the males search of a mating partner, the ehaviour of both sexes during the copulation, their duration and the different state of torpidity and activity of d and ®. The results are discussed in connection with the relevant literature. Einleitung Die Paarungszeit der europäischen Fledermäuse beginnt mit der Auflösung der Wochen- stuben im Spätsommer. Bei vielen Arten erstreckt sie sich über den Aufenthalt im Winterquartier bis in das Frühjahr. Uterus und Vagina der ??2 befinden sich von Beginn der Paarungszeit an im Zustand des Proöstrus und verbleiben den Winter über in diesem. Bei den dd befinden sich den ganzen Winter über reife Spermien im Nebenhoden (Racey 1972; Kurzer 1981). Findet im Herbst oder Winter eine Begattung statt, dann verbleiben die Spermien im weiblichen Genitaltrakt in einem Ruhezustand. Erst im Frühjahr erfolgt die Ovulation und — nach Reaktivierung der Spermien — die Befruchtung der Eizelle. Lediglich bei der Landflügelfledermaus (Miniopterus schreibersi) wird auf einem anderen Weg das gleiche Ziel erreicht. Bei den dd von Miniopterus befinden sich während des Winters keine Spermien mehr in den Nebenhoden, Winterkopulationen kommen bei dieser Art nicht vor (Gustavson 1979). Gleich nach der Begattung im Herbst erfolgt auch die Befruchtung der Eizelle. Die Embryonalentwicklung wird dann aber auf einer frühen Entwicklungsstufe unterbrochen (Keimruhe) und setzt erst im Frühjahr wieder ein. Beide Fortpflanzungsstrategien stellen eine Anpassung der Fledermäuse an das Leben in der gemäßigten Klimazone dar. Eine ähnliche Fortpflanzungsstrategie weisen aber auch einige tropische Glattnasen-Fledermäuse auf, die auf diesem Weg erreichen, daß ihre Jungen zu Beginn der Regenzeit geboren werden (FENToN 1983). Paarungen im Winterquartier sind bei einer ganzen Reihe von Fledermausarten beob- achtet worden, so bei der Teichfledermaus (Myotis dasycneme), der Kleinen Bartfleder- maus (Myotis mystacinus), der Fransenfledermaus (Myotis nattereri), der Mopsfledermaus (Barbastella barbastellus), dem Braunen Langohr (Plecotus auritus), besonders häufig aber bei der Wasserfledermaus (Myotis daubentoni). ROER und EGsBaAzk haben das Paarungsverhalten von M. daubentoni ım Winterquartier bereits 1969 ausführlich beschrieben. Unsere Arbeit ergänzt diese Untersuchung und belegt Verhaltensweisen fotografisch. Weiterhin werden Vergleiche mit dem Paarungsver- halten anderer Säugetiertaxa angestellt. U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5203-0133 $ 02.50/0 Z. Säugetierkunde 52 (1987) 133-140 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 134 E. Grimmberger, H. Hackethal und Z. Urbanczyk ne RER $ & Shane BER ER SET ETEN Abb. 1. In 35 bis 50 m Tiefe unter der Erdoberfläche liegt hier das Fledermausreservat „Nietoperek“. Als Ein- und Ausflugmöglichkeiten können die Fledermäuse neben dem Haupteingangsstollen die wie hier auf den Feldern befindlichen zerstörten Panzertürme nutzen. (Foto: E. GRIMMBERGER) Ergebnisse Im Fledermausreservat Nietoperek (Woiwodschaft Gorzow) in Westpolen überwintern neben 11 anderen Fledermausarten jeweils auch mehrere 1000 Wasserfledermäuse. Bei diesem Quartier (Abb. 1) handelt es sich um ein System von 35-50 m unter der Erdober- fläche gelegenen und über 30 km langen Betontunneln und Kammern mit mehreren Ausgängen (BAGROWSKA-UÜRBANCZYK und UÜRBANCZYK 1983). Durch fast stufenlos ınein- ander übergehende unterschiedliche Temperatur- und Luftfeuchtigkeitsbereiche bestehen für die Fledermäuse ideale Überwinterungsbedingungen. Sie können praktisch jedes Mikroklima auswählen. Da kaum Spalten vorhanden sind, hängen die Tiere meist gut sichtbar freı an der Wand oder in Mauerlöchern (Abb. 2). Insgesamt wurden in den Jahren 1975-1986 49 Paarungen von Myotis daubentoni registriert. Die früheste Paarung wurde am 13.9. 1985 festgestellt. Da keine regelmäßigen Kontrollen erfolgen konnten und bei der Größe des Quartiers auch unterschiedliche Abschnitte erfaßt wurden, ist die Paarungs- häufigkeit in den einzelnen Monaten schwer zu vergleichen. Die maximale Paarungsaktıivi- tät (etwa 70 % aller Paarungen) lag in der Zeit von Oktober bis November. Die spätesten Paarungen wurden am 15. 3. 1986 und am 28. 3. 1986 festgestellt. Die Temperatur lag in den Quartierabschnitten mit Paarungsaktivität zwischen 7,5-9,2°C, die Luftfeuchtigkeit um 100 %. Wie bereits ROER und EGSBAEK (1969) beschrieben, fliegen die paarungsaktiven dd im Winterquartier umher und finden ein artgleiches ? offenbar nach der Methode Versuch und Irrtum. So konnten wır am 27.1. 1985 beobachten, wie ein aktives d in einem aus M. daubentoni, M. myotıs und P. aurıtus bestehenden Cluster nach einem 2 suchte. Es kroch dabei zwischen und über den schlafenden Tieren herum, biß in deren Nackenfell oder in die Ohren, zwängte den Kopf unter das jeweils untersuchte Tier und kontrollierte die Genitalregion, in kurzen Pausen dazwischen leckte es das eigene Genitale. In kurzer Zeit wurden so alle Tiere des Clusters unabhängig von ihrer Artzugehörigkeit im Winterschlaf gestört. Sie äußerten dabei die typischen, langgezogenen schrillen Laute im Winterschlaf gestörter Fledermäuse, ver- Paarungsverhalten der Wasserfledermaus im Winterguartier 135 Abb. 2. Cluster aus über 40 Myotis daubentoni in „Nietoperek“. (Foto: E. GRIMMBERGER) stummten aber rasch wieder, wenn sich das d einem anderen potentiellen Partner zuwandte. Hat das d ein 2 gefunden, nimmt es die Paarungsstellung ein (Abb. 3). Das d umfaßt in der Regel mit seinen Unterarmen die des ?, sein Rumpf ruht auf dem Rücken des 2, sein Kopf reicht bis in dessen Nackenregion. Die Ohren halten beide Geschlechter seitlich abgespreizt, nicht wie sonst im aktiven ungestörten Zustand aufgerichtet. Sowohl das & als auch das 2 halten sich mit den Hinterfüßen an der Unterlage fest. Zu Beginn betrommelt das d das schlafende ® mit den Unterarmen, tritt es mit den Hinterfüßen, reibt seinen Kopf an dessen Nackenregion oder beißt auch in das Nackenfell oder die Ohren. Offenbar erst wenn auch das ? aktiv zu werden beginnt, erfolgt die Intromission des Penis unter dem Uropatagium des 2. Da die Vespertilionidae in Ruhe das Uropata- gium auf den Bauch schlagen, muß das ? dieses vor der Einführung des Penis etwas abspreizen. Es folgten dann Phasen, in denen das d für einige Sekunden frequente Beckenbewegungen ausführte, dabei biß es z.T. in das Nackenfell oder die Ohren des ®, wobei dieses mit aufgerissenem Maul und angehobenem Kopf laut zeterte (Abb. 4). Danach saß das Paar einige Minuten ruhig, das & stützte dabei Oberkörper und Kinn auf das 2. Da das d dann wieder in gleicher Weise aktiv wurde, entstand der Eindruck einer Paarung mit mehreren Fjakulationen. Die Dauer des gesamten Paarungsablaufes lag etwa zwischen 15-30 Minuten. Bei den von uns beobachteten Paaren waren die dd immer voll aktiv und flugfähig. In einigen Fällen flogen sie nach Anleuchten ab, in anderen Fällen ließen sie sich selbst durch intensives Fotografieren nicht stören. Wurden Paare von der Wand abgenommen, war der erigierte Penis der 8 sichtbar, die dd flogen von der Hand 136 E. Grimmberger, H. Hackethal und Z. Urbanczyk Abb. 3. Paarungsstellung von Myotis daubentoni. Beachte die Haltung des Männchens und die bei beiden Geschlechtern seitwärts abgespreizten Ohren. (Foto: E. GRIMMBERGER) sofort ab. Die ?? waren in allen Fällen noch mehr oder weniger torpid, ein Abfliegen von der Hand war ihnen nicht möglich, sie konnten dagegen an der Wand entlang kriechen. Bemerkenswert war, daß bei den Beobachtungen am 27.1. 1985 Paarungen nur ın einem kleinen Abschnitt des Stollensystems festgestellt wurden, in anderen mikroklima- tisch ähnlichen Abschnitten jedoch nicht. Paarungsverhalten der Wasserfledermaus im Winterquartier 157. Abb. 4. Während der Paarung zetert das Weibchen wiederholt mit offenem Maul und angehobenem Kopf. Die geöffneten Augen des Männchens weisen gegenüber den geschlossenen des Weibchens auf die unterschiedlichen Aktivitätsgrade hin. (Foto: E. GRIMMBERGER) Diskussion Bei M. daubentoni erfolgen Paarungen ab August (KLAWITTER 1980); sie können dann über den ganzen Zeitraum des Aufenthaltes im Winterquartier bis in den April hinein beobachtet werden. Unsere Beobachtungen belegen Paarungen von September bis Ende März mit einem Maximum im Oktober und November. EGsBazk stellte ebenfalls wieder- 138 E. Grimmberger, H. Hackethal und Z. Urbanczyk holte Paarungen im März und ım ersten Aprildrittel fest (ROER und EGsBAER 1969), auch STEINBORN und VIERHAUS (1984) beschrieben Paarungen im März. Die der Wasserfleder- maus sehr ähnliche nordamerikanische Art Myotıis lucıfugus zeigt ebenfalls bereits ab Ende August im späteren Winterquartier Paarungsaktivität. Der Paarungsmodus ist bei dieser Art zu diesem Zeitpunkt jedoch ein gänzlich anderer, da die dd rufend an bestimmten Stellen des Quartiers hängen und von den 2% aufgesucht werden, mit denen sie sich dann z.T. paaren. Bei den späteren Paarungen in der Zeit des Winterschlafes zeigt M. lucıfugus dann ein ganz ähnliches Verhalten wie M. daubentoni (FENToN 1983). Bemerkenswert erscheint an unseren Beobachtungen, daß die Paarungen in Abschnitten des Winterquartiers mit relativ hohen Temperaturen (7,5-9,2°C) erfolgten, während in anderen Teilen des Quartiers Wasserfledermäuse beı wesentlich niedrigeren Temperaturen (z. T. unter 3°C) Winterschlaf hielten. Die Frage, ob die Lautäußerungen der 2% bei der Paarung andere dd im Sinne einer Stimmungsübertragung ebenfalls zu Paarungsaktivitä- ten stimulierten, muß offen bleiben. Die Tatsache, daß wir in mikroklimatisch ähnlich günstigen Bereichen an einer Stelle mehrere Paarungen sahen, an anderen Stellen jedoch keine, läßt daran denken. Eine Stimmungsübertragung durch Lautäußerungen beschrieb z.B. NEUWEILER (1969) bei dem Flughund Preropus g. giganteus, bei welchem es dann zu Massenkopulationen kommt. Während RoER und EGsBAEk (1969) annehmen, daß das & ein ® in erster Linie taktil erkennt, indem es bei beiden Geschlechtern die Anogenital- region mit erigiertem Penis abtastet, hatten wir den Eindruck, daß die Geschlechterfin- dung vorwiegend olfaktorısch durch Nasogenitalkontakt des d beim 2 erfolgt. Eine solche Kontaktaufnahme kommt praktisch bei allen Säugetiermännchen bei der Suche nach einem brünstigen ® bzw. während des Paarungsvorspiels vor. Da das 9 alle Fledermäuse eines gemischten Cluster durch seine Suche auch zu Lautäußerungen veranlaßt, könnten auch diese zumindest zum Erkennen artgleicher Tiere führen. Ob zusätzlich geschlechtsspezifische Unterschiede in den Lautäußerungen von im Winterschlaf gestörten Tieren einer Art bestehen, bedarf einer weiteren Klärung. Die Tatsache, daß bei gleichgeschlechtlichen Partnern die Paarungseinleitung bald abgebrochen wird (ROER und EGsBAER 1969), könnte auf solche Unterschiede zurückgehen. Zur Paarungsstellung schrieb KocH (1865): „Die Begattung verrichten die Fleder- mäuse, wie die meisten Thiere der Ordnung Primates von vorn, wobei sie sich mit den Vorderextremitäten umklammern.‘“ Inzwischen ist seit langem bekannt, daß diese wıe beı fast allen Säugetieren a tergo erfolgt. Eine Ausnahme bilden hier lediglich der zu den Kloakentieren (Monotremata) gehörende Australien-Kurzschnabeligel (Tachyglossus acu- leatus), der auf der Seite liegend Bauch an Bauch kopulieren soll (EwER 1976), die Wale sowie einige höhere Primaten mit ıhrer auch im Sexualverhalten größeren Variationsbreite. Unsere Beobachtungen des Paarungsablaufes bei M. daubentoni decken sich weitge- hend mit der Beschreibung der Paarung bei M. mystacinus und M. daubentoni durch van NIEUWENHOVEN (1956) sowie der bei M. Iucifugus durch FEnTon (1983) und bei Nyctalus noctula durch KrEıman und Racery (1969). Typisch sind neben der bereits beschriebenen Körperhaltung ein angedeuteter oder tatsächlicher Nackenbiß (z.T. auch in Kopfhaut oder Ohren) des d beim ? sowie schrille Lautäußerungen des d mit emporgehobenem, geöffnetem Maul. Ein Nackenbiß des d (besser „Nackengriff“, da in der Regel ohne Verletzungen) kommt als stammesgeschichtlich altes Merkmal (TEMBROcK 1983) sowohl bei Echsen als auch in vielen Säugetierfamilien, z.B. Soricidae, Pteropidae, Viverridae, Mustelidae, Felidae, Gliridae vor (EwER 1976). Ob der bei Fledermäusen zu beobachtende Nackenbiß ein homologes Verhalten zum Nackenbiß bei anderen Säugetierfamilien darstellt, erscheint nicht sicher. Möglicherweise dient das Beißen in das Nackenfell bei Fledermäusen eher der Stimulierung des 2, zumal der Nackenbiß nicht so gerichtet wie z.B. bei Raubtieren ausgeführt wird. Die lange Dauer der Kopulation von etwa 15 Minuten oder länger wurde von KrEıman und Racey (1969) auch für Nyctalus noctula erwähnt. BRown (1985) stellte Paarungsverhalten der Wasserfledermaus im Winterquartier 139 bei Antrozous pallidus sogar fest, daß das Paar mehrere Stunden in der Paarungsstellung verblieb. Eventuell erfolgen bei der Wasserfledermaus in diesem Zeitraum mehrere Ejaku- lationen. Einmalig bei Säugetieren dürfte der Unterschied im Wachheits- und Aktivitätsgrad zwischen d und ® bei der Paarung im Winterquartier sein. Während das d voll aktıv ıst, befindet sich das 2 noch in der Aufwachphase aus dem Winterschlaf in einem semitorpi- den Zustand. Das schließt die bei Säugetieren ebenso einmalige Situation ein, daß das ? keine Möglichkeit hat, sich der Begattung durch ein beliebiges d zu entziehen, und daß bei den Winterpaarungen zwischen den dd keine Konkurrenz um die 22 auftritt, d.h. alle adulten dd haben gleiche Paarungschancen, soweit diese nicht durch ihre unterschiedli- chen Energiereserven limitiert sind. Das bei fast allen Säugetieren anzutreffende mehr oder weniger ausgedehnte und oft sehr komplexe Werbeverhalten bzw. Paarungsvorspiel des d dient sowohl dazu, einen artgleichen Partner anzulocken, als auch ıhn dann in Paarungsbereitschaft zu versetzen, was wiederum den Paarungserfolg sichern hilft. Das präkopulatorische Verhalten des Fledermaus-d im Winterquartier beschränkt sich dagegen offenbar auf die Suche nach einem @ und auf das (partielle) Erwecken desselben. Da das Sperma ohnehin lediglich im weiblichen Genitaltrakt deponiert wird, zwischen Begattung und Ovulation mit anschlie- ßender Befruchtung der Eizelle jedoch Monate verstreichen, ist der Wachheits- bzw. Aktivitätsgrad des ? für den Erfolg der Paarung offenbar von untergeordneter Bedeutung. In diesem Zusammenhang ist auch darauf hinzuweisen, daß das Sexualverhalten von Säugetieren im wesentlichen von Zentren im Hypothalamus, limbischen System und Mittelhirn gesteuert wird, bei den dd die Funktion übergeordneter cortikaler Zentren jedoch von wesentlich größerer Bedeutung ist als bei ??. Erektion und Ejakulation sind zwar auch bei Ausfall des Cortex möglich, zum Vollzug des normalen Paarungsablaufes müssen dd aber über eine weitgehend uneingeschränkte Funktion des Cortex (und somit der Vigilanz) verfügen (BAstock 1969). Offenbar stört bei Fledermausweibchen eine Einschränkung der cortikalen Funktion als Folge des semitorpiden Zustandes das Paarungsverhalten nicht. Zusammenfassung Es wird über das Paarungsverhalten von Myotis daubentoni im Winterquartier berichtet. Im einzelnen wird auf mikroklimatische Bedingungen im Winterquartier, die Partnersuche des d, das Verhalten beider Geschlechter während der Paarung, die Dauer der Paarung sowie auf die unterschiedlichen Wachheits- bzw. Aktivitätsgrade beider Geschlechter eingegangen. Als eine Ausnahme ım Paarungs- verhalten bei Säugetieren wird hervorgehoben, daß die ?? bei der Paarung ım Winterquartier im Gegensatz zu den dd noch mehr oder weniger torpid und somit teilweise inaktiv sind. Die Ergebnisse werden anhand der Literatur diskutiert. Literatur BAGROWSKA-UÜRBANCZYK, E.; URBANCZYK, Z. (1983): Structure and Dynamics of a Winter Colony of Bats. Acta theriol. 28, 183-196. BAsTock, M. (1969): Das Liebeswerben der Tiere. Jena: VEB Gustav Fischer. Brown, P. E. (1985). Copulation without intromission in the Pallıd Bat. Seventh International Bat Research Conference, 19.-24. August 1985 (Abstracts), Aberdeen. EWER, R. F. (1976): Ethologie der Säugetiere. Berlin, Hamburg: Paul Parey. FEnTon, M. B. (1983): Just Bats. Toronto, Buffalo, London: University ot Toronto Press. GusTAarson, A. W. (1979): Male reproductive patterns in hibernating bats. J. Reprod. Fert. 56, 317-331. KLAWITTER, J. (1980): Spätsommerliche Einflüge und Überwinterungsbeginn der Wasserfledermaus (Myotıs daubentoni) ın der Spandauer Zitadelle. Nyctalus (N.F.), Berlin 1, 227-234. KLEIMAn, G. D.; Racey, P. A. (1969): Observations on noctule bats (Nyctalus noctula) breeding ın captivity. Lynx 10, 65-77. Koch, C. (1865): Das Wesentliche der Chiropteren mit besonderer Beschreibung der in dem Herzogthum Nassau und den angränzenden Landestheilen vorkommenden Fledermäusen. Wies- baden: Julius Riedner, Verlagshandlung. 140 E. Grimmberger, H. Hackethal und Z. Urbanczyk KULZER, E. (1981): Winterschlaf. Stuttgarter Beiträge zur Naturkunde-Serie C, Nr. 14. NEUWEILER, G. (1969): Zum Sozialverhalten von Flughunden (Pteropus g. giganteus). Lynx 10, 61-64. NIEUWENHOVEN, ]. P. van (1956): Ecological observations in a hibernation-quarter of cave-dwelling bats in South-Limburg. Naturhistor. Genootschap in Limburg 9, 1-55. Raczy, P. A. (1972): Viability of bat spermatozoa after prolonged storage in the epididymis. ]. Reprod. Fert. 28, 309. x RoER, H.; EGsBAER, W. (1969): Über die Balz der Wasserfledermaus (Myotis daubentoni) (Chirop- tera) ım Winterquartier. Lynx 10, 85-91. STEINBORN, G.; VIERHAUS, H. (1984): Wasserfledermaus — Myotis daubentoni (LEISLER in KuHL, 1817). In: Die Säugetiere Westfalens. Ed. by R. SCHRÖPFER, R. FELDMANN und H. VIERHAUS. Münster: Westfälisches Museum für Naturkunde Münster, Landschaftsverband Westfalen-Lippe. TEMBROCcK, G. (1983): Spezielle Verhaltensbiologie der Tiere. Band 2. Jena: VEB Gustav Fischer. Anschriften der Verfasser: Dr. ECKHARD GRIMMBERGER, Steinstr. 58, DDR-2200 Greifswald; Doz. Dr. sc. Hans HACKETHAL, Museum für Naturkunde der Humboldt- Universität, Invalidenstr. 43, DDR-1040 Berlin; ZBIBGNIEW URBANCZYK, Os.J.III. Sobieskiego 26 d/142, Polen — 60-683 Poznan Phylogenetic relationships among four arvicolid genera By BARBARA R. STEIN Museum of Natural History, The University of Kansas, Lawrence, Kansas, USA Receipt of Ms. 27. 7. 1986 Abstract Appendicular myology and osteology in four genera of arvicolid rodents, Clethrionomys, Eothenomys, Alticola and Synaptomys, were evaluated for characters useful in constructing hypotheses of phylogenetic relationships. Postcranial morphological data, combined with information on karyol- ogy, biochemistry, stomach, glandular, and external morphology, support a close evolutionary relationship among these taxa. Myological characters supported the monophyly of Clethrionomys, Eothenomys and Alticola and indicated the apomorphic nature of Alticola relatıve to the other two genera. In no instance was myology alone sufficient to distinguish Clethrionomys from Eothenomys. In contrast, osteology identified derived characters among species of Clethrionomys, whereas the genus Synaptomys was apomorphic both osteologically and myologically. Postcranial morphological data also indicated that within Clethrionomys the species gapperi, glareolus and rutılus are closely related; C. rufocanus ıs distinctive and is considered primitive within the genus. The presence of rooted molars is regarded as a derived character among ee taxa and represents a character reversal ın the proposed phylogeny. Heterochrony is hypothesized as a mechanısm that would explain this arrangement. Introduction Within the Arvicolidae (Honacki et al. 1982), a famıly comprising the voles (Microtini) and lemmings (Lemmini), the Clethrionomys-Eothenomys-Alticola group currently repre- sents one of the most unstable areas of classification at the generic level (CoRBET 1978). Although a close evolutionary relationship among these genera has long been hypothesized (MILLER 1896), attempts to document this relationship have not produced a single, generally accepted classification. Owing to the abundance of Pleistocene rodent fossils, many arvicolid classifications have relied predominantly on cranıal and dental characters (CoRrBET 1978). However, Imarzumi (1957) noted great confusion in the classification of Japanese voles resulting from U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5203-0140 $ 02.50/0 Z. Säugetierkunde 52 (1987) 140-156 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Phylogenetic relationships among four arvicolid genera 141 the use of these characters and numerous investigators subsequently have documented variation of these structures both within and between arvicolid populations (PERRIN 1978; Aı1ısHAI 1980; ABE 1973a, b, 1982; CoRBET 1963, 1964; GUTHRIE 1965, 1971; HAIT- LINGER 1965; BATEs and Harrıson 1980; AıroLDI and HorFMAnN 1984; Mıyao 1966; ZEJDA 1960; ZIMMERMANN 1965; PATTERSON 1983; ILYENKO and Krarıvko 1983). Based on this evidence, I have examined appendicular myology and osteology in Clethrionomys, Eothenomys and Alticola (tribe Microtini) as alternative sets of characters on which to base a cladistic analysis of this group. The bog lemming, Synaptomys, was examined for purposes of comparison within the Arvicolidae, because this genus is considered the most generalized member of the more primitive arvicolid trıbe, the Lemmini (Hınton 1926; ELLERMAN 1940). Postcranial myological characters should accurately reflect evolutionary relationships because it is possible to trace the innervation of muscles and, therefore, to establish their homologies (BRYANT 1945; KLINGENER 1964; Woops 1972). Several studies also exist in which postcranial osteology in rodents has been examined. CARLETON’s (1980) work on neotomine-peromyscine relationships employed postcranial osteology in a phylogenetic analysis. BATEs and HARRISON (1980) attempted to identify interspecific variation using long bones. HECHT (1971) examined the scapula, humerus, femur and pelvis in four genera of rodents and drew attention to their morphological differences among species. BEsT (1978) included postcranial measurements in his morphomerric analysis of Dipodomys, as did COTHRAN (1983) in his ground squirrel study. The study presented herein is not a phylogenetic assessment based solely on postcranial morphology. These data have been combined with information obtained from the litera- ture on karyology, biochemistry, glandular, stomach and external morphology to produce a synthetic phylogeny of the taxa under consideration. Material and methods Separate sets of specimens were used for qualitative myological and osteological analyses. In the former, one side on each of 57 specimens representing 10 taxa was dissected. These taxa included five of the seven currently recognized species of Clethrionomys, specimens of Eothenomys melanogaster, Alticola roylei and Synaptomys cooperi. For osteological analyses four species of Clethrionomys, three of Eothenomys and one each of Alticola and Synaptomys were examined. Skeletons were compared under a magnifying lens and differences in morphology noted. The following specımens were examıned: Alcoholic specimens Clethrionomys gapperi: canapa: Manitoba: North Knife River, 5 ku; usa: New Hampshire: Coos Co., 1.25 mi N, 11.5 mi E Pittsburgh, 2 cm; Grafton Co., Franconia Notch State Park, 4.7 mı S, 3.5 mi E Franconia, 1 cm; Wisconsin: Douglas Co., Solon Springs, 3 FMNH; Wyoming: Fremont Co., 17 mi SW Lander, 1 ku; Lincoln Co., 13.5 mı N, 8 mı E Cokeville, 1 Ku. Clethrionomys californicus: usa: California: Humboldt Co., Cutteback, 1 Mvz; Mendocino Co., Mendocino City, 1 Mvz; Oregon: Jackson Co., 3 mi N, 11 mı E Ashland, 2 ummz. Clethrionomys rex: Japan: Rishirı Island, 1 PsMNnH. Clethrionomys rufocanus: KOREA: central, 1 usnm; Hill 1468 m, 3 mı SSE Sumil-li, 1 AmNH; 3 mı ERt. 33 on 22, 1 usnM; Pusan area, 1 usnM; PRc: No. Manchuria: Great Khingan Mts., Yalu Station, China E. Railway, 2 cas. Clethrionomys glareolus: czecHosLovarıa: Velky Tisy, 2 usnM; near Nymburk, 1 USNM; FINLAND: Uusimaa, near Helsinki, Sıpoa, 3 UMMZ. Clethrionomys rutilus: usa: Alaska: St. Michaels, 2 usnm; Tonzona, 1 usnm; Anchorage, 2 UMMZ; Yukon Hwy., mı 30, 1 umMm2z. Clethrionomys sikotanensis: USSR: RSFSR, Sakhalınsk. Obl., Anivsk. rai., 3 KU. Eothenomys melanogaster: prc: Yunnan: Mucheng, Salween, Drng, 5 AMNH; ROC (TAIWAN): Chuei- Feng, Nan Ton Tsıen, 2 uUsNM. 142 Barbara R. Stein Alticola roylei: AFGHANISTAN: Badakhshan Prov.: 15 m E of Eshkashem, 2 FMNH; PAKISTAN: Dir State: 16 mı No. Dir, 8400’, 1 usnm; 20.5 mı No. Dir., 10400’, 1 usnM. Synaptomys cooperi: usa: Kansas: Atchison Co., Atchison Co. State Lake, 2 ku; Jefferson Co., Nelson Evironmental Study Area, 1 ku; Nebraska: Gage Co., 0.25 mi W Homestead Nat’] Mon., 3 KU. Skeletal material Clethrionomys gapperi: CANADA: Alberta: Fort McMurray, 1 mcz; Quebec: Richmond Gulf, 1 mcz; Bonaventure Co., 5.34 mı N, 2 mı E St. Elzear de Bonaventure, 3 cm; 5.67 mı N, 1 mi E St. Elzear de Bonaventure, 1 cm; 4.5 mı N, 0.5 mı E St. Elzear de Bonaventure, 1 cm; 4.5 mi N, 2 mi E St. Elzear de Bonaventure, 1 cm; usa: Idaho: Adams Co., 0.5 mı E Black Lake, 1 mvz; 0.5 mi E Black Lake, 6800’, 2 ku; SW slope Smith Mt., 1 mvz; 1 mı N Bear R. $., SW slope Smith Mt., 5400’, 1 Ku; 3 mi W Payette L., 1 mvz; Bannock Co., 1 Ku; Bannock Co., Pocatello, 1 Ku; Maine: Franklin Co., 3.7 mi NW Kingfield, 8 cm; Massachusetts: Middlesex Co., Pepperell, 4 mcz; Michigan: Chippewa Co., Marquette National Forest, 1 Ku; Marquette Co., Huron Mts., 10 FMNH; New Hampshire: Grafton Co., Franconıa Notch State Park, 4.75 mı S, 3.5 mı E Franconia: 1 cm; Pennsylvanıa: Crawford Co., Pymatuning Swamp, Sedge Linesville, 1022’, 1 cm; Pymatuning Swamp, 4 mı W Linesville, 2 cm; Pymatuning Swamp, 4 mı W Linesville, 1000’, 1 cm; West Virginia: Randolf Co., Blister Pine Run, 1 mı SE Cheat Bridge, P. ©. 3675’, 2 cm; Wyoming: Albany Co., 3 mı EsE Browns Peak, 3 Ku. Clethrionomys cahfornicus: usa: California: Mendocino Co., 3 mı NW of Gualada, St. Orres Creek, 60’, 1 LAcM; Trinity Co., White Rock Ranger Sta., 4800’, 1 mvz. Clethrionomys rutilus: CANADA: Northwest Territory: Tyrell Lake Region; 7 ummz; Beverly Lake Region, 2 uMMZ; Artillery Lake Region, 1 UMMZ; LAPLAND: I USNM; USA: Alaska: 1 ummz; Chandler Lake, 68°12', 152°45’, 2900’, 3 ku; Glacier Bay Nat’l Mon., Torch Bay, 7 Ku; Glacier Bay Nat’l Mon., Thistle Cove, 2 Ku; Glacier Bay, 1 Ku; St. Lawrence Island, Mt. Gambell, 2 psmnA; St. Lawrence Island, 2 mı SW N. E. Cape, 1 usnm; Anakturuk Pass, 2 psMnH; Paxson Lake, 1 UMMz; Paxson, 1 usnM; Fairbanks, wıthin 100 mı radıus of, 2 usnm; Fairbanks region, Piled River Slough, 2 umMZz; Seward Peninsula, Lava Lake, 5 usnMm; Inmachuk River, 1 usnm; Kuzitrin Lake, 1 usnm; Tonzona, 1 usnm; Nulato River, 1 usnm; Cantwell, 2212’; Jack River 4 mı Camp, 2400’, 1 AMNH; Koalak River, 70° N, 3 mvz; College University of Alaska, 1 psMNnH; Brooks Range, Anzktuvuk Pass, 1 PSMNH. Clethrionomys rufocanus: FINLAND: Lappi District, Kilpisjarvi, 5 cm; Japan: Hokkaido: 6 PsSMNH; Hokkaido: Toshina-Schicho, 1 psMmnH; Hokkaido: Toyotamı, 1 PsMNH; Hokkaido: Yukomambetsu, 1 psMNnH; Rıshirı Island: 3 PpsMNH. Clethrionomys rex: JaPan: Rishiri Island: 8 PsMNH. Clethrionomys centralis (= frater): USSR: 1 UMMZ. Clethrionomys glareolus: austria: Niederösterreich Prov.: I. I. A. S. A., Laxenburg, 1 mvz; ENGLAND: Üttoxeter, 1 AMNH; Berkshire: Culham College, 2 mı SW Abingdon, 2 mvz; Devonshire: Torrington, 3 AMNH; FINLAND: Hame District: 2 km N, 3.5 km E Iso-Evo, 6 cm; Oulo District: 4 km N, 4 km W Rukajawı, 2 CM; FRANCE: Cote d’Or, 1 USNM; UssR: Vyborg, 2 AMNH. Eothenomys melanogaster: prc: Szechwan: Kuan Shien, 8 UsNM; Mupin, 2 usnM; Wan Shien, 1 usnM; Tseo-Jia-Geo, 1 USNM; THAILAND: top of Mt. Angka, 1 mcz. Eothenomys smithi (including Rageus): Japan: Fukuoka Pref., 2 psMnH; Honshu Island: Japanese Alps, 3 psMNH; Yamanashı Pref., Mt. Fuji, 1 PsMNH. Aschizomys (Eothenomys) andersoni (includes imaizumi): JAPAN: 1 PSMNH; Mt. Zaoh, 1 PSMNH; Tochigi Pref:, 1 psMNH; Yumagata Pref., Mt. Rao, 1 psMNH; Honshu Island: Natchinotaki Falls, 1 PsmnH. Alticola sp.: ınpıa: Ladak, above Leh, 1 usnm. Synaptomys cooperi: usa: Kansas: Anderson Co., 6 mi $ Garnett, 3 Ku; Barber Co., 3 mı N, 1 mıE Sharon, 2 ku; Douglas Co., 6 Ku; Lawrence, KU campus, 1 Ku; 7 mıSW Lawrence, 3 Ku; 7.5 mı SW Lawrence, 5 ku; 2.5 mı W Lawrence, 1 ku; 1 mı W Lawrence, 3 ku; 1mı SW Lawrence, 1 Ku; 1.67 mı W Lawrence, 1 Ku; 2 mı W Lawrence, 4 ku; Lawrence, 1.5 mi N, 1.5 mi E courthouse, 4 Ku; Meade Co., 14.5 mi SW Meade, 1 ku; 17 miSW Meade, 1 Ku; Nebraska: Dundy Co., 5 mi N, 2 mi W Parks, DER AMNH = Amer. Mus. Nat. Hist., cas = Calıf. Acad. Scı., cm = Carnegie Mus. Nat. Hist., FMNH = Field Mus. Nat. Hist., Ku = Univ. Kansas Mus. Nat. Hist., LACM = L. A. Co. Mus., mcz = Mus. Comp. Zool. Harvard, mvz = Mus. Vert. Zool. Berkeley, psMNH = Puget Sound Mus. Nat. Hist., UMMZ = Univ. Mich. Mus. Zool., usum = U. S. Nat’l. Mus. A character-by-character analysis was employed in constructing the cladogram because I observed relatively little homoplasy among taxa. A computer analysis of the data set using the PAUP algorithm Phylogenetic relationships among four arvicolid genera 143 (SWOFFORD, 1984) confirmed the phylogram. The cricetine genera Neotoma, Peromyscus, Sigmodon and Oryzomys were examined for outgroup comparison and reference was made to RınkERr’s (1954) classic monograph of myology in these taxa. Conditions observed in these genera were considered plesiomorphic and, thus, determined the polarities of my characters (table). Results Myological characters Although a great deal of individual variation was noted, the myology of Eothenomys was not found to differ from that of Clethrionomys in any identifiable way, nor was variation among species of Clethrionomys felt to have any systematic significance. Of the 110 muscles dissected, four in the forelimb and four in the hind limb did exhibit variation that I considered valuable for systematic analyses (fig. 1). In the forelimb, the in- serion of M. occipito- « © © \ . . . X N S Ss scapularis is confined to the N g g Sg 0 « & N AS S cranial half ofthe vertebral © & g & IS & 5 N € © border in Clethrionomys, Eothenomys and Alticola whereas in Synaptomys and in the cricetine genera the insertion extends along the scapular root to the base of the scapular spine. Second, the insertion of M. omocervicalis ıs con- fined to the metacromion process ın Clethrionomys, Eothenomys and in the cricetines examined but it extends to include the ac- romion ın Synaptomys and in Alticola. Third, Synap- Fıg. 1. Cladogram of the four arvicolid genera examined and their tomys and the cricetine outgroup based on the results of myological dissections (from STEIN genera possess a primarıly 1986) radial insertion of M. biceps brachii versus an insertion that is primarily ulnar in Clethrionomys and in Eothenomys and one which is half radial and half ulnar in Alticola. Fourth, Synaptomys possesses only two points of insertion for M. abductor pollicis longus. All other genera examined, with the exception of Sigmodon, possess three points. In the hind limb, the insertion of M. femorococcygeus is restricted to the knee joint capsule in Synaptomys and ın the cricetine taxa but extends along the distal end of the posterior femoral shaft in Clethrionomys, Eothenomys and Alticola. In addition, the tendons of the two heads of M. rectus femoris are not well defined in Clethrionomys and Eothenomys but are distinct in Synaptomys and in the cricetine genera. Their condition in Altıcola ıs intermediate between these two extremes. Third, in Clethrionomys, Eothenomys and Alticola, M. caudofemoralis arises from the deep surface of M. femorococcygeus. In Synaptomys and in the cricetines the muscle arises directly from the sacral and caudal vertebrae. Lastly, Alticola was the only genus examined that possesses a tendon from M. flexor digitorum brevis to Digit V. Complete muscle descriptions and figures for all muscles dissected are provided ın STEIN (1986). . omocervicalis . oCCipitoscapularis . biceps brachii . abductor pollicis Iongus . femorococcygeus . caudofemoralis . rectus femoris > DH uUR ae 22 SESBZU Sense . flexor digitorum brevis 144 Barbara R. Stein Osteological characters In contrast to the results of my dissections, several osteological characters served to identity individual species of Clethrionomys. Clethrionomys glareolus, which proved indistinguishable from C. rutılus in all other respects, had a greater trochanter of the humerus that was relatively larger than that noted for all other species examined (BATESs and HARRISON 1980). Additionally, the humerus of C. rutılus was recognized as having a relatively larger head than those in the other species although this difference was not outstanding. Moreover, the ridge of the humerus adjacent to the lateral condyle was square in outline as compared with a smoother profile in the remaining taxa. Clethrionomys gapperi, C. glareolus and C. rutilus were united in that all possessed a bifid spine of the axiıs vertebra compared to a thin, flattened spine observed in the other species examined. The one specimen of Alticola available to me also possessed a bifid spine, but this character must be examined in additional material before a definitive statement about its condition in that genus can be made. Eight osteological characters, in addition to four myological ones, supported the monophyly of Clethrionomys, Eothenomys and Alticola (fig. 2). The ventral tip of the atlas in these three taxa was vertical, whereas in Synaptomys and in the cricetines ıt was angled posteriorly (fig. 2a, b). The shape of the clavicle and the acromion process of the scapula \) Fig. 2. Eight osteological charac- k ) : ters which support the monophy- ly of Clethrionomys, Eothenomys N , and Alticola. a: ventral tip of the atlas vertical; c, e: clavicle rela- N\ tively straight; g: acromıon pro- cess of scapula Bose i: del- toid crest of humerus smooth and round in profile; k: notch a > by pubic ramus; m: head of the ns tipped towards the greater trochanter and shaft of the femur PEN relatively thin; o: tibia/fibular fu- sıon relatively more extensive. Fig- \ ures b, d, f, h, j, I, n and p show the corresponding plesiomorphic o p conditions Phylogenetic relationships among four arvicolid genera 145 were also distinctive in these genera. The clavicle was relatively straight, curving slightly at the ends in all three, whereas in Synaptomys and in the cricetines it curved both in anterior/ posterior and ın dorsal/ventral planes, appearing to twist along its course (fig. 2c-f). In general, the acromıon process of the scapula curved gently like a hook at the distal end ın the Microtini, whereas in Synaptomys it appeared spatulate (fig. 2g, h). The deltoid crest of the humerus in the former three genera was smooth and round in profile contrasting with a larger, more “beak-like” curvature in Synaptomys and in the cricetines (fig. 2i, j). In the hind limb osteological differences were noted in several elements. The majority of varıation in the pelvis related to differences in shape between males and females. Overall, sexual dimorphism ıs not considered significant in this family with the exception of pelvic shape differences (RossoLIMo 1962; Aımı 1980; OGnev 1950; MosKVvITINA and SucHKoVA 1976; KAnEKO 1968, 1969; BRowNn and Twıcc 1969) and my observations support this hypothesis. A notch formed by the public ramus was distinct in most species but was relatively weak in many specimens of C. rufocanus and Eothenomys and in some individuals of C. rutiılus (fig. 2 k, |). The head of the femur was angled more towards the greater, versus the lesser, trochanter of that bone and extended above it in Clethrionomys, Eothenomys and Alticola (fig. 2m, n). In Synaptomys and in the cricetines the head was tipped towards the lesser trochanter and the greater trochanter extended above it. The shaft of the femur was thinner in the former three genera relative to its condition in the other taxa examined. Lastly, the degree of fusion of the distal ends of the tibia and fibula was most extensive in the Microtini (fig. 20, p). CARLETON (1980) recognized this character to be significant phylogenetically in his study of neotomine-peromyscine relationships. The genus Synaptomys was apomorphic for a number of osteological, as well as myological, characters (fig. 3). The transverse processes of the atlas were unusually large and protruding relative to the other genera and the dorsal surface of that bone was distinctly notched (fig. 3a, b). The carotid tubercle on the ventral surface of the sıxth cervical vertebra curved posteriorly as asmooth shaft which tapered distally, in contrast to the more angular-shaped tubercles of the other genera (fig. 3c, d). For most individuals of Synaptomys examined, the height of the scapula was small relative to the overall length, with the bone tending to be square in outline, the borders thickened (fig. 3e, f). This condition contrasted sharply with all other species which had deep scapulae, domed cranially and with delicate borders. The only other difference in the forelimb of Synaptomys was noted in the shape of the nail. In this genus the nail was wedge-shaped rather than rounded, and the tip of the pollex bulbous as though the digital pad had been repositioned distally. No functional explana- tion for this phenomenon is obvious at the present time. In the hind limb of Synaptomys only one outstanding apomorphy was noted. The ischial tuberosity of the pubis was a large, flattened triangular area whereas in all other genera it was relatively indistinct. The medial and lateral flanges of the tibia seemed reduced in this genus as well but the difference was not dramatic (fig. 3g, h). I do not report any variation in the shape or size of the calcaneum among these genera because the distal extremities were lacking in a majority of the specimens I examined. I expect that some interspecific or intergeneric differences do exist within this group based on the work of CArLEToN (1980) and Stars (1959). The latter author found this bone to be particularly important taxonomically within the Rodentia. Specifically, within the Arvicolidae, he found variation not only in size but also in the position of the trochlear process, in a peculiar enlargement and twisting of the calcaneal body, and in the presence of an anterior articular surface for the astragalus. He did not, however, specify what these differences were for the genera he examined. 146 Barbara R. Stein Fig. 3. Characters exhibiting an apomorphic condition in Synaptomys. a: transverse processes of atlas large and dorsal surface of atlas distinct- ly notched; c: carotid tubercle of 6th cervical vertebra smooth in outline and taper- ing distally; e: height of N scapula small relative to over- | all length, with thickened | borders, square ın outline; g: ischial tuberosity of pubis large, flat and triangular in shape. Figures b, d, f and h g h show the corresponding plesiomorphic conditions Discussion Information gleaned from the literature pertaining to karyology, biochemistry, stomach, glandular and external morphology have been added to the myological and osteological characters which I observed in order to produce a phyologenetic hypothesis which I feel should closely approximate the evolutionary history of this subgroup of rodents (fig. 4, and table). Descriptions and/or figures of bacula for all groups except Alticola were found. However, sufficient controversy surrounds the nature of this structure as a reliable phylogenetic character within the family that I have not emphasized it greatly (BurT 1936; HAMILTON 1946; DEARDEN 1958; ANDERSON 1960; HooPpEr and HART 1962; HOOPER and Musser 1964; HERSHKOVITZ 1966; Morrıs 1972; Long and FRANK 1968; Aımı 1980; JAMEson 1961; OGnev 1950; Didier 1954). In addition, mammae number does not vary much among these taxa except within the genus Eothenomys (see below), and anatomy of the middle ear has been described only for some Nearctic genera (HooPER 1968). In addition to four myological characters (table) (13-16) and eight osteological charac- ters (20-27), Clethrionomys, Eothenomys and Alticola are united in possessing a bony palate terminating as a thin, transverse shelf and lacking a median bridge sloping dorsally to join the midpoint of the pterygoid fossa (Hınron 1926) (18). Moreover, all three genera possess an intermediate grade I glandular zone of the stomach (CARLEToON 1981) (19) and, compared with Synaptomys, all have a lower incisor that extends posteriorly in the jaw to a Phylogenetic relationships among four arvicolid genera 147 > ©) S a x 2 & N S N S N) S Ö c 2 © A N ‚Oo Na S N) R NS > & @ S 5 > 7 &> S N) S Io < 5 19,;54-56 19, ,,51 38-40 52 43-49 3-12 28,50 3,28-35 36.;41-42 15 13,,34,, 36-37 13-27 Fıg. 4. Final cladogram. See table for a list of characters and their derived states point distal and labial to m; (ELLERMAN 1940) (17). Eothenomys closely resembles Clethrionomys externally as well as in the shape of the skull (ELLERMAN 1940; ALzen 1940; JAMEson 1961), and one author even proposed the genus as a possible antecedent to Clethrionomys (ALLEn 1940). However, the two may be distinguished by the complete absence of rooted molars in Eothenomys, even in animals of very old age (Hınron 1926; ELLERMAN 1940), as opposed to their presence in Clethrionomys (1). Alticola ıs sımilar to Clethrionomys in most respects but differs in the presence of slender and elongate cheek teeth (CoRBET 1978) (30) and a body form that ıs not modified externally for fossoriality (ELLERMAN 1940). Its close relationship to Eothenomys has been recognized by several authors (ELLERMAN and MORRISON-SCOTT 1951; ALLEN 1940), one of whom postulated that it was a rock-dwelling version of this woodland vole (ALLen 1940). All members of the genus Clethrionomys possess a bright, rufous pelage (41), rooted molars (1), a large single, rather than double, opening at the base of the alisphenoid (CARLETON 1985) (42) and an increase in sıze, but a decrease in number of Meibomian (= tarsal) glands compared to Eothenomys (Quay 1954) (36). Within this genus, however, C. rufocanus has differentiated morphologically, ecologically and biochemically (OGnEv 1950; GRAF 1982). The species is distinguished by its large body sıze (46) and by late rooting of ıts molars (Hınrton 1926) (43). OGnev (1950) also noted that its ear ossicles differed appreciably (44) from those of other species in the genus and that its penis (47) was relatively long and narrow. The postorbital process of the skull is well developed in rufocanus (MiLLER 1900) (48) and the root of the third lower molar lies in a separate capsule lingual to the incisor root (HınTon 1926) (45). The distinctive nature of this species is supported further by its displaying lower allozymic affınities to C. rutilus and C. glareolus than these species share with one another and by its possession of two unique allozyme alleles and two minor hemoglobin (Hgb) bands, one of which is unique (NADLER et al. 1978) (49). NADLER et al. (1978) have hypothesized that rufocanus diverged from the rutilus-gapperi stock prior to 13,000 years ago, perhaps in response to the same Pleistocene events that led to the divergence of other mammalian groups such as the long-tailed ground squirrels. Because of its distinctive nature, C. rufocanus has been varıiously assigned to the genera Craseomys and Aschizomys but, most recently, it has been retained within the genus Clethrionomys (Aımı 1980). The species Clethrionomys gapperi, C. glareolus and C. rutilus are considered to be closely related within the genus because each species has a bifid spine on the axis vertebra Barbara R. Stein 148 reurdreumd 19pPIog jeururgy : 061 NvnaaTıgq Areigel gu sed spu9ıx9 : | jfen3un] eur 03 spu9ıx9 :0 JOSIIUT A9MOTT '/] 12 aaındıy u weıdopep UI IINIJYSUOI 07 PISN SIIJIEIEUD JO 3SYT olgPL l 0561 AANDO 9397[d4W09 19p1oqg eure] :Q snojfem 'EE 1unsıp J10U Spray 7:] ıoded sıyJ p>10jo91q Ardıeys “Suof :] 9861 NIELS WUnSIp speay 7:0 STIOWSF SNI994 "N '9] ‘//6] sLaaaoy 10[09 JUO (LIOUS :Q [Te L ze "I0I040W.2F "N WOAF UTLIO :] AreyuaumpnA:] 9861 NIALS UTSLIO JEIGNIDA "wind: 0) SeIOWS>FJOPNEI "N 'G] 976] NOLNIH p>dopAap :0 x9ffod "TE 9urds UO UONAIISUI ou: 91e3uof9 9 A9PuaJs :] 986] NIALS uonaasur aurds 'deas :0 "dessond1990 "W "I 8/6] LAdUO0) peoig [94:0 sıe[oW '0€ UONIISUI seum "wind :z A 131] 03 uopu9 :] seupn Z/] ‘Terper z/L:1] ıoded sıy3 A 1131] 0) uopua ou :Q "Adaq "IIp 'Xa] "N '67 9861 NIALLS uonAasur ferpes "und :Q myseag sdoarg "WEI pıprq aurds : ] Pa | ıoded sıyy pouanef 9urds :Q SIxV '87 ı9ded sıyı se] :o soguejz fergt], ‘ZI ı9ded sıyy JAISUIIXI 9JOUL "[94 2] sepn3uers Jelp:] ‘0861 NOLITNVO ss9]:O LOSE EZ ı9ded sıyı p>dopAa>p Afyeam :0 Ay1soasgng fery9s] '[] MOAIEU [O1] 9urpno arenbs “moxeu :] ı9ded sıyJ peoig [91:0 ıyeys “anwm,] '97 soded sıyı Juno YIOOWS “IPIM !Q epndesg ‘oJ "18913 % peay "mag afsue "ws: Afyroous p9AınD :] aoded sıyy 19859] 9 peay "Mag afdue "ws :Q sI9Jury90N “NUm] 'Gz ı9ded sıyy Jepnäue :g PADqNI pno1ey 'G yu9sa1d you : p>ymou :] ıoded sıyy Juasge Y9I0U :Q snweısigng 'pZ ı9ded sıyy 8 9PIM :0 wnsıoppru ‘sen 'g ofıfoad ur oows :] s799e7 1sed "pnnoıd pur Se] :] 1oded sıyy 9NYIJ-Jooy:Q 138919 [op “snaa9wunyg 'E7Z ıoded sıyı [feus :0 Vo1d 'suen ‘seny '/ II[-JOOUY :J ejppred 1e "und "119sur :j ı9ded sıyı oyeingeds :Q UOTWOADK “eindeag '7z 986] NIELS ana '3sod "119sur:0 SN93A9909010W2F "W "9 1ysrens Afoane[o4:] snoqjng ‘padeys- 93PaM :] ıoded sıyy sauepd 7 ur paAım9 :Q IPPALII "IZ 9861 NIALS p>puno1 :Q x9jfod 70 [ren 'S jean19A ‘901d 'zuaA :] uon1osur jo 'sıd7z:] ı9ded sıyı ‘ue poddn '501d 'IuaA :0 senv '07z 9861 NIALS uon1osur Jo 'sıd e :0 ‘3uof 'jjod 'pge 'W ‘+ III ?peı3 "pounssgur :g "IOSUT "WONEIIW Pur [ELWOADE :] II >peı3 "paunsaur :Z 9861 NIALS UONAIISUT [EIWOIIEIIW !Q SITEITAIIIOWO "N '€ ] >peı3 "pounsgur:] Srewstd :] 1861 NOLITAVY sejnpuepsiway :Q »UOZ "pur[s ‘y9ewosg '6] 9761 NOLNIH 9yeprdsna:9 uoneins1yuo9 JefoM 'T ‘ısod yfoys 'sue.n “ury2 :J Or61 NvwaaTıg SS9OOA:] 9761 NOLNIH sırd 'Tef ‘9spıı "pow /M :Q 9ejed Auog 'g] ‘9761 NO.LNIH P22004 :0 YIMOAS IeJoW "| IIINOS EIS A9IILIEUD IIINOS 9IeIS A9IIELIEUTD) Hınrton 1926 Hınton 1926 OGnev 1950 MirLLEr 1900 NADLER et al. ons 0: rel. short, rounded base 1: long & narrow, base flat O0: weakly devel 1: well develope 1: root lingual to incisor rt. O: none unique O: root labial to incisor rt. O: relatively small 1: relatively large 48. Postorbital proc. 45. Lower m3 46. Body sıze 49. Allozymes 47. Penis CARLETON 1981 CARLETON 1981 Quay 1954 HooPER and 1: inc. size, dec. number 2: further inc. size, dec. no. 1: intermed. depth 0: group I type 2: deep 1: emargınate 0: small, numerous 0: shallow 0: smooth 34. Stomach, incis. ang. 35. Stomach, bord. fold 36. Meibomian glands 37. Glans penis Phylogenetic relationships among four arvicolid genera 1449 (28) and distinctive G-banding oc © 0 a x So 2 of chromosomes one and nine [6] EX (al) SIWEREN®= =Q (GAMPERL 1982) (50). Indeed, ) 0020% > Ss 8 ELLERMAN and MORRISON- 2 Q IE = = a ö S => = Susi ScoTT (1951) considered these [64 H . IRRE © ae Ssao0nsıse 2 43 species difficult to define be- Sur z . ö o ki iR SL uns (52) in these species is absent in aguanTy Nas all other members of the genus I. EN EN Een en SEI Swesge Ei (BoropkIn and KHLYBORA 8 m Bose > 5 SS Eee 1980; Rausch and RauscH SPOF] - cc fa (e) N E < % Höhenstufen Abb. 2. Verteilung der Beutebelege von W2 und M2 auf die Höhenstufen 40 m Risse (In = 61) A Tagespeilungen er (n = 506) Pw] E © o 20 e a 10 bis 800 900-4099 1100-1299 4300-1499 4500-1699 über 1700 Höhenstufen Abb. 3. Relativer Höhenvergleich der Tagespeilungen und der Beutebelege von W2 und M2 Zum Prüfen der (zweidimensionalen) Verteilungen der Rifßfundorte (Abb. 1) wurden sie auf die Nord-Süd- bzw. West-Ost-Achse der Streifgebiete projiziert, klassiert und mit einer hypothetischen gleichmäßigen Verteilung und mit den Tagespeilungen verglichen. Die Rißfundorte wichen nicht signifikant von einer gleichmäßigen Verteilung ab (alle P>0.16). Der Vergleich der Tagespeilungen (W2: n = 295, M2: n = 263) und der Risse zeigte nur für M2 in Nord-Süd-Richtung einen signifikanten Unterschied (P = 0.020): Im südlichsten Teil seines Streifgebiets machte M2 mehr Beute als nach der Aufenthaltshäufig- keit zu erwarten gewesen wäre (übrıge P>0.21). Abb. 2 veranschaulicht die Verteilung der Risse von W2 und M2 auf die Höhenstufen. Die mittlere Fundorthöhe der Rehe (1024 m) lag hochsignifikant tieter als die der Gemsen (1310 m) (P<0.001). Die Fundorthöhe der erbeuteten Paarhufer für die zwei Luchse getrennt ergibt für M2 eine signifikante Differenz (P< 0.001), jedoch nicht für W2 (P = 174 U. Breitenmoser und H. Haller 60 Risse (n = 64) 50 = Tageslager ) (n = 509) N [®] Fa 40 o% E 30 le )} & 5) u | ET un } eo 4) & [eb] > 10 0.202.088 0-21 22-31 32-45 46-54 5.03 Neigungsstufen in Grad Abb. 4. Relative Verteilung der Beutebelege und der Tageslager von W2 und M2 auf die Neigungs- stufen 0.102), das nur wenige Gemsen rıß. Ein relevanter Unterschied zwischen den beiden Tieren besteht weder für die Höhenverteilung der Tagespeilungen (P = 0.068) noch für die der Beutebelege (alle Risse: P = 0.391, Rehe: P = 0.962, Gemsen: P = 0.161). Hingegen lagen die Risse hochsignifikant (P<0.001) tiefer als die Tagespeilungen (Abb. 3): Die durchschnittliche Fundorthöhe der Beutereste betrug 1150 m ü.M., die der zugehörigen Tageslager der Luchse am Rißstag 1236 m ü.M. (n = 74). Ein (allerdings nicht signifikanter) Unterschied zeigt sich auch zwischen den zu verschiedenen Jahreszeiten für M2 realisierten Beuteserien (Tab. 3): Sommer: Rißhöhe x = 1326 m, Tageslager x = 1328 m (n = 11); Winter: Rißhöhe x = 1239 m, Tageslager x = 1260 m (n = 10). Überreste von Beuten liegen häufig im Bereich des unteren Waldrands, wo sich vor allem die Rehe oft aufhalten. Aus dieser auch von Menschen stark begangenen Zone zieht sich der Luchs tagsüber in die ruhigeren, steilen Hanglagen zurück (vgl. Abb. 7): Von den 61 Rissen von W2 und M2 lokalisierten wir 36,1 % ın Zivilisationsnähe (näher als 500 m horizontal und 200 m vertikal an Dauersiedlungen oder regelmäßig befahrenen Straßen), von 509 Tagespeilungen hingegen nur 13,8% (Verschiedenheit hochsignifikant, P<0.001). Während die Beuten im engeren Untersuchungsgebiet überzufällig häufiger in Zivilisationsnähe lagen als nach dem Geländeangebot (ermittelt durch 281 Rasterpunkte) erwartet (P = 0.025), verhielt es sich mit den Tageslagern gerade umgekehrt (P = 0.002). Das Gelände um die Tageslager wies eine durchschnittliche Neigung von 41° auf, die Fundorte der von den beiden Luchsen erbeuteten Huftiere nur 34° (P<0.001; Abb. 4). Sowohl die mittlere Neigung der Rißorte als auch die der Tageslager war steiler als die durchschnittliche Geländeneigung von 32° (P = 0.004 bzw. 0.003). Bezüglich der Vertei- lung der Risse und der Tageslager auf die Expositionen zeigten sich keine statistisch relevanten Unterschiede. Bei den Tagespeilungen hielten sich die Luchse zu 92 %-97 % ım Wald auf (W2: 92,6 % von 285, M2: 92,7 % von 233 Tagespeilungen). Die erbeuteten Tiere lagen häufiger außerhalb des Waldes: W2: 12 (36,4 %) von 33, M2: 11 (39,3 %) von 28 (Unterschiede zur Verteilung der Tageslager hochsignifikant mit P<0.001). Die beiden Luchse verließen den Wald zum Beutemachen zwar etwa gleich oft, aber nicht gleich weit: Die Risse von M2 fanden wir durchschnittlich 203 m außerhalb des Waldes, die von W2 nur 45 m. Beutebelege innerhalb des Waldes waren für M2 im Mittel 209 m vom Waldrand entfernt, für W2 153 m. Die durchschnittliche Entfernung (Horizontaldistanzen) zwi- Zur Nahrungsökologie des Luchses in den schweizerischen Nordalpen 175 schen Tageslager und Fundort der Beute am Rifßtag betrug für alle Luchse 332 m (minimal O m, maximal 1850 m, n = 72). Der Unterschied zwischen W2 (x = 410 m, n = 31) und M2 (x = 154 m, n = 25) war signifikant (P = 0.012). Luchse bewegen sich zum Jagen den Hangwaldgürtel entlang durch ihr Streifgebiet. Dabeı werden zentrale Gebietsteile etwa einmal pro Monat, periphere Räume wesentlich seltener besucht (zur Raumnutzung vgl. HALLER und BREITENMOSER 1986). Dadurch gelangt der Luchs immer wieder in Gebiete, in denen er längere Zeit nicht gejagt hat. Die mittlere Distanz von einem erbeuteten Paarhufer zu den Rissen der letzten 10 Tage war für W2 5721 m (n = 43), für M2 9704 m (n = 15), zu denen der vergangenen 100 Tage 8345 m (n = 145) bzw. 14 359 m (n = 103) (vgl. auch Tab. 4). Im Niedersimmental konnten W2 und M2 (zufällig) kurz nacheinander in dasselbe Gebiet gelangen. Bei der Betrachtung der zeitlichen und räumlichen Entfernung einer Beute zum letzten Riß des anderen Luchses zeigte sich eine hochsignifikante negative Korrelation (r = —0.376, P = 0.009, n = 39): Je weiter zurück der zuletzt gefundene Riß des Artgenossen lag, desto näher war der Platz der Erbeutung. Dabeı war die Korrelation stärker, wen die beiden betrachteten Paarhufer- risse derselben Art angehörten (r = —0.568, P = 0.021,n = 13), und schwächer, wenn die Risse die jeweils andere Paarhuferart betrafen (r = —0.260, P = 0.101,n = 26). In 10 Fällen, in denen wir höchstens 10 Tage und 10 km vom letzten Riß des anderen Luchses entfernt ein erbeutetes Huftier fanden, handelte es sich jedesmal um die andere Art („gleich“ :,,un- gleich“ = 0:10, erwartete Verteilung = 4,7:5,3, Differenz signifikant mit P = 0.013). Erbeutungsfrequenz, Nutzung und Beutewahl Die vier Beuteserien (Tab. 2 und 3) vermitteln einen Eindruck von der Häufigkeit, mit der ein Luchs Beute reißt. In den Beuteserien sind alle während den Phasen der intensiven Überwachung gefundenen Risse berücksichtigt sowie vier Nahrungsbelege (2 Rehe, 1 Feldhase, 1 Eichhörnchen), die anhand der Kotanalysen nachgewiesen werden konnten und mit Sicherheit nicht mit einer der gefundenen Beuten identisch sind. Die Serien 2 bis 4 erstreckten sich jeweils über einen Zeitraum von drei Monaten, die erste Serie (Tab. 2a) nur über 55 Tage: Nach dem Fang von W2 (6.2. 1984) stellte sich heraus, daß es zwei Jungtiere führte, was Gelegenheit bot, die Nahrungswahl einer Familie (= Weibchen mit Jungluchsen) zu betrachten. Eines der Jungtiere (W3) konnte am 29. 2. 1984 ebenfalls mit einem Sender ausgerüstet werden. Es (und offenbar auch der zweite Jungluchs) trennte sich am 30.3. 1984 von seiner Mutter; das am 31.3. gerissene Reh nutzte W2 alleın. Die überwachten Jungtiere (W3 und W4) waren von Beginn ihrer Selbständigkeit an in der Lage, ebenfalls Paarhufer zu erbeuten. Zu größeren Beutetieren kehrt der Luchs mehrmals zurück. Das erfolgt oft schon ın der Dämmerung oder kurz nach Anbruch der Dunkelheit: Bei 10 Fängen an Rissen kamen die Luchse 7 Mal vor 2000 Uhr, zweimal zwischen 2100 und 2300 Uhr und nur einmal nach Mitternacht. Etwa ein Drittel der größeren Beuten (23 von 64 noch zu beurteilenden Fällen) waren mit Schnee, Laub oder Gras zugedeckt. Von einem vollständig genutzten Paarhufer bleiben nur Fell, Kopf, stärkere Knochen und der Verdauungstrakt übrig. Muskelfleisch, innere Organe (mit Ausnahme von Magen und Darm) und kleinere Knochen wie Rippen werden verzehrt (Abb. 9). Die in einer Nacht aufgenommene Nahrungsmenge war sehr unterschiedlich und lag zwischen einem und mehreren Kilo Fleisch. Ein adulter Luchs vermag ein Reh oder eine Gemse in 2 bis 3 Nächten vollständig aufzufressen (vgl. M2 in Tab. 3), vor allem dann, wenn er in unmittelbarer Nähe der Beute lagern kann. Selbst wenn er sich tagsüber zurückziehen muß, verweilt er während der Nacht oft längere Zeit am Riß. Am 7.4. 1983 hatte W1 eine Gemse erbeutet und ihr Tageslager 500 Meter entfernt in einem steilen, felsigen Hangwald gewählt. Am Abend um 2010 Uhr kehrte die Luchsin zur Beute zurück und fraß während einer Stunde (Abb. 8). Danach hielt sie sich in der unmittelbaren Umgebung auf und ging noch mehrmals an den 176 U. Breitenmoser und H. Haller Tabelle 2 Beuteserien von W2 ım Winter Beutetier Nutzung Distanz zur letzten Beute Beleg! Art Sex Alter Tage Prozent Meter Tage a. Serie 1: W2 mit zwei Jungtieren vom 5. 2. 1984-31. 3. 1984 2 257 _ 100 364 4 769 1718 Du 2 Eichhörnchen Gemse Reh Reh Reh Gemse Reh Reh Gemse Reh Reh 9 025 5 920 195 1439 4 829 493 7 568 12 476 1 861 2 050 42. u2. n2. „2. 2: 22. 2. a2: 2: 49: 3: ED} rd: ae 2). Sau sms sis een De Dr you | wDrerme Pet D + ooDDynDy | nme PNOVVHr-PE£DHrTFHFEN | WMNVG | BBBB5BB5 5 5 5 5 5 5 5 5 5 b. Serie 2: W2 vom 1.1. 1985-31. 3. 1985 Reh 2 Reh Reh Reh Reh Feldhase Reh Reh Reh Reh Reh Reh Feldhase Reh Gemse ee) 3 545 ZI SSELEQSı So X= © ERDE a ae a Sr > RD \ “2 : (I di Pferdeantilopen (Hippotragus equinus) äsen unter Wasser methode - für Hippotragus sicherlich eine Besonderheit dar. Im „Book of Antelopes“ von SCLATER and THomas (1899/1900) findet sich lediglich das Zitat: “The Roan Antelope is . amuch more regular drinker than the Sable. A troupe - ıf undisturbed, comes every evening to about the same spot on same favourite sluit of standing water”. Literatur HALTENORTH, TH.; DILLER, H. (1977): Säugetiere Afrıkas und Madagaskars. München: BLV. SCLATER, PH.; THoMmAs, O. (1899/1900): The Book of Antelopes IV. London: R. H. Porter. Anschrift des Verfassers: Dr. rer. nat. WOLFGANG GEWALT, Direktor des Zoologischen Gartens Duisburg, Mülheimer Str. 273, D-4100 Duisburg 1 BEKANNTMACHUNG The Fifth International Theriological Congress will be held from 22. to 29. August, 1989 in Rome, Italy. For detailed annoncement write to: The secretariat of the 5th International Theriological Congress c/o Dipartimento de Biologia Anımale e dell’Uomo Universitä di Roma “La Sapıienza” Viale dell’Universitä 32 1-00185 Roma (Italy). BIIETEIBESERESTIUTNIEGEN Brüm, V.: Vergleichende Reproduktionsbiologie der Wirbeltiere. Berlin-Heidelberg- New York-Tokyo: Springer-Verlag 1985. 387 S., 109 Abb. DM 58,-. ISBN 3-540-13816-1 Das Buch basiert auf dem inhaltlich erweiterten Manuskript einer Vorlesung, und es wendet sich an Biologiestudenten im Hauptstudium. Die Titel der 9 Kapitel lauten: 1. Einführung - 2. Die verglei- chende Anatomie der Urogenitalsysteme - 3. Die Bildung der Gameten — 4. Befruchtung und frühe Entwicklung - 5. Die vegetative Regulation der Reproduktion — 6. Reproduktionsverhalten — 7. Die innere Brutpflege der Submammalier — 8. Die innere Brutpflege der Säugetiere — 9. Aufßsere Brut- pflege. Diese Überschriften lassen bereits erkennen, daß vergleichenden Aspekten und biologischen Besonderheiten ein hoher Stellenwert eingeräumt wird, und darin liegt m.E. der Wert dieses Buches. Gemindert wird er allerdings schon dadurch, daß ım Hinblick auf Taxonomie und systematische Gruppierung z.T. recht BE neioncl verfahren wird. Wenn Brüm darüber hinaus z.B. beim Vergleich der Bauprinzipien männlicher Urogenitalsysteme aufgrund unterschiedlicher Komplexität eine Reihung ZLepidosiren-Protopterus-Holostei einschließlich (!) Polypterus-Teleostei vornimmt und dabeı von einer Entwicklungsreihe spricht, werden phylogenetisch denkende Zoologen wohl kaum folgen können. Umfang und Tiefe der verschiedenen Themen variieren sehr. Während z.B. die vergleichende Anatomie der Urogenitalsysteme sehr ausführlich beschrieben wird, werden andere Bereiche recht stiefmütterlich abgehandelt. Der Stoffwechsel der Steroidhormone fehlt so gut wie ganz, seine Bedeutung für das Reproduktionsverhalten ebenfalls, über Epıphyse und Gonadenfunktion ließe sich wahrlich mehr und Neueres sagen, veraltet sind auch Aussagen über Reproduktionszyklen, die Biotechnologie in der Reproduktion ist nicht, auch nicht in ihren biologischen Grundlagen, erwähnt. Als generellen Ver her ich empfunden, daf im Text Quellenangaben fehlen. Der interessierte Leser — und dazu dürften auch die Studierenden im Hauptstudium zählen — hat so kaum die Möglichkeit, sich über spezifische Probleme detaillierter zu informieren. Ungewöhnlich und ırritie- rend ıst der Gebrauch von Anführungszeichen: Sie werden einmal — wıe üblich - zur Kennzeichnung verfremdet verwendeter Wörter benutzt, daneben aber auch beı der Einführung wissenschaftlicher Termini, die man z.B. besser durch Fettdruck hätte hervorheben können. E. Haase, Kiel FEDDERSEN-PETERSEN, DorRIT: Hundepsychologie. Wesen und Sozialverhalten. Mit einem Vorwort von Konrad Lorenz. Stuttgart: Franckh 1986. 104 S., 41 Abb., DM 26,-. ISBN 3-440-05589-2 Dieses Buch hat gefehlt und nützt hoffentlich dem Hund und seiner Beziehung zum Menschen. Auf gut 100 Seiten liefert die Autorin eine Übersicht fast aller Verhaltensbereiche des Hundes und legt dabei besonderen Wert auf Gebiete, die für die Erziehung und für die Beziehung zum Menschen bedeutsam sind. Abstammung, Haustierwerdung und Vergleiche mit Wolf und Schakal werden auf den ersten dreißig Seiten behandelt. Im Kapitel über die „Sprache des Hundes“ wird vornehmlich das körperliche und akustische Verhaltensinventar vorgestellt. Bei der „Verständigung zwischen Hund und Mensch“ findet man neben grundsätzlichen Informationen wichtige Aussagen und Überlegungen der Autorin zum „Verständnis“ des Hundes für menschliche Gefühle, über die ‚‚Seele‘‘ des Hundes und über die (oft therapeutische) Funktion als Ersatz für eine zwischenmenschliche Beziehung. Frau FEDDERSEN-PETERSEN tritt hier im besten Sinne des Wortes als Anwältin des Hundes auf. Die in jüngerer Zeit häufiger diskutierte Aggressivität wird mit Hinblick auf fehlerhafte Erziehung und falsch verstandene Zucht (Bullterrier als Beispiel anhand einer neueren Dissertation) kritisch erörtert. In einem kurzen Abschnitt über Verhaltenstörungen werden genetische und umweltbedingte Ursachen sowie Möglichkeiten der Behandlung geschildert. bschiehend folgen wertvolle Daten zur Welpen- aufzucht, die für eine erfolgreiche Einbindung des Hundes in seinen menschlichen Sozialverband beachtet werden müssen. Ich habe dieses Buch mit großer Zustimmung gelesen, frage mich aber, ob die Autorin die Gruppe von Hundehaltern erreichen wird, die ihre Tiere völlig vermenschen, da das Buch in einer recht anspruchsvollen Sprache geschrieben ist. Dringend empfohlen sei es allen, die beruflich mit Hunden umgehen, und natürlich jenen, die ihre Verantwortung in der Beziehung zu diesem „besonderen Haustier“ (F.-P.) begriffen haben. Zwei Anmerkungen zum Schluß: Der Begriff „Wesen“ wird von der Autorin immer in Parenthese gesetzt und als „wissenschaftlich eigentlich unhaltbar‘‘ angesehen. Eigenartigerweise erscheint er trotzdem auf der Titelseite. Offenbar mochte der Verlag hier der Autorin nicht folgen. Die Zeichnungen der Autorin sind eine erfreuliche Bereicherung für dieses gelungene Buch. M. KöHncke, Hamburg Deutsche Gesellschaft für Säugetierkunde 60. Hauptversammlung in Stuttgart, 28.September bis 2. Oktober 1986 Kurzfassungen der Vorträge und Posterdemonstrationen. Herausgegeben von Fritz Dieterlen, Stuttgart. Zusammenstellung: Pia Wilhelm, Stuttgart. 1986. 99 Seiten. Kartoniert 24,-— DM Ein Hauptziel der Deutschen Gesell- schaft für Säugetierkunde ist, auf ihren Jahrestagungen über Säugetiere arbeitende Wissenschaftler verschie- denster Fachrichtungen zusammen- zuführen, den Gedanken- und Erfah- rungsaustausch anzuregen, um so Erkenntnisse aus den einzelnen Forschungsgebieten zu integrieren. Da auf einer Jahresversammlung nicht alle Fachrichtungen gleichermaßen vertreten sein können, wurden - wie schon in den Vorjahren — wieder Schwerpunkte gesetzt: Erstens, die Paläontologie und, damit verbunden, Fragen der Systematik und Evolution der Säugetiere. Zweitens, ein Themenkreis, der vor- wiegend physiologische und etholo- gische Probleme bei Säugetieren um- faßt, und drittens, ökologische Themen, besonders zu Fragen des Artenschutzes und der Wildbiologie. Sehr erfreulich ist, daß eine größere An- zahl von Beiträgen den Fledermäusen gewidmet ist- der bei uns am stärksten bedrohten Säugetierordnung. Zu beziehen durch jede Buchhandlun Verlagsbuchhandlung PAJL Paul Parey Spitalerstraße 12 SER 2000 Hamburg 1 Erscheinungsweise und Bezugspreis 1987: 6 Hefte bilden einen Band. Jahresabonnement Inland: 298,- DM zuzüglich 13,80 DM Versandkosten, Gesamtpreis 311,80 DM einschließlich 7% Mehrwert- steuer. Jahresabonnement Ausland: 298,— DM zuzüglich 24,— DM Versandkosten. Das Abonne- ment wird zum Jahresanfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, Spitalerstraße 12, D- 2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesellschaft für Säugetier- kunde“ erhalten die Zeitschrift unberechnet ım Rahmen des Mitgliedsbeitrages. Z. Säugetierkunde 52 (1987) 3, 129-200 Georg Wilhelm Rieck / Konrad Zerobin Zuchthygiene Rind Pareys Studientexte Nr. 46. Von Prof. Dr. med. vet. Georg Wilhelm Rieck, emerit. Ordinarius für Erbpatho- logie und Zuchthygiene, ehemals Direktor des Instituts für Zuchthygiene und veterinärmedizinische Genetik der Justus-Liebig-Universität Gießen und Prof. Dr. med. vet. Konrad Zerobin, Direktor des Instituts für Zuchthygiene der Uni- versität Zürich. 1985. 149 Seiten mit 39 Abbildungen und 21 Tabellen. Balacron broschiert DM 29,— Das junge Fachgebiet Zuchthygiene schließt die Lücke tierärztlicher Tätigkeit, die von dem Fach Tierzucht und Tierhaltung einerseits und den klinischen Fächern Gynäkologie und Andrologie andererseits nicht oder nur gelegentlich abgedeckt wird. Dementsprechend spiegelt der Aufbau dieses Studientextes das Basiskon- zept des akademischen Unterrichts des Faches Zuchthygiene in der Veterinär- medizin und Tierzucht wider. Seine Besonderheit liegt in der Auswertung und syn- optischen Zusammenfassung der Fakten, die zur Prophylaxe und schließlich zur Kontrolle der Fruchtbarkeit im Herdenverband beitragen können. Damit wird eine Schwerpunktbildung der Information über dieses für die tierärztliche Großtierpraxis so wichtige Gebiet erleichtert. Hier werden nicht nur die Studenten der Veterinär- medizin angesprochen, sondern auch die Studierenden der Landwirtschaft, um sie mit der tierärztlichen Problematik auf dem Gebiet der Zuchthygiene vertraut zu machen. Auch Absolventen höherer landwirtschaftlicher Lehranstalten können die- sen Studientext zur Weiterbildung nutzen. Aus dem Inhalt: Begriff und Lehrinhalt des Fachgebietes Zuchthygiene — Zucht- hygienische Literatur - Ökonomische Bedeutung der Fruchtbarkeit des weiblichen Rindes - Biologische und betriebswirtschaftliche Voraussetzung einer ungestörten Fertilität — Die Fruchtbarkeit beeinflussende Faktoren -— Maßnahmen zur Reduzie- rung von perinatalen Kälberverlusten - Erhöhung des Kälberaufkommens - Besa- mung und Bestandsfruchtbarkeit — Organisation und gesetzliche Grundlagen der Besamung - Auswahlkriterien von Jungbullen auf Besamungstauglichkeit - Zyklussynkronisation (ZS) - Genetik der Fruchtbarkeit -— Spezielle Genetik der Fer- tilität — Herdensterilität (H. St.) - Ursachen - Anhang: Tierzuchtgesetz vom 20. April 1976 (Auszug). PAUL P/REY Berlin und Hamburg Vol. 52 (4), 201-264, August 1987 ISSN 0044-3468 @2127Ar Be = ITSCHRIFT FÜR SAUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY To our Readers — An unsere Leser > Klima, M.; Bangma, G. C.: Unpublished drawings 0 eibryos from the Hill Collection and some problems of marsupial ontogeny. — Unveröffentlichte Zeichnungen von Marsupialia-Embryonen aus der Hill-Sammlung und einige Probleme der Ontogenese vön Marsupialia” 201 Matthias, U.; Witte, G. R.: Untersuchungen zur Genese des Maulwurfshaufens bei Talpa europaea (L.). — Studies on the formation of the molehill of Talpa europaea (L.) 212 Schanz, F.; Erkert, H. G.: Resynchronisationsverhalten der circadianen Aktivitätsperiodik von Galagos (Galago senegalensis, Galago crassicaudatus garnettii). — Re-entrainment of circadian activity rhythms in galagos (Galago senegalensis, Galago crassicaudatus garnettii) 218 Rothe, H.; König, A.; Darms, K.; Sieß, Margaretha: Analysis of litter size in a colony of the common marmoset (Callithrix jacchus). — Analyse der Wurfgröße in einer Kolonie von Weißbüscheläffchen (Callithrix jacchus) 227 Corti, M.; Merani, Maria Susana; de Villafahe, Gloria: Multivariate morphometrics of Vesper mice (Calomys): preliminary assessment of species, population and strain divergence. — Multivariate Morphometrie an Vesper- mäusen (Calomys): vorläufige Abschätzung von Arten, Populationen und Zuchtstammunterschieden 236 Gemmeke, H.; Radtke, Margot; Niethammer, J.: Zur innerartlichen Proteinvariation bei der Waldmaus (Apodemus sylvaticus). — On the intraspecific variation of some proteins in the wood mouse (Apodemus sylvaticus) 242 Gill, Ayesha; Petrov, B.; Zivkovie, S.; Rimsa, Desanka: Biochemical comparisons in Yugoslavian rodents of the families Arvicolldae and Muridae. — Biochemische Vergleiche von jugoslawischen Rodentia der Familien Arvicolidae und Muridae 247 Korpimäki, E.; Norrdahl, K.: Low proportion of shrews in the diet of small mustelids in western Finland. — Geringer Anteil von Spitzmäusen in der Nahrung kleiner Musteliden aus West-Finnland 257 Wissenschaftliche Kurzmitteilungen Fuchs, W.: Haselmäuse (Muscardinus avellanarius) in Nestern freibrütender Singvögel. — Dormice (Muscardinus avellanarius) in nests of open nesting song-birds 261 Fischer, M. S.: Die Oberlippe der Elefanten. — The upper lip of elephants 262 Buchbesprechungen 263 % Verlag Paul Parey Hamburg und Berlin HERAUSGEBER/EDTEORS P. J. H. van BreeE, 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. Thenıus, Wien - W.VERHEYEN, Antwerpen SCHRIFTLEITUNG/7EDITORTANIZOREI@E 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 ın dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, insbesondere die der Übersetzung, des Nachdrucks, des Vortrags, der Entnahme von Abbildungen und Tabellen, der Funk- und Fernsehsendung, der Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei- ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch im Einzelfall grund- sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines Beitrages ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber- rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen. Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of the fırst page of an article in this journal indicates the copyright owner’s consent that copies of the article may be made for personal or internal use, or for the personal or internaluse of specific clients. This consent is given on the condition, however, that the copier pay the stated percopy fee through the Copyright Clearance Center, Inc., 21 Congress Street, Salem, MA 01970, USA, for 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 Mit einer Beilage des Gustav Fischer Verlages und einer Beilage des Verlages Paul Parey. Fortsetzung 3. Umschlagseite © 1987 Paul Parey. Verlag: Paul Parey, Hamburg und Berlin. Anschriften: Spitalerstr. 12, D-2000 Hamburg 1; Lindenstr. 44-47, D-1000 Berlin 61. —- Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., D-2240 Heide/Holst. An a Leser der Zeitschrift für Säugetierkunde ei der Entwicklung der Weltwirtschaft seit Anfang der 80er Jahre wurden die Anschaffungsetats der wissenschaftlichen Bibliotheken international teilweise mehrfach drastisch verringert. Ein Ende dieses Trends ist gegenwärtig nicht erkennbar - erst recht nicht eine Umkehr. Die Bibliotheken müssen, der Not gehorchend, mit einer wesentlichen Reduzierung der Zahl der abonnierten Zeitschriften reagieren. Damit sinkt auch die Verkaufsauflage der einzelnen Zeitschriften, die von Abbestellungen betroffen sind. 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 oz Ki werden. mit Ihrem Bibliothekar sprechen und ihm deutlich machen, daß die schnelle Verfügbarkeit der Zeitschrift innerhalb Ihrer Bibliothek für die Effektivität Ihrer Forschungsarbeit sowie tung ist. Wir hoffen auf einen Bro unserer gemeinsamen Bemühungen. Die Hunausoruen. SE | | Verrac Pauı Parey a BR R ni Tan er ae To the Reader of tel International Be of Matisiakan Be Ba ilerons and m the en Se will follow. / ‘ [Fr ’ 2 „ ; " ? # 2 7 h a Fi \ - fi h » “ Unser kat an Sie ganz persönlich geht dahin, daß Sie möglichst bald Ar “ für die Ausbildung des m Nachwuchses von ausschlaggebender Bu developmer Er es economy since the a ah ne acquisition | | budgets of senifte libraries all over the world often were drastically reduced. At present there is no end of this, tendency i in sight - a change for the better even less. Thelibrariesre com to answer with a substantial reduction of their journal subscriptions. Asaresult also the edition-runs of individual journals affected by cancellations decrase. Bydecreasing | edition-runs, the cost per copy increases and consequently the retail price - ‚more a 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 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 him 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 VERLAG PAUL PAREY Unpublished drawings of marsupial embryos from the Hill Collection and some problems of marsupial ontogeny By M. Kııma and G. C. BANGMmA Institute of Anatomy, Johann-Wolfgang-Goethe-University, Frankfurt/M. and Hubrecht Laboratory, International Embryological Institute, Utrecht Receipt of Ms. 22. 8. 1986 Abstract Some 80 year-old original drawings of intrauterine embryos, neonates and pouch young of marsupials by A. Cronin are presented which belong to the Hill Collection, Utrecht, but were never published. They comprise 11 drawings of the brush-tailed phalanger Trichosurus vnlpecula, eight drawings of the koala Phascolarctos cinerens, two drawings of the common vombat Vombatus ursinus and one drawing of the red-necked pademelon Thylogale thetis. The external morphology of the early developmental stages of these specimens is compared with the inner morphogenesis of some organs, particularly of the fore-lımb, shoulder girdle and sternal apparatus. Functional and evolutionary correlations are discussed. Past and present of the Hill Collection JAMES PETER Hırr was born on the 21st of February 1873 at Kennoway in Scotland. In 1889 Hırı entered the University of Edinburgh to study Science. In 1892, although he had not yet graduated, he was appointed Demonstrator of Zoology and Botany at the Biology Department of Sydney University, Australia. Once there, he started to collect all kinds of animals for his courses and his own research; this marks the beginning of his collection. After his graduation to B. Sc. ın 1898 Hırı stayed in Australia until 1906, when he returned to England to become Professor of Zoology at the University College of London. In 1921 he was appointed Professor of Embryology and Histology at the same College. He retired in 1938 but continued his scientific work until his death on the 24th of May, 1954. Hırr’s bibliography comprises 58 papers, the greater part of which particularly deals with the early development of marsupials and monotremes, ın which he took a special interest. His research had one central theme, which he expressed as follows: “the data of embryology, properly interpreted, can and do furnish striking and demonstrative evidence of genetic relationships’” (Hırr 1932). An important part of his collection was assembled during his Australian period. In later years it was extended with species from Africa and South America, particularly by means of purchases, the bills of which are still to be found among the original Hırr documents. Hırr also personally collected material during expeditions, e.g. in Brasil in 1913. Although all classes of vertebrates are represented ın his collection, the great majority of species belongs to the mammalian orders of the mono- tremes, marsupials and primates. At the time of his death Hırr’s collection was partly in his own residence and partly at the London University. The fear that careless use and neglect would sooner or later entail the loss of the collection prompted Dr. CATHERINE J. Hırr, Hırr’s daughter and herself an embryologist, to find ways to prevent this. Dr. P. D. Nıruwkoor, who was at that time Director of the Hubrecht Laboratory, agreed to house the collection in the Hubrecht Laboratory on the condition that it would be completely catalogued. The Hubrecht U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5204-0201 $ 02.50/0 Z. Säugetierkunde 52 (1987) 201-211 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 202 M. Klima and G. C. Bangma Laboratory already possessed another large embryological collection, assembled by Pro- fessor A. A. W. HuBrREcHT (1853-1915) mainly in the Indonesian Archipelago. It took Dr. Hırı ten years to catalogue her father’s collection. After she had completed this work, the collection was transferred to the Hubrecht Laboratory in 1967. The Central Embryological Collection is still subdivided into two separate entities, the Hill Collection and the Hubrecht Collection. It consists of material stored in alcohol and in paraffin (ca. 2000 flasks), of microscopical slides (ca. 80 000), and of original catalogues, note-books, drawings, photographs and other documents. The material pertains to the normal development of more than 600 vertebrates species. In addition some special collections related to experimental-morphological investigations have been incorporated, e.g. the collections of SPEMANN, MANGOLD, ROTMAN, RAVEN and GRÜNEBERG (collection of mouse mutants). All this material is expressly available for morphological, descriptive and comparative embryological investigations. A concise catalogue of the Central Embry- ological Collection has been edited by Dr. E. C. BOTERENBROOD (1977). A copy of this catalogue is available from Dr. G. C. BanGma, Curator of the collection. Currently an increasing number of scientists use the collection for such studies. In this respect it is noteworthy that in 1911 both HUBrRECHT and Hırr stood at the birth of the Institut International d’Embryologie, now called the International Society of Develop- mental Biologists (ISDB). It ıs satisfactory to see that after so many years the collections of these noted embryologists still serve the original aims of this Society and even provide new inspiration to present-day embryologists. Drawings of marsupial embryos by A. Cronin Apart from the written entries, the notebooks of J. P. HıLı contain some photographs and many simple drawings of marsupial embryos, neonates and pouch young. However, only a few of these have actually been used in J. P. Hırr’s publications, for example simple outline drawings of Perameles nasuta (Hırı 1898) and Thylacıs obesolus (Hırı 1901). A simple outline drawing of Dasyurus quoll was published, simultanously showing the Fig. 1. Brush-tailed phalanger Trichosurus vulpecula, intrauterine embryos. 1: MS 253, 5 mm ?, 2: MS 250, 5.0 mm Unpublished drawings of marsupial embryos 203 Fıg. 2. Brush-tailed phalanger Trichosurus vulpecula, intrauterine embryos. 1: MS 257, 7.5 mm, 2: MS 259, 6.0 mm, 3: MS 257 ?, 7.5 mm, 4: MS 265, 8.5 mm, 5: MS 269, 10.25 mm, 6: MS 273, 11.5 mm, 7: MS 283, 14.0 mm, 8: neonate MS 286, 15.0 mm, 9: neonate MS 289, 15.0 mm 204 M. Klima and G. C. Bangma arrangement of the foetal membranes (Hırr 1900). A number of very sımplified drawings of intrauterine embryos, neonates and pouch young of Trichosurus vulpecula are contained in the work of Fraser and Hiırr (1915). The only publication containing complete and detailed illustrations of the external shape of neonates and pouch young ıs the publication of Hırı and Osman Hırr (1955) on Dasyurus quoll. Obviously the authors were aware of the unusually good quality of these illustrations, because they wrote: “The beautifully executed pencil drawings of the pouch young... were made by Mr. A. Cronın of Sydney.” However, it must be said that the reproductions in the quoted paper are not satisfactory, and that many delicacies were lost due to high-contrast printing. This is shown by comparison with the original drawings, 17 altogether, which are in the Hill Collection. There are also other pencil drawings by A. Cronin in the Hill Collection, which to our knowledge have never been published. In reference to the Marsupialia there are some drawings of the ovary, the uterus, the foetal membranes and the placental structures and some additional drawings of complete ıintrauterine embryos, neonates and pouch young. Besides the 17 original drawings of the eastern dasyure Dasyurus quoll, mentioned above, there are 11 drawings of the brush-tailed phalanger Trichosurus vulpecula, two drawings of the koala Phascolarctos cinereus, eight drawings of the common vombat Vombatus ursinus and one drawing of the red-necked pademelon Thylogale thetis. Only a few drawings bear a signature. Nevertheless, their distinctiveness leaves no doubt that all of them were drawn by A. Cronin. The years of their production are uncertain, because they bear no date. However, we may assume that it must have been around the turn of the century, when J. P. Hırr was ın Australia collecting the embryological material. This assumption ıs also supported by the following note, taken from the publication of Hırr and Osman Hırr (1955): “... The material at our disposal was collected by one of us (J. P.H.)..... during the years 1895-1905, when he was Demonstrator of Biology in the University of Sydney.” Additionally, some of the drawings bear handwritten notes by J. P. Hırı with annual dates between 1897 till 1902. Also going by the condition of the drawing paper one would judge their age to be between 80 and 90 years. The pencil drawings, measuring only 5-15 cm in length, have been carried out in a minute pencil-technique which reveals many details. The Fig. 3. Koala Phascolarctos cinereus, intrauterine embryos. 1: MS 346, 7.5 mm, 2: MS 347, 9.0 mm Unpublished drawings of marsupial embryos 205 Fıg. 4. Koala Phascolarctos cinerens, intrauterine embryos. 1: MS 345, 4.5 mm, right lateral, 2: left lateral view, 3: MS 348, 11.0 mm, 4: MS 350, 13.5 mm, 5: MS 351, 17.0 mm, 6: MS 352, 18.0 mm appearance of the objects indicates that the drawings were made from fresh material or immediately after fixation. As most objects are very tiny ı.e. about 5-15 mm long, there ıs no doubt that they had to be drawn with the aid of a magnifying glass or a microscope. The drawings of marsupial embryos, neonates and pouch young by A. Cronın show a high degree of technical perfection and at the same time attest to the high artistic level of the artist. It ıs regrettable that they have not been published after their completion and therefore have remained without significance for science. Not only would they have enriched the store of knowledge at that time, they would also have been superior to the then technically imperfect photography. If we publish the drawings today, about 80 to 90 206 M. Klima and G. C. Bangma years after they were made, it is not for historical reasons only. They are in their way still much more instructive than some of the recent, technically sophisticated photographs. Additionally, there are not many published usefull depictions of marsupial embryonic stages and therefore we consider it appropriate to present the drawings to a larger public. The most important reason for the publication of the drawings however, is that they depict individuals of which we have extensive series of microscopical slides. The drawings allow a comparison between the histological results and the exact rendering of the external morphology of the individuals examined. In this paper we attempt such a comparison with regard to the morphogenesis of the sternal-shoulder apparatus. This topic is treated in more detail in another contribution as well (KLıma, submitted). External morphology of the intrauterine embryos, neonates and pouch young of marsupials Compared with the placentals, marsupials are born in a quite unusual condition. On the one hand they may be characterized as immature embryos, while on the other hand they are endowed with some highly specialized features. This led some authors to the convic- tion that they reflect genuine larval stages (LECHE 1890). However, this opinion gained little approval. In any case “marsupials at the time of birth are true embryos, anatomically and physiologically. All species of marsupials give birth of tiny young, ranging in weight from 10 mg to about 750 mg. An individual neonate always has less then 0.2% of the mother’s body weight (1% for the litter as a whole). Placentals in contrast are relatively much larger at brith, with individual neonates in small species comprising 2-3 % of the mother’s body weight (20% or more for the litter).”” (Quoted from LILLEGRAVEN 1984, slightly changed with statements from Brock 1964; TYNDALE-BiscoE 1973; EISENBERG 1981; MırLLar 1981; RusseL 1982.) The most important features of marsupial neonates were summarized by Hırı and Osman Hiırr (1955) in their investigation on Dasyurus as follows: “Notable at bırth are 1. the relatively advanced state of the fore-limbs, 2. the presence of deciduous claws upon the manual digits and their later replacement by definitive claws, 3. the presence of the oral shield as a specialisation upon the muzzle, 4. the unique cervical swelling, 5. special features of the tongue and larynx, with the intranarial epiglottis, 6. the reptilian state of the lungs, 7. the advanced condition of the stomach and duodenum compared with the gut, 8. the state of development of brain and sense organs, the olfactory parts being especially forward in development, together with those parts of the nervous system necessary for controlling the movements of the fore-limb, sucking and respiratory movements, 9. the indifferent state of the gonads.” Many of these characteristics must, with no doubt be regarded as special adaptations to the unusual process of parturition in marsupials. Thus, despite its small size and immature developmental state the neonate has to achieve a remarkable performance at birth. “The life of the pouch young is a hazardous one, beginning with the long journey into the pouch on day 1. This blind and naked creature must grasp and crawl its way from the vagına to the marsupium, an average distance of 5 cm (in Didelphis), with no help from the mother other than the prebirth preparation which consists of licking and cleaning the genital and pouch area and the removal of the amnionic membranes” (HunsAacker 1977). In this process the neonate’s fore-limbs play a decisive role. Compared with the body and the hind-limbs they are overproportionately large (Figs. 1-5). They are provided with pointed claws. They move alternately and by doing so drag the body in a wriggling manner together with the passively hanging minute hind-limbs. The whole distance from the vulva to the teat in the mother’s pouch is managed solely through the power of the fore-limbs. The body as well as the hind-limbs behave passive ın the process. Unpublished drawings of marsupial embryos 207 A rapid development of the anterior body half and the fore-limbs, respectively, may be detected even in the earliest developmental stages (Trichosurus Fig. 1: 1, 2; Phascolarctos Figs. 3: 1, 2 and 4: 1, 2; Vombatus Fig. 5: 1). Basıcally it is possible to observe a certain developmental predominance of the fore-limbs over the hind-limbs in the developmental process of all mammals. However, this is nowhere as pronounced as in the marsupoals. At birth the fore-limbs perform some kind of crawling movement (HARTMAN 1920; McCrapy 1938; HırL and Osman Hırr 1955; and others). This requires a powerful musculature and a solid skeletal support. Both have been demonstrated by the histological investigations. Among the muscle-groups this holds mainly for the pectoral musculature (CHEnG 1955). The skeleton of the shoulder girdle forms a broad breast-shoulder arch, which requires a large space between the fore-limbs. This indeed corresponds to the external shape of both the late intrauterine embryos and the neonates. Between the fore- limbs a broad chestwall extends, which is sometimes well padded with a distinet muscle bulge. The distance between the fore-limbs is significantly large as ıs easily observed from the front in both intrauterine embryos and neonates (Trichosurus Fig. 2: 7, 9; Phascolarctos Fig. 4: 5, 6; Vombatus Fig. 5: 2). The bases of the extremities are located in the lateral trunk area. This lateral trunk position is typical reptilian. Most certainly it is advantageous for the crawling movements. The extremities will shift their positions from lateral to medial only postnatally in the pouch young, after the breast-shoulder arch has been reduced. Gradually they come to show the characteristically narrow “mammalian posi- tion” under the trunk. The fore-limb digits are provided with sharp, bent claws. There are not yet definitive Fig. 5. Common vombat Vombatus ursinus, intrauterine embryos. 1: MS 360, 9.0 mm, 2: MS 361, 15.5 mm. 3: Red-necked pademelon Thylogale thetis, intrauterine embryo MS 408, 17.0 mm 208 M. Klima and G. C. Bangma organs, but only provisional, non-permanent structures, which are usually termed “larval claws’” (Hırr 1900) or “deciduous claws’”’ (Hırr and Osman Hırr 1955). They develop in late stages of intrauterine embryos and reach their maximum development at the neonate state (Zrichosurus Fig. 2: 7, 8, 9; Phascolarctos Fig. 4: 5, 6; Vombatus Fig. 5: 2; Thylogale Fig. 5: 3). In the pouch young they are initially reduced and later replaced by definitive claws. Their organic function ıs clear, namely to keep a firm hold in the mother’s abdominal hair during the crawling movements at birth (BArTon 1806; Owen 1841; BEARD 1897; Hırı 1898, 1900, 1910, 1918; HARTMAN 1920, 1923, 1928, 1952; LANGWOR- tHY 1928; McCrapy 1938; Hırn and Osman Hırr 1955; a.a.). The transıtional appearance of claws during the short time period around birth reveals obvious parallels with the temporary appearance of the breast-shoulder arch. This is also a provisional structure, which comes into being ı in the late stages of intrauterine development and reaches its maximum development in the neonate, to be reduced immediately after birth. Both structures belong to the same adaptive system which allows typical marsupial birth ın all its special aspects. Early development of the shoulder girdle and sternal apparatus in marsupials During marsupial embryogenesis the following structures can be recognized in the shoulder girdle and sternal apparatus: the scapula, the metacoracoid, the procoracoid, the first rıb, the paired sternal elements, the unpaired sternal element and the clavicle. All these elements together form a compact, uninterrupted arch ın both, intrauterine embryos and neonates. Immediately after birth, this arch ıs reduced in the pouch young. Then the solid connection between the left and right body halves ceases to exist, leaving only a loose connection vıa the clavicle. The metacoracoıd becomes the processus coracoıdeus scapulae. The procoracoid becomes the praeclavium. The unpaired sternal element fuses with the paired ones into the uniform sternal manubrium. The first rıb takes its usual position in the thorax. Thus the shoulder girdle and sternal apparatus as a whole show all the typical features of the adult marsupials as early as ın the pouch young. The embryonic shoulder girdle and sternal apparatus of the marsupials show strong sıimilarities with the conditions in primitive monotremes. However, in the pouch young and in adult marsupials they generally reflect the conditions in the phylogenetically progressive placentals (cf. BROoM 1897, 1899, 1902, 1912; Watson 1917; Hanson 1920; RoMER 1922; McCrapy 1938; CHEnG 1955; McKennA 1961; KLıma 1968, 1973, 1975, 1985). The phylogenetically determined condition of the shoulder girdle and sternal apparatus during embryogenesis as a relatively primitive structure consisting of many elements is used functionally. In nearly all of the investigated marsupials the reduction of existing elements ıs delayed until after birth (Dasyurus, Trichosurus, Phascolarctos, Vombatus, Petrogale, Thylogale, Wallabia, Macropus). The only exceptions are the peramelids 7hy- lacıs and Perameles (Kııma, submitted). Accordingly, the neonates posses a solid arch, running straight from one side of the body to the other. This breast-shoulder arch serves many tasks: 1. Some of its parts, particularly scapula and sternum, offer an attachment surface for the mighty musculature which moves the fore-lımbs. 2. In its extensive glenoıd cavity, which ıs formed by the scapula and the metacoracoid, the enormous humerus head is solidiy anchored. 3. Its medial parts, the sternum, the first rıb and the metacoracoid, form a pressure-resistant structure that prevents the humeri from shifting medially when the huge pectoralis musculature contracts. The morphogenesis of the shoulder girdle and sternal apparatus in marsupials takes place in the following manner: the paired structures inıtially originate in the lateral body wall and later slowly shift medially towards each other. Shortly before they reach the Unpublished drawings of marsupial embryos 209 4 ern 3 6 Fıg. 6. Ventral view of the shoulder girdle and sternal apparatus. 1: Brush-tailed phalanger Trichosurus vulpecula, intrauterine embryo 12.0 mm, 2: intrauterine embryo shortly before birth 13.0 mm, 3: pouch young 17.5 mm. 4: Koala Phascolarctos cinereus, intrauterine embryo 12.25 mm, 5: neonate 16.5 mm, 6: pouch young 24.0 mm. Abbrev.: a=acromion; cl=clavicle; g= glenoid cavity; icc= unpaired sternal element; mc = metacoracoid; pe = procoracoid; pel = praeclavium; pes = coracoid processus; rı,,,,; = rıbs; s = scapula; st = sternum; stl = paired sternal element medial plane, an unpaired structure is generated between them. It ıs indeed the inıtially unpaired chondral element of the interclavicle of ancestral forms which in the marsupials and placentals becomes the unpaired sternal element (KLımA 1973, 1985). In the final stages of intrauterine development this process ıs readıly visible insotar as the paired structures start to merge with the unpaired ones in the medial plane (Trichosurus Fig. 6: 1; Phascolarctos Fig. 6: 4). In due course the continually merging elements give rise to the 210 M. Klima and G. C. Bangma above mentioned continuous breast-shoulder arch. This reaches its maximum development in the intrauterine embryos and neonates immediately before and at birth (Trichosurus Fig. 6: 2; Phascolarctos Fig. 6: 5). Immediately after birth the metacoracoid begins to reduce, the sternum and first rıb decrease in size, and the breast-shoulder arch is dissolved. So even in the youngest pouch young there is no longer any solid connection of the body halves in the area of the shoulder girdle and sternal apparatus (Trichosurus Fig. 6: 3; Phascolarctos Fig. 6: 6). The fast reduction of the breast-shoulder arch is related to the further course of ontogenesis. While the pouch young develops further in the pouch its fore-limbs are barely used. Along with this decreasing functional significance, the initially overproportional size ıs reduced and the proportions will gradually change in favour of the hind-lımbs (SHArman and PıLTon 1964). Acknowledgements The support of the Hubrecht Laboratory, International Embryological Institute, Utrecht and of the Deutsche Forschungsgemeinschaft, Bonn, made it possible to carry out the investigations on the marsupial embryos from the Hill Collection. We are most grateful to Mrs. Dr. E. C. BOTERENBROOD, Dr. J. FAßBEr and Prof. Dr. P. D. NıEuwkoor, Utrecht, and to Mrs. R. GunTRuMm and Mrs. G. LEIBE, Frankfurt, for their valuable help. Zusammenfassung Unveröffentlichte Zeichnungen von Marsnpialia-Embryonen aus der Hill-Sammlung und einige Probleme der Ontogenese von Marsnpialia Bisher unpublizierte Originalzeichnungen von intrauterinen Embryonen, Neugeborenen und Beutel- jungen von Marsupialıa aus der Hill-Sammlung in Utrecht sind vorgestellt worden. Es handelt sich um etwa 80 bis 90 Jahre alte Bleistiftzeichnungen, Werke von A. CronIn aus Sydney. Es sind 11 Zeichnungen von Fuchskusu (Trichosurus vulpecula), 8 Zeichnungen von Koala (Phascolarctos cinereus), 2 Zeichnungen von Nacktnasenwombat (Vombatus ursinus) und 1 Zeichnung von Rothals- filander (Thylogale thetis). Die externe Morphologie dieser frühen Entwicklungsstadien ist verglichen worden mit der inneren Morphogenese einiger Organe, insbesondere der vorderen Extremität, des Schultergürtels und des Brustbeins. Funktionelle und evolutive Zusammenhänge werden diskutiert. Literature BArTon, B. S. (1806): Facts, observations and conjectures relative to the generation of the Opossum of North America. In a letter... to Mons. Roume of Paris. Philadelphia: T. and G. Palmer 1806. Reprinted in Annals of Philosophy, Vol. VI, 349-354, 1923 (cited after Hırı and Osman Hırı 1955). BEARD, ]J. (1897): The birth-period of Trichosurus vulpecula. Zool. Jb. (Abt. Anat.) 2, 77-96. Brock, M. (1964): The blood forming tissues and blood of the newborn opossum (Didelphis virginiana). 1. Normal development through about the one-hundredth day ot life. Ergeb. Anat. Entwickgesch. 37, 237-366. BOTERENBROOD, E. C. (1977): Concise catalogue of the central embryological collection of the Hubrecht Laboratory. Issued by Hubrecht Laboratory, Utrecht, Netherlands. BRooMm, R. (1897): On the existence of a sterno-coracoidal articulation in a foetal marsupial. J. Anat. Phys. 31, 513-515. — (1899): On the development and morphology of the marsupial shoulder girdle. Trans. Roy. Soc. Edinb. 39, 749-770. — (1902): On the early condition of the soulder-girdle in the polyprotodont marsupials Dasyurus and Perameles. J. Linn. Soc. Zool. 28, 449454. — (1912): The morphology of the coracoid. Anat. Anz. 41, 625-631. CHENG, CZE-CHing (1955): The development of the shoulder region of the opossum, Didelphis virginiana, with special reference to musculature. J. Morph. 97, 415471. EISENBERG, J. F. (1981): The Mammalian Radiations: An analysis of trends in evolution, adaption, and behavior. Chicago: Univ. Chicago Press. FRAsER, E. A.; Hırr, J. P. (1915): The development of the thymus, epithelial bodies and thyroid in the Marsupialia. Part I. Trichosurus vulpecula. Phil. Trans. (B) 207, 1-85. Hanson, F. B. (1920): The problem of the coracoid. Anat. Rec. 19, 327-345. HARTMan, C. G. (1920): Studies in the development of the opossum Didelphys virginiana L. V. The phenomena of parturition. Anat. Rec. 19, 1-11. Unpublished drawings of marsupial embryos 211 — (1923): Breeding habits, development and birth of the opossum. Rep. Smithson. Inst. Washington for 1921, 347-363. — (1928): The breeding season of the opossum (Didelphis virginiana) and the rate of intra-uterine and postnatal development. J. Morph. Physiol. 46, 143-200. — (1952): Possums. Austin: Univ. of Texas Press. Hırr, J. P. (1898): The placentation of Perameles (Contributions to the embryology of the Marsupia- lia, I). Quart. J. micr. Sci. N. S. 40, 385446. — (1900): On the Foetal Membranes, Placentation and Parturition of the Native Cat (Dasyurus viverrinus). Anat. Anz. 18, 364-373. — (1901): Contributions to the embryology of the Marsupialia. II. On a further stage in the placentation of Perameles. III. On Quart. J. micr. Sci. 43, 1-22. — (1910): Contributions to the embryology of the Marsupialia. IV. The early evelopment of the Marsupialia, with special reference to the native cat (Dasyurus viverrinus). Quart. J. micr. Sci. 56, I57: — (1918): Contributions to the embryology of the Marsupialia. V. Some observations on the early development of Didelphis aurita. Quart. J. micr. Sci. 63, 91-139. — (1932): The developmental history of vertebrates (Croonian Lecture). Philos. Trans. Royal Soc. London, ser. B, 221, 45-178. Hıirr, J. P.; Hırı, W. C. Osman (1955): The growth-stages of the pouch-young of the Native Cat (Dasyurus viverrinus) together wıth observations on the anatomy of the new-born young. Trans. Zool. Soc. London 28, 349-425. HUNSAKER, D. (1977): Ecology of New World Marsupials. In: The Biology of the Marsupials. Ed. by D. HunsAakeEr. New York, San Francisco, London: Academic Press. 95-156. Krıma, M. (1968): Early development of the human sternum and the problem of homologization of the so-called suprasternal structures. Acta anat. (Basel) 69, 473484. — (1973): Die Frühentwicklung des Schultergürtels und des Brustbeins bei den Monotremen (Mammalıa: Prototheria). Adv. Anat. Embryol. Cell. Biol. 47, 1-80. — (1975): Ancestrale Rudimente in der Morphogenese des Manubrium sterni beim Menschen. Verh. Anat. Ges. 69, 793-797. — (1985): Development of shoulder girdle and sternum in mammals. In: Vertebrate Morphology. Fortschritte der Zoologie, Bd. 30. Ed. by Duncker and FLEISCHER. Stuttgart, New York: G. Fischer. 81-83. — (submitted): Early development of the shoulder girdle and sternum in marsupials (Mammalıa: Metatheria). Adv. Anat. Embryol. Cell. Biol. LANGWORTHY, O.R. (1928): The behaviour of pouch-young opossums correlated with the myeliniza- tion of the tracts in the nervous system. J. comp. Neur. 46, 201-240. LEcHE, W. (1890): Zur Charakteristik der extra-uterinen Entwicklung der Beuteltiere. Verh. Biol. Vereins, Stockh. 2, 112-116. LILLEGRAVEN, J. A. (1984): Why was there a ““marsupial-placental dichotomy”’? In: Mammals: notes for a short course. Organızed by P. D. GinGERIcH and C. E. BAapcLey, Univ. of Tennessee, Dept. of Geological Scı. Studies in Geology 8, 72-86. — (1985): Use of the term “trophoblast” for tissues in therian mammals. J. Morph. 183, 293-299. McCrapvy, E. JR. (1938): The embryology of the Opossum. Amer. Anat. Mem. 16, 1-233. McKenna, M.C. (1961): On the shoulder girdle of the Mammalian subclass Allotheria. Amer. Mus. Novitates (New York) 2066, 1-27. MiLLaR, J. S. (1977): Adaptive features of mammalian reproduction. Evolution 31, 370-386. Owen, R. (1841): Article “Marsupialia”. In: Cyclopaedia of Anatomy and Physiology. Vol. 3. Ed. by Topp. London, 1839-1847. (Cited after Hırn and Osman Hırı 1955). RoMER, A. S. (1922): The comparison of mammalian and reptilian coracoids. Anat. Rec. 24, 3947. Russe, E. M. (1982a): Parental investment and desertion of young in marsupials. Am. Nat. 119, 744-748. — (1982b): Patterns of parental care and parental investment in marsupials. Biol. Rev. 57, 423-486. SHARMAN, G. B.; Pırron, P. (1964): The life history and reproduction of the red kangaroo (Megaleia “rufa). Proc. zool. Soc. Lond. 142, 29-48. TYNDALE-BiscorE, H. (1973): Life of Marsupials. New York. American Elsevier Publishing Comp. Watson, D. M. S. (1917): The evolution of the tetrapod shoulder-girdle and fore-limb. J. Anat. (Lond.) 52, 1-63. Authors’ addresses: Prof. Dr. M. Kııma, Institute of Anatomy, Johann-Wolfgang-Goethe-Univer- sity, Theodor-Stern-Kai 7, D-6000 Frankfurt/M., Germany and Mrs. Dr. G. C. BAnGMA, Hubrecht Laboratory, International Embryological Institute, Uppsala- laan 8, NL-3584 CT Utrecht, Netherlands Untersuchungen zur Genese des Maulwurfshaufens bei Talpa europaea (L.) Von U. MarTtHıas und G. R. WITTE Fachbereich Biologie und Chemie der Universität Kassel Eingang des Ms. 27. 8. 1986 Abstract Studies on the formation of the molehill of Talpa europaea (L.) A male individuum of Talpa europaea was kept in the laboratory for several weeks in order to study his behaviour when building a molehill. It was seen that the mole brings the loose earth to the surface in numerous walks. He pushes the soil from below under the upper part of the hillock where he had established some caves previously. When the capacity of these caves is used up he creates new room by pressing the soil at the side or above. In this way the molehill is growing by pressure of soil from below under the existing earth-layer. Einleitung Talpa europaea ıst in Mitteleuropa weit verbreitet und besiedelt nahezu alle terrestrischen Ökosysteme. So finden sich seine Haufen (Erdhügel), als leicht erkennbarer Vorkommens- Nachweis, beispielsweise auf Feuchtwiesen mit hohem Grundwasserstand, auf Trockenra- sen, in Laub- und Mischwäldern sowie auf intensiv bewirtschafteten Acker- und Grün- landflächen. Aufgrund seiner subterranen Lebensweise ist er auf das Anlegen eines unterirdischen Gangsystems angewiesen, das aus tunnelartigen Gängen und damit ın Verbindung stehenden Wohn- bzw. Nestkesseln besteht. Beim Bau dieser unterirdischen Hohlräume ist es notwendig, den beim Graben anfallenden Abraum - soweit er nicht durch Verpressen in die Gangwände beiseite geschafft werden kann - aus dem Gangsystem zu entfernen und zu den sogenannten Maulwurfshaufen aufzuschichten. Die äußere Form dieser Haufen ist wohlbekannt. Ihre Größe schwankt beträchtlich. So können kleine Haufen mit einem Durchmesser von etwa 10 cm großen Burgen mit einem Durchmesser von mehr als 1 m und etwa 0,5 m Höhe gegenübergestellt werden. Diese „Riesenhügel“ lassen sich häufig in Feuchtgebieten mit hoch anstehendem Grundwasser finden. Sie sind meist von zahlreichen Gängen durchzogen und enthalten in der Regel ein bis mehrere Nester (vgl. ArmsBy et al. 1966). Über das Anlegen von Maulwurfshügeln liegen unterschiedliche Vorstellungen vor. Frühere Autoren (z.B. Apams 1903) sind davon ausgegangen, daß Talpa europaea die beim Graben anfallende Erde mit Kopf und Schultern vor sich herschiebt und dann ın Form von schornsteinartigen Erdzylindern hinausdrückt. Spätestens seit Hısaw (1923) ist bekannt, daß Maulwürfe die Erde mit einer Vorderhand, ähnlich wie ein Bulldozer, vor sich herschieben. Dabei geht man auch in jüngeren Publikationen (siehe z.B. Buch 1980) nach wie vor davon aus, daß das wegzuräumende Erdmaterial in großen Portionen und in Form von „Erdschornsteinen‘ herausgedrückt und somit von oben auf den Haufen aufgelagert wird. Eigene Beobachtungen haben uns dazu veranlaßt, diesen Bau-Modus zumindest als Regelfall in Zweifel zu ziehen und eine entsprechende experimentelle Untersuchung durchzuführen. U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5204-0212 $ 02.50/0 Z. Säugetierkunde 52 (1987) 212-218 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Untersuchungen zur Genese des Maulwurfshaufens bei Talpa europaea (L.) 2,18 Material und Methode Ein männlicher Maulwurf wurde ım Frühjahr 1986 über einen Zeitraum von 3 Wochen in einer Käfiganlage gehalten, die über eine Schlauchverbindung an einen Erdbehälter von ca. 120 | Fassungs- vermögen angeschlossen war (Abb. 1). Der Erdbehälter hatte die Form eines Pyramidenstumpfes. Oben war er mit einem kurzen Rohrstück (Dränageschlauch, 45 mm Innen-@®) an einen darüber stehenden Tisch angeflanscht. Dieser Tisch war bis zu einer Höhe von 6 cm mit schwarzer Erde (Torfstruktur) bedeckt und mit Gräsern und Wiesenkräutern bepflanzt. Dem in die Anlage einge- brachten Maulwurf stand zum Graben in der ersten Woche gelber Löß im Erdbehälter zur Verfügung. Danach wurde der Löß gegen eine rötliche (eisenhaltige) Erde ausgetauscht. In der dritten Woche wurde brauner Lehm dargeboten. Nach dem Einsetzen des Maulwurfs in die Anlage wurden dessen Aktivitäten beobachtet und die Genese eines Maulwurfshaufens fotografisch festgehalten. Zur exakten Dokumentation der Bau-Aktivitäten wurde an der Schlauchverbindung zwischen Erdbehälter (Pyramidenstumpf) und Tisch ein Mikrophon mit Tonbandanlage angeschlossen. Der Maulwurf wurde einmal täglich gefüttert. Über Nacht sowie an Tagen, an denen keine Beobachtungsmöglichkeiten bestanden (Feiertage, Wochenenden), wurde das Tier vom Erdbehälter abgetrennt und hatte dann lediglich seine Kompaktkäfiganlage (WırrE 1981) zur Verfügung. Nach dem Absperren wurde ın dem durchwühlten und mit Gängen durchzogenen Erdbehälter Erde nachgefüllt und möglichst stark verdichtet, um hierdurch den Maulwurf zu weiterem Graben und Auswerfen der anfallenden Lockererde zu veranlassen. Nach Abschluß der Beobachtungszeit wurde der entstandene Maulwurfshügel (durch vertikale Schnitte) scheibchenweise in 5 cm Abständen abgetragen, um Aufschlüsse über seinen inneren Aufbau und die Einschichtung der unterschiedlich gefärbten Erden zu erhalten. Die in Abb. 3 wiedergegebenen Schnitte wurden nach fotografischen Dokumenten gezeichnet. Ergebnisse Wird ein Maulwurf mit seiner Käfiganlage an den Erdbehälter angeschlossen, so beantwor- tet er diese Erweiterung seines Aktionsraumes sofort (bzw. nach Beendigung der gerade stattfindenden Ruheperiode) mit Eindringen in dieses neue Gebiet. Obwohl das Erdreich beim Einfüllen in den Pyramidenstumpf stark verdichtet wurde, hatte der Maulwurf beim Eingraben in den Erdvorrat keinerlei Schwierigkeiten. Durch seitliches Verpressen wurde der erste begehbare Gang geschaffen. Zum weiteren Ausbau des Gangsystems war es dann jedoch nötig, den beim Graben anfallenden Abraum aus dem Gangsystem hinauszuschie- ben. In derartiger Weise agierte unser Versuchsmaulwurf erstmals nach zweitägigem Aufenthalt im Erdbehälter. Dabei war zu beobachten, daß er nach dem Passieren des kurzen Schlauchstutzens, der zur Erdoberfläche führte, zunächst äußerst vorsichtig das aufliegende Bodenmaterial mit Kopf und Rückenpartie über dem Zugang mehrfach anhob bzw. lockerte und gleichzeitig durch intensives Schnüffeln über mehrere Minuten das Umfeld erkundete. Bei dieser Aktion unternahm der Maulwurf mehrere kurze Ausflüge (bis zu einer Länge von 20 cm) in die seitliche Umgebung. Hierbei blieb er jedoch vollkommen unter einer dünnen schützenden Bodenschicht dem Blick verborgen. Auch der Zugang seines Gangsystems wurde niemals bis zur Obertläche freigelegt. Nach diesem Erkundungsgang verschwand der Maulwurf in dem Erdbehälter, begann zu graben und transportierte wenige Minuten später erstmals Erde aus der Tiefe heraus. Die ersten Erdaushübe wurden direkt am Zugang unter die Torfschicht gedrückt, die sich hierdurch deutlich aufwölbte. Danach kam der Maulwurf erneut nach oben, verteilte den Abraum durch seitliches Eindrücken ın den entstehenden Hügel und schuf so Raum für weiteres Erdmaterial. Anschließend verschwand er wieder nach unten und transportierte in kurzen Abständen mehrfach Erde zum Haufen, bis dessen Lagerkapazitäten erschöpft waren. Hierauf folgten wiederum Arbeiten am Hügel, die durch seitliches und nach oben gerichtetes Einpressen der herangebrachten Erde Platz für weiteren Abraum schufen. Zu Anfang benötigte der Maulwurf jeweils etwa 1 bis 2 Minuten, um Raum (,Kavernen‘“) für etwa 5 bis 10 Erdaushübe anzulegen. Im weiteren Verlauf, mit größer werdendem Haufen verlängerten sich die Zeiten für das Anlegen solcher Erdlagerplätze auf bis zu 5 Minuten. 214 U. Matthias und G. R. Witte Danach konnten in den Kavernen bis zu 20 Erdaushübe untergebracht werden. Das Anlegen von Lagerkapazitäten im Hügelinneren war in aller Regel mit einem Erkunden der näheren Umgebung verbunden. Erkennbar wurde dies am intensiven Schnüffeln und an gelegentlichen kurzfristigen Ausflügen in oberflächennahe Tunnelgänge (vgl. Wırre 1985). Bei den Bauarbeiten am Maulwurfshaufen wurde deutlich, daß das von uns beobachtete Tier das herausgeschaffte Erdmaterial stets nur unter bereits bestehende Erdschichten drückte und hierdurch zur Vergrößerung des Haufens beitrug. Es kam so zu der von außen beobachtbaren Aufwölbung der zu Anfang ebenen schwarzen Bodenoberfläche. Durch weiteres von unten her erfolgtes Hineinpressen von Erde riß der dunkle Torfboden auf, und die zum Graben zunächst angebotene gelbe Erde wurde sichtbar. Mit weiterer Bautätigkeit nahm das Volumen des gelben Erdaushubes von unten heraus zu und fiel schließlich über die Ränder der zur Seite abgedrängten Torfschicht. Nach dem Auswech- seln des gelben Löß gegen die rote Erde war trotz intensiver Bautätigkeit zunächst lediglich ein Anwachsen der Haufengröße zu beobachten. Erst nach einer weiteren und beträchtli- chen Oberflächenvergrößerung kam es schließlich zum Aufbrechen der gelben Erdschicht und zum Austreten von roter Erde. Auch beim letzten Erdwechsel (Austausch der roten gegen braune Erde) trat trotz intensiver Bauaktivitäten zunächst keine braune Erde an der Oberfläche zu Tage. Erst kurz vor Abschluß der Untersuchung, nachdem der Hügel bereits beträchtlich angewachsen war, wurde lediglich an einer Stelle eine geringe Menge des braunen Materials sichtbar. Innerhalb der drei Wochen, die der Maulwurf zum Haufenbau zur Verfügung hatte, erreichte der Hügel bei einer Höhe von etwa 20 cm einen Durchmesser von ca. 50 cm. Das Tabelle Aktivitäten des Maulwurfs beim Haufenbau Datum Aktivität gelbe Erde .4. Durchwühlen des Erdbehälters, ohne Erdauswurf . Durchwühlen des Erdbehälters, ohne Erdauswurf .4. Durchwühlen des Erdbehälters, Erkunden der oberen Bodenschicht und erstmaliges Aus- bringen von Erdmaterial (Zahl und Dauer der Erdtransporte: 98 in 140 mın) .4. Fortsetzung des Haufenbaus (Zahl und Dauer der Erdtransporte: 85 ın 90 mın) .4. Durchwühlen des Erdbehälters, ohne Erdauswurf .5. inaktiv, kein Erdauswurf rote Erde . Durchwühlen des Erdbehälters, ohne Erdauswurf . Erneute Aktivitäten beim Haufenbau (Zahl und Dauer der Erdtransporte: 86 ın 115 mın) . Fortsetzung des Haufenbaus (Zahl und Dauer der Erdtransporte: 68 in 105 mın) . Durchwühlen des Erdbehälters, ohne Erdauswurf . Weiterer Bau am Haufen (Zahl und Dauer der Erdtransporte: 52 ın 90 min) . inaktiv, kein Erdauswurf . inaktiv, kein Erdauswurf braune Erde . inaktiv, kein Erdauswurf .5. Erneute Aktivitäten beim Haufenbau (Zahl und Dauer der Erdtransporte: 44 in 45 mın) . Fortsetzung des Haufenbaus (Zahl und Dauer der Erdtransporte: 36 in 20 min) . Fortsetzung des Haufenbaus (Zahl und Dauer der Erdtransporte: 64 in 100 min) .5. Fortsetzung des Haufenbaus (Zahl und Dauer der Erdtransporte: 85 in 120 min) . inaktiv, kein Erdauswurf .5. inaktıv, kein Erdauswurf . inaktiv, kein Erdauswurf Untersuchungen zur Genese des Maulwurfshaufens bei Talpa europaea (L.) 2A Pflanztisch \ Erdbehälter Käfiganlage Abb. 1. Aufbau der Versuchsanlage verbaute Erdvolumen betrug ca. 15 I. Zum Erreichen dieser Haufengröße benötigte der Maulwurf insgesamt 13 Stunden und 45 Minuten. Die „Arbeitszeiten“ waren auf 9 Tage verteilt; die einzelnen Grab-Aktivitätsschübe waren zwischen 20 und 140 min lang (Tab.). Hügelbau-Aktivitäten wurden in dem hier beschriebenen Beispiel jeweils lediglich einmal täglich festgestellt. Die restliche Zeit verteilte sich im wesentlichen auf Nahrungsaufnahme und Ruheperioden (vgl. BuscHInGER und WITTE 1976; WITTE 1981). Während der gesamten 3wöchigen Hügelgenese wurden insgesamt 628 Erdtransporte gezählt. Das Volumen eines einzelnen Transportganges muß dabei etwa 1-2 Eßlöffeln entsprechen, um ein Erdvolumen von 15 | zu erreichen. Bei dem schichtweisen Abtragen des fertigen Maulwurfshaufens wurde deutlich, daß Talpa innerhalb des Erdhügels mehrere zu den Seiten hin orientierte Gänge angelegt hatte (Abb. 2). Während der Bauphasen sind diese Gänge sicherlich mehrfach verlagert, zum Anlegen von Kavernen und zum Einpressen von Abraummaterial benutzt worden. Die Vertikal-Schnitte des Hügels ließen erkennen, daß der Maulwurf aus der Tiefe hochtrans- portierte Lockererde in der Regel von unten in den Haufen einschichtet. Besonders deutlich wird dies beispielsweise beim 5. Anschnitt der Abb. 3. Dort ıst zu erkennen, daß der gelbe Löß des ersten Erdangebotes zunächst von roter (zweites Erdangebot) und später von der braunen Erde (drittes Erdangebot) immer weiter nach oben geschoben wurde. Für die Unterbringung von Abraum scheint die Anlage eines Gangsystems ab einer bestimmten Haufengröße unvermeidlich zu sein. Jedoch dienen diese Gänge sicherlich auch zur Erkun- dung der näheren Umgebung, vielleicht auch zur besseren Ventilation und im vorliegenden Fall sogar dem Einbrin- gen von Nestmaterial: Während der Untersuchungszeit war mehrfach zu beobachten, daß unser Tier, ohne sich selbst an der Oberfläche zu zeigen, am Rande des Haufens trockene Grashalme an den Wurzeln abriß und ın den Erdbehälter eintrug. Nach dem Öffnen der Erdpyramide A en Maul. zu Versuchsabschluß konnten wir dort ein frisches, voll wurfshaufen (in der Mitte der ausgebautes Maulwurfsnest vorfinden (Abb. 4). kreisrunde Zugang von unten) 50cm 216 U. Matthias und G. R. Witte 2. Anschnitt (10 cm) 1. Anschnitt ( 5 cm) 3. Anschnitt ( 15 cm) 4. Anschnitt (20 cm) 6. Anschnitt (30 cm) TÜR 7. Anschnitt (35 cm) 8. Anschnitt (40 cm) 9. Anschnitt (45 cm) & 3. Erdaushub (braun) . Anschni 10cm „ 2. Erdaushub ( rot ) 1. Erdaushub ( gelb ) et oz & vorgegebene Erdschicht D— 50 cm = Abb. 3. Anschnittprofile des Maulwurfshaufens in Abständen von 5 cm. Die geschwärzten bzw. gerasterten Flächen entsprechen den vorgefundenen Erdschichten. Weiß gebliebene Flächen ım Haufen geben die Anschnitte des Gangsystems wieder (vgl. Abb. 2) Diskussion Entgegen anderslautenden Beschreibungen in der Literatur wurde bei unseren Beobach- tungen unter Laborbedingungen festgestellt, daß Talpa enropaea den beim Graben anfal- lenden Abraum immer nur in kleinen Portionen (20-30 ml) nach oben brachte und ım unteren Bereich des Hügels in vorher geschaffenen Kavernen einlagert. MELLANBY (1971) wie auch GODFREY und CROWCROFT (1960) erläutern in ihren Monographien dagegen, daß Maulwürfe die Erde in großen Portionen (,Erdschornsteine‘‘) durch ıhr Gangsystem und schließlich auch durch den Haufen bis in das Freie drücken. Beide berufen sich auf SKoCZEN (1958), der sehr detailliert beschreibt, zu welchen Leistungen Maulwürfe fähig sind. So wurde von ıhm ein Maulwurf beobachtet, der innerhalb von 20 min 6 kg Erde ausgeworten hatte. In nassen, nicht-sandigen Gebieten Untersuchungen zur Genese des Maulwurfshaufens bei Talpa europaea (L.) 2 Abb. 4. Maulwurtfsnest in dem geöffneten Erdbehälter wurden von ihm Erdschornsteine gesehen, die den Haufen um 10-15 cm überragten und einen Durchmesser von 5-6 cm aufwiesen. SKOCZEN schätzt die Gesamtlänge der durch den Gang gedrückten ‚„Schornsteine‘“ auf insgesamt 40 cm und ein Gewicht von 800 g, das der Maulwurf mit einer Hand vor sich herschiebt. Dies entspricht in etwa dem 10-12fachen seines Eigengewichtes. Berücksichtigt man ferner noch, daß beim Schieben des Erdreichs durch den Gang starke Reibungswiderstände auftreten, wird deutlich, zu welchen enormen Kraftleistungen Maulwürfe fähig sind. Es ist aber nicht anzunehmen, daß Talpa sein Leistungspotential unter Normalbedingungen immer voll ausschöpft. Leider finden sich bei SKOCZEN neben der Nennung von Leistungsmaxima keine Angaben über durchschnittliche Verhaltensnormen beim Haufenbau. Wir vermuten aufgrund des hier geschilderten Experimentes und aufgrund zahlreicher, mehrjähriger Beobachtungen im Freiland und im Labor (vgl. Wırre 1985), daß der Haufenbau in der Regel in zahlreichen Transportschritten mit jeweils entsprechend niedrigen Transportvolumina stattfindet. Die Bildung großer Erdschornsteine ist sicherlich als Sonderfall anzusehen und vom Bodenzustand wie auch vom individuellen Verhalten des Maulwurfs abhängig. 218 U. Matthias und G. R. Witte Zusammenfassung Über einen Zeitraum von mehreren Wochen wurde ein männlicher Maulwurf im Labor gehalten und beim Bau eines Maulwurfshaufens beobachtet. Es wurde festgestellt, daß Talpa europaea die wegzu- räumende Erde mit zahlreichen Transportgängen nach oben schafft und unter dem Haufen in vorher geschaffenen Erdsammelplätzen einlagert. Sind die bestehenden Lagerkapazitäten erschöpft, werden durch seitliches und nach oben gerichtetes Verpressen neue Ablageräume geschaffen. Der Maulwurfs- haufen gewinnt somit durch Einpressen und Unterschichten des Abraums von unten heraus an Größe. Das von anderen Autoren beschriebene Herauspressen größerer Erdportionen über die Haufenober- fläche hinaus ist nicht als Regelfall anzusehen. Literatur Apams, L. E. (1903): Contribution to our knowledge of the mole (Talpa europaea L.). Proc. Manch. Lit. Phil. Soc. 47, 1-39. Armssy, A.; QUILLIAM, T. A.; SOEHNLE, H. (1966): Some observations on the ecology of the mole. ]. Zool. (Lond.) 149, 110-111. Buch, J. (1980): Muldvarpen. Kopenhagen: Rhodos. BUSCHINGER, A.; WITTE, G.R. (1976): Untersuchungen zur Periodik der lokomotorischen Aktivität des europäischen Maulwurfs (Talpa europaea L.) unter natürlichen Bedingungen. Bonn Zool. Beitr. 27, 7-20. GODFREY, G.; CROWCROFT, P. (1960): The life of the mole. London: Museum Press. Hısaw, F. L. (1923): Observations on the burrowing habits of moles (Scalopus aquaticus machrinoi- des). J. Mamalogy 4, 79-88. MELLANBY, K. (1971): The mole. London: Collins. SKOCZEN, $. (1958): Tunnel diggin by the mole (Talpa europaea Linne). Acta theriol. 2, 235-249. WITTE, G.R. (1981): Erfahrungen mit der Käfighaltung von Maulwürfen (Talpa europaea L.). Zool. Garten (N. F.) 51, 193-215. WITTE, G.R. (1985): Talpa europaea (Talpidae) - Graben und Erdtransport. Encyclopaedia Cinema- tographica, Biol. 17/17, 3-19. Anschrift der Verfasser: Dr. Uwe MATTHIAS und Prof. Dr. GÜNTER R. WITTE, Fachbereich Biologie/ Chemie, Universität Kassel, Heinrich-Plett-Str. 40, D-3500 Kassel Resynchronisationsverhalten der circadianen Aktivitätsperiodik von Galagos (Galago senegalensis, Galago crassicaudatus garnettü)' Von F. ScHanz und H. G. ERKERT Institut für Biologie III der Universität Tübingen, Tübingen, FRG Eingang des Ms. 27. 7. 1986 Abstract Re-entrainment of circadian activity rhythms in galagos (Galago senegalensıs, Galago crassicandatus garnettü) Studied was the re-entrainment behaviour of the circadian activity rhythm in the nocturnal Prosimian species Galago senegalensis and Galago crassicandatus garnettii after sudden 8-hrs advance and delay shifts of an entraining LD-cycle of 12:12 and 8:16 hrs (56:0,07 Ix). In both species a distinet asymmetry effect occurred. G. senegalensis re-entrained more rapidly after a delay shift than after an advance shift of the LD-Zeitgeber, whereas G. crassicaudatus responded in the opposite way. As in ' Mit Unterstützung der Deutschen Forschungsgemeinschaft (ER 59/10). U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5204-0218 $ 02.50/0 Z. Säugetierkunde 52 (1987) 218-226 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Resynchronisation der circadıanen Aktivitätsperiodik von Galagos 210 both species the spontaneous period of free-running activity rhythm predominantly is shorter than 24 hrs, such a difference in the re-entrainment behaviour of the circadian activity rhythm is not in accordance with present hypotheses trying to explain the cause of different asymmetry effects. Therefore it is supposed that only a hypothesis considering the period length of the circadıan rhythm as well as the characteristics of the phase-response curve will lead to a reasonable explanation of species specific differences in the expression of asymmetry effects. Einleitung Die Aktivitätsperiodik licht- und dunkelaktiver Arten wird durch ein endogenes Timing- System gesteuert, dessen Spontanperiode ın der Regel von 24h abweicht und deshalb durch Zeitgeber mit der Umweltperiodik synchronisiert werden muß. Hauptzeitgeber ist dabei das Licht bzw. der tagesperiodische Wechsel der Beleuchtungsstärke. Im synchroni- sierten Zustand haben die circadianen Rhythmen eine bestimmte, stabile Phasenbeziehung zum Zeitgeberzyklus. Nach einer plötzlichen Verschiebung dieser Umweltperiodik benö- tigt der Organismus mehrere Übergangszyklen, um wieder eine stabile Phasenlage zu ihr zu erreichen. Dabei folgt die circadiane Aktivitätsperiodik einer Vorverlagerung (advance- shift) bzw. Rückverschiebung (delay-shift) des LD-Zeitgebers in der Regel orthodrom durch eine sukzessive Vorverlagerung bzw. Rückverschiebung der Aktivitätsphase. Die Anzahl der Übergangszyklen, die nach 6- bis 8stündigen Zeitverschiebungen zur Resyn- chronisation notwendig sind, um wieder die ursprüngliche Phasenlage einzunehmen, hängt unter anderem von der Schwingungsbreite und der Schwingungsform des Zeitgebers ab (AscHorr et al. 1975; AscHorrF 1978; ERKERT 1976). In beiden Fällen soll ein schwächerer Zeitgeber zu einer verlängerten Resynchronisationszeit führen. Unabhängig davon dauert die Resynchronisation nach advance- und delay-shifts in der Regel unter- schiedlich lang (AscHorr 1978). Dieses Verhalten des circadianen Systems, d.h. die Abhängigkeit der Resynchronisationsgeschwindigkeit von der Richtung der Zeitgeberver- schiebung, wird als Asymmetrieeffekt (AscHorF 1969) oder Richtungseffekt (ERKERT 1976) bezeichnet. WEVER (1966) postuliert eine Abhängigkeit des Asymmetrieeftekts von der Circadıan- periode und ermittelt in seinem mathematischen Modell circadianer Rhythmen eine Periodenlänge von 24,5 h, bei der die Resynchronisationszeiten nach gleich großen advance- und delay-shifts des Zeitgebers übereinstimmen sollen. Auch ERKERT (1976) vermutet einen ursächlichen Zusammenhang zwischen dem Richtungseffekt und der Spontanperiodenlänge t. Er nımmt an, daß Arten, bei denen t vorwiegend >24 h liegt, nach einer Rückverschiebung des Zeitgebers schneller resynchronisieren als nach einer Vorverlagerung und sich Arten mit t <24 h entgegengesetzt verhalten. AscHorr (1978) diskutiert zwar die Möglichkeit eines Zusammenhangs zwischen der Spontanperioden- länge und/oder den Charakteristika der Phasenresponse-Kurve, nımmt dann aber eher eine kausale Beziehung zwischen der Ausprägung des Asymmetrieeffekts und der licht- oder dunkelaktiven Lebensweise der betreffenden Arten an. Danach sollen dunkelaktive Spezies nach Rückverschiebung und lichtaktive Arten nach Vorverlagerungen des LD-Zeitgebers schneller resynchronisieren als nach Zeitgeberverschiebungen in umgekehrter Richtung. Das Resynchronisationsverhalten der circadianen Aktivitätsperiodik (CAP) des süd- amerikanischen Nachtaffen (Aotus trivirgatus), der als einzige dunkelaktive Simiae-Art gilt, entspricht insofern allen drei Vorstellungen, als er nach delay-shifts schneller resyn- chronisiert und eine zwischen 24,3 und 26,3 h gelegene Spontanperiode aufweist. Bei dem ebenfalls dunkelaktiven Senegalgalago (Galago senegalensis) liegt die Spontanperiode dagegen meist unter 24h (ERKERT et al. 1984). Dementsprechend sollte er nach den Vorstellungen von WEVvER (1966) und ERKERT (1976) nach Vorverlagerung des LD- Zeitgebers schneller resynchronisieren als nach Rückverschiebungen, wohingegen nach AscHoFF (1978) das Gegenteil zu erwarten wäre. Außerdem sollte sich nach delay- und 220 F. Schanz und H. G. Erkert advance-Verschiebungen eines schwächeren Zeitgebers, wie ıhn ein LD 8:16 für den Nachtaffen darstellt, die Resynchronisationsdauer unabhängig vom Asymmetrieeffekt verlängern. Um dies zu überprüfen, wurde das Resynchronisationsverhalten der circadia- nen Aktivitätsperiodik von 5 Senegalgalagos nach 8stündiger Vorverlagerung und Rück- verschiebung eines LD-12:12 und eines LD 8: 16 mit 56:0,07 Ix ermittelt. Um zu klären, ob für die Galagidae einheitliche Resynchronisationsmechanismen des circadianen Systems charakteristisch sind, wurden die 8stündigen Phasenverschiebungen des LD 12:12 Zeitge- bers zusätzlich an dem Rıiesengalago (Galago crassicandatus) durchgeführt, dessen Spon- tanperiode t wie die des Senegalgalagos ebenfalls unter 24 h liegt. Material und Methode Als Versuchstiere dienten 5 männliche Galago senegalensis und 3 Galago crassicandatus (2 8,1). Alle Versuche wurden in einem schall- und lichtisolierten Konstantraum bei 25 # 1>C und 65 + 5% relative Luftfeuchte durchgeführt. Die Tiere lebten einzeln in 80 x 80 x 50cm (Galago senegalensis) und 95 x 95 x 105 cm (Galago crassicandatus) großen Drahtkäfigen, die mit Kletterstan- gen und einem Schlafkasten ausgestattet waren. Die Käfige waren freischwingend in hölzernen Beleuchtungskästen nach ERKERT (1970) und KrAcHT (1979) aufgehängt, die mıt mehreren Tages- lichtleuchtstoffröhren (Osram 20 Watt/15) bestückt waren und eine völlige optische und akustische Isolation der Tiere ermöglichten. Die Galagos wurden täglich während der Lichtzeit mit kleinge- schnittenem Obst, Karotten und Tenebriolarven gefüttert. Zusätzlich erhielten sie 2X pro Woche Haferschleimbrei, dem Vitamin-Mineral-Aufbaupräparate beigemengt waren. Zur Erfassung der lokomotorischen Aktivität diente eine elektroakustische Registriermethode nach Bay (1978). Der durch bi- oder quadrupede Fortbewegung, Putzen, Fressen und Orientierungs- bewegungen des Tieres im Drahtkäfig induzierte Körperschall wurde dabei von einem speziellen Körperschallmikrophon (E. Holthausen, Merula TVJ/XL) aufgenommen. Das Mikrophonsignal triggerte über eine Verstärkerschaltung eine konstantfrequente Folge von Rechteckimpulsen. Diese wurden dann von einem elektronischen Zähler (Demel V 224) aufsummiert, über einen Kleincompu- ter (Apple Ile) abgerufen und in 5-min-Intervallen auf Diskette abgespeichert. Zur weiteren Verarbei- tung und graphischen Darstellung der Daten diente ein zweiter Apple Ile Computer mit angeschlosse- nem Plotter (Watanabe). Alle Resynchronisationsversuche wurden mit gleichbleibenden Beleuchtungsstärken von 56 + 21x ın der L-Zeit und 0,07 + 0,03 1x in der D-Zeit durchgeführt. Vor Beginn der einzelnen Resynchronisationsversuche wurden die Galagos jeweils mehrere Wochen lang durch einen LD 12:12 synchronisiert. Damit ließen sich Nachwirkungen vorangegangener Versuchsbedingungen (‚‚afteref- fects“, PITTENDRIGH und DAANn 1976; MOoORE-EDE et al. 1982) weitgehend ausschließen. Finem advance-shift des LD 12:12 von A® = +8 h folgte im ersten Versuch nach 26 Tagen ein gleichlanger delay-shift von AD =-8 h des LD-Zeitgebers. Um den Einfluß der Reihenfolge solcher Zeitgeberver- schiebungen auf die Resynchronisationszeit zu testen, wurde der LD 12:12 ın einem ergänzenden Versuch bei 2 Galago senegalensis (VT 10, VT 12) zuerst um 8 h zurück und später wieder nach vorne verschoben. Zur Ermittlung der Auswirkungen eines „schwächeren Zeitgebers‘“ auf den Resynchro- nisationsverlauf wurde den Senegalgalagos dann in einem weiteren Versuch nach genügend langer Adaptationszeit ein LD 8:16 um 8 h vorverlagert und 26 Tage später wieder zurückverschoben. Bei Galago crassicandatus wurde zuerst ein 8stündiger delay-shift eines LD 12:12 und 26 Tage später ein gleich großer advance-shift dieses Zeitgebers durchgeführt. Ergebnisse Bei allen Versuchstieren hängt die Resynchronisationsdauer von der Richtung der Zeitge- berverschiebung ab. Auch bei Galago senegalensis und Galago crassicandatus treten also deutliche Asymmetrie- oder auch Richtungseffekte auf. Als Beispiel dafür zeigt Abb. 1 links die Aktivitätsperiodik eines Galago senegalensis-3 (VT 12). Nach der zuerst vorge- nommenen 8stündigen Vorverlagerung des LD 12:12 benötigt dieses Tier 9 Übergangs- zyklen, um wieder in die ursprüngliche Phasenlage zum Zeitgeber zu kommen, nach der späteren Rückverschiebung des LD’s dagegen nur 6. Tabelle 1 faßt die ermittelten Daten aller Senegalgalagos zusammen. Während sich die durchschnittliche Resynchronisations- Resynchronisation der circadıianen Aktivitätsperiodik von Galagos DD Galago senegalensis LD 12:12 LD 12:12 0) 0 \ \ 15 \ 15 \ \ \ \ 30 \ 30 45 45 aA RGENS SEE 7 neeTE 5 17 hCET 5 Abb. 1. Verlauf der Resynchronisation der circadianen Aktivitätsperiodik eines Senegalgalagos (VT 12) nach 8stündigen Phasenverschiebungen eines LD 12:12 (links, Mitte) und eines LD 8:16 (rechts) von jeweils 56:0,07 Ix. Beim ersten Versuch (links) wurde der LD-Zeitgeber zuerst vorverlagert und später zurückverschoben, beim zweiten Versuch (Mitte) erfolgten die Zeitverschiebungen in umge- kehrter Reihenfolge. Der Aktivitätsverlauf ist jeweils in 30-min-Werten wiedergegeben. Schraffierte Flächen kennzeichnen die Lichtzeit Tabelle 1 Spontanperiodenlänge im LL (56 Ix) und Dauer der Resynchronisation der circadianen Aktivitätsperiodik von Galago senegalensis (nach 8stündiger Vorverlagerung (AP = +8 h) und Rückverschiebung (A® = -8 h) eines LD 12:12 (links) und eines LD 8:16 (rechts) von jeweils 56:0,07 1x) Resynchronisationszeit (d) LL 0.07 lx t(h) 2353 24,0 DEE 235 23,6 23,012. 9,2 zeit nach der 8stündigen Vorverlagerung des LD 12:12 auf 9,2 + 1,6 Tage beläuft, beträgt sie nach der 8stündigen Rückverschiebung 5,0 # 0,7 Tage. Der Unterschied ist signifikant (Wilcoxon-Test, p<0,05). Das Verhältnis zwischen der Resynchronisationsdauer nach DR F. Schanz und H. G. Erkert Vorverlagerung und Rückverschiebung des LD-Zeitgebers, das als Index für die Größe des Asymmetrieeffekts verwendet werden kann, beträgt hier im Durchschnitt 1,85:1. Abb. 2 zeigt das Resynchronisationsverhalten bei den Riesengalagos. Auch bei ihnen tritt ein deutlicher Asymmetrieeffekt auf. Im Gegensatz zu Galago senegalensis sind jedoch die Resynchronisationszeiten der Riesengalagos nach der Rückverschiebung des Zeitgebers länger als nach der Vorverlagerung. Die durchschnittliche Resynchronisationszeit aller 3 Versuchstiere beträgt 6,3 Tage nach den advance-shifts und 10 Tage nach den delay-shifts. Daraus errechnet sich ein Asymmetrie-Index von 0,63. 8 (0) h 8 omse Ba 7 aaa” Gamma Ta Tome "1 Een (0) w 2 z o &4 o “2 Q OR .r D ORSOERE ° ® 0 5 [e) x. & o a [) Bo & our U) Ba a00on : ; 22 £ Abb. 2. Resynchronisationsverlauf der circadianen % = | Aktivitätsperiodik von 3 Galago crassicandatus (D 10 Es EVER AV 2 O,V IS) Tnach 8stündiger Vorverla- l gerung (A) und Rückverschiebung (D) eines Zeit- +2 | geber -— LD 12:12 (56:0,07 1x). Wegen starker e | Maskierung der Aktivität durch hohe Beleuch- tungsstärken während der Lichtphase der verscho- benen LD-Zyklen wurde für dıe Darstellung des Resynchronisationsverlaufes nach dem advance- 15 shift der Aktivitätsbeginn und nach dem delay- LD 12:12 shift das Aktivitätsende herangezogen Bei den beiden Senegalgalagos, bei denen die im ersten Versuch vorgenommenen Zeitgeberverschiebungen zu einem späteren Zeitpunkt in umgekehrter Reihenfolge wie- derholt wurden, bleibt der gezeigte Asymmetrieeffekt erhalten. Während VT 12 (Abb. 1, Mitte) hier die gleiche Resynchronisationsgeschwindigkeit hat wie zuvor, braucht VT 10 dabei zur Resynchronisation nach dem advance-shift 10 anstatt 12 Übergangszyklen und nach dem delay-shift 4 anstatt 5. Der Asymmetrie-Index der beiden Versuchstiere ändert sich dabei fast nicht. Daraus geht hervor, daß für den von Galago senegalensis gezeigten Asymmetrieeffekt weder die Reihenfolge der Zeitgeberverschiebung verantwortlich sein kann, noch eine zu kurze Synchronisationsphase zwischen der Vorverlagerung und Rückverschiebung des LD’s. Er ist deshalb als charakteristisch für das Resynchronisations- verhalten der circadianen Aktivitätsperiodik dieser Art anzusehen. Die Änderung des Lichtzeit: Dunkelzeitverhältnisses von 12:12 auf 8:16 h hat zwar andere Resynchronisationszeiten zur Folge, doch bleibt der Asymmetrieeffekt erhalten. Der Aktivitätsrhythmus von VT 12, der als Beispiel im rechten Diagramm von Abb. 1 dargestellt ist, benötigt zur Resynchronisation nach dem 8stündigen advance-shift des LD 8:16 8 Tage und nach dem delay-shift 5 Tage. In Abb. 3 ıst der Resynchronisationsverlauf der circadianen Aktivitätsperiodik aller 5 Versuchstiere nach der jeweils 8stündigen Vorverlagerung und Rückverschiebung des LD 12:12 und LD 8:16 zusammengefaßt. Da das Ende der Aktivitätsphase der Galagos im LD 8:16 eine leicht positive Phasenwinkel- differenz aufweist, d.h. schon etwas vor dem Lichtwechsel auftritt, beziehen sich die in der letzten Spalte der Tabelle 1 angegebenen Resynchronisationszeiten auf den Zeitpunkt, ab dem das Aktivitätsende wieder eine stabile Phasenlage zum Beginn der verschobenen Resynchronisation der circadianen Aktivitätsperiodik von Galagos 223 Lichtzeit zeigt. Im Durchschnitt beläuft sich die Resynchronisationszeit aller Versuch- stiere nach dem advance-shift des LD 8:16 auf 7,4 + 0,8 Tage und nach dem delay-shift auf 5,6 # 0,5 Tage. Das bedeutet eine Verringerung des Asymmetrie-Indexes von 1,8 auf 1,3. Sie geht im LD 8: 16 primär auf eine signifikante Erhöhung der Resynchronisationsge- schwindigkeit nach dem advance-shift zurück (p< 0,05). Die gegenüber dem LD 12:12 auftretende Verlängerung der durchschnittlichen Resynchronisationsdauer um 0,6 Tage nach dem delay-shift ist dagegen statistisch nicht signifikant. A D> A D> 8 (0) h 8 8 (0) h 8 0 07 I m Emo am —— nn 1 We Bomann Dr vo Aometo mes TEn em MiernmeErmErERCE me: ame EEE n Bob 7 of z 2 > koeooa Br I! r a Bl ur 8%0 ng. a bo 8 ‚Bu.0 a 88 .d 5 avod on sein a9 v fr’ som . N: a0 98» ie. a2 wo .o Rap u. Roß oo 5 ler d4 e .; 10 86 a a Ds Bo na. 15 LD 12:12 LD 8:16 Abb. 3. Resynchronisationsverlauf der circadianen Aktivitätsperiodik von 5 Galago senegalensis Männchen (4 VI 7,0 VT 9,0 VT 10,» VT 11, v VT 12) nach 8stündiger Vorverlagerung (A) und Rückverschiebung (D) eines Zeitgeber - LD 12:12 (links) und im LD 8:16 (rechts) (56:0,07 Ix). Wegen starker Maskierung der Aktivität durch hohe Beleuchtungsstärken während der Lichtphase der verschobenen LD-Zyklen wurde für die Darstellung des Resynchronisationsverlaufs nach dem advance-shift der Aktivitätsbeginn und nach dem delay-shift das Aktivitätsende herangezogen Diskussion Senegal- und Riesengalago zeigen bei der Resynchronisation der circadianen Aktivitätspe- rıodik nach Phasenverschiebungen des LD-Zeitgebers einen deutlichen Asymmetrie- oder Richtungseffekt. Er ıst bei Galago senegalensis sowohl im LD 12:12 als auch im LD 8:16 durch raschere Resynchronisation nach 8stündiger Rückverschiebung als nach entspre- chender Vorverlagerung des Zeitgeberzyklus gekennzeichnet. Ein solches Resynchronisa- tionsverhalten einer Art mit unter 24 h gelegener Spontanperiode entspricht weder der von WEVER (1966) noch der von ERKERT (1976) angenommenen Kausalbeziehung zwischen der Ausprägung des Asymmetrieeffektes und der Spontanperiodenlänge. Vielmehr scheint es die AscHorr’sche Annahme zu bestätigen, daß lichtaktive Arten nach Vorverlagerungen und dunkelaktive nach Rückverschiebungen des LD-Zeitgebers schneller resynchronisie- ren als nach jeweils entgegengesetzter Zeitgeberverschiebung. Der beim Riesengalago gefundene Asymmetrieeffekt läßt sich jedoch damit nicht vereinbaren und spricht dagegen eher für die vermutete Beziehung zur Spontanperiode. Tabelle 2 gibt eine Übersicht über die bei einigen licht- und dunkelaktiven Säugetierar- 224 F. Schanz und H. G. Erkert Tabelle 2 Vorwiegend über (+) oder unter (-) 24 h gelegene Spontanperiodenlänge T bei einigen licht- und dunkelaktiven Säugetierarten und Ausprägung des jeweiligen Asymmetrie- oder Richtungs- effekts bei der Resynchronisation der circadianen Aktivitätsperiodik nach positiven und negati- ven Phasenverschiebungen des LD Zeitgebers Die Angaben für den Menschen (Kreın 1975) beziehen sich auf die Resynchronisation der Kerntem- peratur und anderer Parameter nach Transmeridianflügen in westlicher und östlicher Richtung Species Spontanperiode Zeitgeber- tr größer (+)/ Autoren vorw. über verschiebung kleiner (-) (+)/ unter nach Ab+ als (-) 24 h) lichtaktiv Homo sapıens Kreın u. Wegmann (1975) Callıthrix jacchus ERKERT et al. (1986 dunkelaktıv Mesocricetus auratus PITTENDRIGH und DAAN (19762), Ponr (1978) Pon (1968, 1978) ASCHOFF (1978, 1979) ERKERT (1976) ERKERT (1970) Kracht (1979) Kracht (1979) KracHhr (1979) KracHr (1979) Kracht (1979) KLEINKECHT (1986) vorl. Arbeit vorl. Arbeit ERKERT et al. (1984) ERKERT und THIEMANN- JÄGER (1983), THIEMANN- JÄGER (1986) Glis glıs Rattus rattus Rousettus aegyptiacus Eidolon helvum Molossus molossus Molossus ater Phyllostomus discolor Sturniva llium Glossophaga soricina Petaurus breviceps Galago senegalensis Galago crassicandatus gar. Microcebus murinus Aotus trivirgatus Var ar | on nn on mn mn nn nn Mm MO MO 00 00 ON ON ++ I +4+++++ +1 A®+ = Vorverlagerung, AB- = Rückverschiebung des LD; tr = Resynchronisationszeit ten ermittelten Asymmetrieeffekte und zeigt zugleich, ob die jeweilige circadiane Spontan- periode vorwiegend über oder unter 24,0 h liegt. Ein Vergleich dieser Daten mit den Annahmen der drei genannten Hypothesen läßt erkennen, daf keine von ihnen ausreicht, um die gefundene Abhängigkeit der Resynchronisationsgeschwindigkeit von der Richtung der Zeitgeberverschiebung in jedem Fall zu erklären. Entscheidend dürften demnach andere Zusammenhänge und Mechanismen sein. Darauf deutet auch der Befund hin, daß beim Senegalgalago im LD 8:16 der schon im LD 12:12 gezeigte Asymmetrieeffekt zwar ebenso erhalten bleibt wie bei Ronsettus aegyptiacus (ERKERT 1976) und Aotus trivirgatus (ERKERT und THIEMANN 1981), daß aber dieses von 1:1 abweichende Lichtzeit-: Dunkel- zeit-Verhältnis bei ihm nicht, wie bei den beiden anderen Arten als „schwächere Zeitge- ber“ zu einer generellen Verlängerung der Resynchronisationszeiten führt, sondern gleiche (nach Rückverschiebung) oder sogar kürzere Resynchronisationszeiten (nach Vorverlage- rung des LD) zur Folge hat als nach entsprechenden Phasenverschiebungen des LD 12:12. Einer von PITTENDRIGH entwickelten Modellvorstellung über den Mechanismus der Synchronisation circadıaner Aktivitätsrhythmen zufolge (PITTENDRIGH 1960; PITTEN- DRIGH und Daan 1976; MOORE-EDE et al. 1982) spielt die Phasenresponse eines aus zwei miteinander gekoppelten Oszillatoren bestehenden endogenen Schrittmachersystems auf Licht, bzw. Beleuchtungsänderungen für die synchronisierende Wirkung von Beleuch- tungszyklen eine entscheidende Rolle. Zur Etablierung einer langfristig stabilen Phasenbe- Resynchronisation der circadianen Aktivitätsperiodik von Galagos 20 ziehung der biologischen Periodik zum Zeitgeberzyklus soll es nach diesem Modell nur dann kommen, wenn das Licht und/oder die Beleuchtungsänderungen des LD den Organismus während der frühen und späten subjektiven Nacht - dies entspricht bei dunkelaktiven Arten den ersten und letzten Stunden der Aktivitätszeit, bei lichtaktiven Arten der Ruhezeit - so treffen, daß die dadurch ausgelösten phasenspezifischen positiven und negativen Phasenverschiebungen gerade die Abweichung der Spontanperiode 1 von der Zeitgeberperiode T kompensieren. Wendet man dieses Modell auf die Resynchronisa- tion der CAP nach Phasenverschiebung des Zeitgebers an, so ergibt sıch folgendes: Nach einer positiven oder negativen Phasenverschiebung des LD-Zeitgebers treffen das Licht und/oder die Beleuchtungsänderungen plötzlich auf anders reagierende Phasen des circa- dianen Zyklus. Die dabei ausgelöste Nettoverschiebung der endogenen Periodik reicht deshalb in der Regel nicht aus, um die durch die Zeitgeberverschiebung stark vergrößerte Phasendifferenz zwischen beiden Rhythmen sofort auszugleichen. Dazu sind mehrere Übergangszyklen nötig. Während dieser Resynchronisationszeit wird der Organismus dann zu jeweils unterschiedlichen Phasen seiner Circadianperiode vom Licht, bzw. der Beleuchtungsänderung getroffen und reagiert deshalb vermutlich zum Teil auch mit unterschiedlich großen Phasenverschiebungen. Dies könnte dann die ım Verlauf der Resynchronisation häufig zu beobachtende Änderung der Resynchronisationsgeschwin- digkeit bewirken (Abb. 2, 3). Nach diesen Vorstellungen würde also die Resynchronisa- tionsdauer dann nicht nur von der Größe und Richtung der Zeitgeberverschiebung, sowie von der Größe des zu kompensierenden Periodenunterschiedes und damit letztlich von der Spontanperiodenlänge abhängen. Entscheidend wären auch das Lichtzeit-: Dunkelzeit- Verhältnis, das mitbestimmt, welche Phasen des circadianen Systems vom Lichtwechsel getroffen werden, und vor allem das Phasenresponse-Verhalten des jeweiligen circadıanen Schrittmachersystems, das sich im artspezifischen Verlauf der Phasenresponsekurve mani- festiert. Eine befriedigende Erklärung der bei den Galagos und anderen Säugetierarten gefundenen interspezifischen Unterschiede im Resynchronisationsverhalten und der Aus- prägung des Asymmetrieeffekts (Tab. 2), sowie der unterschiedlichen Auswirkungen geänderter Lichtzeit-Dunkelzeit-Relationen des LD-Zeitgebers auf die Resynchronisa- tionsgeschwindigkeit erscheint somit möglich, wenn nicht nur von der Spontanperioden- länge ausgegangen wird, sondern auch von der jeweiligen Phasenresponsekurve, deren Verlauf wiederum eine enge Beziehung zur Spontanperiode aufweist (PITTENDRIGH und Daan 1976). Obwohl bei der Diskussion der Ursachen für das Auftreten des Asymme- trieeffektes bei der Resynchronisation verschiedentlich auch schon die Möglichkeit eines Zusammenhanges mit der Phasenresponsekurve in Erwägung gezogen wurde (ERKERT 1976; AscHorFr 1978; PoHL 1978; ERKERT et al. 1984), stehen konsequente Analysen dieses Phänomens auf der Basıs konkreter experimenteller Daten über die Spontanperiodenlänge und die Phasenresponse-Charakteristika bislang leider immer noch aus. Solche Analysen wären aber auch insofern wichtig, als sie zugleich einen Beitrag zur Überprüfung der Relevanz des Phasenresponse-Modells für die Synchronisation circadianer Rhythmen durch LD-Zeitgeber leisten könnten. Danksagung Herrn PD Dr. CH. WELKkER, Kassel, danken wir sehr herzlich für die Überlassung der Riesengalagos. Zusammenfassung Bei 5 Galago senegalensis und 3 Galago crassicandatus wurde das Resynchronisationsverhalten der circadianen Aktivitätsperiodik (CAP) nach 8stündigen Phasenverschiebungen zeitgeberwirksamer LD-Zyklen von 560,07 Ix untersucht. Im LD 12:12 traten bei beiden Arten deutliche Asymmetrie- effekte auf. Während die Riesengalagos nach der Rückverschiebung des LD durchschnittlich 10 und nach der Vorverlagerung 6,3 Tage zur Resynchronisation benötigten, brauchten die Senegalgalagos dafür 5,0 bzw. 9,3 Tage. Nach entsprechenden Phasenverschiebungen eines LD 8:16 änderten sich 226 F. Schanz und H. G. Erkert die Resynchronisationszeiten bei den Senegalgalagos auf 5,6 und 7,4 Tage. Da die Spontanperioden- länge der CAP bei beiden Arten unter 24h liegt, entspricht das Auftreten gegensätzlicher Asymmetrieeffekte keiner der bisher diskutierten Hypothesen über die Ursache des Einflusses der Richtung der Zeitgeberverschiebung auf die Resynchronisationsgeschwindigkeit. Ausgehend vom PırrenprıcH’schen Phasenresponse-Modell der Synchronisation ciırcadianer Rhythmen wird disku- tiert, daß sich sowohl das Zustandekommen artspezifischer Asymmetrieeffekte als auch der Einfluß des Lichtzeit: Dunkelzeit-Verhältnisses 24stündiger LD-Zyklen auf die Resynchronisationszeit erklä- ren lassen dürfte, wenn dabei außer der Spontanperiodenlänge auch die Phasenresponse-Charakteri- stika der betreffenden circadianen Systeme berücksichtigt werden. Literatur ASCHOFF, J. (1969): Desynchronization and Resynchronization of Human Circadian Rhythms. Aerospace Med. 40, 844. — (1978): Problems of Re-entrainment of Circadıan Rhythms: Asymmetry Effect, Dissociation and Partition. In: Environmental Endocrinologie. Ed. by J. AssenmAcHER and D. $. FARNER. Berlin- Heidelberg-New York: Springer. 185-195. — (1979): Circadıan rhythms: Influences of Internal and External Factors on the Period Measured in Constant Conditions. Z. Tierpsychol. 49, 225-249. ÄASCHOFF, J.; HOFFMANN, K.; PoHL, H.; WEVER, R. (1975): Reentrainment of Circadian Rhythms after Phase-shifts of the Zeitgeber. Chronobiologia 2, 23-78. Bay, F. (1976): Untersuchungen zur Lichtsteuerung der Aktivitätsperiodik von Fledermäusen (Myotis myotis Borkh, 1797 und Phyllostomus discolor A. Wagner, 1843). Diss. Univ. Tübingen. ERKERT, $. (1970): Der Einfluß des Lichtes auf die Aktivität von Flughunden (Megachiroptera). Z. vergl. Physiol. 67, 243-272. ERKERT, H. G. (1976): Der Einfluß der Schwingungsbreite von Licht-Dunkel-Cyclen auf Phasenlage und Resynchronisation der circadianen Aktivitätsperiodik dunkelaktiver Tiere. J. interdiscipl. Cycle Res. 7, 1, 71-91. ERKERT, H. G.; NAGEL, B.; SCHANZ, F.; THIEMANN-JÄGER, A. (1984): Vergleichende Untersuchun- gen zur Chronobiologie nichtmenschlicher Primaten. Verh. Dtsch. Zool. Ges. 77, 217. Kıemım, K. E.; WEGMAnNN, H. M. (1975): Das Verhalten des menschlichen Organısmus beim Zeitzonenflug. Fortschr. Med. 93, 1497-1502. Kracht, $. (1979): Anpassungen circadianer Systeme — vergleichende Untersuchungen an neotropi- schen Fledermäusen. Diss. Univ. Tübingen. MOooRE-EDE, M. C.; SULZMAN, F. M.; FULLER, C. A. (1982): The Clocks that Time Us - Physiology of Circadıan Timing Systems. Cambridge, London: Harvard Univ. Press. PITTENDRIGH, C. $. (1960): Circadian Rhythms and the Circadıan Organization of Living Systems. Cold Spr. Harb. Symp. quant. Biol. 25, 159-184. PITTENDRIGH, C. $.; Daan, $. (19762): A Functional Analysıs of Circadıan Pacemakers in Nocturnal Rodents. I. The Stability and Lability of Spontaneous Frequency. J. comp. Physiol. 106, 223-252. PITTENDRIGH, C. S.; Daan, $. (1976c): A Functional Analysıs of Circadıan Pacemakers ın Nocturnal Rodents. V. Pacemaker Structure: A Clock for all Seasons. J. comp. Physiol. 106, 333-355. PoHr, H. (1968): Einfluß der Temperatur auf die freilaufende circadiane Aktivitätsperiodik bei Warmblütlern. Z. vergl. Physiol. 58, 364-380. — (1978): Comparative Aspects of Circadian Rhythms in Homeotherms. Re-entrainment after Phase-shift ot the Zeitgeber. Int. J. Chronobiol. 5, 493-517. THIEMANN-JÄGER, A. (1986): Charakteristika der circadianen Aktivitätsperiodik von Nachtaffen (Aotus trivirgatus, Humboldt 1811). Diss. Univ. Tübingen. WEVER, R. (1966): The Duration of Re-entrainment of Circadian Rhythms after Phase Shifts of the Zeitgeber. J. Theoret. Biol. 13, 187-201. WEVER, R. (1979): The Circadian System of Man. Results of experiments under Temporal Isolation. New York: Springer. Anschrift der Verfasser: F. ScHanz und Prof. Dr. Hans G. ERKERT, Institut für Biologie III, Universität Tübingen, Auf der Morgenstelle 28, D-7400 Tübingen, FRG Analysis of litter size in a colony of the common marmoset (Callithrix jacchus) By H. RoTHE, A. Könıc, K. Darms and MARGARETHA SIESS Institute of Anthropology, University of Göttingen Receipt of Ms. 27.7. 1986 Abstract In our C. jacchus colony the incidence of triplet litters is slightly higher than that of twin births. The observation of litter size incidence as an annual event, as has been done in the literature masks important details of litter sıze distribution. The wild-caught females of our colony gave birth to nearly twice as many twins than triplets, and we could not recognize a tendency to produce more triplets in conjunction with increasing length of captivity in our laboratory. The most conspicuous changes in the relative frequencies of litter size compared to wild-caught females occurred in the F,-generation. The relative frequency of quadruplets and triplets increased enormously, whereas that of twins and singletons decreased considerably. In the following generations (F, to F4) we observed again a definite increase in twin sets, with an associated decrease in triplet sets. Therefore the often repeated statement, that a refined management will lead to an increase of triplet deliveries, apparently is neither sufficient nor correct. Introduction As a rule marmoset females are multiparous (survey of the literature see ROTHE 1979). However, this observation originates from laboratory kept colonies, whereas very little information on this aspect in free ranging marmoset populations is available (e.g. Hus- RECHT 1984). Many marmoset colonies which are kept and bred under laboratory conditions, show an increase of triplet births, aphenomenon which has been considered by some authors to be due to the generally better living conditions, i.e. nutrition, medical care etc., of primates in captivity than would be possibly true for free living individuals (see for example Hiıppıeston 1978). KIRKwOOoD et al. (1983) observed a sudden and enormous increase of triplet births in Saguinus oedipus after the authors had reduced the ratio of carbohydrates in favour of protein in the daily diet. However, we do not know as yet whether the increase of triplet births in captive marmosets results from an augmentation of the ovulation rate or from a reduction of the prenatal death rate of ova/zygotes. From other anımal species, on the contrary, we have better information on this aspect. Voss (1950, 1952) (sheep), LÜDIKE-SPANNENKREBS (1955) (rabbit), and Boyz (1956) (pıg) observed an intensified follicular activity and an increased production of ova in captive individuals in comparison with their respective wild counterparts. Moreover, with some domestic anımals and some wild species, e.g. Clethrionomys glareolus (BRAMBELL 1948) a reduced prenatal mortality of ova/zygotes could be detected under laboratory conditions (see also ScHILLING 1952; MICHAELIS 1966; SADLEIR 1969); and, according to BoyE (1956) some uterine conditions, for example, size, weight, vascularisation, seem to have a considerable effect on fetal mortality. To what extent the genetical constitution of the marmoset females plays a role on the disposition to produce more triplets/quadruplets with increasing age, remains to be U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5204-0227 $ 02.50/0 Z. Säugetierkunde 52 (1987) 227-235 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 228 H. Rothe, A. König, K. Darms and Margaretha Sieß investigated. From human females for example we know, that the age of the mother and the parity are important factors associated with the occurrence of dizygote twins/triplets (VoGEL and MoTursky 1979). Material and methods 202 litters of the common marmoset, including five abortions (definition see Könıc et al., in press) from 44 breeding females of our colony which has been in existence since 1968, were analysed. Eight females were wildcaught (wc), 15 F,-, 15 F>-, 4 F3- and 2 F,-generation (Table 1). 8 of the 44 females are still living and breeding in our laboratory. Further details on the anımals are shown in Table 1. Results 46.04 % (n=93) of all identifiable litters (n=202) of our colony of common marmosets were triplets (Tr); twin (Iw) births occurred 80 times (=39.6 %); singletons (Si) 18 times (8.94%), and quadruplets (Qu) 11 times (=5.45 %). Abortions and premature births occurred in 5.94 % (n=12) of the births, whereby in 7 cases the size of the respective litter could not be ıdentified, the remaining five consisted of one triplet, two twins, and two singletons. No less than three abortions occurred in one female (F45, Gr. F, Table 2). A break down of the data on litter size according to the females’ generation shows the following results. The wild-caught females (n= 8, 53 litters) had fewer triplets (n=16, 30.19%) than twins (n=30, 56.60 %). Quadruplets occurred only once (F15, Gr. A1). Quite different results come from the F,-females (n=15, 83 litters). The percentage of quadruplet births amounted to 9.64 (n=8), of triplets to 61.45 (n=51), whereas twın deliveries were clearly in the minority (n=18, 21.69 %). Only slight differences are apparent between the relative frequency of singletons of wild-caught and F} females (wcF: 11,32 %, n=6; FıF: 7.23 %, n=6). The females of the F,-generation (n=15, 36 litters) differ widely from the F,-females as to litter size distribution. They had only 47.22 % (n=17) triplets, whereas the percentage of twın births was 44.44 (n=16). In the F3-generation the frequency of trıiplets dropped down to 26.9% (n=7), and the percentage of twin births increased to 56.69 (n=15). The F3-females have a relatively high frequency of singletons (=15.38 %, n=4). It must be stressed however, that ın relation to the number of females, the number of litters ıs substantially lower in the F,-generation than in the F,-generation. For that reason we cannot decide wether the striking differences in the percentage of the different litter sizes are characteristic for F>-females or are artifacts arısing from the differences in the number of litters of the two generations. The same restriction is true for the F3-females (n=4), although, all things considered, we have observed a strong tendency to have twin births. From the F,-females we have had only four deliveries up to now from two females, so that we cannot as yet comment on the distribution of litter sıze in that generation. The compilation of the data according to consanguinity of the breeding females of our colony reveals some more interesting results (Table 4). The most conspicuous, of which, is the familial accumulation of quadruplets. No less than six (=54.54%) of the eleven quadruplet sets occurring since 1968 were observed in four F,-sisters, daughters of one wc- founder female (F12, Gr. B, Table 1) of our marmoset population. If one includes a great- great-grand-daughter of F12, 63.63 % (n=7) of all quadruplets we have had hitherto, have occurred in clan B. Three Qus were born in clan C], two by the mother and one by one of her daughters (Table 4). That is, more than 90% of Qus of our C. jacchus females are restricted to two clans. Only one female (F15, wc, Table 1) which had Qus was not related to the other Qu-mothers. The direct comparison of the litter size of clan-mothers and of their respective Table 1. Survey of all groups and breeding females No. of Genera- Number of litter Com- female tion 1 9EE105 EDEN 16 7 E A920 ments ” 2 0015 0012 0011 0004 0014 0441 0439 0440 EIDOND -NnSngun ® ze) @I@® OOo DDr-uUVDDWm NDNNDNM Q DOW UDNDWD>NDWW-+N vuwumuun-wuub Na WwWH+N WW N WW WW WW WW DWDWN WW rWFREDND 155) no [857 ww 3 3 2 3 22 2 2 8 2 1 2 3 3 4 Do>5 RW a: following birth of litter 20, group was dissolved, female 0011 received another male companion | b: brother and sister were breeding c: father died shortly after birth ot litter 4, from then on the eldest son bred with the mother d: after birth of litter 8, alpha female was paired with a new male e: male partner of female 0076 was only 9.5 months when paired unintentionally f: female 0302 was presumably F1-generation g: litter 8 were presumably twins h: female 0357 died due to complications during birth of litter 1 ı: female 0356 was often in poor condition j: father and daughter were breeding k: female 0257 had been hand reared l: female 0123 was already pregnant when paired m: female 0231 was only 28 months old when litter 1 was born n: brother and sister were breeding we = wildcaught; A = abortion; ? = litter size not identifiable; 1 = singleton; 2 = twins; 3 = triplets; 4 = quadruplets 230 H. Rothe, A. König, K. Darms and Margaretha Sieß Table 2 Frequency of quadruplets, triplets, twins, singletons, and abortions Quadruplets Triplets Twins Singletons absolute frequency 11 93 80 18 relative frequency 5.45 46.04 39.60 8.91 abortions no. of fetuses ıdentifiable — 1 2 2 abortions no. of fetuses not ıdentifiable relative frequency of abortions daughters reveals the following results. F12, mother of clan B, had 13 litters with a balanced relation of Tr- and Tw-deliveries (n=6 each, 46.15 %). This female had only one Sı. Her five daughters on the contrary mainly had Trs (n=21, 53.85 %) and Qus (n=6, 15.37 %), but relatively few Tws (n=8, 20.51 %). Only slıght differences exist in the frequency of Sıs of the mother (= 7.69%) and her daughters (=10.26 %). If one compares the sıze of the litters of the clan-mother with those of all breeding females which are related to her, only minor differences can be observed in the frequency of Trs (46.15 % and 47.5 %); and the differences in the portion of Tws also is reduced. That is to say, mother and daughters have more differences when one compares the sıze of their litters, than the mother and the group of all her breeding female descendants. Clan mother F11 (Gr. C) has nearly as many Trs (=47.32%) as F12. F11 has relatively fewer Tw-deliveries (n=7, 36.84%) and more Sıs (n=3, 15.79 %). Her daughters showed remarkable differences to their mother in the distribution of litter size: 16 Tws, (26.09 %); 17 Trs (73.91 %). Sıs were not born. In clan C too, with the exception of the relative frequency of Sıs (n=4, 6.56 %), the group of all female descendants of the clan mother F11 does not deviate so far the litter sıze distribution of the founder female as does the group of her daughters. Litter size distribution in clan C shows greater differences from the distribution of the total population of breeding females of our C. jacchus colony than holds true for clan B. Female F269 (clan CD) resembles in the litter size distribution her great-grand-mother (F12, clan B), except for the frequency of singletons (n=2, 16.67 %). Also in this group of related females the daughters deviate more strongly from the litter size distribution of their mother than does the group of all female descendants from their founder female F269. Except ot slight differences in the litter size distribution, clan CD reflects nearly exactly the colony distribution. Founder female F59, clan K, daughter of F12, shows an equal frequency of Tr- and Tw- births (n=4, 57.14 % rsp. n=3, 42.86 %), with the daughters resembling their mother ın this respect. However, the number of litters ıs very low (n=5). Clan V (F123) shows a nearly identical distribution of litter size with that of clan mother F269 (Gr. CD), a daughter of F123. As a whole, both clans differ mainly only ın the frequency of Qus. In the other three clans (CE, F302; CJ, F349; CL, F356) the number of litters is too low to be described separately (see Table 4). The analyse of litter sıze according to parity reveals following results. The wc-females (n=8) predominantly have Tws (n=45) even when they have lived for many years in captivity. At first one can observe an increase of Tw-deliveries to the disadvantage of TIrs and Sıs. In the F,-generation we detected a great spread of the different litter sizes along the sequence ot deliveries (15 females, 82 litters). The Trs dominate from the start of the sequence, but there is no further increase of Trs with increasing parıty whereas this is true for Qus, mainly to the detriment of Sis and Tws. Also in the F,-generation (15 females, 35 litters) we did not observe a continuous increase of Trs in relation to parity but instead a 'soJeuw1sF JuePUIISIP ZUIP99Ig [fe pur 9yyour uej9 = A !sIsaydnep pur Jsyıouı uepp = AI “I9yJ0W UEII JO SOeWIaF JuePU99SaP Surpaa1g [fe = III “19y2ow ueJ9 Jo SIOJySnep = ]J !9yIow urf9 = ] 'SUOI[SUIS = S !surm = ML !sıopdın = 1], ‘sıapdnıpenb = U Or S’ZS SUR EA SL SG @8 OI BE az A BERDISSETE 008 0°07 e8 STLTESEEVEEI ESS OEREEEELT I6 69 615 SI I S € 1% I I RL BE E97 AT) AI Ol sl 6ch 8rS SEE TLV88 OST ORIG el [4 G S 9 USERS CE SET6ICH, IE 013 Soc 685 FSsl e S 9 Dr] 8 le) BE EKE ER ZN 0°09 0'Or 8’9E vlt GIVACDN, 6 [ 9ZIS -ON1T ur) sapewaz PaYep2ı ur 3Z15 19991] Jo Aduanbaıy sAep2ı pue 3Injosqy ZA SUIMI = MJ !s20[dın = 1], Aydnespfim = 9m !aaqumu »ınjosge = N '[geipmuapt st 9Z1S 19T = 1 fopgerpmuopt 30U ST 9ZIS Jan] = Tu 096€ r0’9r c6 el! vb < 0’Sc 0°0S Ze | se’sı ES c6 97 € 95°S yrvr cclr cıI U@Z Zae 19 [Rei ce Tl 39S coE I % % N ! Lu suopfäurg ssopdıı], sıo[fdnıpendy suonIogqy N SIWTN SIPAPGN Sofewa] N /UoneIuss uUoNEI9U93 ‚sajewsF 3} 03 uonep.ı ur 9Z1S 19921] Jo Aduanbaıy sAryej31 pue aynjosqy EU 292 H. Rothe, A. König, K. Darms and Margaretha Sieß Table 5 Number of groups and females; generation; number of litters and litter size No. of Natal Gene- Number of litter female Group ration 2a az a TO 20 [p) = ° [=] ae) 1057 BE 3 DEN a DEN 2 N 3:33 132714, 3838 12) BI DEP @ & E € E M M IS us) = @) A De Clan CE @) es w 2 Clani@E 2 ® rm a SUSE C] Clan CJ we = wildcaught; A = abortion; 1 = singleton; 2 = twins; 3 = triplets; 4 = quadruplets; ? = litter sıze not identifiable. rather great fluctuation of litter size occurred, which as a whole was shown by a decrease of Trs in favor of Sıs and Tws. The increase in parity of the F;-females (n=4, 26 litters) are characterized by a further decrease of the frequency of Tr-births, however, not continuously, and by an increase of Tws. A comparison of the litter with increasing parity of the consanguineous breeding females of our colony does not indicate a continuous increase in the number of Tr- deliveries, but shows a great fluctuation along the total sequence of births. Captive born females of C. jacchus show a higher frequency of Trs in the first lıtter Litter size in Callıthrix jacchns 299 onwards, than do wc-females, but there is no further increase in the incıidence of Trs during their individual lives. The analysıs of the question whether females which were born as a Tr would also give birth predominantly to Trs of Qus gave the following results. With 32 females of our colony we know the size of the litter in which they were born: 15 in Tr-, 14 in Tw-sets and 3 were singletons. 11 of the 15 Tr-females (=73.33 %) predominantly had Trs as well, whereas the rest (=26.67 %) did not correspond in this respect. In the females born in Tw- sets we could not find a tendency to give birth to Tws, they had just as many Tws as Trs (=42.86 % each). The singleton-females also, did not show any correspondance between the size of their litters and the fact they were singletons. Discussion The analysis of litter size of our C. jacchus females has shown, that as a whole the incidence of Tr-litters is slightly higher than that of Tw-births (see however Grıst 1973; HEARN et al. 1975; PhHırLLıps 1976; HıppLeston 1977; HamPpTon et al. 1978; OGDen et al. 1978; McIntosH and LOoKkER 1982). When one observes the litter size incidence as an annual event, as has been done in the literature, one sees a relative increase in number of the Tr- and Qu-deliveries and a decrease in the number of Tw-sets (HıppLeston 1976; POOLE and Evans 1982; Burt and Prant 1983). This pattern was also observed ın our colony (see Fig.). A more precise investigation, however, illuminated the fact that such a global quadruplets 100 MM triptets | | = twins |] singletons [6] S er oe LIITLTLLELITIIILIIIIIIIID 88 —— — ZENRIINRANN I\\ \\\ 1 RRRRRUUNN ü = PITEIIEITTTI EENN RISISIS vr S S © SSSY SFTIIIITLTIITTTILIIIITITITIIID => U TIITIETITIIIITIITTTETTITITITITIT IN zZ u ©9 PELITIILIIITIIITIITITITTTITTTD NEITIIIILIT 1968 _ wo _ (10) 1986 year Ei g. Annual variation of litter size in a colony of common marmosets Im investigation, masks important details of litter size distribution. The eight wc-females of our colony gave birth to nearly twice as many Tws than Trs, and, moreover, we could not recognize a tendency to produce more Trs in conjunction with increasing length of captivity in our laboratory. On the contrary, they had a rather balanced frequency ot Tr- and Tw-sets. The most conspicuous changes in the relative frequencies of Tr-/Qu- and Tw-/Si-deliveries compared to wc-females occurred in the F,-generation, that is, in the daughters of the wc-females. The relative frequency of Qus increased by more than 300 %, that of Trs by more than 100% whereas the frequency of Tws decreased by 150 %, and that of the Sis by about 50 % (Table 3). However in the following generations, F, to F4, we could detect major differences in litter size distribution compared to that in the F}- 234 H. Rothe, A. König, K. Darms and Margaretha Sieß generation, with a tendency to produce more Tws than Trs, than that already observed in the wc-females. This means, that in comparison to F,-females we had observed a definite increase in T'w-, with an associated decrease in Tr-sets. A comparable result was described by Puırıps (1976), although the author did not make it quite clear wheather the “parents” were wild-caught or captive born. In his colony Tw-sets predominated (66.67 %, 159 litters). The often repeated statement, that a refined management (e.g. HıppLeston 1978) will lead to an increase of Tr-deliveries, apparently is neither sufficient nor correct. Then, this would mean that the wc-females must have the greatest increase of Qu-/Tr-deliveries, due to some kind of “flushing effect”’; however, this is most probably not the case. Even if there is increased ovulation rate due to better nutrition, medical care etc., the implantation rate would not be increased nor would the prenatal death rate/resorption rate of ova and/or zygotes be affected. This result as well as the observation that a once reached level in the frequency of Tr-sets will not be further increased ın the course of a female’s individual life, corresponds with results from the breeding of domestic animals. In domestic animals it has been found that a permanent and substantial supply of food has no positive effect on litter size, even the contrary may be the case, but an abrupt change, from a spare to an optimal feeding results in the phenomenon known as “flushing’ - an increase in the number of ova being released (see for example Kırkwoonp 1983 for S. oe. oedipus). The enormous increase of Trs in the fırst litters of the group of F,-females may in fact be a consequence of better colony management including a reduction of any stress-producing factors, or it may even be due to better adaptation to life under laboratory condition. At least the distribution of the Qus does not exclude the assumption that litter size has a management independent familiar disposition, but the change of the relative frequencies of Tr- and Tw-deliveries ın the following generations, however, cannot be explained at the moment. The data of other C. jacchus colonies should also be analysed under this aspect in order to ascertain the true relationship between litter size and other factors. Acknowledgements We are greatly indebted to Miss. T. GATEsMAN, German Primate Center, for her help in translating this article into English and to Miss U. RADESPIEL for technical assistance. Zusammenfassung Analyse der Wurfgröße in einer Kolonie von Weißbüscheläffchen (Callithrix jacchus) 202 Würfe von 44 züchtenden Weibchen des Weißßbüscheläffchens wurden analysiert. Acht Weibchen waren Wildfänge, 15 gehörten der F,-, 15 der F;-, vier der F3- und zwei der F,-Generation an. 46,04% (n=93) aller Würfe, deren Größe zweifelsfrei festgestellt werden konnte, waren Drillinge, 39,6% (n=80) Zwillinge, 8,94% (n=18) Einlinge und 5,45% (n= 13) Vierlinge. Wildfangweibchen hatten weniger Drillinge (30,19%) als Zwillinge (56,60 %). Bei den F,-Weibchen stieg der Anteil der Drillinge auf 61,45%, während Zwillingsgeburten klar in der Minderheit waren (21,69%). F>- Weibchen hatten nur 47,22% Drillinge und 44,44% Zwillinge. In der F,-Generation sank die Häufigkeit der Drillingsgeburten auf 26,9% und der Anteil der Zwillingsgeburten stieg auf 56,69 %. Bei den F4-Weibchen gab es bis jetzt nur vier Geburten von zwei Weibchen, so daß wir über die Verteilung der Wurfgröße in dieser Generation noch nichts sagen können. Nach ihrer Wurfgröße unterscheiden sich Mutter und Töchter stärker als die Mutter und all ihre züchtenden weiblichen Nachkommen zusammen. Aus Drillingsgeburten stammende Weibchen warfen hauptsächlich Dril- linge, während wir bei Weibchen aus Zwillingswürfen keine Neigung zu Zwillingsgeburten feststellen konnten. Sie hatten ebensoviele Zwillinge wie Drillinge (je 42,86 %). References BoyE, H. (1956): Vergleichende Untersuchungen über die arterielle Gefäßversorgung des Uterus von Wild- und Hausschweinen. Z. Tierzüchtg. Züchtbiol. 67, 259-296. BRAMBELL, R., (1948): Prenatal mortality in mammals. Biol. Rev. 23, 370407. Litter size in Callithrix jacchus 235 Burr, D. A.; PLAnT, M. (1983): Observations on marmoset breeding at Fisons. Anımal Technology 34, 29-36. Grıst, S. M. (1974): The common marmoset (Callıthrix jacchns) — a valuable experimental anımal. ]. Institute Anımal Technicians 27, 1-7. Hampton, $S. H.; Gross, M. J.; HamPToNn, Jr. J. K. (1978): A comparison of captive breeding performance and offspring survival in the family Callitricidae. In: Primates in medicine. Vol. 10: Marmosets in experimental medicine. Ed. by. E. I. Goldsmith and J. Moor-Jakowski. Basel: Karger. S. 88. HEaARNn, J. P. (1983): The common marmoset (Callıthrix jacchus). In: Reproduction in New World Primates. Ed. by J. P. Hrarn, Lancaster: MTP Press. 183-212. HEARN, J. P.; BURDEN, F. J. (1979): “Collaborative” rearing of marmoset triplets. Lab. Anim. 13, 131-133. HEaARNn, J. P.; Lunn, $. F.; BURDEN, F. ]J.; PiLCHER, M. M. (1975): Management of marmosets for biomedical research. Laboratory Animals 9, 125-134. Hıppıeston, W. A. (1976): Large scale production of a small laboratory primate - Callithrix jacchns. In: The laboratory animal in the study of reproduction. Ed. by T. H. ANTIKATZIDES, $. ERICHSEN and A. SPIEGEL. Stuttgart: G. Fischer. 51-57. HippLeston, W. A. (1977): Large scale production of a small laboratory primate: Callithrix jacchus. (unpubl. Ms.) Hıppıeston, W. A. (1978): The production of the common marmoset, Callithrix jacchus, as a laboratory anımal. In: Recent Advances in Primatology, 2. Ed. by CHivErs and LANE-PETTER. London: Academic Press. 173-181. HUBRECHT, R. C. (1984): Field observation on group size and composition of the common marmoset (Callıthrıx jacchus jacchus) at Tapacura, Brazil. Primates 25, 13-21. KIRKWOOoD, J. K.; EPSTEIN, M. A.; TERLECKT, A. J. (1983): Factors influencing population growth of a colony of cotton-top tamarıns. Laboratory Animals 17, 3541. LÜDIKE-SPANNENKREBS, R. (1955): Studien über die Anzahl der Eizellen von Wildkaninchen und verschiedenen Hauskaninchenrassen. Z. mikrosk. — anat. Forschung 61, 454-486. McInTosH, G. H.; LOOKER, J. W. (1982): Development of a marmoset colony in Australia. Lab. Anı. Scı. 32, 677-679. MicHazLIs, H. (1966): Der weibliche Genitalcyclus der Rötelmaus Clethrionomys glareolus (Schreber, 1780) und Zusammenhänge mit der vorgeburtlichen Sterblichkeit. Z. wiss. Zool. 174, 290-376. OcDen, J. D.; WoLFE, L. G. (1979): Reproduction of wild-caught marmosets (Sagninus labiatus labiatus) under laboratory conditions. Lab. Anım. Scı. 29, 545-546. ÖOGDen, J. D.; WOLFE, L. G., DEINHARDT, F. W. (1978): Breeding Saguinus and Callıthrix species of marmosets under laboratory conditions. In: Primates in medicine. Ed. by E. I. GoLDsMITH and ]. Moor-JankowsK1, Basel: Karger. 1978. 79. Pritrips, I. R. (1976): The reproductive potential of the common cotton-eared marmoset (Callıthrix jacchus) i in captivity. J. med. Primatol. 5, 49-55. BoorEs a B.; Evans, R. G. (1982): Reproduction, infant survival and productivity of a colony of Common marmosets (Callıthrix jacchus jacchus). Laboratory Anımals 16, 88-97. ROTHE, H. (1979): Das Fthogramm von Callıthrix jacchus Erxleben, 1777 (Primates, Ceboidea, Callitrichidae). Eine morphaktische Analyse des Verhaltens mit besonderer Berücksichtigung des sozialen Umfeldes. Unpubl. Ms., Univ. Göttingen. SADLEIR, R. M. F. S. (1969): The ecology of reproduction in wild and domestic mammals. London: Academic Press. ScHILLıng, E. (1952): Studien zur Kausalität vorgeburtlicher Sterblichkeit. Z. Tierzüchtg. Züchtbiol. 60, 263-231. STEVENSON, M. F. (1979): The behaviour and ecology of the common marmoset (Callithrix jacchus jacchus) in its natural environment. In: Biology and Behaviour of Marmosets. Ed. by H. ROTHE, H.-J. WoLters and J. P. HEarn. Göttingen: Selbstverlag Rothe. 1979. 298. VocGeL, F.; MoTuLsky, A. G. (1979): Human Genetics. Berlin: Springer. Voss, G. (1950): Messend-zoologische Untersuchungen an Ovarien verschiedener Schafrassen. Zool. Anz. 145 (Klatt-Festschrift), 1028-1056. — (1952): Rassenanalytische Untersuchungen an Eierstöcken von Schafen. Zool. Jb. Abt. Allg. Zool. 72, 438-467. Authors’ addresses: Priv. Doz. Dr. HARTMUT ROTHE, ANDREAS KÖönIG, KURT DARMS, MARGARETHA Sıess, Institut für Anthropologie der Universität Göttingen, Bürgerstraße 50, D-3400 Göttingen, FRG Multivariate morphometrics of Vesper mice (Calomys): preliminary assessment of species, population and strain divergence By M. Corrı, MARIA SUSANA MERANI and GLORIA DE VILLAFANE Department of Animal and Human Biology, University of Rome and Microbiology, Parasitology and Virology Faculty of Medicine, University of Buenos Aires Receipt of Ms. 15. 8. 1986 Abstract Investigated the morphometric varıation of three species of the South American genus Calomys (C. callidus, C. laucha, C. musculinus). Canonical Varıiate Analysis (CVA), performed on 13 measure- ments from the left mandible, discriminates categorically the three species (p<0.01) and allows for a- posteriori identification of C. laucha and C. musculisnus. These two morphologically very similar species are the vectors of the viral Argentine Haemorrhagic Fever. CVA performed on 4 natural populations and 2 laboratory strains from C. muscnlinus distinguishes successfully the groups (p<0.001) showing a pattern of phenetic relationships congruent with the origins of the strains. These results suggest that multivariate morphometrics is a powerful tool to investigate on natural popula- tions which are genetically determined to become viral infected. Introduction Systematics of the South American genus Calomys ıs still not completely clear. Some authors recognize different species in the genus (THoMmAs 1916; TATE 1932; ELLERMAN 1941; CABRERA 1961; HERSHKOVITZ 1962; Honackt et al. 1982). Out of the numerous putative species described, cytogenetic survey has been carried out and banding patterns have been established for 3 species only, namely C. musculinus (FORCONE et al. 1980), C. laucha (VıruLLo et al. 1983), and ©. callidus (VıruLro et al. 1984). C. musculinns shows a 2n = 38 karyotype, whereas C. laucha has 2n = 64 and C. callidus 2n = 48. C. musculinus and C. laucha are morphologically very sımilar and live ın sympatry in Buenos Aires province and in the South of Cordoba province (Fig. 1). Massoıa et al. (1968) found some differences in the dental features and in tail length between C. musculinus and C. laucha which however are not useful to distinguish juvenile and/or oldest stages of the two species. Distinctness ın karyological and allozyme patterns (GARDENAL et al. 1977) suggests that the two species are reproductively ısolated and no hybridization occurs. C. callidus belongs to a group of forms which are morphologically larger (CABRERA 1961) and lives in the Mesopotamia and in the South of Santa Fe province where it overlaps the distribution area of the former two (Fig. 1). These three species are the best known within the genus for their enormous importance as vectors of the Junın virus, the ethiological agent of Argentine Hemorrhagic Fever (Paronpi et al. 1959; SaBartrını et al. 1977; SaBaTtını and CoNTIGIanI 1982). C. laucha and ©. musculinus have been identified as the real vectors of this arenavirus, and they present viral persistence (SABATTINI et al. 1977). In overlapping areas C. callidus can be infected by these species and becomes a vector itself of the disease (VIDELA et al. 1985). U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5204-0236 $ 02.50/0 Z. Säugetierkunde 52 (1987) 236-242 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Multivarıate morphometrics of Vesper mice (Calomys) 237, 70 60 50 70 60 50 Fig. 1. Distribution area of the three species of Calomys based on data from literature and observations by the authors. C. callidus 7; C. laucha ©; C. musculinus ©. Numbers indicate capture localities as follows: 1 = Pergamino Circulo 3; 2 = Pergamino Circulo 4; 3 = El Palmar; 4 = Laguna Larga; 5 = Chapadmalal; 6 = Laboulaye In the present paper we use a multivarıate morphometric approach to assess phenetic relationships among these three species ın order to determine the best suitable functions for an a-posteriori classification of virus infected specimens. Moreover, we want to test whether multivariate morphometrics could detect any intra-specific differentiation among populations which could be consistent with their different sensibilities to become persist- ently infected by the virus. Multivariate morphometrics are an appropriate tool to investigate geographical varıiation and racıal affinities (GouLp and JOoHnston 1972; REYMENT et al. 1984; THORPE 1976), to discriminate closely related species (Corri et al. 1985) and to ıdentify house mouse laboratory strains (FEstınG 1972, 1976; LEAMmY 1975, 1977) and patterns of growth (ATCHLEY et al. 1985). Most of these studies use the mandible as source of data because it represents a pool of highly heritable traits (FEsting 1972, 1976; LEeAmy 1975, 1977; 'THORPE et al. 1982; Cortt et al. 1985; Hausser 1985), ıs usually preserved in museum collections and for the accuracy and rapidity in recording pro- cedures. Material and methods The analysis was performed on 100 specimens belonging to the three species, i.e. one population from C. callidus, 2 from C. laucha, 4 natural populations and two laboratory strains from ©. musculinus (Tab.; Fig. 1). The specimens were sampled during ecological and virological surveys carried out through capture and recapture in epidemiological and non epidemiological areas. Only males were collected from the field not to alter the number of females in reproductive stage. Only one sex was therefore analysed in the present report. This does not particularly perturb the assessment of divergence and relative similarities between the groups since the species exhibit sexual dimorphism and separate analyses should have been performed for the two sexes. Only adults were used to minimize the effect of allometric variation associated with growth. 13 measurements on the mandible of 100 specimens were recorded (Tab.), following the rapıd and 238 M. Corti, Maria Susana Meriani and Gloria de Villafane Table Species, populations and geographic location of the Calomys samples examined Species Population Geographic location Number of Specimens Calomys callidus El Palmar 323075, 58=W 11 Prov. Entre Rıos. Parque Nacional El Palmar Calomys laucha Pergamino 995045400322 Circulo 3 Prov. of Buenos Aires Pergamino 33>.50’ S, 60>.32 W Circulo 4 Prov. of Buenos Aires Calomys musculinus Laguna Larga 30>.25’ S, 64>.25’ W Cuartel 6 Prov. of Cordoba Chapadmalal SS 0013738. Prov. of Buenos Aires Laboulaye 34>.5' S, 63>.24'’ W Prov. of Cordoba Pergamıno 33>.507 5 60> 32 Circulo 3 Prov. of Buenos Aires vi /H CNEA Commision Nacional de strain Energia Atomiıca. Buenos Aires vi strain Istituto de Virologia Cordoba accurate method described by FestınG (1972). To avoid lateral asimmetry only left mandibles were used. All data were converted to logarithms to render the relationship between them linear. To assess distinction and phenetic relationships between species/populations Canonical Variate Analysıs (CVA) was performed. CVA ordinates groups so that the between-group variation is maximized relatively to within-groups varıation. Two different CVAs were performed, the first on the 7 natural populations of the three species, the latter on the 4 C. musculinus populations plus the two laboratory strains. Mahalanobis distances between group centroids were used to compute UPGMA phenograms (SNEATH and SokAL 1973). CVA was performed through the 7M program of the BMDP (1983) package. All data are available upon request from the authors. Results On the basıs of CVA performed on the 7 natural populations the three species can clearly be distinguished. The separation of species and populations on the group means is given in Figure 2. All the between-species F-values are highly significant (P< 0.01). The classifica- tion matrix shows an average percent of correct classıfıcation of 95.97 % (C. callıdus = 100 %; C. laucha = 95.24 %; C. musculinus = 92.68 %). The first canonical axis separates the species in two distinct groups, one with the big C. callıdus and one with the smaller C. musculinus and C. laucha, according to classical taxonomy. The second canonical axis clearly discriminates C. musculinus from C. laucha and no intermediate forms between them are observed. CVA performed on the four natural populations and the two laboratory strains of C. musculinus discriminates categorically (p<0.001) the sıx groups. The first, second and third canonical axes absorb 49.27 %, 20.43 %, 12.24 % of the total variation respectively, to explain comulatively the 81.24 % of the total dispersion (Fig. 3). The classification matrix shows a percentage of correct classification of 83.6 % (min. 66.2 %, max. 100 %). The UPGMA computed on Mahalanobis distances between group means shows the pattern of phenetic similarıty between populations and laboratory strains (Fig. 4). Multivariate morphometrics of Vesper mice (Calomys) 239 Fıg. 2. Canonical Variate Analysis of the three species of Calomys. In this 3 dimensional plot the vertical axis is the third and final canonical variate. First axis evidences differences in size between C. callidus (W) and C. musculinus (©) and C. laucha (@) Discussion It is generally assumed that the species of the genus Calomys, on the basıs of their size, are to be classıfied in two distinct groups, one with the big forms, including C. callidus, C. callosus, C. venustus (VITULLO et al. 1983, 1984; ReıG 1984; RoLDaNn et al. 1985), and one with the smaller ones, ıin- cluding C. musculinus, C. laucha, C. hummelineki (Tuomas 1916; Husson 1960; HERSHKOVITZ 1962; Massoıa et al. 1968). The phenetic distinction of C. callıdus from C. laucha and C. mus- culinus along the first canonıcal variate seems to fit well with the foregoing statements (Fig. 2). Second canonical varıate de- scribes more evidently changes between C. musculinus and C. laucha which do not seem related with size (Fig. 2). Categorical morphometric distinction be- tween these two full species is completely congruent with kary- ological (FoORcoNE et al. 1980; VıruLLo et al. 1983, 1984) and alloeenzymatic evidences (GAR- Fig. 3. Canonical Variate Analysis of the four natural popu- DENAL et al. 1977). This conclu- lations and the two laboratory strains of C. musculinus. The sion is supported by lack of inter- vertical axis is the third Canonical variate. Note the distinct- »1:___ ness of the Pergamino (1) population in respect to the close inediate forms due v 5 ybridiza relationship of Chapadmalal (2) and Laboulaye (3). vı (4) dom in Syaandte conditions asıM and vi/H (5) strains are more related to their common Pergamino (Tab., Fig. 1). ancestor Laguna Larga (6) 240 M. Corti, Maria Susana Meriani and Gloria de Villafane Whilst karyology, alloenzy- vi/H mes and multivarıate mor- phometrics contribute indepen- dently to fully identify the LAGUNA LARGA species, they do not allow to ın- fer any phylogeny for the group. Further studies are necessary to elucidate whether the distinction VI CHAPADMALAL LABOULAYE in big and small Calomys species PERGAMINO is a morphological convergenee —— ——— —— or a cladıstic divergence. 4 2 0 Figures 3 and 4 summarize the Fig. 4. This UPGMA clustering has been computed on results of the analyses performed sr Mahalanobis distances between group means and shows the on the four populations and the phenetic relationships between the four natural populations two laboratory strains from C. and the two laboratory strains of C. musculinus musculinus. The isolate position of the Pergamino population (Fig. 3) and its high distance from the other groups (Fig. 4) is evident. Pergamino is the center of the endemic area of the Argentine Haemorrhagic Fever and probably this population responds differently to the virus in the wild. Relative distinction and close clustering of the Laguna Larga population with the two laboratory strains (Fig. 3, Fig. 4) show a pattern of phenetic similarity which is congruent with the well known origin of these strains. “vi” strain (Tab.) originated from wild anımals from Laguna Larga some 20 years ago and has been maintained in laboratory conditions for over 50 generations (DE VILLAFANE 1981). “vi /H” strain (Tab.) derived from F 40 vı strain and was bred at the same room temperature and light system, but it is evident that differences in husbandry and mating systems were sufficient to establish a subline divergence within few generations. Strikingly, the morphometric divergence between these sublines is even stronger than the one observed in the two natural populations from Chapadmalal and Laboulaye (Fig. 3, Fig. 4) which are geographically far apart (Fig. 1). The close morphometric similarity between Laboulaye (periendemic to the virus- infected area) and Chapadmalal (a non virus-endemic area) populations, might suggest that the latter could become persistently infected by the virus ın the future. As a matter of fact, the endemic area progressively expanded northward in the past decade (WEISSENBACHER and DAMoNTE 1983) and a southward expansıon can be expected unless some mechanism would oppose it. Our results suggest that mandible multivariate morphometrics represent a powerful tool to investigate on the dynamics of virus expansion, for species identification and for assessing population and subline divergences. Acknowledgements We thank Aprıana Kajon for her valuable help and generous assistance in recording data. We especially thank ERNESTO CAPannA for continuous help and encouragement. This work was supported by Fundacion Emilio Ocampo, Secreteria de Ciencia y Tecnica, Argentina, and Scuola di Dottorato dı Ricerca in Biologia Evoluzionistica dı Roma, Italy. Zusammenfassung Multivariate Morphometrie an Vespermäusen (Calomys): vorlänfige Abschätzung von Arten, Populationen und Zuchtstammunterschieden Die Variation von 3 Arten des südamerikanischen Genus Calomys (C. callidus, C. laucha, C. musculinus) wurde morphometrisch untersucht. Dazu dienten 13 Maße an linken Mandibeln. Mit diesen konnten kanonische Analysen durchgeführt werden. Deutlich waren die Angehörigen der 3 Arten voneinander zu trennen, vor allem wurde eine Identifikation von C. laucha und C. musculinus Multivariate morphometrics of Vesper mice (Calomys) 241 möglich. Diese beiden morphologisch sehr ähnlichen Arten sind Träger des virusbedingten argentini- schen Blutfiebers. Zusätzlich konnten durch kanonische Analysen 4 natürliche Populationen und 2 Laborzuchtstämme der Art C. musculinus deutlich unterschieden werden und eine Nähe der Zucht- stämme zu den Populationen ihrer Abstammung. Diese Ergebnisse deuten an, daß multivariate Morphometrie ein wichtiges Hilfsmittel darstellt, um natürliche Populationen zu untersuchen, die genetisch determiniert sind, durch Viren infiziert zu werden. References ATCHLEY, W. R.; PLUMMER, A. A.; Rıska, B. (1985): Genetics of mandible form in the mouse. Genetics 111, 555-577. BMDP (1983): Biomedical computing programs, Health Science Computing Facility. Univ. Califor- nase, A. CABRERA, A. (1961): Catalogo de los mamiferos de America del Sud. Rev. Mus. Arg. Cs. Nat. Zool. Vol. 4, 309-732. CorTI, M.; Loy, A.; AzzaroLı, M. L.; Capanna, E. (1985): Multivariate analysis of osteometric traıts ın Italian moles (genus Talpa). Z. Säugetierkunde 50, 12-17. ELLERMAN, J. R. (1941): The families and genera of living rodents. London: British Museum (Nat. Hist.). Vol. 2, 1-690. Festing, M. (1972): Mouse strain identification. Nature 238, 351-352. FestinG, M. (1976): Phenotipic varıability of inbred and outbred mice. Nature 263, 230-232. FORCONE, A.; Luna, M.; KrAVETZ, F. O.; Lısantı, J. A. (1980): Bandas C y G de Calomys musculinus (Rodentia, Cricetidae). Mendeliana 4, 57-65. GARDENAL, C. N.; JUAREZ, N. T.; GUTIERREZ, M.; SAaBBATInı, M. $. (1977): Contribuciön al conocimiento de tres especies del genero Calomys (Rodentia, Cricetidae). Physis, Secc. C 36, 169-178. GouLp, S. J.; JoHNsSTonN, R. F. (1972): Geographic varıation. Ann. Rev. Ecol. Syst. 3, 457-498. Hausser, J. (1984): Genetic drift and selection: their respective weights in the morphological and genetic differentiation of four species of shrews in Southern Europe (Insectivora, Soricidae). Z. zool. Syst. Evolut.-forsch. 22, 302-320. HERSHKOVITZ, P. (1962): Evolution of neotropical cricetine rodents (Muridae) with special reference to the phyllotine group. Feldiana Zool. 16, 1-524. Honack1, J. H.; Kınman, K. E.; Koeppr, J. W. (1982): Mammalian species of the world: a taxonomic and geographical reference. Lawrence, Ka.: Allen Press and Ass. Systematic Coll. Husson, A. M. (1960): A new species of the rodent Baiomys from Aruba and Curazao. Stud. fauna Curazao 10, 33-40. LEAMmY, L. (1975): Component analysis of osteometric traits in randombred house mice. Syst. Zool. 24, 176-190. LEAmY, L. (1977): Genetic and environmental correlations of morphometric traits in randombred house mice. Evolution 31, 357-369. MassoIA, E.; FORNES, A.; WAINBERG, R. L.; FRONZA, T. G. (1968): Nuevos aportes al conocimiento de las especies bonaerenses del genero Calomys (Rodentia, Cricetidae). Rev. Inv. Agrop., Inta, Argentina, 4, 63-92. PARODI, A. $.; DE LA BARRERA, J. H.; RUGIERO, M. C.; GRENWAY, D. S.; YERGA, M.; METTLER, M.; BoxAacA, M. C.; FRIGERIO, M. (1959): Los reservorios del virus de la Fiebre Hemorragica epidemica de la provincia de Buenos Aires. Prensa Medica Argentina 46, 554-556. Reıc, ©. A. (1984): Significado de los metodos citogeneticos para la distincion y la interpretacion de las especies con especial referencia a los mamiferos. Rev. Mus. Arg. Cs. Nat. 13, 19-44. REYMENT, R. A.; BLACKITH, R. E.; CAMPBELL, N. A. (1981): Multivarıate morphometrics. London: Academic Press. RoıLvan, E.R. S.; VıTuLLo, A. D.; MERANI, M. $S.; von LAWZEWITESCH, 1. (1985): Cross fertilization in vivo and in vitro between three species of Vesper Mice, Calomys (Rodentia, Cricetidae). J. Exp. Zool. 233, 433-442. SABATTINI, M. N.; GONZALES DE Rıos, L. E.; Draz, G.; VEca, N. R. (1977): Infeccion natural y experimental de roedores con virus Junin. Medicina, Buenos Aires, 37, 149-159. SABATTINI, M. N.; ConTIgIanı, M. S. (1982): Ecological and biological factors influencing the maintenance of arenavirus in nature with special reference to the agent of Argentinean hemorrhagic fever. Acad. Bras. C 251-262. SNEATH, P. H. A.; SokAL, R. R. (1973): Numerical taxonomy. San Francisco: Freeman. Tate, G. H. H. (1932): The taxonomic history of certain South and Central American cricetid rodents: Neotomys with remarks upon its relationships; the cotton rats (Sigmodon and Sygmomys) and the fish eating rats (Ichthyomys, Anetomys, Rheomys, Neusticomys and Daptomys). Am. Mus. Novitates. 583, 1-10. THomas, ©. (1916): On the grouping of South American Muridae that have bee referred to Phyllotis, Euneomys and Eligmodontia. Ann. Mag. Hist. ser 8, 17, 139-146. 242 M. Corti, Maria Susana Meriani and Gloria de Villafane THORBE, R. $. (1976): Biometric analysıs of geographic variation and racial affinities. Biol. Rev. 51, 407-452. THORPE, R. $.; CorTI, M.; CAPannA, E. (1982): Morphometric divergence of Robertsonian popula- tion-species of Mus: a multivariate analysis of size and shape. Experientia 38, 920-923. VIDELA, G.; Kayon, A. E.; CARBALLAL, G.; MERANI, M. S.; WEISSENBACHER, M. C. (1985): Sensibilidad de Calomys callidus lactante a la infeccion con la cepa XJ-Clon 3 de virus Junin. IX Reun. Cient. Soc. Arg. Virologia, Abs., 12. VILLAFANE, G. DE (1981): Reproducion y crecimiento de Calomys musculinus murillus (Thomas, 1916). Hist. Nat. 1, 237-256. Vıru11o, A. D.; Hovara, V. L.; MERANI, M. $. (1983): Comparacion de patrones de bandeo en especies de genero Calomys (Rodentia, Cricetidae). IX Con. Latinoamericano Zool., Arequipa, Perü, Abs., 13. VıruLLo, A. D.; Kayon, A. E.; PERCICH, R.; ZULETA, G.; MERANI, M. S. (1984): Caracterizacion citogenetica de tres especies de roedores (Rodentia, Cricetidae) de la Republica Argentina. Rev. Zool. Mus. Arg. Cs. Nat. 13, 491498. VıruLLo, A. D.; Hopvara, V. L.; MERANI, M. $.; WEISSENBACHER, M. C. (1985): Efecto de la infeccion persistente con virus Junin sobre la fertilidad de Calomys musculinus. IX Reun. Cient. Soc. Arg. Virologıa. WEISSENBACHER, M. C.; DAMONTE, E. B. (1983): Fiebre hemorrhagica argentina. Adel. Microbiol. Enf. Infecc. 2, 119-171. Authors’ addresses: Marco CoRrrı, Dipartimento di Biologia Anımale e dell’Uomo, Universitä di Roma “La Sapienza”, via A. Borelli 50, I-00161 Roma, Italy; MAarıa Susana MERANI and GLORIA DE VILLAFANE, Catedra de Microbiologia, Parasitologia, Virologia y Inmunologia, Pıso 11, Facultad de Medicina, Universidad de Buenos Aires. Paraguay 2155, Capital Federal, Argentina Zur innerartlichen Proteinvariation bei der Waldmaus (Apodemus sylvaticus) Von H. GEMMEkE, MARGOT RADTKE und J. NIETHAMMER! Zoologisches Institut der Universität Bonn Eingang des Ms. 26.11.1986 Abstract On the intraspecific variation of some proteins in the wood mouse (Apodemus sylvaticus) Compared the alleles of the three polymorphic protein locı for transferrine, postalbumine, and glucose-6-phosphate dehydrogenase ın more than 200 wood mice from different populations distri- buted from Sweden to Northwest Africa. Most of these alleles varied irregularly without any clear geographic pattern. An exception are the transferrines a and c. Allele c occurs in specimens of eastern Austria, northern Yougoslavia, Italy and Sardinia while allele a is found in individuals of Austria and the rest of the populations not yet mentioned. The allelic patterns add arguments for the possible source of some isolated populations. For example the wood mice from Tunisia probably derived from Southwest Europe, not from Italy or the Eastern Mediterranean. ! Mit Unterstützung durch die Deutsche Forschungsgemeinschaft. U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5204-0242 $ 02.50/0 Z. Säugetierkunde 52 (1987) 242-247 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Zur innerartlichen Proteinvariation bei Apodemus sylvaticus 243 Einleitung Elektrophoretisch nachweisbare Proteinunterschiede entsprechen in der Gattung Apode- mus recht gut den vermuteten phylogenetischen Abständen (GEMMERE 1980). Auch wo morphologische Unterschiede dazu nicht ausreichen, können ähnliche Arten wie Apode- mus sylvaticus und A. flavicollis aufgrund bestimmter, elektrophoretisch trennbarer Pro- teine sicher bestimmt werden (GEMMERE und NIETHAMMER 1981). Vor allem bei Apode- mus sylvaticus variieren einige der erfaßten Proteine auch innerartlich erheblich (GEMMERE 1980, 1981). Bisher wurden mit Methoden der numerischen Taxonomie Gesamtabstände zwischen den Populationen berechnet. Dabei bleibt jedoch ungeklärt, wie weit die Allele für einzelne Loci geographisch variieren. Auf diese Frage ist bei der Waldmaus nur NIETHAMMER (1982) kurz eingegangen. In der vorliegenden Arbeit soll sie ausführlicher behandelt werden. Dazu haben wir die drei innerhalb von Apodemus sylvaticus bei GEMMERE (1980) am stärksten varıiierenden Loci ausgewählt und behandeln die Verbrei- tung ihrer Allele. Von einer Betrachtung der Verbreitung einzelner Allele erhoffen wir unter anderem Hinweise auf die Herkunft isolierter Populationen und haben aus diesem Grunde Waldmäuse aus Schweden, von Mallorca und aus Nordafrika einbezogen. Material und Methode Über 220 Waldmäuse, deren Herkunft aus der Tabelle hervorgeht, wurden lebend nach Bonn gebracht und dort für gelelektrophoretische Vergleiche der Locı für Transferrin (Tf), Postalbumin (Pa) und Hexose-6-Phosphatdehydrogenase (Hexose-6-PD) verwendet. Einzelheiten finden sich bei GEMMERE (1980). Die Bezeichnungen der Allele sind beibehalten. Die gesperrt gesetzten Populatio- nen sind bisher bezüglich der varıablen Loci nicht publiziert, die übrigen teilweise oder ganz bei GEMMERE (1980, 1981) veröffentlicht. Ergebnisse Die Tabelle gibt die Häufigkeiten der Allele der drei Proteine bei 221-228 Waldmäusen an, die von 21 Orten oder Gebieten stammen. Manchmal wurden weiter entfernte Fundstellen zusammengefaßt, so mehrere Orte östlich der Rhone und aus den Cevennen unter Südfrankreich. Die meisten Allele streuen über das ganze Untersuchungsgebiet, ohne daß sich geogra- phische Beziehungen erkennen lassen. So fehlt die sonst meist vorherrschende Variante b von Postalbumin nur in Portugal und Tunesien, wo sie bei mehr Material aber vielleicht noch zu finden ist. Die seltenere Variante a wurde in einem geschlossenen Block von Wangerooge bis Südfrankreich gefunden, einmal aber auch außerdem auf Mallorca. Allel c fehlt nur im Norden und sonst dort, wo wenig Material vorliegt. Bei Hexose-6-PD herrscht Allel a ın Italien und am Neusiedlersee vor, seltener aber auch an einigen anderen, weit voneinander entfernten Orten (Tunesien bis Bad Driburg). Allel b kommt fast überall vor, c fehlt in Sardinien, auf Mallorca und in Süditalien. Eine gewisse Regelmäßigkeit lassen die Transferrine a und c erkennen (s. Abb.). Nebeneinander kommen sıe bisher nur am Neusiedlersee vor, sonst vikarlieren sie: c wurde von Rovinj, Krk, dem Monte Gargano und Sardinien nachgewiesen, a von sämtlichen übrigen Orten. Dagegen ist die Variante b mit meist geringerer Häufigkeit regellos verteilt. Die bisherige Verteilung läßt vermuten, daß das Allel a ın Nordafrika und im ganzen westlichen Europa von Portugal und Südfrankreich bis Schweden vorkommt, Allel ce hingegen im mittleren und östlichen Südeuropa von Sardinien bis mindestens Jugoslawien. Da bei dieser Vertei- lung klimatische Ursachen als Selektionsgründe unwahrscheinlich sind, bleibt nur die 244 H. Gemmeke, Margot Radtke und J. Niethammer Tabelle Variabilität von Transferrin (Tf), Postalbumin (Pa) und Hexose-6-Phosphat-Dehydrogenase (Hexose-6-PD) in verschiedenen Populationen von Apodemus sylvaticus Population Transferrin Postalbumin Hexose-6-PD b Schweden - 40 Mellum Wangerooge Bad Driburg Hennef Bonn (linksrheinisch) Chamonix/Schweiz Südfrankreich Spanıen Portugal Tunesien Mallorca Berchtesgaden ESSR Bayerischer Wald Melk an der Donau Neusiedlersee Rovinj/Jugoslawien Krk/Jugoslawien Monte Gargano/ltalien 2 Sardinien 10 20 Die Allele sind wie bei GEMMERE (1980, 1981) bezeichnet. Die Häufigkeiten sind in % angegeben, wobei die Summe wegen unterschiedlicher Abrundung nicht immer genau 100 ist. n gibt die Zahl der untersuchten Individuen an. Variable Anzahlen finden sich dann, wenn für die verschiedenen Loci nicht alle Tiere eindeutige Banden zeigten. (n total: 221-228). Annahme, daf die beiden Allele eine historisch bedingte Primärgliederung der Waldmaus widerspiegeln. Unter den aufgeführten Populationen finden sich einige von Inseln: Mellum, Wanger- ooge, Krk, Mallorca und Sardinien und zwei aus isolierten Randgebieten: Schweden und Tunesien. Diese Populationen eignen sich zur Prüfung der Frage, wie weit ihre Allele Hinweise auf die Herkunft der Population geben können und wie weit ihre genetische Variabilität eingeschränkt und ihre genetische Struktur verändert ist. Ein erster Vergleich zeigt: Alle Allele von Inselpopulationen kommen auch auf dem benachbarten Festland vor: die von Mellum und Wangerooge bei Bad Driburg, die von Krk bei Rovinj, die von Mallorca ın Südfrankreich und Spanien und die von Sardinien am Monte Gargano in Italien. Diese Übereinstimmungen stehen in Einklang mit der fast trivialen Hypothese, wonach die Inselpopulationen jeweils von denen des benachbarten Festlandes abstammen. Nicht ganz trivial ist die Übereinstimmung von Sardinien mit dem Monte Gargano, nicht aber mit Südfrankreich, Mallorca oder Tunesien. Auf den Inseln kommen entweder ebenso viele Allele vor wie auf dem benachbarten Festland oder weniger. Nur für Mellum ist eine Beschränkung der genetischen Variabilität im Vergleich zum Festland gesichert. Von Amrum und Krk liegen für diese Fragestellung zu wenige Tiere vor. Auf Sardinien (GEMMERE 1980) und Mallorca erscheint die genetische Variabilität nicht eingeschränkt. Von Sardinien sind zwar wenige Allele bei den drei betrachteten Proteinen gefunden worden, doch entspricht das den Verhältnissen am Monte Gargano. Die in Schweden gefundenen Allele stimmen nahezu perfekt mit denen von Mellum Zur innerartlichen Proteinvariation bei Apodemus sylvaticus 245 vereinfacht 600 km Verteilung der Transferrin-Allele bei Apodemus sylvaticus in Europa und Nordafrika. Kreise = Allel a, Quadrate = Allel b, Striche = Allel c. Allel d (nicht eingetragen) nur am Neusiedlersee. Die Grenze zwischen den vikariierenden Allelen a und b ist als durchgezogene, die nördliche Grenze des Artareals als gestrichelte Linie eingetragen überein und sınd auch sonst überall in Mittel- und Westeuropa verbreitet. Dies spricht eher für eine Herkunft der südskandinavischen Waldmäuse aus Norddeutschland (SIrvVonEN 1982). Solange die Proteinzusammensetzung von Waldmäusen aus Nordosteuropa nicht bekannt ist, kann allerdings der zweite mögliche Weg — der Landweg von Osten her um den Bottnischen Meerbusen in einer nacheiszeitlichen Wärmephase - nicht ausgeschlossen werden. Die Allele der Waldmäuse Tunesiens finden sich auch in Südwesteuropa, etwa in Spanien und Portugal oder Spanien und Südfrankreich. Gleichzeitig unterscheiden sie sich von denen sardischer und süditalienischer Waldmäuse. Zwar sind ım östlichen Mittelmeer- gebiet, das auch als Ursprungsland der ın den Atlasländern isolierten Populationen in einer feuchteren Phase in der Vergangenheit in Frage kommt, bisher keine Populationen untersucht. Doch läßt die bisher bekannte Transterrinverteilung in Europa dort Abwei- chungen vermuten wie auch die Tatsache, daß bei Waldmäusen in Nepal ein eigenes, in Europa nicht festgestelltes Transterrinallel gefunden wurde. 246 H. Gemmeke, Margot Radtke und J. Niethammer Diskussion Die meisten der hier untersuchten Allele scheinen über das ganze Gebiet hinweg aufzutre- ten. Häufigere Allele wurden an mehr Orten gefunden als seltenere. Dies spricht dafür, daß die selteneren Varianten bei mehr Material wahrscheinlich auch an den meisten Orten entdeckt würden und daß die beobachteten Frequenzunterschiede eher zufällig sind. Der Zufall kann zweı Quellen haben: einmal den Zufall bei der Auswahl der Stichprobe, der mit Signifikanzberechnungen abgeschätzt werden kann. Dann der Zufall, der tatsächliche Unterschiede zwischen Kleinpopulationen bewirkt. Zwar ist Apodemus sylvaticus eine häufige, verbreitete und euryöke Art mit ziemlich großem Aktionsradius der Individuen. Trotzdem wechseln auch bei ihr regelmäßig bewohnte Gebiete mit spärlich oder gar nicht besiedelten kleinen Räumen, in denen die Populationen zwischen verschiedenen Jahren erheblich schwanken können. Da bei Signifikanzberechnungen der Unterschiede zwischen Populationen wegen des noch geringen Materialumfangs meist höchstens schwache Siche- rung zu erreichen ıst, kann der Einfluß beider Quellen auf Unterschiede schlecht gegenein- ander abgeschätzt werden. Da geographische Beziehungen deutlich nur beim Transferrin und klimaparallele Beziehungen gar nicht erkennbar sind, spricht die geographische Verteilung der Allele eher für selektive Neutralität. In der Tabelle fehlen die Allele von drei Waldmäusen aus Nepal, weil bei ihnen die Zugehörigkeit zu A. sylvaticus zweifelhaft war (GEMMEKE und NIETHAMMER 1982). Sie besitzen alle einheitlich ein eigenes, in Europa nicht nachgewiesenes Transferrin, außerdem Postalbumin b und Hexose-6-PD c, in Europa häufige Allele. Sollten sie zu A. sylvaticns gehören, würde das ebenfalls einen geographischen Bezug bei den Transferrinen, nicht aber bei den anderen beiden Allelen bedeuten. Die Waldmaus ist im letzten Glazial aus Mittel- und Nordeuropa verschwunden, dürfte aber in Südeuropa überdauert haben. Von hier aus hat sie in der Nacheiszeit die nördlicheren Gebiete neu besiedelt. Die Verteilung der vikariierenden Transferrine a und c spricht dafür, daß Mitteleuropa und Schweden von Südwesteuropa aus, nicht aber von Italien oder Osteuropa, neu besetzt wurden. Am Neusiedlersee lebt wahrscheinlich eine Mischpopulation beider Herkünfte. Zu erwarten wäre, daß mit der Wiederbesiedlung die genetische Variabilität nach Norden abgenommen hat. Dem könnte die geringe Allelenzahl in Schweden entsprechen. Aber bereits in Deutschland ist die Allelenzahl mindestens so groß wie ın den mutmaßli- chen glazialen Refugien (Iberische Halbinsel, Süditalıien). Die gute Übereinstimmung zwischen tunesischen und südwesteuropäischen Waldmäu- sen bildet ein weiteres Beispiel für einen Austausch terrestrischer und nicht flugfähiger Säugetiere im westlichen Mittelmeergebiet trotz der Tatsache, daß hier ım Pleistozän wahrscheinlich keine Festlandverbindung bestanden hat. Weitere Beispiele wären die westmediterrane Heckenmaus (Mus spretus) und die auf Westeuropa und die Atlasländer beschränkte Hausspitzmaus (Crocidura russula). Zusammenfassung Die polymorphen Loci für die Proteine Transferrin, Postalbumin und Glukose-6-Phosphatdehydro- genase wurden bei Apodemus sylvaticus von Schweden bis Nordafrika untersucht. Die Mehrzahl der Allele variiert geographisch regellos. Eine Ausnahme bilden die Transferrine a und c, die sich weitgehend ausschließen (Abb.). Die wahrscheinliche Herkunft von Insel- und Randpopulationen kann vielfach eingeengt werden. So dürften die Waldmäuse aus Nordwestafrika aus Südwesteuropa stammen, nicht aber aus Italien oder Osteuropa. Zur innerartlichen Proteinvariation bei Apodemus sylvaticus 247 Danksagung Für die Beschaffung von Waldmäusen danken wir den Herren Dr. M. KAsır NAUROZ (Tiere von Mellum), Prof. Dr. R. ScHRÖPFER, Osnabrück (Tiere aus Tunesien) und Dr. J. A. ALcovERr, Barcelona (Tiere von Mallorca). Literatur GEMMEKE, H. (1980): Proteinvarıation und Taxonomie in der Gattung Apodemus (Mammalıa, Rodentia). Z. Säugetierkunde 45, 348-365. GEMMEKE, H. (1981): Genetische Unterschiede zwischen rechts- und linksrheinischen Waldmäusen (Apodemus sylvaticus). Bonn. zool. Beitr. 32, 265-269. GEMMEKE, H.; NIETHAMMER, J. (1981): Die Waldmäuse Apodemus sylvaticus und A. flavicollis vom Monte Gargano (Süditalien). Z. Säugetierkunde 46, 162-168. GEMMEKE, H.; NIETHAMMER, J. (1982): Zur Characterisierung der Waldmäuse (Apodemus) Nepals. Z. Säugetierkunde 47, 33-38. NIETHAMMER, J. (1982): Zur Arealgeschichte europäischer Nagetiere. Verh. Dtsch. Zool. Ges. 1982, 145-157. SIIVONEN, L. (1982): The history of the Fennoscandian mammal fauna. Acta Zool. Fennica 169, 7-10. Anschriften der Verfasser: Dr. HUBERT GEMMEKE, Institut für Nematologie und Wirbeltierkunde, Toppheideweg 88, D-4400 Münster; Dipl.-Biol. MARGOT RADTKE und Prof. Dr. JocHEN NIETHAMMER, Zoologisches Institut der Universität, Poppelsdorfer Schloß, D-5300 Bonn Biochemical comparisons in Yugoslavian rodents of the families Arvicolidae and Muridae By AyzsHaA GiLı, B. PETROv, S. Zıvkovı6 and DEsankA RımsA Department of Biology, University of Nevada, Reno, Institute for Biological Research “Sinisa Stankovic”, Beograd Receipt of Ms. 27.7.1986 Abstract Twenty eight proteins from six species of Arvicolidae and three species of Muridae were examined electrophoretically. Anımals of the species Microtus arvalıs, M. epiroticus, Clethrionomys glareolns, Pıitymys subterraneus, P. felteni, Dinaromys bogdanovi, Apodemus flavicollis, A. sylvaticus and A. agrarıns were collected ın various regions of Yugoslavia. The NEı genetic distances approximately correspond to previous conceptions of the relationship of the species. The distance between Microtus and Pıtymys species is negligible, Microtus to Clethrionomys ıs 0.60, and the average distance from Dinaromys to the remaining arvicolids is 1.21. Thereby the special position of Dinaromys within the voles ıs again accentuated. Apodemus sylvaticus and A. flavicollis are not distinguishable ın Yugoslavia with the locı examined. Their distance to Apodemus agrarıns, on the other hand, is great. Introduction General agreement has not been reached on the classıfication of some of the species dealt with in this study, and differing classifications have been proposed (Hınton 1926; GRoMov and PoLyakov 1977; CORBET 1978; CHALINE and MEın 1979; HonAckt et al. 1982). CoRBET (1978: 94) notes that: “Only a very comprehensive revision using all the U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5204-0247 $ 02.50/0 Z. Säugetierkunde 52 (1987) 247-256 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 248 Ayesha Gill, B. Petrov, S. Zivkovic and Desanka Rimsa available characters is likely to result in a stable arrangement.” The use of biochemical characters may aid substantially. The purpose of this paper is to apply the results of electrophoretic analysis to the classification of certain species of the Arvicolidae and Muridae of the Yugoslavian fauna. The systematic value of electrophoretic data and the advantages of using it as a systematic tool have been stressed by Avıse (1974). Using electrophoretic analysis in the study of the Arvicolidae in parts of western Europe and North America, GRAF and coworkers (GRAF and ScHoLı 1975; GRAF and MEYLAN 1980; GRAF 1982) have found a general correspond- ence between classıfications based on classical systematics and on genetic distances derived from electrophoretic data, wıth some differences. This paper presents a preliminary survey of anumber of Yugoslavian microtine species and species of the genus Apodemus. Joint analysis of samples collected at the same sites of the two families allows comparison of patterns of intra-famıly varıation. The microtines, including voles and lemmings, are a distinet group assigned to the family Arvicolidae by Honackt et al. (1982). About half the microtine species found in Yugoslavia are rep- resented in this study, including Microtus arvalıs, the common vole; M. epiroticus (M. subarvalıs), which is distinguished from M. arvalıs on the basıs of ıts karypotype (MEYER et al. 1972) and of morphological and other characteristics (RuZıc et al. 1975; PETROV and Ruzıc 1982); Clethrionomys glareolus, the bank vole; two species of the genus Pitymys, P. subterraneus and P. felteni; and Dinaromys bogdanovi, Martino’s vole endemic to the mountains of southwestern Yugoslavia. Apodemus ıs a fairly distinct genus within the Muridae. It consists of about 13 species (Honackt et al. 1982) and is the dominant group of murid rodents in the Palaearctic (CorgEr 1978). Three of the five Apodemus species found ın Yugoslavia are represented in this study, A. flavicollis, the yellow-necked mouse; A. sylvaticus, the wood mouse; and A. agrarius, the striped field mouse. Materials and methods Individuals of the families Arvicolidae (n=55) and Muridae (n=21, all in the genus Apodemus) were live-trapped at 11 localities in Yugoslavia. The number of specimens trapped at each locality is given ın parentheses. Microtus arvalis (M. a.): Debrc (4); Brezovica, Sara Mts. (6); Knjazevac (2); Kanjıza (13). Microtus epiroticus (M. e.): Skopje (2). Clethrionomys glareolus (C. g.): Brezovica, Sara Mts. (1); Zljeb Mts. (3); Travna Gora, Ribnica (5); Ig, Ljubljana (1); Djerdap (2). Pitymys subterraneus (P. s.): Ig, Ljubljana (2); Debrc (2); Crni Vrh, Zagubica (1); Kanjıza (1). Pitymys felteni (P. f.): Brezovica, Sara Mts. (1). Dinaromys bogdanovi grebenscikovi (D. b. g.): Zljeb Mts. (4= ? + litter of 3). Dinaromys bogdanovi bogdanovi (D. b. b.): Brezovica, Sara Mts. (5= + litter of 4). Apodemus flavicollis (A. f.): Beska, Srem (7); Ig, Ljubljana (3); Djerdap (1). Apodemus sylvaticus (A. s.): BeSka, Srem (1); Crni Vrh, Zagubica (2); Djerdap (1). Apodemus agrarius (A. a.): Beska, Srem (5); Debre (1). Methods of tissue sample preparation and horizontal starch gel electrophoresis were essentially as described in SELANDER et al. (1971) with slight modifications. For Lithium hydroxide gels, 11.0 % Electrostarch Lot 307 (Electrostarch Co., Madison, WI, USA) was used and 12.4% starch was used for all other gels. The buffer and stain systems for proteins screened in this study (Table 1) are described in SELANDER et al. (1971) except for stain a. 50ml 0.2M Tris HCl, pH 8.0, 250mg D- sorbitol, 1Omg NAD, 15mg NBT, 2 mg PMS; stain b. 37.5 ml H;O, 12.5ml 0.2M Trıs HCpE 8.0, 5ml 0.1M MgCh, 75mg glucose, Iml G-6-PDH, 50mg ATP, 10mg NAD, 10mg NADP, 1Omg MTT, 5mg PMS; and stain c. 50ml 0.2M Tris HCl, pH 8.0, 2ml 0.25M MnCLh, 40 mg leucylalanine, 10mg snake venom, 20mg peroxidase, 1Omg O-dianisidine-di-HCl. Alcohol dehy- drogenase (ADH) does not have to be stained specifically and is seen on many dehydrogenase gels. It was read on gels stained for glycerol-3-phosphate dehydrogenase (GPD). Biochemical comparisons in Yugoslavian Arvicolidae and Muridae 249 Table 1 Enzymes and nonenzymatic proteins screened Enzyme or protein name Abbreviation Tissue Buffer system! Stain system! Alcohol dehydrogenase ADH Liver NE, DEIES:O 6 Glycerol-3-phosphate GPD Liver 5, IC, plEl 8.0 6 dehydrogenase Sorbitol dehydrogenase SORDH Liver SZENE pEIRS:O a Lactate dehydrogenase LDH Kidneys 55. KO, DSL) 10 Malate dehydrogenase MDH Liver ARE, pEII63 11 Malic enzyme ME Liver A ICPEILES 11 Isocitrate dehydrogenase ICD Kidneys SELE,pEIS:0 8 Glucose-6-phosphate G-6PD Kidneys LTE IPEIES:O 4 dehydrogenase Superoxide dismutase SOD Kidneys 9. TME 7 or 2 1@IpEIT8:O 7% Glutamate-oxaloacetate GOT Liver EOS 5 transaminase or 5. TC, pH 8.0 5 Hexokinase HK Kidneys SENEIPEIES:O b Peptidase PEP Liver 4. TC, pH 6.3 c Glucose phosphate ısomersse GPI Liver SIE PEINS:O 14 Hemoglobin HB Hemolysate 1. Tris HC1 3 Albumin ALB Plasma 2: OE 3 Transferrin ARSE: Plasma 2 ONE! 3 General proteins GP Hemolysate 1. Tris HC1 9) ! Given in SELANDER et al. (1971) except for stains a, b, and c, described in Materials and methods. Mathematical methods: Gene frequencies, measures of genetic varıation, NEr’s (1978) unbiased genetic distance and a dendrogram based on the latter were derived from input on single individual genotypes (electromorphs), using the computer program BIOSYS-1 (SwoFFORD and SELANDER 1981). This program can handle missing data, basing pairwise genetic distances only on locı present in both species. All the allozymes were not detected over the range of species studied, and there were differences in enzyme activity among species. Allozymes not present or which could not be reliably discerned in particular species are left blank (Table 2). Loci not included are GPI-1 for Microtus arvalis, SOD-3 for M. epiroticus, PEP-1 for Clethrionomys glareolus and HK-1 for the two Pitymys species, leaving 27 locı analysed for each of these five species. Dinaromys bogdanovi bogdanovi was also analysed with 27 locı, missing GPI-2, and D. b. grebenscikovi with 24, missing GPI-2, ME, HK-1 and HK-4. G-6-PD, PEP and GP-2 are blank for all the Apodemus species. A. flavicollis also lacks ICD-2, ME, HK-4 and ADH, leaving 21 locı. There are 22 locı analysed for A. sylvaticus, which also lacks ME, HK-1 and HK-4, and 24 locı for A. agrarius, which lacks ICD-2. The unweighted pair group method with recomputation of coefficients by arıthmetic averaging (UPGMA) was used in the BIOSYS-1 cluster analysis (SNEATH and SoKAL 1973). A second measure, of genetic sımilarıty rather than genetic distance, was also calculated. Jaccard’s association coefficient, Sp=a/(a+ u), where a = the number of matched electromorphs (1:1) and u =the number mismatched (0:1 and 1:0) (SNEATH and SorkAL 1973) was calculated for species pairs, based on loci for which both species were scored. S; depends only upon the presence (1) or absence (0) of alleles as indicated by bands on the starch gels (electromorphs), not on allele frequencies as the genetic distance does. Negative matches are excluded. Jaccard’s coefficient was found to be less sensitive to sample sıze differences than other proposed measures of similarıty (NADLER et al. 1982). Both Ner’s unbiased genetic distance and Jaccard’s coefficient are appropriate for small sample sizes. Evidence justifying the use of one or a few samples for electrophoretic protein comparisons of different evolutionary units has been presented by Avıse (1974), SarıcH (1977), Neı (1978), and GoORMan and Renzı (1979). 250 Ayesha Gill, B. Petrov, S. Zivkovic and Desanka Rimsa Results and discussion Arvicolidae The microtines have been recognized as a sharply defined group among the rodents, whether they have been given family status (CHALInE and MEIn 1979; Honackt et al. 1982) or subfamily status (Simpson 1945; CoRBET 1978). One of the most unstable areas of classification at the generic level ıs the Microtus/Pitymys/Arvicola group (CoRBET 1978), and opinion differs as to whether the latter two should be retained as distinct genera or considered subgenera of Microtus (GROMoVv and PoLyakov 1977; Honackt et al. 1982). Biochemical similarıties of the species in this study are based on electrophoretic data (Table 2) representing a broad range of glucose-metabolizing enzymes, other enzymes, and nonenzymatic proteins. Twenty eight presumptive locı were scored for the arvicolids and 25 for species of Apodemus, but all the allozymes were not detected over the range of species studied (see Materials and methods). Allozymes not present ın a particular species are left blank ın Table 2. The results given here are an initial survey of some of the protein varıants present in these species and do not represent estimates of population frequencies. The interrelationship among the species based on their biochemical sımilarities (Table 2) is depicted in a dendrogram (Fig. 1). The Microtus and Pitymys samples form one major cluster at an average genetic distance D=0.58 from the Clethrionomys glareolus cluster. The Dinaromys bogdanovi samples are distinct from the other microtines with D=1.21. The distance of Apodemus agrarius from A. sylvaticus and A. flavicollis ıs almost 1.00, and D=1.30 between the families of arvicolids and murids. The goodness of fit of the dendrogram to the cophenetic matrix derived from the Neı (1978) unbiased genetic distances is given by a cophenetic correlation of 0.913. Neı showed that the number of individuals used for estimating average heterozygosity and genetic distance can be very 1407 1.26°%711512770:9877°0:8417207.057.0:5672. 0122702822 031222.9:00 Clethrionomys glareolus Microtus epiroticus | Microtus arvalis | Pitymys subterraneus Pitymys felteni Pitymys subterraneus Dinaromys bogdanovi Apodemus flavicollis Apodemus sylvaticus mm O— Em m Apodemus agrarius 7402 1126 115122220:98. 70,84 .07.02 7.05608.0422 07230 80817:0:00 Genetic Distance Dendrogram of unbiased genetic distances (NEı 1978) derived from data of Table 2 Biochemical comparisons in Yugoslavian Arvicolidae and Muridae 2a Table 2 Allele frequencies of species of Microtus, Clethrionomys, Pitymys, Dinaromys, and Apodemus in Yugoslavia Species” No of Individuals Loci and alleles** I. Glucose-metabolizing enzymes 10022090222,1.005217.005221200 0.04 252 Ayesha Gill, B. Petrov, S. Zivkovic and Desanka Rimsa Table 2 (continued) Species” No of Individuals Loci and alleles”“* HK-1 1.00 Il. Other enzymes 1.007 1.007 31200 Biochemical comparisons in Yugoslavian Arvicolidae and Muridae 255 Table 2 (continued) Species” No of Individuals Loci and alleles** III. Nonenzymatic proteins 020387 2053 1.007.092 29:67 0.08 0.33 0.227506 KLO0ZEROI 1.00 GP-1 IR OO FEO0 FEB OOFEEEO IE 1:09 GP-2 a 029602. 0.502 =21:.00,7205327°.0:502 2 1:0077 1:00 b 0.04 0.50 0.67. 0.50 * Abbreviations for locı are given in Table 1, for species in Materials and methods. - ** Alleles are listed in order of increasing mobility; a is the slowest. small if the genetic distance ıs large and the average heterozygosity low. For average heterozygosities of the order of 0.06 and genetic distances greater than 0.15, the expected magnitude of the bias is not important even if a single individual ıs sampled from each of the species compared (NEı 1978). Several populations of ©. glareolus, M. arvalıs, and P. subterraneus were sampled, and the varıability within these species is evident (Fig. 1). Due to the sample sizes, however, the estimates of intraspecific varıability are tentative. There was no evidence of clınes or other patterns in the varıability within any of the species. Three major clusters within the Arvicolidae mirror the classical systematic classification: 1) Clethrionomys, 2) Microtus/ Pıtymys group, and 3) Dinaromys (Fig. 1). Within the second group there are further clear subdivisions. The four M. arvalıs samples cluster tightly (D between 0.01 and 0.07). GRAF (1982) found very sımilar genetic distances between eight populations of western European M. arvalıs (D between 0.00 and 0.08). The M. epiroticus sample ıs distinct from the M. arvalıs cluster (D=0.24). The karyotype of M. epiroticus ıs characterızed by 2n=54, while for M. arvalis 2n=46 (MEyer et al. 1972; Zıvxovic et al. 1976); hybrids between them are sterile (PETROV and Ruzıc 1982). The single individual of P. felteni ıs clustered among those of P. subterraneus but there ıs considerable biochemical varıation within the Pıtymys samples, D ranging from 0.19 to 0.37 (Fig. 1). The close biochemical affınıty of P. subterraneus and P. felteni ıs also apparent in their association coefficient, S7=0.68, which is almost as high as that between the subspecies of D. bogdanovi, S7=0.69 (Table 3). The Pıtymys samples form a subcluster separate from the Microtus samples (Fig. 1), with an average distance of D=0.42 between the subclusters. While the genetic distances, based on allelic frequencies, indicate a difference between Microtus and Pitymys, this is not the case when one looks only at presence or absence ot alleles as represented in Jaccard’s coeffiecient of association (Table 3). M. epiroticus ıs more closely related to the Pıtymys species than M. arvalıs ıs; M. arvalıs is just as similar to P. felteni as it is to M. epiroticus. The relationships among M. arvalıs, P. subterraneus, and C. glareolus tound in this study 254 Ayesha Gill, B. Petrov, S. Zivkovic and Desanka Rimsa Table 3 Jaccard’s coefficient of association, showing the biochemical similarity between pairs of certain rodent species of Yugoslavia M.e. Ins: Dip C.g. 0.558 0.578 0.548 0.649 0.618 0.684 Abbreviations for species are given in Materials and methods. are quite similar to those found by Grar (1982), with C. glareolus being a little more closely associated with P. subterraneus than with M. arvalıs (Table 3). The monospecific genus Dinaromys, Martino’s vole, was earlier referred to the genus Dolomys, but evidence indicates it is a separate, recent genus (CoRBET 1978). Animals of this genus live in small isolated populations at altitudes from sea level to 2200 m (PETRoOVv and ToporovIC 1982). KOENIGSWALD (1980, cited in Honackt et al. 1982) believes, on the basıs of the internal structure of the molar enamel, that Dinaromys probably has no close relationship with any other living arvicolid. The distinctiveness of the endemic relict species D. bogdanovi ıs strongly supported by the biochemical data (Fig. 1, Table 3). It ıs almost as distinct from members of its own family (D=1.21) as it is from the Apodemus species in the family Muridae (D=1.30). The species has been divided on morphological grounds into two subgroups, the bogdanovi group in the northern part of its distribution and the grebenscikovi group in the southern part (Toporovıc 1956; PETROV and Toporovı6 1982). While the genetic variability of D. bogdanovı ıs low (Table 2), as is typical of isolates, most of the varıation which does occur is between the groups, not within them, supporting their classification into separate subgroups (Fig. 1). Muridae Three species of the genus Apodemus are included in this study. There are taxonomic problens concerning the genus (CoRBET 1966; Williams et al. 1980). A. sylvaticus and A. flavicollis are difficult to distinguish morphologically and often occupy the same deciduous woodland habitats (Durıc and Tvrrkovı6 1972). CoRBET (1978) suggested that the two species may hybridize, and some investigators believe that introgression has occurred between the species in southern Europe, whereas others think they occur together without interbreeding (GEMMERE 1980). A. agrarius ıs easily distinguished morphologically. The genetic distance found between A. sylvaticus and A. flavicolls was small (D between 0.28 and 0.46) compared to their great distance from A. agrarius (D=0.99), in agreement with morphological (DuLı6 and Tvrrkovi6 1972; NIETHAMMER and KrAPP 1978), chromosomal (VUJoSEVIE ET ar. 1984), and earlier electrophoretic data (GEMMEKE 1980). A. sylvaticus and A. flavicollis showed the strongest association of any of the taxa, with an association coefficient of 0.85 (Table 3). EnGeL et al. (1973) and GEMMERE (1980) found greater biochemical differences between A. sylvaticus and A. flavicollis populations in western Europe. The dissimilarıty of A. agrarins from the other two Apodemus species is much greater than that between other rodent congeners reported in the literature (AvısE 1974). Biochemical comparisons in Yugoslavian Arvicolidae and Muridae 2.55 There was no heterozygosity found in the four A. sylvaticus individuals sampled. They were all homozygous for the same allele at 15 locı, but one anomalous individual from Beska was homozygous for different alleles than the other three at seven other locı. Because of the high degree of homozygosity the effect of this one unusual individual is magnified in the dendrogram and it is distinct from the others (D = 0.46). Its electrophore- tic pattern for these variant locı differed from that of A. flavicollis except for the hemoglobins. Because of the unusual electrophoretic results for this individual, its species identity was double checked and verified. Moreover, the specimens of A. sylvaticus and A. flavicollis are from regions of Yugoslavia in which there is no problem in distinguishing them morphologically. GEMMERE (1980) found geographical varıation in western Euro- pean A. sylvaticus, and both the present study and that of WıLLıams et al. (1980) suggest geographical varıation in Yugoslavian A. sylvaticus. Electrophoretic analysis is not useful for investigating differences between families because the number of shared alleles become minimal. It is indicative of the distinctiveness of D. bogdanovi within the microtines and of A. agrarins within the genus Apodemus that the genetic distance between the two families is not much greater than the distances of these two species from other members of their respective groups (Fig. 1). Acknowledgements This work was conducted when the senior author (AEG) was a National Academy of Sciences Exchange Scholar with the Yugoslav Council of Academies. She wishes to thank the National Academy of Sciences, USA, and expecially KAARA ETTEsvoLD and the staff of the Section on USSR and Eastern Europe for arranging this research visit and for their generous support and help. Her great appreciation goes to her hosts in Yugoslavia, the Serbian Academy of Sciences and Arts and the Institute for Biological Research “Sini$a Stankovic” in Beograd for their support and warm hospital- ity. It was a privilege and a true pleasure to be able to live and do research ın Yugoslavia. We thank J--D. Grar for his helpful comments. Zusammenfassung Biochemische Vergleiche von jugoslawischen Rodentia der Familien Arvicolidae und Muridae 28 Proteine wurden bei 6 Arten der Arvicolidae und 3 Arten der Muridae elektrophoretisch untersucht. Es handelt sich um die Arten Microtus arvalıs, M. epiroticus, Clethrionomys glareolus, Pitymys subterraneus, P. felteni, Dinaromys bogdanovi, Apodemus flavicollis, A. sylvaticus und A. agrarıns von denen Tiere ın verschiedenen Gebieten Jugoslawiens gesammelt wurden. Die nach Neı berechneten genetischen Abstände entsprechen annähernd den bisherigen Vorstellungen über die abgestufte Verwandtschaft der Arten. Der Abstand zwischen den beiden Microtus- und Pıitymys- Arten ist gering, der zu Olethrionomys beträgt 0,60, der mittlere Abstand von Dinaromys zu den übrigen Arvicoliden 1,21. Damit wird die Sonderstellung von Dinaromys innerhalb der Wühlmäuse erneut betont. Apodemus sylvaticus und A. flavicollis sınd in Jugoslawien mit den untersuchten Loci nicht unterscheidbar. Ihr Abstand zu Apodemus agrarins ist hingegen groß. References Avıs£, J. C. (1974): Systematic value of electrophoretic data. Syst. Zool. 23, 465481. CHALINE, J.; MEın, P. (1979): Les Rongeurs et l’evolution. Parıs: Doin. CoRBET, G. B. (1966): The terrestrial mammals of western Europe. London: Foulis Co. 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. Duuisc, B.; Tvrrkovi6, N. (1972): Sur la possibilite de la differenciation des populations de Mulot (Apodemus sylvaticus Linne, 1758) et de Mulot forive (Apodemus flavicollis Melchior, 1834) dans les iles de l’Adriatique (1). Rapp. Comm. int. Mer. Medıt. 21, 101-104. ENGEL, W.; VoGEL, W.; VoıcuLsscu, I.; Ropers, H.-H.; ZEnZEs, M. T; BENDER, K. (1973): Cytogenetic and biochemical differences between Apodemus sylvaticus and Apodemus flavicolls, possibly responsible for the failure to interbreed. Comp. Biochem. Physiol. 44B, 1165-1173. GEMMEKE, H. (1980): Proteinvariation und Taxonomie in der Gattung Apodemus (Mammalıa, Rodentia). Z. Säugetierkunde 45, 348-365. GORMANn, G. C.; Renzı, Jr., J. (1979): Genetic distance and heterozygosity estimates in electrophore- tic studies: Effects of sample size. Copeia 1979, 242-249. 256 Ayesha Gill, B. Petrov, S. Zivkovic and Desanka Rimsa GRAr, ].-D. (1982): Genetique biochimique zoogeographie et taxonomie des Arvicolidae (Mammalıa, Rodentia). Revue suisse Zool. 89, 749-787. GRAF, J.-D.; MEyLAn, A. (1980): Polymorphisme chromosomique et biochimique chez Pitymys multiplex (Mammalia, Rodentia). Z. Säugetierkunde 45, 133-148. GRAF, J.-D.; ScHoLL, A. (1975): Variations enzymatiques et relations phyletiques entre neuf especes de Mierotmae (Mammalıa, Rodentia). Revue suisse Zool. 82, 681-687. GRoMov, I. M.; Poryakov, I. Ya. (1977): Fauna SSSR. Mlekopitayushchie, Vol. 3, 1-504. Polevki (Microtinae). Moscow and Leningrad: Nauka. Hiınron, M. A. C. (1926): Monograph of the voles and lemmings (Microtinae) living and extinct. Trust. Brit. Mus. London, Vol. 1. Honackı, J. H.; Kınman, K. E.; KoeppL, J. W. (1982): Mammal species of the world. Lawrence, Kansas: Assoc. Syst. Coll. MEYER, M. N.; OrLov, V. N.; SKHOLL, E. D. (1972): O nomenklature 46- ı 54-chromosomnykh polevok tipa Microtus arı valis Pall. (Rodentia, Cricetidae). Zool. Zhur. 51, 157-161. NADLER, C. F.; HOFFMANN, R. S.; VORONTSov, N. N.; KoeppL, J. W.; DEUTSCH, L.; SUKERNIK, R. I. (1982): Evolution in ground squirrels. II. Biochemical comparisons ın Holaretic populations of Spermophilus. Z. Säugetierkunde 47, 198-215. Neı, M. (1978): Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89, 583-590. NIETHAMMER, J.; Krapp, F. (1978): Handbuch der Säugetiere Europas. Wiesbaden: Akademische Verlagsges. DB Ruzıc, A. (1982): Microtus epiroticus Ondrias, 1966 — Südfeldmaus. In: Handbuch der Säugetiere Europas. Wiesbaden: Akademische Verlagsges. Band 2/1, 319-330. PETRov, B.; Toporovic, M. (1982): Dinaromys bogdanovi (V. et E. Martino, 1922) — Bergmaus. In: Handbuch der Säugetiere Europas. Wiesbaden: Akademische Verlagsges. Band 2/1, 193-208. Ruäic, A.; Perrov, B.; Zıvkovi, S.; Rımsa, D. (1975): On the species independence of the 54- chromosome vole Microtus epiroticus Ondria 1966 (Mammalia, Rodentia), its distribution, ecology and importance as a pest in the west part of Balkan Peninsula. J. for Scient. Agric. Res., Beograd, 28, 153-160. SARICH, V. M. (1977): Rates, sample sizes, and the neutrality hypothesis for electrophoresis in evolutionary studies. Nature 265, 24-28. SELANDER, R. K.; SMITH, M. H.; Yang, $. Y.; JoHNnson, W. E.; GENTRY, J. B. (1971): Biochemical polymorphism and systematics in the genus Peromyscus. I. Variation in the Old-field Mouse (Peromyscus polionotus). Stud. Genetics 6, 49-90. Sımpson, G. G. (1945): The principles of classification and a classification of the mammals. Bull. Amer. Mus. Nat. Hist. 85, 1-350. SNEATH, P. H. A.; SokaL, R. R. (1973): Numerical taxonomy. San Francisco: W. H. Freeman. SWOFFORD, D. L.; SELANDER, R. B. (1981): BIOSYS-1: aFORTRAN program for the comprehensive analysis of electrophoretic data in population genetics and systematics. J. Heredity 72, 281-283. Toporovi6, M. (1956): Varıiability of the endemic genus Dolomys Nehring, (Microtinae, Rodentia). Arch. biol. scı., Beograd, 8, 93-109. Vujosevic, M.; Rımsa, D.; Zıvkovi6,S. (1984): Patterns of G- und C-bands distribution on chromosomes of three Apodemus species. Z. Säugetierkunde 49, 234-238. WırLıams, $. L.; HAFNER, J. C.; Doran, P. G. (1980): Glans penes and bacula of five species of Apodemus (Rodentia: Muridae) from Croatia, Yugoslavia. Mammalıa 44, 245-258. Zivkovic, $.; Rımsa, D.; Ruzıd, A.; PErrov, B. (1976): Cytogenetical characteristies, taxonomic status and distribution of the voles with 46 and 54 chromosomes of the Microtus arvalis group in Yugoslavia (Rodentia, Mammalıa). Arch. Biol. Scı., Beograd, 26, 123-134. Authors’ addresses: AyesHa E. GırL, New York State Dept. Health, Wadsworth Laboratories, Albany, New York 12201, USA; Boris PETROV, SLOBODAN ZIVKOVIC, DESANKA Rımsa, Institute for Biological Research “S. Stankovic”’,Y-11000 Beograd, Yugoslavia Low proportion of shrews in the diet of small mustelids in western Finland By E. KorrımÄkI and K. NORRDAHL Receipt of Ms. 15. 8. 1986 Abstract Scats and guts of stoat and least weasel collected after a decline of microtines in western Finland showed that these mustelids did not use shrews as their alternative prey, probably because of their distaste for shrews. Therefore it seems that small mustelids did not cause the synchronous decline phases in microtine and shrew populations observed in the transition zone of Fennoscandian microtine cycles. In this zone the most likely reason, especially for spring and summer declines, may be the predation by avian predators which frequently feed on shrews in addition to microtines. Introduction Based on the density varıations of small rodents, Fennoscandia can be divided into three regions: 1. a southern zone (from 55> to about 59 >N) with quite stable vole populations, 2. a transıtion zone (about 59>-61 >N) and 3. a zone north of 61 >N characterızed by pronounced fluctuations in microtine populations (Hansson and HENTTONEN 1985). According to KorPImÄkI (1986a, b) the transıtion zone extends from the southern coast of Finland to 63 >N in the western part of the country. In the northern zone, shrews have a low phase synchronously with cyclic microtines (Hansson 1984; HENTTONEN 1985), but both small mammal groups are non-cyclic in the southern zone (Hansson 1984). In the transition zone, microtines and shrews have synchronous decline phases, although the other phases are not regularly synchronous (Korpımäkı 1986a). The above authors suggested that the synchrony could be best explained by predation. According to HENTTONEN (1985), the most probable group of predators contributing the synchronous lows in the northern zone is small mustelids, especially the least weasel Mustela nivalıs. In the present paper, we will study, whether thıs suggestion is true in the transition zone. If small mustelids cause synchronous lows in microtine and shrew populations, they should use shrews as their alternative prey when microtines decrease. Therefore we collected data on the food of the stoat Mustela erminea and least weasel in southern Ostrobothnia, western Finland, after a severe decrease of vole populations. Material and methods In the large field plain of Alajoki (63>05’N, 22>55’E), 2 guts and 172 scats of the stoat (a total of 201 prey items) and 3 guts and 30 scats of the least weasel (33 prey items) were sampled between autumn 1983 and spring 1985. A majority of samples was collected in winters 1983-1984 and 1984-85 during snow-tracking. Teeths, bone fragments, and hair and feather remains were used for the identification of vertebrate prey. To find remains of all prey items in each scat, smallest scats were totally used for identification and from greater scats hair and feather remains were taken from several parts of the scat. Remains of more than one vertebrate prey could be found in 1 gut and 27 scats (16.3 %) of the stoat, but each gut and scat of the least weasel included only one vertebrate prey. Since only a low U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5204-0257 $ 02.50/0 Z. Säugetierkunde 52 (1987) 257-260 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 258 E. Korpimäki and K. Norrdahl proportion of the food samples of the stoat contained more than one prey item, correction factors (ERLINGE 1981) were not used. Scale patterns, cross-sections and medulla types were examined from hair remains, and their idenfication was made according to Day (1966) and DEBROT et al. (1982). Hairs collected from small mammals trapped in the study area were also used as reference material. Feather remains were ıdentified according to Day (1966). Small mammals at Alajoki were snap-trapped between 1977 and 1985. Captures were performed in May and early June (spring catches) and late August and early September (autumn catches), at four sample quadrats (in a cultivated field, an abandoned field, a pine forest and a spruce forest) each year (a total of 13 404 trap nights). At each quadrat, 50-60 Finnish galvanızed metal snap-traps were set in small mammal runs at distances of 10 m apart for four days (sometimes only three days) and were checked once a day (see KorPIMÄRI 1984, 1985; NORRDAHL 1985 for further details). Although these traps are relatively efficient, they are too robust for smaller species of shrews (HENTTONEN 1985; KorrımÄkı 1986a), and the trappability of harvest mouse Micromys minutus with them is also low (KoskeLa and Vıro 1976). Therefore pitfall traps made by deep buckets were also used in 1984-1985. According to our observations, they are more efficient than snap-traps for collecting both shrews (see also e.g. Aurak 1967; PanKAKoskI 1979), harvest mice (KoskELA and Vıro 1976) and microtines. Results After a peak in winter 1982-1983, both Microtus vole (M. agrestis and M. epiroticus) and bank vole (Clethrionomys glareolus) populations were low in autumns 1983 and 1984; shrews were clearly more abundant than these voles (Table 1). Based on pitfall trappıngs, harvest mice were more frequent Tablod than voles but less frequent than shrews in autumn 1984. The most important prey of the stoat was harvest mouse, followed by Microtus spp. and water vole Ar- vicola terrestris (Table2). Stoats fre- quently took also passerines, mus- krats (Ondatra zibethica), house Microtus spp. mice (Mus musculus), and Norway Bank vole rats (Rattus norvegicus) but only ex- Harvest mouse > Be: a: ceptionally shrews (one individual, 0.5 %), although these were the No. of trap nights most abundant group of small mam- mals in the field (Table 1). The most common prey items of the least weasel were Microtus voles and harvest mouse but shrews were not at all recorded in its diet. Density indices for small mammals in autumn trap- pings at Alajoki in 1983-84 (Trapped ind./100 trap nights) Snap traps Pitfall traps 1983 1984 Discussion Stoats and weasels seemed to catch both murids and microtines in the similar relation as their availability indicated, but they evidently avoided to feed on shrews. The most probable reason is the well-known distaste of both mustelids for shrews (see e.g. KınG 1980; ERLINGE 1981 with references). This result gives circumstantial evidence for the suggestion by KorPpımäkI (1986a) that these small mustelids should not cause the decrease of shrew populations in the transition zone. Birds of prey, which frequently feed on shrews in addition to microtines (see KorrımÄkı 1986a: Table1l) may play a more important role in simultaneous declines of microtines and shrews in this zone, especially ın spring and summer, although we cannot completely exclude the effects of other extrinsic factors in this case (see KorPIMÄRI 1986a). The above suggestion must be treated with caution, however, since there are at least Low proportion of shrews in the diet of small mustelids 259 Table 2 Diet composition of the stoat and least weasel at Alajoki between autumn 1983 and spring 1985 as indicated by scat and gut analyses (The figures show the proportions of various prey cate- gories as percentages by number) Prey categories Stoat Weasel Microtus spp. Bank vole Water vole Muskrat Norway rat House mouse Harvest mouse Sciurids Lagomorphs Shrews Passerines Others (insects, plants etc.) No. of prey items three factors that may confuse the results. 1. The number of food samples in the present case ıs still relatively low. 2. According to observations during snow-tracking (KORPIMÄRI and NORRDAHL unpubl.), stoats can sometimes kill shrews in winter, but they do not feed on them, probably because of distaste. The similar response has been reported also for the weasel (Rusına 1960). Therefore gut and scat analyses do not necessarily reflect exact effects of small mustelids on prey populations. 3. The food data on weasels concerns mainly males, since females only rarely hunt above the snow surface in winter (KORPIMÄRI and NORRDAHL unpubl.). According to ERLINGE (1975), some aberrant weasel females kept in captivity readily fed on shrews, although they usually showed distaste for shrews. We are well aware that the low proportion of shrews observed in the diet of small mustelids may reflect only the situation in the transition zone, where the number of alternative prey other than shrews is higher than ın the northern zone. Occasıonally greater proportions have been also recorded in American (HamıLTon 1933; ArLpous and Man- WEILER 1942) and Russian studies (OGNnEvV 1935; VERSHININ 1972), but not in Western (Day 1968; Kına 1980) and Northern European studies (ERLINGE 1975, 1981). Probably, small mustelids may feed on shrews only in areas where other, more preferred alternative prey groups are scarce (for example in the northern zone). Acknowledgements We thank Mıkko Hast, Ossı HEMMINKI, Tımo Hyrsky and OLLI NORRDAHL for assistance in the field, Assoc. Prof. SEPPoO SuULKAvA for valuable comments on the manuscript, and the Emil Aaltonen Foundation and the Academy of Finland for financial support in connection with the present study. Zusammenfassung Geringer Anteil von Spitzmäusen in der Nahrung kleiner Musteliden aus West-Finnland Exkremente und Mägen von Hermelin und Mauswiesel wurden in Zeiten nach einem Zusammen- bruch von Wühlmaus-Populationen (Microtini) in West-Finnland gesammelt und auf Nahrungszu- sammensetzung untersucht. Dabei zeigte sich, daß diese kleinen Musteliden keine Spitzmäuse (Soricidae) als Ersatznahrung genutzt hatten, wahrscheinlich aus Abneigung vor diesen möglichen Beutetieren. Hermeline und Mauswiesel haben daher wahrscheinlich die synchrone Abnahme der 260 E. Korpimäki and K. Norrdahl Wühlmaus- und Spitzmaus-Populationen, die in den Übergangszonen der fennoscandischen Wühl- mauszyklen festgestellt worden ist, nicht verursacht. In dieser Zone sind wohl in stärkerem Maße Raubvögel verantwortlich zu machen, besonders für Frühjahrs- und Sommerzusammenbrüche. Sie fressen außer Wühlmäusen ım allgemeinen auch Spitzmäuse. References Aıpous, $. T.; MANWEILER, J. (1942): The winter food habits of the short-tailed weasel in Northern Minnesota. J. Mammalogy 23, 250-255. AuLAK, W. (1967): Estimation of small mammal density in three forest types. Ekol. Polska (A) 15, 733-788. Day, M. G. (1966): Identification of hair and feather remains in the gut and faeces of stoats and weasels. J. Zool. Lond. 148, 201-217. — (1968): Food habits of British stoats (Mustela erminea) and weasels (Mustela nivalıs). J. Zool. Lond. 150, 485-497. DEBROT, $.; Fıvaz, G.; MERMOD, C.; WEBER, J.-M. (1982): Atlas des poils de mammiferes D’ Europe. ERLINGE, S. (1975): Feeding habits of the weasel Mustela nivalıs ın relation to prey abundance. Oikos 26, 378-384. — (1981): Food preference, optimal diet and reproductive output in stoats Mustela erminea in Sweden. Oikos 36, 303-315. Hamırrton, W. J. (1933): The weasels of New York. Am. Midl. Nat. 14, 289-344. Hansson, L. (1984): Predation as the factor causing extended low densities in microtine cycles. Oikos 43, 255-256. Hansson, L.; HENTTONEn, H. (1985): Gradients in density varıations of small rodents: the importance of latitude and snow cover. Oecologia (Berl.) 67, 394-402. HENTTONEN, H. (1985): Predation causing extended low densities in microtine cycles: further evidence from shrew dynamics. Oikos 44, 156-157. Kınc, C. M. (1980): The weasel Mustela nivalıs and its prey in an English woodland. J. Anım. Ecol. 49, 127-159. KoRrPIMÄRI1, E. (1984): Population dynamics of birds of prey in relation to fluctuations in small mammal populations in western Finland. Ann. Zool. Fennicı 21, 287-293. — (1985): Prey choice strategies of the kestrel Falco tinnunculus in relation to available small mammals and other Finnish birds of prey. Ann. Zool. Fennici 22, 91-104. — (1986a): Predation causing synchronous decline phases in microtine and shrew populations in western Finland. Oikos 46, 124-127. — (1986b): Gradients in population fluctuations of Tengmalm’s owl Aegolius funereus in Europe. Oecologia (Berl.) 69, 195-201. KoskELA, P.; Vıro, P. (1976): The abundance, autumn migration, population structure and body dimensions of the Harvest Mouse ın Northern Finland. Acta Theriol. 21, 376-387. NORRDAHL, K. (1985): The population fluctuations of small mammals in Suomenselkä and southern Ostrobothnia, western Finland, in 1969-1984. Suomenselän Linnut 20, 57-68 (in Finnish with English summary). Ocnev, $. I. (1935): Mammals of U.S.S.R. and adjacent countries. Israel Program for Scientific Translatıons, Jerusalem. PAnKAKOsKI, E. (1979): The cone trap — a useful tool for index trappıng of small mammals. Ann. Zool. Fennicı 16, 144-150. Rugına, M. A. (1960): Some features of weasels (Mustela nivalıs L.) ecology based on observations in the Moscow region. Byull. Mosk. o-va Ispyt. Prir. Otd. Biol. 65, 27-33. (Translated from the Russian by British Lending Library, Boston Spa, Yorks, trans. no. RTS 2292.) VERSHININ, A. A. (1972): The biology and trapping of the ermine in Kamchatka. Byull. Mosk. o-va Ispyt. Prir. Otd. Biol. 77, 16-26. Authors’ addresses: ERKKI KoRPIMÄRTI, kp. 4, SF-62200 Kauhava, Finland, and Kar NORRDAHL, Mäntytie 8, SF-62300 Härmä, Finland NZISSIEINSICHHIENE ESTER SKEE RAN TATETIEREUINI GEN Haselmäuse (Muscardinus avellanarius) in Nestern freibrütender Singvögel Von W. Fuchs Eingang des Ms. 7. 1. 1987 Angeregt durch die Beobachtungen von BERTHOLD und QUERNER (1986) über Haselmäuse in Nestern freibrütender Singvögel möchte ich über ähnliche Fälle berichten. Sie stammen aus der Umgebung von Einsiedeln/Schweiz und liegen im Schnitt um 1000 m ü.M. Fall 1: 4 Beobachtungen zwischen 26.4. und 1.6. 1951; bewohntes Nest einer Rabenkrähe (Corvus corone) in einer Tanne. Eine Haselmaus ist „Untermieter‘‘ im Krähenhorst. Sie verläßt den Horst, der ca. 7m über dem Waldboden liegt, jedesmal, wenn ich den Horst berühre (den ich zwecks Beobachtung der Krähen besuche). Das Ein- schlupfloch zum kugeligen Nest liegt auf der Unterseite des bewohnten Krähenhorstes. Fall 2: 12.5. 1952. In einem wohl vorjährigen und von den Elstern (Pica pica) nicht mehr bewohnten Nest an einem Waldrand, etwa 3 m über dem Waldboden, hat sich eine Haselmaus eingenistet. Die Haselmaus verläßt ıhr Nest schon, als ich die etwa 20 cm dicke Esche schüttle, in der sich das Nest befindet, um zu kontrollieren, ob das Elsternnest besetzt seı. Fall 3: 18.6. 1952. Eine Haselmaus hat ihr Nest in die Mulde eines verlassenen Singdrosselnestes (Turdus philomelos) ın einer Tanne eines Jungwuchses mitten in einem Wald gebaut. Höhe über dem Boden ca. 4 m. Fall 4: Diese Beobachtung stammt aus der Gegend von Schwyz/Schweiz und liegt etwa 1300 m ü.M. Nach einer stürmischen Winternacht, in der es auch regnete, fand ich auf einer mehr als 1m hohen Schneedecke in einem etwa 3m hohen Tannendickicht ein heruntergefallenes Nest der Singdrossel (Turdus philomelos), in dessen Mulde ein „Mäuse- nest‘““ gebaut war. Darin fand ich eine schlafende Haselmaus. Danach bedarf die Angabe von STORcH (1978, S. 270) „Freie Singvogelnester werden nur selten ausgebaut“ wohl einer Überprüfung. Literatur BERTHOLD, P.; QUERNER, U. (1986): Die Haselmaus (Muscardinus avellanarius) in Nestern freibrü- tender Singvögel. Z. Säugetierkunde 51, 255-256. STORCH, G. (1978): Muscardinus avellanarins (Linnaeus, 1758) — Haselmaus. In: Handbuch der » Säugetiere Europas. Hrsg. von J. NIETHAMMER und F. Krapp. Wiesbaden: Akademische Verlags- ges. Bd. 1, 259-280. Anschrift des Verfassers: \WENDELIN Fuchs, Fliederweg 2, CH-6438 Ibach/Schweiz U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5204-0261 $ 02.50/0 Z. Säugetierkunde 52 (1987) 261 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Die Oberlippe der Elefanten Von M. S. FISCHER Zentrum der Morphologie, Dr. Senckenbergische Anatomie, Frankfurt am Main Eingang des Ms. 8. 4. 1987 Der Rüssel der Elefanten war Gegenstand vieler Studien. Boas und PauLLi (1908) haben seine Anatomie beschrieben und einzigartig illustriert. Die Untersuchung der ontogeneti- schen Bildung dieses Organes ist eine der wenigen noch zu füllenden Lücken. Als kurze, vorausgeschickte Mitteilung einer umfangreichen Untersuchung der Kopfentwicklung will ich hier auf eine bislang unbekannte Tatsache hinweisen. Bei einem sehr jungen Fötus von Loxodonta afrıcana (Fadenlänge Rüsselspitze — Schwanzspitze = 130 mm, SSL 60 mm) tritt noch eine selbständige, vom Rüssel getrennte Oberlippe auf (Abb. 1). Größere Föten (Fadenlänge 310 mm, SSL 147 mm und Faden- länge 440 mm, SSL 220 mm) zeigen bereits die für Elefanten charakteristische Verschmel- zung der Oberlippe mit der ventralen Rüsselbasis. Bei diesen Exemplaren tritt die von FRIANT (1933) bei einem Fötus (Fadenlänge 750 mm) beschriebene buccale Tasche (,„vesti- bule bucal“) auf. Diese Bildung wird von Frıant als Überrest einer Oberlippe gedeutet. Loxodonta africana. Fötus von 130 mm Fadenlänge (Rüsselspitze bis Schwanzspitze). Sowohl in der Lateralansicht (a) als auch in der Ventralansicht (b) erkennt man die Mundspalte und deren Begren- zung durch eine selbständig ausgebildete Oberlippe. (Aufnahme U. TRAUTMANN) Mit dem vorliegenden Befund muß somit der beispielsweise von EALEs (1926), FRIANT (1933) und FRADE (1955) geäußerten Ansicht, die Elefanten besäßen keine selbständig ausgebildete Oberlippe, widersprochen werden. Gleichzeitig kann gezeigt werden, daß sich die Bildung des Rüssels zunächst auf die Verlängerung der nasalen Anteile zu beschränken scheint. Ob bei dem vorliegenden Fötus an der Ventralseite des Rüssels schon Teile der Oberlippe beteiligt sind, wie es nach Boas und Paurti (1908) charakteristisch für Adultformen anderer rüsseltragender Säugetiere ist, können erst weitere Untersuchungen zeigen. U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5204-0262 $ 02.50/0 Z. Säugetierkunde 52 (1987) 262-263 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Die Oberlippe der Elefanten 263 Danksagung Ich bedanke mich sehr bei Herrn Prof. Dr. J. $S. PErry, Cambridge, und Dr. R. M. Laws, London, für die Überlassung des äußerst wertvollen Materials sowie bei Prof. Dr. P. LuckeErr für die Beschaffung. Frau U. TRAUTMANN danke ich für die photographische Arbeit. Literatur Boas, J. E. V.; Paurtı, S. (1908): The elephant’s head. Jena: G. Fischer. EALzEs, N. B. (1926): The anatomy of a foetal African elephant, Elephas africanus (Loxodonta africana). Trans. R. Soc. Edin. 54 (3), 491-551. FRADE, F. (1955): Ordre des Proboscidiens. In: Trait€ de zoologie. 17 (1). Dir. par P.-P. Grass£. Paris: Masson et cie. 715-783. FRIANT, M. (1933): La regression de la levre superieure au cours de l’ontog£nie individuelle chez l’Elephant. C. R. Acad. Sci. Paris 196, 878-879. Anschrift des Verfassers: Dr. MAarTın S. FiscHER, Zentrum der Morphologie, Dr. Senckenbergische Anatomie, Klinikum der Johann-Wolfgang-Goethe-Universität, Theodor- Stern-Kai 7, D-6000 Frankfurt am Main BIIESEBESPIRESTIUNGEN SıeLy, R. M.; SMITH, R. H. (eds.): Behavioural Ecology: ecological consequences of adaptive behaviour. 25th Symp. British Ecological Society, Reading 1984. Oxford: Blackwell Scientific Publications 1985. 530 pp., 132 fıgs. £ 19.50. Nachdem in den letzten 10 Jahren besonders in England die Verhaltensökologie intensiv betrieben worden ist, war dieses Forschungsgebiet für ein Symposium der British Ecological Society überfällig. Auf ihrem 25. Symposium beschäftigten sich jetzt 27 Referate und 40 Poster mit den ökologischen Konsequenzen anpassungsfähigen Verhaltens. Im Mittelpunkt stand die Frage, wie das Verhalten des Individuums die Dynamik der Population beeinflussen kann. Diese Thematik ist von besonderem Interesse, da bisher Populationsbiologen in ihren Arbeiten nicht die Verhaltensmechanismen und Ethologen nicht die Populationsdynamik berücksichtigten. Auf dem Symposium fanden daher die Funktionskreise Ernährung, Raumnutzung, Fortpflanzung und Sozialisation besondere Beachtung. Neben sehr ausführlichen Übersichtsreferaten, die über 20 % des Berichtsbandes einnehmen und neben Beispielen aus der Ornithologie und Entomologie sind für die Säugetierbiologie folgende Beiträge relevant: 1.In einer sehr umfassenden Diskussion über intraspezifische Ernährungsspezialisationen wird u.a. das Ergebnis eines Fütterungsexperiments mit Rötelmäusen (Clethrionomys) gebracht, das zeigt, daß? die Ausnutzung einer Nahrung auch davon abhängt, in welcher Reihenfolge die einzelnen Nahrungsqualitäten verzehrt werden. 2. Unter den Soricinae finden sich zwei Strategien, Hungerperioden zu überstehen: die kleinen Arten (bis 5 g) erhöhen ihre Tagesaktivität, vornehmlich zur Nahrungssuche, die großen Arten (bis 12 g) reduzieren dagegen ihre Tagesaktivität. 3. An Beispielen der amerikanischen Wühlmaus Microtus townsendii wird aufgezeigt, daß verschiedene Raum- und Nahrungsnutzung die Dichte von Wühlmaus-Popula- tionen beeinflussen. So kann dieses Verhalten ausschlaggebend sein, welche Tiere den Populations- überschuß bringen, die extrinsischen Faktoren aber das Schicksal dieses Überschusses bestimmen. Die Berücksichtigung des Raum- und Nahrungsnutzungsverhaltens schaffe eine Verbindung zwischen den intrinsischen und den extrinsischen Regulationshypothesen. 4. Durch radio-tracking wurde beim Grauhörnchen ($. carolinensis) gefunden, daß die saisonale Variation der Streifgebietsgröße mehr mit der sexuellen Aktivität als mıt dem Nahrungsangebot in Zusammenhang steht. 5. Es wird der Einfluß der Sozialstruktur auf die Population von Carnıvora und Primates diskutiert und dabei hervorgeho- ben, daß das soziale Verhalten auf der Stufe des Individuums die Populationsdynamik beeinflussen kann, auch dann, wenn dichteabhängige Einwirkungen fehlen. 6. Es wird eine Erklärung versucht für die Beobachtung, daß Gruppenterritorien bei einzelnen Carnivora-Arten eine sehr unterschiedliche Größe besitzen können. Denjenigen Arten, die mit zunehmender Gruppengröße ihr Territorium vergrößern („expansionists““) stehen diejenigen Arten gegenüber, die ein relativ kleines, ökonomisch 264 Buchbesprechungen zu verteidigendes Territorium behaupten und eine maximale Gruppengröße nicht überschreiten („contractors“‘). Die Lösung wird in der Nutzung unterschiedlicher Nahrung und deren Verteilung gesehen. 7. Eine vergleichende Studie zweier Kaninchenpopulationen in zwei unterschiedlichen Habitaten (Kalkhügel, Sanddünen) hatte das Ergebnis, daß an beiden Standorten die Aggressions- handlungen vornehmlich intrasexuell auftraten und die Männchen um den Weibchenbesitz konkur- rierten. Die dichteabhängige Sterblichkeit und die Populationsstabilität standen auf den Sanddünen in Zusammenhang mit der Konkurrenz um Nahrung, auf den Kalkhügeln mit der Konkurrenz um Höhlen und Partner. 8. Schließlich werden Beispiele dafür angeführt, daß bei sozial lebenden Säugetieren arttypische Fortpflanzungssysteme die Demographie beeinflussen können. Wenn Grup- pen anwachsen, aber aufgrund von Dominanzverhältnissen der Fortpflanzungserfolg sinkt, dann vollzieht sich die Populationsregulation auf dem Niveau der Gruppe, nicht auf dem der Population; das kann sowohl einen Einfluß auf die Populationsstabilität als auch auf die Konkurrenzstrategie der Gruppenmitglieder haben. Der größte Teil der Beiträge ist als Diskussionsgrundlage für ethologische, ökologische und besonders für verhaltensökologische Seminare sehr geeignet. Die Übersichtsreferate bieten sich an, um mit den Zielen der Verhaltensökologie vertraut zu machen. R. SCHRÖPFER, Osnabrück HART, B. L.: The behavior of domestic anımals. New York, Oxford: Freeman 1985. 390 pp., 75 fıgs. £ 33.95. ISBN 0-7167-1595-3 In this book the American veterinarian B. L. HART presents many results of comparative ethology on domestic anımals under a certain perspective. In one respect his purpose is to stimulate ethological investigations on this special and extremely easily available anımal group, and on the other hand the author wants to make reactions and special behavioral traits of domestic anımals understandable to their owners and keepers. In a short introduction the author very generally outlines the importance of domestic anımals for human beings, start and development of domestication, as well as ancestry of the most important species. Comparative ethological, experimental, and physiological methods are pointed out to be the basıc approaches for the study of anımal behavior. The diverse chapters in the main part of the book are associated with two sections. In the first one, there are rather common descriptions of social organisation, communication and aggression, sexual behavior, maternal care, and feeding. The second section mainly deals with the significance of genetics, early experience, learnıng, and hormonal constitution influencing or probably causing special behavioral traits. Concluding remarks on contributions of behavioral science to issues in anımal welfare furtheron are given in the last chapter. The descriptions are essentially concerned with dog, goat, sheep, cattle, pig, horse, chicken, and turkey. Results on laboratory mammals are also included. The book is well written and attractively illustrated. Many problems are discussed more ın depth, some others are only mentioned or even ignored. Numerous literature citations demonstrate the in- depth discussion of certain problems and enable readers to further information. However, several statements don’t represent modern knowledge, and certain very important Sowjet studies as well as numerous investigations of European scientists unfortunately remain completely unmentioned. These results could really have helped in answering many questions. Comparisons of behavioral elements between wild living ancestral forms and domestic animals are only mentioned sparsely and in a very overall manner. Possible influences on behavior caused by morphological peculiarities of the domestic anımals’ body or caused by “disproportioning” of organs remain disregarded, too. This is surprising because, after all, behavior is dependent on structure and function as well. Disregarding these gaps, however, the book can stimulate further studies in many ways, but it cannot be valued as a synopsis of modern knowledge on the comprehensive theme indicated by the title. D. Kruska, Kiel Meilensteine der Genetik Eine Einführung - dargestellt an den Entdeckungen ihrer bedeutendsten Forscher. Von Prof. Dr. Ivar Jonansson. Unter Mitarbeit von Prof. Dr. Arne Müntzıng, Lund. Aus dem Schwedischen über- tragen von Prof. Dr. Hans-OTTo GRAVERT, Kiel. 1980. 252 Seiten mit 146 Einzeldarstellungen in 89 Abbildungen, 8 Übersichten und 12 Tabellen. Ganz auf Kunstdruckpapier. 21,5X13,5cm. Kartoniert 58,- DM. | Ivar Johansson, 1979 als 87jähriger mit der Hermann-Nathusius-Medaille der Deutschen Gesellschaft für Züchtungskunde geehrt, hat noch selbst den Durchbruch des Mendelismus miterlebt und als einer der bedeutenden Repräsentanten der Genetik schon frühzeitig die Chancen erkannt und genutzt, die sich aus der ftschen Anwendung genetischer Erkenntnisse in der Tierzucht ergeben. Wie sehr den Verfasser über sein spezielles Arbeitsgebiet hinaus die Genetik mit ihrem vielfältigen Einfluß auf unser aller Leben fasziniert, beweist der vorliegende geschichtliche Abriß. In zwölf Kapiteln werden die Vererbungstheorien vor Mendel, der ee die biochemische Genetik, die Zyto- genetik und die Populationsgenetik in ihrer Entstehung wie in ihrer heutigen Bedeutung behandelt. Dabei geht es nicht nur um die jeweiligen Theorien, sondern auch um ihre praktischen Anwendungs- a hrlıkeiten bei Mensch, Tier und Pflanze. Besonders eindrucksvoll zeigt sich die Entwicklung der Genetik an den Pionierleistungen einzelner herausragender Forscherpersönlichkeiten, die auf ihren Gebieten bahnbrechende Entdeckungen ge- macht haben und in großer Zahl mit dem Nobelpreis ausgezeichnet worden sind. Die Liste der berühmten Namen reicht, um nur einige Beispiele zu nennen, von Darwin, Galton, dem Wegbereiter der Biometrik, und Mendel über Correns, Tschermak und de Vries, die Wiederentdecker der Mendel- scher: Gesetze bis zu den Drosophila-Forschern Morgan und Muller, berücksichtigt die Populations- genetiker Hardy, Weinberg, Wright und Fisher wie die Molekulargenetiker Watson und Crick, Jacob, Monod, Nirenberg, Ochoa, Lederberg und Luria. Unterstützt durch zahlreiche instruktive Abbildungen, macht Johanssons Einführung in die Genetik breitere, biologisch interessierte Kreise mit den Mechanismen der Vererbung vertraut. Das Buch weckt somit Verständnis für genetische Fragen, zeigt zugleich aber auch die Grenzen der genetischen Manipulierbarkeit. Chromosomen Organisation, Funktion und Evolution des Chromatins. Von Prof. Dr. WALTER NAGL, Kaiserslautern. „Pareys Studientexte“, Nr. 23. 2., neubearbeitete und erweiterte Auflage. 1980. 228 Seiten mit 102 Ab- bildungen und 12 Tabellen. 23,5x 15,5 cm. Balacron broschiert 29,— DM Mit dieser übersichtlichen Gesamtdarstellung des heutigen Wissensstandes über die strukturelle und funktionelle Organisation der Chromosomen wendet sich Prof. Nagl an alle, die sich für die Regulation von Entwicklung und Evolution sowie für die mögliche Manipulierbarkeit von Lebewesen, vor allem des Menschen, interessieren. Beschrieben wird ein faszinierendes Forschungsobjekt mit seinen Rück- wirkungen auf viele Teildisziplinen wie Genetik, Evolutionsforschung, Entwicklungsbiologie oder Krebsforschung. Darüber hinaus bietet das Buch mit Begriffsdefinitionen und eehilrenden Literaturangaben dem Dozenten und Lehrer eine schnelle und aktuelle Information; als Leitfaden ermöglicht es dem Studierenden eine gründliche Einarbeitung in die Materie. Bere PR AUL PAREY - HAMBURG UND BERLIN Erscheinungsweise und Bezugspreis 1987: 6 Hefte bilden einen Band. Jahresabonnement Inland: 298,— _ DM zuzüglich 13,80 DM Versandkosten, Gesamtpreis 311,80 DM einschließlich 7% Mehrwert- steuer. Jahresabonnement Ausland: 298,- DM zuzüglich 24,— DM Versandkosten. Das Abonne- ment wird zum Jahresanfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhandlung oder beı der Verlagsbuchhandlung Paul Parey, Spitalerstraße 12, D- 2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesellschaft für Säugetier- kunde“ erhalten die Zeitschrift unberechnet ım Rahmen des Mitgliedsbeitrages. Z. Säugetierkunde 52 (1987) 4, 201-264 GRZIMEKS ENZYKLOPÄDIE 5 Bände mit insgesamt 3.280 Seiten und 3.500 Abbildungen. Lexikon-Großformat je Band 18,4x25 cm. Wählen Sie zwischen 2 attraktiven Ausstattungsvarianten: Die Erscheinungsfolge: September 1987 —- Band IV Pantherkatzen, Robben, Hunde, Wale, Hasen, Elefanten, Nashörner, Pferde, Zebras u.a. Februar 1988 —- Band I. Primaten, biologische Anthropo- logie. April 1988 - Band. Die Säugetiere: Ursprung, Bau und Leistungen. Beuteltiere, Igel, Maul- würfe, Fledermäuse u.a. Juli 1988 - Band III. Nagetiere, Mader, Bären, Klein- kalizen. Oktober 1988 - Band V. Kamele, Giraffen, Flufspferde, Wildrinder, Ziegen, Schafe, Schweine u.a. der edlen Luxus-Ausgabe mit Leder- rücken, Goldschnitt und farbiger Rückenprägung. Subskriptionspreis pro Band DM 168,-, späterer Laden- preis DM 198,- je Band. Gesamtpreis DM 840,-. Subskriptionsersparnis DM 150,- 1B der Leinen-Ausgabe aus hochfeinem Naturleinen mit farbigem Schutz- umschlag. Subskriptionspreis pro Band DM 128,-, späterer Ladenpreis DM 148,- je Band. Gesamtpreis DM 640,-. Subskriptionsersparnis DM 100,-. Die Subskription endet jeweils mit Fertig- stellung des letzten Bandes. Über 100 Autoren aus der ganzen Welt. Aus redaktionellen Gründen erscheint Band IV zuerst, und zwar im September 1987. NER Die völlig neue Enzyklopädie der Säugetiere mit den jüngsten Erkenntnissen der Feldforschung, Okologie und Evolutionsbiologie. Ausführliche zoologische Daten in vergleichend-tabellarischer Form. Weitere Informationen im Buchhandel oder vom Kindler-Verlag, Rauchstr. 9-11, 8000 München 80 Vol. 52 (5), 265-328, Oktober 1987 ISSN 0044-3468 C 21274 F =.IITSCHRIFT FÜR SAUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Organ der Deutschen Gesellschaft für Säugetierkunde Böer, M.: Beobachtungen zur Fortpflanzung und zum Verhalten des Drill ( Mandrillus leucophaeus Ritgen, 1824) im Zoo Hannover. — Observations on reproduction and behaviour of captive drills (Mandrillus leucophaeus Ritgen, 1824) in Hannover Zoo 265 Haaf, T.; Weis, H.; Schmid, M.: A comparative cytogenetic study on the mitotic and meiotic chromosomes in hamster species of the genus Phodopus (Rodentia, Cricetinae). — Vergleichende zytogenetische Untersuchun- gen an den Hamstern Phodopus sungorus und P. roborovskii (Rodentia, Cricetinae) 281 Rödel, R.: Alter und Linsengewicht von Meriones shawi grandis (Cabrera, 1907). — Age and eye-lens weight in Meriones shawi grandis (Cabrera, 1907) 291 Stahnke, Adelheid: Verhaltensunterschiede zwischen Wild- und Hausmeerschweinchen. — Behavioral comparison of wild and domestic cavies 294 Richardson, P. R. K.: Food consumption and seasonal variation in the diet of the aardwolf Proteles cristatus in southern Africa. 1. Nahrungsbedarf und saisonale Variabilität in der Ernährung des Erdwolfes Proteles cristatus in Südafrika 307 Buchbesprechungen 825 Verlag Paul Parey Hamburg und Berlin HERAUSGEBERZ/EDTLORS P. J. H. van BreEeE, Amsterdam - W. FIEDLER, Wien — H. Frick, München - W. HERRE, Kiel - H.-G. Krös, Berlin - H.-J. Kuhn, Göttingen - B. Lanza, Florenz - T. C. S. MORRISON-SCOTT, London - J. NIETHAMMER, Bonn — H. REICHSTEIN, Kiel - M. Rönss, Hannover -— D. STARCK, Frankfurt a. M. - F. Strauss, Bern — E. THENIUs, Wien — W.VERHEYEN, Antwerpen SCHRIFTLEITUNG/EDITORTATZORFI@E 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 Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei- ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ist auch im Einzelfall grund- sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines Beitrages ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber- rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen. Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of the first page of an article in this journal indicates the copyright owner’s consent that copies of the article may be made for personal or internal use, or for the personal or internaluse of specific clients. This consent is given on the condition, however, that the copier pay the stated percopy fee through the Copyright Clearance Center, Inc., 21 Congress Street, Salem, MA 01970, USA, for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective, or for resale. For copying from back volumes of this journal see “Permissions to Photo-Copy: Publisher’s Fee List” of the ETE: Fortsetzung 3. Umschlagseite © 1987 Paul Parey. Verlag: Paul Parey, Hamburg und Berlin. Anschriften: Spitalerstr. 12, D-2000 Hamburg 1; Lindenstr. 44-47, D-1000 Berlin 61. — Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., D-22409 Heide/Holst. Beobachtungen zur Fortpflanzung und zum Verhalten des Drill (Mandrillus leucophaeus Ritgen, 1824) im Zoo Hannover! Von M. BöER Zoologischer Garten der Landeshauptstadt Hannover Eingang des Ms. 22. 9. 1986 Abstract Observations on reproduction and behaviour of captive drills (Mandhillus lencophaeus Ritgen, 1824) in Hannover Zoo Studied zoobiological consequences of reproductive and behavioural patterns in a captive drill family. Tight social bonds existed among maternally related individuals. Births occurred 179, 182 and 183 days respectively after observed copulation. Maximal labial tumescence in estrous females lasted 3-7 days. Copulations were observed 2-8 times a day in 48-72 h. Menstrual cycles varied between 2542 days. First post partum menses in lactating females occurred after four months. During another 4 months females did not conceive despite of regular labial tumescence and copulations. With a mean interbirth interval of 19 months and a lactation period of 16 B females lactated the first half of pregnancy. A young female gained sexual maturity at the age of 3,5 years. 7-8 months old drills show the adult black coloured face mask. At this time maternal milk ıs no longer the main food resource. Weaning, independance of the mother and integration into social hierarchy are completed after 16 months. Attempts of charging, herding and copulation in subadult males as well as aggressive attacks of subadult females toward adult females of all reproductive stages indicating intragroup tendencies for expansion promote important processes in the dynamics of the group leading to social lability. Taking away young drills from their parental group and socializing them with new partners of fitting age as zoobiological steps substitute possible emigration of individuals of this age class from a drill troup ın the field, prevent severe fights with the leading male and ensure social stability of the breeding unit under zoo conditions. Einleitung Ziel der Untersuchungen war es, die in Hannover gewonnenen Erfahrungen bei der Haltung und Fortpflanzung des in der Wildbahn vom Aussterben bedrohten und in den Zoologischen Gärten nur selten gezeigten Drill darzustellen. Die Arbeit soll insbesondere zuchtrelevante Informationen über diese bisher wenig bearbeitete Waldpavianart geben. Die Daten zur Fortpflanzung und Jugendentwicklung und zum Sozialverhalten subadulter Individuen sollen richtungweisend für zukünftige Management-Maßßnahmen im Rahmen des internationalen Zuchtbuches sein. Material und Methode Datenerfassung In der Zeit vom 1. 4. 80 bis zum 31. 10. 85 wurden in der im Zoologischen Garten von Hannover gehaltenen Drillgruppe Geburts- und Paarungstermine notiert und mit entsprechenden Daten voran- gegangener Jahre aus der betriebsinternen Tierkartei ausgewertet. Zwischen dem 14. 8. 84 und dem ! Herrn Prof. Dr. LOTHAR DITTRicH zum 55. Geburtstag gewidmet. U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5205-0265 $ 02.50/0 Z. Säugetierkunde 52 (1987) 265-281 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 266 M. Böer 19. 9. 85 wurde das Verhalten der Gruppe an 29 Tagen mit auffallenden Aktivitäten (Geburten, Paarungen, aggressive Auseinandersetzungen, ausgedehnte Spielphasen) jeweils für die Dauer von 30 bis 120 Minuten durchgehend oder auf zwei bis drei Beobachtungsperioden verteilt mittels Handdik- tiergerät oder direkt auf Protokollblättern qualitativ und quantitativ erfaßt und ausgewertet. Konnten die Aktivitäten mehrerer Drills aufgrund zu hoher Verhaltensfrequenzen nicht mehr gleichzeitig verfolgt werden, wurde nur ein Tier beobachtet (focal anımal sampling), oder es wurden lediglich die Aktivitäten zwischen zwei Drills (dyad interactions) verfolgt. Außerhalb dieser Beobachtungsphasen wurden bei 5 Säuglingen zwischen dem 1. und 16. Lebensmonat 385 Saugvorgänge beobachtet, zeitlich erfaßt und ausgewertet sowie die Veränderungen der Gesichtsfärbung schriftlich oder bildlich festgehalten. Fotografiert wurde mit einer Minolta SRT 303 und einem 400 Asa-s/w-Film. Wegen der Anfälligkeit der Drills für Störnoxen wurde auf die Verwendung eines Blitzgerätes verzichtet, was zum Teil zu verminderten Kontrasten der Aufnahmen führte. Tiermaterial Die Zusammensetzung der Drillgruppe im Beobachtungsverlauf, individuelle Daten sowie Verwandt- schaftsverhältnisse der Tiere untereinander sind in einer Übersicht (Tab. 1a, b) dargestellt. Tabelle 1a Individuelle Daten von 15 im Zoo Hannover gehaltenen Drills Name Geschlecht geboren an Herkunft/Elten in Hannover seit Abgang Transfer Hexe (H) OR ca. 1958 Wildfang 21.03.62 21.3.84 Tod Sonja (S) 0.1 ca. 1974 Wildfang ? lailezz Alexander (A) 1.0 ca. 1970 MWildfang ? 4.07.78 Katharina (K) 0.1 ca. 1970 Mildfang ? 04.07.78 Tschita (T) 0.1 ca. 1972 MWildfang ? 19.04.82 Jochen (J) 1.0 ZINSEN DER Geburt 03.12.79 Zoo Arnhem/Nl Tom (To) 120 10-062 802 EA ak Geburt 20.03.84 Zoo Armnhem/Nl Sue (Su) 0.1 03.0 2:81, ZA xES Geburt 06.02.86 Zoo Osnabrück Ronny (R) 1.0 30128820, Nexak Geburt 01.11.84 Zoo San Diego Kurt (Ku) 10 ZU EIORSSEENDES Geburt 17.07.86 Zoo Wuppertal Hanna (Ha) 0.1 25er STE Geburt 06.02.86 Zoo Osnabrück Kottan (Ko) 1.0 30.09.8348 AxK Geburt 23.08.85 Tod Pamela (P) VS ee SS Geburt 21.01.56 Tod Margot (M) 0.1 107093857 2 AUX2S Geburt Pongo (Po)* 1.0 ca. 1975 MWildfang ? 04.09.85 06.02.86 Zoo Osnabrück * : wurde nicht in der Zuchtgruppe gehalten Haltung und Fütterung Bis 1982 wurden die Drills in einfachen, mit Sitzbrettern und Kletterstangen ausgestatteten Gitter- und Drahtkäfigen gehalten. Danach zogen sie in den zu einer modernen, mit Glas gegenüber dem Zuschauer abgegrenzten Innenanlage umgebauten ehemaligen Gorillakäfig des Tropenhauses (BÖER 1983) um. Fortpflanzung und Verhalten des Drill im Zoo Hannover 267 Tabelle 1b Aufbau der Gruppe im Beobachtungsverlauf ' 14 Aazi sozialer Rang der Weibchen Transfer Transfer nach Gewöhnung an Po Gestorben Von einer ca. 30 m? großen Grundfläche führen freischwingende kunststoffdurchtränkte Hanf- seile, die einseitig am Deckengitter befestigt sind, in die Nähe eines in 2 m Höhe beidseitig waagerecht aufgehängten, etwa 3 m langen Fichenastes, der Platz für zwei adulte Weibchen mit Kindern bietet. Andererseits gelangen die Drills vom Gehegeboden über schräg vom Zuschauer nach oben wegfüh- rende Kletterbäume auf eine rückwärtige, 1,30 m höher gelegene und etwa 20 m? große Ebene, die die Decke des zwischen Innen- und Außenanlage laufenden Ganges für Tierpfleger bildet. Dieser Gang ist an mehreren Kontaktzonen mit Stahlgeflecht gegenüber den Affengehegen im Inneren des Hauses abgegrenzt. Als Außenanlage dient zur Zeit noch der ehemalige Gorilla-Außenkäfig (BöER 1983). 268 M. Böer Hydraulisch und mechanisch zu bedienende Schieber erlauben ein Umsetzen der Drills in benachbarte Gehege oder den Außenkäfig. Die obere Ebene ist mittig durch eine Konstruktion senkrecht verlaufender Holzstämme unter- teilt, die Jungtiere und Weibchen, nicht aber dem adulten Mann, ein Hindurchschlüpfen ermöglicht. Auch bei naßkaltem Wetter mit Temperaturen um den Gefrierpunkt werden die Drills für 30 bis 60 Minuten, meist am frühen Nachmittag, in den Außenkäfig gelassen. Neben der einmal täglich angebotenen Hauptfütterung, die aus weit verstreutem Obst, Gemüse sowie Calzium- und Eisenzusätzen besteht, werden unregelmäßig 2- bis 3mal täglich Mischungen aus Sämereien und frische, belaubte oder unbelaubte Äste angeboten. Dieses Futterschema hat sich bestens bewährt und verhindert weitestgehend intraapezifische Aggressionen um Nahrungsressourcen (RÖLL 1985). Drills fressen in der Wildbahn neben pflanzlicher Kost auch Süßwasserkrabben (Sudanonautes) (VOELKER 1985; VOELKER und SacHs 1977), Pilzmücken (Fungivoridae), vermutlich auch Milben (Acarına), Geradflügler (Orthopteria) sowie Schnecken (Pseudachatina spp. und Lignus solimanus) (GARTLAN und STRUHSAKER 1972). Diese Nahrungsquellen werden im Zoo durch gekochtes Fleisch und rohe oder gekochte Eier, die 1- bis 2mal in der Woche angeboten werden, substituiert. Die Drills trinken Wasser aus dem 1 qm großen und 10 cm tiefen Badebecken, das sie tagsüber häufig zum Waschen und Spielen aufsuchen. Schwarzer Tee, Kräutertee sowie gelegentlich Milchbrei wird ın Tränkepfannen gereicht, die unter entsprechend dimensionierten Gitteraussparungen in den Käfig geschoben, jedoch von den Drills nicht vollständig hineingezogen werden können. Unmittelbar am Gitter ist im Umgang mit diesen Pavianen äußerste Vorsicht geboten, da sie ohne Vorwarnung ihre Arme blitzschnell durch die Maschen stoßen und meist das Gesicht des Betreuers zu treffen suchen. Ergebnisse Lokomotion Obwohl adulte Drills die Fortbewegung auf dem Boden entsprechend ihrer Lebensweise in Kamerun (GARTLAN 1972) gegenüber der auf mobilen Objekten (Äste, Seile) und in der Vertikalen (Gitter) bevorzugen, und nie biped oder quadruped hangelnd beobachtet wurden, weist die Art verschiedene Formen arboricoler Lokomotion auf. Insbesondere zeigen Jungtiere im Alter zwischen 12 und 24 Monaten eine erstaunliche Geschicklichkeit bei der Bewegung über dem Boden. So beobachtet man gelegentlich Handstandlaufen auf labilen Ästen (bis zu 9 Armschritte hintereinander). Drills dieser Altersklasse machen sich anscheinend mit dem Verlauf kombinierter Distanzsprünge zwischen Boden, senkrechten Objekten und dem Deckengeflecht des Geheges, die über mehrere Meter gehen, zunächst durch mehrmaliges Verfolgen der geplanten Sprungbahn mit den Augen vertraut. Sekundenlang sitzen die kleinen Drills beobachtend da, ehe sie losspringen. Fehl- sprünge sah ich nie. Junge Drills hangeln oft ein- oder beidarmig bzw. ein- oder beidbeinig beim Spiel zwischen Ästen und Seilen, um mit den freien Extremitäten an Objekten und Partnern manipulieren zu können (Abb. 1). Muttertiere sind jederzeit in der Lage, selbst 1 bis 1,5 kg schwere Jungtiere, die aus oberen Ebenen auf sie zuspringen, schnell und sicher zu greifen. Ein fest schlafendes Jungtier, das plötzlich durch spielende Artgenossen vom Ast gestoßen wurde, verhinderte seinen Absturz aus 3 m Höhe, indem es sich mit blitzschnel- len kurbelnden Armbewegungen noch den Zugriff zu demselben Ast, auf dem es geschla- fen hatte, sichern konnte, wobei es nicht mehr als 20 cm gefallen war. Verhalten am Futter Einige Drills legen sich zunächst einen Futtervorrat in den weit dehnbaren Backentaschen an, bevor sie dann mit prallen Backen die später aufgenommene Nahrung sofort zerkauen und abschlucken. Dieses Verhalten ermöglicht rangniedrigen Gruppenmitgliedern eine weitestgehend vollständige Mahlzeit, selbst wenn sie bereits wenige Minuten nach Fütte- rungsbeginn vom Futter vertrieben werden. Spontanes Verlassen des Futterplatzes bei gefüllten Backentaschen und Aufsuchen eines sicheren Platzes verhindert Aggressionen Fortpflanzung und Verhalten des Drill im Zoo Hannover 269 ranghoher Tiere und trägt sicher- lich zum Verbleib subdominanter Individuen im Harem bei. Im Sinne der Feindvermei- dung mag dieses Verhalten auch als Adaption an die Präsenz von Beutegreifern in der Wildbahn ei- ne Rolle spielen. Männliche Jungtiere werden bis zum Alter von 15 bis 16 Monaten am Fut- terplatz in unmittelbarer Nähe des Männchens geduldet. Auch das gelegentlich von rangniedrigen Tieren beobachtete Hervorwürgen von Nahrung in die Backentaschen mit anschlie- ßendem bedächtigen Kauen und erneutem Abschlucken kann bei akuter Feindvermeidung eine langsame und verdauungsphysio- logisch sinnvolle Nahrungsauf- bereitung an einem sicheren Ort gewährleisten. Dieses Verhalten wurde nur gelegentlich in zeitli- chen Zusammenhang mit sozia- len Auseinandersetzungen in der Gruppe gezeigt, erfolgte jedoch häufiger ın neutralen Situationen. n adulle I nu nun Ba: Abb. 1. Lange Finger- und Zehenglieder ermöglichen Jung- adulte Weibchen erbrachen gele- zieren griffsichere Lokomotionen auf über dem Boden be- gentlich auch nach langer, unge- findlichen Objekten störter Nahrungsaufnahme Fut- terpartikel, um sie erneut zu kauen und dann abzuschlucken, ohne daß ein zeitlicher Zusammenhang mit möglichen Stressoren deutlich wurde. Solitärverhalten Insbesondere Jungtiere spielen regelmäßig ın flachem Wasser, wobei sie bis zum Nabel untertauchen. Bewegungen des Solitärspieles sind das Hin- und Herschleudern der hinte- ren Körperhälfte bei beidarmigem Bodenkontakt — die Bewegungsrichtung wird blitz- schnell geändert - sowie Schleuderbewegungen des Kopfes. Sozialverhalten Die wohl für erwachsene gruppenführende Männchen typische Körperhaltung des oft minutenlangen Abstützens auf den Ellenbogen bei gleichzeitigem Hochhalten des bunten Hinterteiles konnte erstmals im Alter von 17 Monaten bei einem jungen Männchen unmittelbar nach Kopulationsversuchen mit einem jungen Weibchen für die Dauer von 11 Sekunden beobachtet werden. Als sich das weibliche Tier entfernte, verharrte der junge Mann in dieser Stellung regungslos weiter, was nur noch schwer als reine Spielaufforde- rung zu verstehen war. Ein signifikanter zeitlicher Bezug zu besonderen sozialen Situationen wie Fortpflan- 270 M. Böer zung oder Aggression in der Gruppe konnte für das Auftreten dieses Verhaltens bei dem gruppenführenden Männchen nicht festgestellt werden. Ausnahmslos wurde der rotblaue Gesäßspiegel von allen Gruppenmitgliedern abgewandt und immer zur Peripherie des Geheges ausgerichtet. Häufigste Form des Kontaktverhaltens ist der sehr kurzzeitige und daher in der Gruppe nicht zu quantifizierende Blickkontakt zweier Individuen, der in der Regel ohne bemer- kenswerte Kopf- oder Körperhaltung ausgeführt wird. Zwei beieinanderstehende Drills tippen gelegentlich mit dem Finger auf den Nasenrücken des Partners oder nehmen auch durch gegenseitig oder einseitig ausgeführtes Belecken der Lippen des Partners freundlich miteinander Kontakt auf. Neben der sozial integrativen Funktion hat das Allogrooming unter Drills die Aufgabe, dem Komfortbedürfnis nachzukommen. Jungtiere strecken ihren Müttern in unregelmäßig wechselnder Reihentolge insbesondere Kinn-, Hals- und Schul- terregion entgegen, die belaust werden sollen. Rangniedrige hungrige Tiere, die sich mitunter nicht in Anwesenheit dominanter Gruppenmitglieder an besonders begehrte Futtermittel heranwagen, „groomen“ sich gegenseitig, offensichtlich in einer Kontliktsituation stehend und Erregung abbauend. Ein weit aufgerissenes Maul kennzeichnet das Spielgesicht. Der Partner wird durch Kopfschleudern oder mit flach auf den Boden gestreckter Hals- und Kopfregion, gespreiz- ten, auf dem Boden aufliegenden Unterarmen und hocherhobenem Hinterteil angestarrt und so zum Spiel aufgefordert. Drills drohen inter- und intraspezifisch durch vertikales Kopfnicken mit starrem Blick auf den Gegner. In aggressiver Stimmung können die Haare auf dem Schädeldach aufgerichtet und gelegentlich die Caninı entblößt werden. In defensiver Stimmung zeigen Drills ihr Gebiß fast vollständig, aber geschlossen. Das Maul öffnet sich bei großer Angst leicht, das Drohgesicht kann dann von schrillen hochfrequenten Schreien begleitet sein. Als Abwehrmaßnahme rangniedriger Tiere gegenüber dominanten werden häufig Fuß- tritte beobachtet. Eine sozial wirksame Beißfhemmung entwickelt sich erst zwischen dem 10. und 16. Lebensmonat. Spielende Jungtiere können sich so heftig ineinander verbeißen, daf$ es zum Angsturinieren der Kämpfer kommt. Schwerste Formen der Aggressionen sind das Beißen in die Genitalregion sowie das Beifsschütteln. Diese Verhaltensweisen gehen mit großen Verletzungen einher, werden jedoch bei stabilem Gruppengefüge nicht gezeigt. Drills imponieren durch quadrupedes Querpräsentieren mit angehobenem Schwanz und gestreckten Ellenbogen- und Kniegelenken vor dem Gegner. Der Imponierende schaut den Partner an. Die Körperachsen beider Tiere befinden sich im Winkel zwischen 45° und 135° zueinander. Querpräsentieren löst Aggressionen ranghöherer Drills aus. Kopulationsbewegungen und Aufreitversuche außerhalb der Östrusphasen der Weibchen müssen als Imponiergehabe des adulten Männchens und ranghoher Weibchen gegenüber subdominanten gewertet werden. Jungtiere, die älter als 5 Monate sind, weichen Erwachsenen aus. Gegenseitiges Ausweichen ist ein wichtiger Indikator für die soziale Rangordnung unter den Weibchen (Rörr 1985). Der erwachsene Mann besitzt gegenüber allen Gruppenmitgliedern das Vorlaufsrecht. Er umgeht lediglich Muttertiere, die gerade ihr Kind säugen, selbst an Engpässen wie z. B. aufsteigenden Kletterästen, und fordert sie dabei nie zum Ausweichen auf. Subdominante Weibchen werden während des Östrus vom Männchen gegenüber ranghöheren Weibchen beschützt und in dessen unmittelbarer Nähe geduldet. Gruppendynamische und soziobiologische Aspekte Alle Gruppenmitglieder müssen in der Lage sein, gegenüber dem adulten Männchen in Situationen sozialer Spannung einen Abstand von 4 bis 5 m einzuhalten oder ihm zumindest optisch (hinter Objekten) ausweichen zu können. Adulte Weibchen nehmen Fortpflanzung und Verhalten des Drill im Zoo Hannover 2! meist Distanz von 2 bis 3 m zueinander ein. Heranwachsende Jungtiere müssen nichtver- wandten adulten Drills ihrer Gruppe gelegentlich auf 6 bis 8 m Distanz oder gar darüber hinaus ausweichen, um Aggressionen zu vermeiden. Sind Ausweichreaktionen aufgrund vorgegebener Gehegeabmessungen nicht möglich, zwingen agonistische Verhaltensweisen sowie abnehmende Beißhemmung in der Gruppe den Tiergärtner zur Isolation jung- adulter Tiere. Quantitative Auswertungen sozialer Verhaltensweisen in der Gruppe lassen engste soziale Bindungen zwischen verwandten Tieren, also innerhalb einer Mutterlinie, sowie häufige agonistische Verhaltensweisen zwischen Mitgliedern verschiedener Mutterlinien erkennen. Eine wichtige Aufgabe des gruppenführenden Männchens scheint daher die Befriedung der Gruppe durch aggressive Unterbindung agonistischer Aktivitäten zwischen verschiedenen Mutterlinien zu sein. Andererseits unterstützen Jungtiere sowohl eigene Geschwister sowie auch die Kinder anderer, meist ranghöherer Mutterlinien bei der Lokomotion an unwegsamen Gehegebe- reichen. Das ranghöchste Weibchen der Gruppe ermöglichte sogar Töchtern rangniedriger Mütter und in einem Falle einem verwaisten, nichtverwandten Drillmädchen soziale Sicherung durch Betreuung (Tragen) oder vollständige Adoption, die wie folgt ablief: Adoption eines 8% Monate alten Kindes (P) durch das ranghöchste Weibchen (K) am 9. September 1985: 1. Phase: Häufiges Abbeißsen des sich nähernden Kindes mit Körperverletzung (Rißwunde an der Innenseite des Unterschenkels). 2. Phase: Duldung des Körperkontaktes erstmalig 4 Stunden nach Abtrennen der Mutter. Aufnahme des Waisen in die Arme nach 5 Stunden. 3. Phase: Tragen und Säugen des Adoptivkindes am 2. Tag nach Abtrennen der Mutter. Der Sohn des ranghöchsten Weibchens (K) war 17 Tage vor dieser Adoption gestorben. Verhalten gegenüber Menschen Der erwachsene Mann (A) zeigt sich in Anwesenheit nicht regelmäßig am Gitter erschei- nender Personen ausnahmslos aggressiv und versucht, durch die Maschen zu schlagen. Gelegentlich zeigt er dieses Verhalten auch gegenüber vertrauten Tierptlegern. Erwachsene Weibchen verhalten sich individuell sehr unterschiedlich. Ranghohe Tiere sind nicht immer aggressiver als rangniedrige Tiere gegenüber Personen. Jungtiere verhalten sich sowohl gegenüber Zuschauern am Glas als auch gegenüber Tierpflegern an der rückwärti- gen verdrahteten Kontaktzone freundlich und neugierig und fordern auch zum Spiel auf. Mit 1 bis 1% Jahren kann sich aus dem Spiel allmählich Aggression entwickeln. Das Interesse am Zuschauer läfst in diesem Alter allmählich nach. Fortpflanzungsphysiologie Geburten in der Gruppe fanden ausnahmslos nachts statt und wurden nicht beobachtet. Geburtsvoranzeichen war lediglich das ungewöhnlich langsame Aufnehmen von Futter- partikeln in den letzten 4 Wochen vor der Geburt. Geburten erfolgten in drei Fällen 179, 182 und 183 Tage nach beobachteten Kopulationen bei drei verschiedenen Weibchen (K, S, T). Havıpıan et al. (1979) geben für den Drill in einem beobachteten Fall eine Tragzeit von 173 Tagen an. Paarungen konnten in Hannover an zwei bis drei aufeinanderfolgenden Tagen auftreten, selbst wenn die Genitalschwellung der Weibchen nach einer meist fünf bis sieben Tage anhaltenden Plateauphase bereits wieder zurückging. Andererseits ist auch die Annahme einer einige Tage vor der letzten maximalen Genitalschwellung auftretenden Ovulation bisher nicht widerlegt. Während des Östrus verpaaren sich Drills sieben- bis achtmal täglich, je nach Partner, sozialer Situation und Dauer der Bindung der Partner. In Hannover konnten Kopulationen bei allen tragenden Weibchen in zeitlichem 272. M. Böer Zusammenhang mit unregelmäßig auftretenden Brunstschwellungen während der gesam- ten Tragzeit beobachtet werden. Ein junges Weibchen (Su) gebar im Alter von 3 Jahren und 10 Monaten ihr erstes Kind, war demnach spätestens mit 3 Jahren und 4 Monaten geschlechtsreif. Menstruationsblutungen traten ın einem Fall bereits 127 Tage post partum bei einer laktierenden Mutter auf. Bereits 21 Tage post partum konnten Paarungen beobachtet werden, wobei die Weibchen keine Genitalschwellung aufwiesen. Frühestens vier Monate nach Einsetzen der Regel kam es zur Konzeption. Der hormonelle Status während der Laktation unterdrückt vermutlich das zyklische Geschehen vollständig zunächst für vier Monate und verhindert in den darauffolgenden vier Monaten Konzeptio- nen, nicht jedoch die für den Mann attraktiven Genitalschwellungen. 12 Sexualzyklen wurden bei drei Weibchen jeweils zwischen zwei Geburten beobachtet und varıierten in der Länge zwischen 25 und 42 Tagen. Die durchschnittliche Zwischentragezeit liegt bei 19 bzw. bei 17 Monaten, wenn man auch vorausgegangene Totgeburten berücksichtigt (Tab. 2). (Kinder konnten bereits 7% Monate nach erfolgter Tabelle 2 vorausgegangener Totge- burt, jedoch auch erst 30% Monate nach Geburt und Aufzucht des älteren Ge- Zwischentragezeit des Drill (in Tagen), ermittelt an 19 Geburten von 4 Weibchen Katharina Sonja Tschita Hexe schwister zur Welt kommen.) 564 319* 409 512 Geht man von einer 6 933 228* 619 542 Monate langen Tragzeit 631 679 1165 aus, wird ein Weibchen 438 312% 571 bei Berücksichtigung der 566 ermittelten Zwischentra- gezeit in der Regel 13 Mo- nate post partum erneut Durchschnittliche Zwischentragezeit : 19 Monate erfolgreich gedeckt. Be- (17 Monate bei berücksichtigten Totgeburten(*)) fruchtungen erfolgen dem- nach in der Regel erst ge- gen Ende der 15 bis 16 Monate langen Laktation (s. a. Abb. 3), die Weibchen geben also etwa in der zweiten Hälfte der Tragzeit keine Milch mehr, was hinsichtlich des fötalen Größenwachstums in dieser Zeit energetisch sinnvoll erscheint. Bisher liegen keine Daten zur Dauer der Fruchtbarkeit der Weibchen vor. Ein ım Alter von mindestens 5 Jahren 1962 nach Hannover gelangtes Weibchen (H) brachte bis 1970, also etwa bis zu seinem 14. Lebensjahr, regelmäßig Junge zur Welt, wurde jedoch erst ım Alter von ca. 28 Jahren wegen Altersschwäche eingeschläfert (DITTRICH 1984). Daten aus den Zoologischen Gärten von Paignton, England, sowie New Orleans, Louisiana, ergeben, daß männliche Drills mit 61 Jahren erstmalig erfolgreich kopulieren können. Ein in Hannover im Juni 1980 geborener Mann (To) kopuliert mit östrischen Weibchen regelmäßig seit dem Frühjahr 1985. Die deutlich ausgebildete Färbung der Anal- und Genitalregion, des Nacken- und Schulterfettpolsters sowie der Gelbfärbung ım Bereich des Sternaldrüsenfeldes und die Ausprägung der Nasenrückenwülste sind frühe- stens mit 6 Jahren erreicht. Gewichtsbestimmungen liegen nicht vor, doch dürften Männchen dieser Altersstufe etwa das 1Y;fache des Gewichtes 5- bis 6jähriger Weibchen haben. Bis zum Alter von 4% bis 5 Jahren werden junge Männer von erwachsenen Weibchen nicht anerkannt und dürfen mit diesen nicht unter Ernstbezug kopulieren (DE Jong, mündl. Mitt., 1985). Aggressive Verhaltensweisen adulter Weibchen gegenüber heran- wachsenden, neu in die Gruppe integrierten Männchen können über mehrere Jahre hinweg Fortpflanzung und Verhalten des Drill im Zoo Hannover 273 normales Kopulationsverhalten der Männchen verhindern, selbst wenn diese nach einiger Zeit - inzwischen herangereift — sozial akzeptiert und zur Paarung aufgefordert werden. Dabei kann es zu irreversiblen Verhaltensstörungen kommen, derart, daß ein männlicher Drill nur noch masturbiert (DE Jong, mündl. Mitt., 1985). Masturbierende Männchen beider Mandrillus-Arten könnten sich jedoch in anderen Fällen auch regelmäßig mit Weibchen verpaaren. Außerhalb der Brunst lassen adulte Weibchen den Sexualkontakt mit ihren bis zu 3 Jahre alten Söhnen zu. Ältere Söhne durften auch außerhalb der Östrus weder ihre Mütter noch andere Weibchen ihres Elternverbandes begatten. Emigrationstendenzen heranwachsender Drills Ein heranwachsendes Weibchen (Su) wurde mit etwa 2% Jahren in zeitlichem Zusammen- hang mit ersten vollständigen Brustschwellungen sozial sehr unausgeglichen. Aggressionen gegenüber anderen adulten Weibchen, meist in Koalition mit der eigenen Mutter (S), führten zu erheblichen Beißereien und einer lang anhaltenden Labilität des Gruppengefü- ges. A war stets bemüht, Beißereien unter seinen Weibchen zu verhindern, wobei er sehr genau Su als auslösendes Moment unruhiger Phasen erkannte und sie an die Peripherie des Geheges trieb. Nach vielen vergeblichen Aufforderungen gelang es dem in der Brunst befindlichen jungen Weibchen, den aggressiv gestimmten A zur Paarung zu bewegen. Typische freundliche Verhaltensweisen der Kontaktnahme zweier Sexualpartner wurden dabeı nicht beobachtet. Unmittelbar nach einer Kopulation wurde Su von A erneut in eine Gehegeecke getrieben. Zwischen Vater und Tochter konnte nur eine Kopulation beobach- tet werden, die allerdings am Ende des zweiten Trächtigkeitsmonats lag, wenn man den Geburtstermin berücksichtigt. Während der Trächtigkeit sowie in den ersten 912 Monaten nach der Geburt verhielt sich Su bis zum Eintritt einer erneuten Brunst in der Gruppe friedlich. Dann wurde sie wiederholt aggressiv gegenüber T. Nachdem sie sich in der Nacht nach dem ersten Tag einer maximalen Genitalschwellung eine ca. 25 cm lange, weitklaffende Hautwunde ım Bereich des Rückens zugezogen hatte, die ihr vermutlich nur von dem mit sehr langen Incisivi ausgestatteten A hatte beigebracht werden können, wurde es von der Gruppe isoliert und einem anderen erwachsenen Männchen (Po) nach Verheilung der genähten Wunde zugesellt. Mit diesem Männchen verpaarte es sich bei der ersten Gelegenheit während einer Östrusphase, nachdem beide Partner mehrere Wochen lang am Sichtgitter Möglichkeiten der sozialen Kontaktnahme hatten. S verpaarte sich im Alter von ca. 4 Jahren mit A 5 Monate nach dessen Ankunft in Hannover. Vier heranwachsende Männchen (J, To, R, Ku) zeigten erstmals im Alter von einem Jahr Hüteverhalten gegenüber den Weibchen sowie aggressives Verhalten gegenüber dem Vater A, zunächst im Spiel, später jedoch mit Ernstbezug. Sie trieben die Weibchen und störten die Ruhephasen des Vaters mit zunehmendem Alter immer häufiger. Beißereien der Weibchen untereinander um Rückzugsplätze oder Vorlaufsrechte als Folge dieses Hüte- verhaltens sowie Droh- und Angriffsversuche seitens A gegen den jeweils ältesten Sohn führten zu einer Phase sozialer Instabilität. To wurde ım Alter von 21 Jahren erheblich durch A verletzt (Durchtrennung der Achillessehne durch Beißschütteln). Es konnte nach 10wöchiger Behandlung in der Isolation geheilt an einen anderen Zoo abgegeben werden. Die anderen jungen Männchen (J, R, Ku) wurden früher, im Alter von 13, 23 bzw. 28 Monaten, aus der Gruppe genommen, sobald Tendenzen zur Labilität des Sozialgefüges bemerkbar wurden. 274 M. Böer Saugverhalten und Laktation Bereits wenige Stunden nach der Geburt saugen junge Drills mit 100 bis 120 kräftigen Pumpzügen in der Minute Milch. Säugephasen sind von Ruhepausen, in denen die Neugeborenen schlafend und über die Nase atmend die mütterlichen Zitzen mit dem Mund fest umschließen, nur dann deutlich abzugrenzen, wenn am gut sichtbaren Gesichts- profil des Kindes Saugbewegungen erkennbar sind. Nur selten hielten sich Mütter mit ihren Neugeborenen an für den hinter der Glas- scheibe stehenden Beobachter günstigen Standorten im Gehege auf, so daß Saugakte sichtbar wurden. Selbst wenn Jungtiere bereits älter als 10 Monate waren, wandten sich die Mütter während des Säugens ihrer Kinder häufig vom Besucherraum ab oder bevorzugten dabei die schwer einsehbare hintere obere Plattform. Bei fünf jungen Drills konnten daher zwischen dem ersten und 16. Lebensmonat nur 385 Saugvorgänge zeitlich registriert und für eine Auswertung herangezogen werden. Während sich der Mund des Kindes in den ersten 48 Lebensstunden kaum von den mütterlichen Zitzen löst, werden diese bereits in der zweiten Hälfte der ersten Lebenswo- che überwiegend nur noch zum Trinken oder während des Schlafens mit dem Mund gefaßt (Abb.2). Die Kinder nuckeln dann auch ım Fell der Mutter oder lutschen am Daumen, wenn sie nıcht hungrig sind. Beim Säugen wechseln die Kinder alternierend bis zu 8mal beide Zitzen, die einzelnen Saug- akte verkürzen sıch ständig. Man kann annehmen, daß die kleinen Drills so lange trinken, bis die Zitzenzisternen leer sind. Der Kurvenverlauf (Abb. 3) weist eine Depression im 3. und 4. Lebensmonat auf. Vermutlich erreichen die Milchdrüsen wäh- rend dieser Zeit ıhr Produktions- maximum, d.h. einen so hohen Milchfluß, daß die Kinder in re- latıv kurzer Zeit zur benötigten Milchmenge gelangen. Zwischen dem 8. und 9. Le- bensmonat werden die Saugzei- ten deutlich kürzer. Zu diesem Zeitpunkt sind die letzten unpig- mentierten Bereiche der Ge- sichtshaut verschwunden. Die Kinder sind vollständig schwarz im Gesicht. Sofern sie in diesem Alter die Mutter verlieren, beste- Abb. 2. Schlafendes Jungtier, die Zitze der Mutter mitdem hen bereits recht gute Über- Mund festhaltend lebenschancen bei fester Nah- rung. Dennoch saugen die Kin- der in der Regel bis zum Ende des 16. Lebensmonates, dann noch für ca. 30 sec/h, bevor sie etwa 3 bis 4 Monate vor Geburt des nächsten Geschwisters von der Mutter beim Saugversuch mit den Händen abgeschlagen oder sanft abgebissen werden. Fortpflanzung und Verhalten des Drill im Zoo Hannover 275 1000 800 600 500 400 300 200 150 1oo 50 Abb. 3. Länge der Zitzenkontakte während des Milchtrinkens (s/h) im Verlaufe der ersten 16 Lebensmonate bei jungen Drills (n = 5). Abszisse: Alter in Monaten; Ordinate: Zeitdauer des Saugens in Sekunden/Stunde Die Milchversorgung zwischen dem 9. und 16. Lebensmonat scheint allerdings noch wichtig, um eine normale Entwicklung des Kindes zu sichern. Die mit knapp 9 Monaten verwaiste P lief im Alter von 11 Monaten mit eingeknickten Hinterläufen sowie leicht aufgekrümmtem Rücken durchs Gehege. Milchbrei, angereichert mit Vitaminen, Spuren- elementen und Kalzıum, führte innerhalb von 4 Wochen zur Besserung ihres Zustandes. Entwicklung einiger morphologischer Merkmale und Verhaltensweisen bis zum Alter von 18 Monaten Das Haarkleid der Neugeborenen ist cremeweiß gefärbt, die unpigmentierten Hautpartien sind unbehaart und blafrosarot, vom Scheitel bis zur Schwanzspitze erstreckt sıch im Bereich der Medianen dorsal ein etwa 0,5 cm breiter dunkelgrauer Fellstreifen. Sitzschwie- len und Schwanzunterseite des fest am Bauch der Mutter klammernden Neugeborenen leuchten als deutlich sichtbares rosarotes ‚„„I‘“ zwischen den Hinterextremitäten der sıch fortbewegenden Mutter auf. Schon ım Alter von 4 Tagen können die Kinder neben der Mutter sitzen, am Daumen lutschen und zeigen bei Annäherung eines adulten Tieres Abwehrgesicht und Abwehr- schreien. Nach 12 Tagen lauten sie bereits unbeholtfen umher, die Bewegungen erscheinen unkoordiniert und taumelnd, die Extremitäten erfüllen noch nicht vollständig die Stütz- funktion. Sie entfernen sich bis zu 1,50 m für wenige Minuten von der Mutter. In diesem Alter werden erstmals Abwehrbeißen und -schlagen gegen die Mütter, die ıhre Kinder zurückholen wollen, beobachtet. Intensive manuelle, optische sowie olfaktorische Unter- suchungen der Futtermittel werden durchgeführt. Das Haarkleid erscheint jetzt hellgrau, der Stirnschopf sowie die übrige Kopfbehaarung dunkelgrau, die ebenso gefärbte Rücken- linie ist bereits 1 cm breit. Mit 21% Wochen fressen junge Drills Obst und laufen koordiniert, alle Objekte ım Umkreis von 2m um die Mutter herum werden intensiv kontrolliert. Im Alter von einem Monat können kleine Drills schon zum Mitlaufen von ihren Müttern aufgefordert werden. Kleine quadrupede Sprünge sind bereits möglich. Nach sechs Wochen beginnt die Pigmentierung im Bereich der Schläten, die sich zunächst dunkelgrau-blau färben. Die dorsale Medianlinie über der Wirbelsäule weist auf einer Breite von 3 cm die gleiche dunkelgraue Färbung auf, wie auch die Außentlächen der 276 M. Böer Ober- und Unterarme und -beine. Bei männlichen Jungtieren färbt sich die Hautregion cranial und caudal des Ohres deutlicher graublau als bei weiblichen Kindern. Sämtliche Futtermittel werden gefressen, Wasser oder Tee werden getrunken. Die erste Schwarzfär- bung des Gesichtes zeigt sich cranıal der Ohren bei etwa 7 Wochen alten Säuglingen. Drills dieser Altersstufe können bereits einen halben Meter weit springen sowie biped laufen. Richtung und Dauer eines Austluges werden selbständig gewählt, die Mütter folgen ihren Kindern. Im Alter von 2 Monaten versuchen junge Drills vorsichtig und unbeholfen Bäume zu erklettern. Die Ausflüge ohne Mutter erreichen mitunter schon 15 bis 30 Minuten Länge. Das Fell dunkelt jetzt ein. Mit 10 Wochen können die Drills bereits selbständig und schnell an Ästen klettern und erfahren auch durch ältere Jungtiere keine Hilfestellung, während des Spieles unter den Jungtieren wird auf Kinder dieser Altersstufe bei Verhaltsweisen des Drängelns und Rempelns, des Spielbeißens und -schlagens sowie dem spieleri- schen Anspringen keine Rück- sicht mehr genommen. Mit 312 Monaten bewegen sich die Jungtiere ohne Kontrolle der Mütter in allen Ebenen des Geheges, erregen jedoch die Auf- merksamkeit ihrer Mütter, wenn sie an freischwingenden Seilen hangeln. Während der Fütterung halten sie engen Kontakt zu ihren Müttern. Die schwarze Gesichts- färbung erstreckt sich jetzt auf Schläfen, Unterlider sowie Ober- lippe. In diesem Alter beginnen junge Drills auch mit der Auf- nahme rohfaserhaltiger Futterbe- standteile wie z.B. Rinde und Blätter frisch geschnittener Zwei- ge (Abb. 4). Mit 4) Monaten zeigt sich im Gesicht die Schwarzfärbung in Brillenform um die Augen herum. Mit 6 Monaten sind junge Drills vollständig bis auf die noch hellen Hautflecke im Bereich der Backenknochen umgefärbt, die Abb. 4. Jungtier, die Hände seiner mit frisch geschnittenen endlich nach 8 Monaten einge- Asten beschäftigten Mutter beobachtend dunkelt erscheinen (Abb.5). Die Jungtiere suchen ıhre Mütter jetzt lediglich bei Gefahr, zum Säugen oder zum Schlafen auf, andere Aktivitäten werden selbständig durchgeführt. Auch die Nahrungsaufnahme erfolgt nicht mehr in unmittelba- rer Umgebung der Mütter. Galten bis zum Alter von etwa 7 Monaten die Mütter als die jeweiligen Hauptbezugs- partner sozialer Aktivitäten, widmen junge Drills in der folgenden Zeit anderen Gruppen- mitgliedern mehr und mehr ihre Aufmerksamkeit. Sind sie älter als 8 Monate, zeigen sıe soziale Verhaltensweisen der Unterwerfung (Präsentieren) sowie der freundlichen Kon- Fortpflanzung und Verhalten des Drill im Zoo Hannover 277. A valtihh AH Sl S \\ \ N j Br) ns SI x IT III N N S N RN Sun N \ \ Abb. 5. Ontogenetische Entwicklung der Gesichtsfärbung junger Drills, schematisiert. a: 1.-6. Woche; b: 7.-13. Woche; c: 14.-17. Woche; d: 18.-20. Woche; e: 21.-23. Woche; f: ab 24. Woche taktnahme (z.B. allogrooming), insbesondere gegen ranghohe erwachsene Tiere der Gruppe. Mit 91% bis 10 Monaten springen junge Drills aus dem Stand 3 m weit. Sie sind in der Lage, blitzschnell 8 bis 10 m weite Kombinationssprünge auszuführen, die sich aus zwei bis vier unmittelbar hintereinander folgenden Einzelsprüngen zusammensetzen. Senk- rechte Wände und selbst labile Schwingelemente dienen dabei als Lande- und Absprung- stellen zugleich. Diese Beweglichkeit ermöglicht ihnen jederzeit den 10- bis 20mal schwe- reren adulten Tieren, die sich auf dem Boden deutlich langsamer fortbewegen, auszuwei- chen. Provokative spielerische Aggressionen junger Drills veranlassen die bedrängten Adulten gelegentlich zu Gegenangriffen, die zu Körperverletzungen führen können, wenn den Jungtieren nicht rechtzeitig die Flucht in die Regionen über dem Boden gelingt. Mit 11 bis 12 Monaten werden bei männlichen Drills erstmalig sexuelle Verhaltenswei- sen (Aufreiten, Friktionsbewegungen) gegenüber erwachsenen, gleichaltrigen und auch jüngeren Weibchen beobachtet. 16 Monate alte Drills führen ein von ihren Müttern sozial unabhängiges Leben in der Gruppe und sind in die jeweilige Rangordnung ihres Geschlechts integriert. Noch mit 18 Monaten klammerte sich ein männliches Jungtier bei Unruhe an seine Mutter. Körperkon- takt während der nächtlichen Ruhephase zeigen jedoch auch 3 bis 31% Jahre alte Tiere gegenüber Individuen derselben Mutterlinie. Sobald ein Drillweibchen erneut ein Neuge- borenes führt, wird das sich nähernde ältere Jungtier abgebissen und lediglich nachts bei Körperkontakt geduldet. 278 M. Böer Diskussion Jouventin (1975) und MeLren etal. (1981) beschreiben für den Mandrill, GARTLAn (1970) für den Drill das olfaktorische Markieren von Baumstämmen mit der Sternaldrüse. Dasselbe Verhalten konnte ich mehrmals bei einem adulten männlichen Drill (Po) wenige Minuten nach dessem erstmaligen Betreten der Innenanlage im Osnabrücker Zoo beobach- ten. Ich sah dieses Verhalten nıe bei Alexander. Vermutlich‘ wird es nur in selten frequentierten Gebieten in der Wildbahn gezeigt, erleichtert das Zusammenfinden ver- streuter Gruppenmitglieder oder dient der Orientierung in unbekanntem Gelände. Unsere Weibchen forderten das Männchen regelmäßig aktiv und spontan durch Geni- talpräsentieren zur Paarung auf. Sie unterscheiden sich in diesem Funktionskreis nicht so deutlich von weiblichen Tieren anderer Pavianspezies, wie es MELLEN etal. (1981) an ihren im Östrus befindlichen Mandrill-Weibchen, die sich recht passiv vom Männchen treiben ließen, beobachten. Stereotype Aufforderungen zum „Allogrooming“, wie sie für Mandrills beschrieben sind (MELLEN et al. 1981), wurden in Hannover nicht beobachtet. Anstarren mit hochgezogenen Augenbrauen, wie es van HooFr (1967) für die Gattung Papio beschreibt, MELLEN et al. (1981) jedoch nie bei Mandrills beobachteten, wurde gelegentlich als Drohform unter den Weibchen gezeigt. Otten bleibt, ob bei der beobachteten vorn-niedrig-hinten-hoch-Stellung des jungen Männchens gegenüber einem weiblichen Jungtier eine Phase der ontogenetischen Reifung eines schon stammesgeschichtlich vollzogenen Ritualisationsprozesses dieser Verhaltens- weise deutlich wurde. Vielleicht hat sich aus der Spielaufforderung junger Drills ein Verhalten mit neuer soziosexueller Signalfunktion gegenüber intraspezifischer Konkurrenz bei adulten Männchen entwickelt. Auch aus dem Funktionskreis der Nahrungsaufnahme, z.B. aus dem Fressen vom Boden mit vorn tiefgebeugtem Kopf und Vorderkörper, könnte sich diese Verhaltensweise entwickelt haben. Die Drills treffen ihre Futterauswahl jedoch meist nach Greifen eines Nahrungsbrockens mit den Händen und anschließender kurzer oraler Kontrolle. Unsere Beobachtungen deuten darauf hin, daß Mandhrillus leucophaeus eine geringfügig längere Tragzeit als Mandhrillus sphinx aufweist. Die ermittelte Tragzeit für Drills von 179 bis 183 Tagen ist länger als die von Hapıpıan und BERNSTEIN (1979) mit 173 Tagen, von CARMAN (1979) mit 172 bis 176 Tagen für den Mandrill angegebenen Werte, sie liegt jedoch im Rahmen der für die Gattung Papio angegebenen Zeitspanne für Tragzeiten von 154 bis 183 Tagen (ZUCKERMAN und PARkESs 1932). Demnach scheinen Drills die von allen pavianverwandten Arten längste Tragzeit zu haben. Erstaunlich sind zweifellos große Unterschiede in der Geschwindigkeit, mit der sich einige Merkmale und Verhaltensweisen im Verlauf der Ontogenese beı beiden Arten der Backenfurchenpaviane entwickeln: Nach Carman (1979) verlassen kleine Mandhrills ihre Mütter zum erstenmal etwa 2 Wochen später als es nach unseren Beobachtungen junge Drills tun. Die erste Futteraufnahme wurde bei Mandtrills im Alter von 35 Tagen (CARMAN 1979), bei unseren Drills jedoch bereits mit 17 bis 19 Tagen gezeigt. Den Zeitpunkt der Gesichtsumfärbung gibt derselbe Autor mit 73 bis 77 Tagen an, während die Drillkinder in Hannover frühestens mit 210 Tagen vollständig umgefärbt sind. Einer Entwöhnung kleiner Mandrills im Alter von 8 Monaten steht eine solche beim Drill im Alter von 15 bis 16 Monaten gegenüber. Im Alter von 7% bis 8 Monaten scheint trotz nicht beendeter Laktation eine soziale und ernährungsphysiologische Selbständigkeit der Drill-Jungtiere weitestgehend erreicht zu sein. In Portland gelang die vollständige Integration eines handaufgezogenen Mandrills dieser Altersstufe in die Gruppe (LITTLEwooD und SMITH 1979). CaRMAN (1979) sah erste Genitalschwellungen bei jungen Mandrill-Weibchen ım Alter von 3% Jahren, in Hannover zeigte ein junges Drill-Weibchen solche Erscheinungen Fortpflanzung und Verhalten des Drill im Zoo Hannover 279 bereits ein Jahr früher, ein weiteres, aus dem elterlichen Verband genommenes Weibchen hatte bereits mit 24 Monaten erste Östruserscheinungen. HapıDıan und BErnNSTEIN (1979) sahen erste zyklische Erscheinungen bei einem Drill/Mandrill-Hybriden im Alter von 29 Monaten. Eine weitergehende Interpretation einander abweichender Befunde dieser beiden Arten setzt umfangreicheres Datenmaterial voraus. Heranwachsende weibliche Jungtiere verlassen spätestens zum Zeitpunkt des Eintritts der Geschlechtsreife, männliche Individuen in der Regel viele Monate früher, den elterli- chen Verband. Aggressive Verhaltensweisen des Vaters drängen die Jungtiere beider Geschlechter an die Peripherie des Aktionsraumes der Gruppe. Ziel der Aktivitäten des adulten Männchens ist zweifellos die durch heranwachsende Jungtiere herbeigeführte Labilität im Sozialgefüge zu verhindern, die in folgenden Situationen entsteht: 1. An Intensität und Häufigkeit zunehmende Imponierversuche, Hüteverhalten (GART- an 1970) und Kopulationsversuche subadulter Männchen gegenüber adulten Weib- chen führen zu großer Unruhe in der Gruppe. Die Aufzucht der Säuglinge kann mitunter empfindlich gestört werden. 2. Drohverhalten und ungehemmte aggressive Übergriffe subadulter Weibchen mit sozia- ler Expansionstendenz gegenüber laktierenden Müttern oder deren gerade entwöhnten Kindern stellt die ungestörte Reproduktion der Gruppe in Frage. Diese Veränderungen des Sozialgefüges und der damit verbundene Stabilitätsverlust in der Gruppe entwickelt sich langsam im Verlaufe von Wochen und Monaten, so daß der aufmerksame Beobachter in der Regel rechtzeitig eingreifen kann, bevor es zu Verletzun- gen kommt. Weitere Auslöser für aggressives Verhalten des Männchens dürften das bei jungen Männern im Alter von ca. 5 Jahren entstehende schwarze Circumanalfeld inmitten der rotblauen Gesäßregion, die dem Vatertier beim Präsentieren der Jungen auffallen muß, sowie die ersten vollständigen Genitalschwellungen junger, lange vertrauter Weibchen sein. Die der Fortpflanzungsbiologie des Drill entsprechenden zuchttechnischen Folgerun- gen können nachstehende Maßnahmen umfassen: 1. Eine zweite, schwingende oder z. T. labile Ebene ın Form einer Konstruktion mitein- ander verbundener Seile und Äste ermöglicht auch in einem kleinen Zoogehege von nur 20 bis 40 qm Grundfläche rangniedrigen Drills oder säugenden Müttern, vor einem nur auf dem Boden schnell beweglichen adulten Männchen jederzeit auszuweichen. So können Individualdistanzen zwischen den Gruppenmitgliedern auch unter begrenzten räumlichen Verhältnissen aufrecht erhalten werden, womit eine wichtige Grundlage zur ungestörten Kinderaufzucht gegeben ist. 2. Weibliche Drills sollten mit etwa 3 Jahren, spätestens jedoch zu Beginn des ersten vollständigen Östrus, junge Männer etwa mit 1 bis 3 Jahren bei Auftreten instabilisie- renden Hüteverhaltens aus der Gruppe genommen werden. 3. Junge Weibchen sınd bereits im Alter von etwa 2 Jahren zu einer sozialen Bindung an fremde adulte Drillmänner fähig, sie können demnach unmittelbar nach Herausnahme aus der Gruppe der Eltern an den zukünftigen Sexualpartner gewöhnt werden. 4. 3- bis 4jährige Männchen sollte man nach Möglichkeit nicht unbedingt mit adulten, sozial potenten Weibchen, die deren Sexualentwicklung hemmen können, vergesell- schaften. 5. Jungtiere sollten frühestens mit 13 bis 15 Monaten von ıhren Müttern getrennt werden. Provozierte Emigrationstendenzen heranwachsender Männer scheinen dem sozialen Mechanismus zugrundezuliegen, der vermutlich in der Wildbahn zu reduzierten Anzahlen männlicher Tiere in den Gruppen (GARTLAN 1970) führt und intrasexuelle Selektion ın Richtung auf sexuellen Dimorphismus bei Mandrillus begünstigen mag. 280 M. Böer Danksagungen Herr Prof. Dr. HELMUT WILkEns, Anatomisches Institut der Tierärztlichen Hochschule Hannover, sowie insbesondere sein wissenschaftlicher Zeichner, Herr E. THEEL, ermöglichten die graphische Darstellung der Drillköpfe. Für Schreibarbeiten und Korrekturen danke ich Frau URsULA REICHELT und Herrn Dr. GÜNTER MERZ. Die Herren RoLanp WoLF und ANDREAS PRELLWITZ sowie ihre Mitarbeiter waren für die Pflege der Tiere verantwortlich. Zusammenfassung Haltungsbedingungen, Fütterungsschema, Freßgewohnheiten sowie einige bemerkenswerte Verhal- tensweisen einer Zuchtgruppe von Drills werden beschrieben. Enge soziale Bindungen fanden sich zwischen verwandten Tieren, d.h. innerhalb der Mutterli- nien. Ein verwaister, noch milchabhängiger junger Drill wurde durch ein fremdes Weibchen, das laktierte, vollständig adoptiert. Drei Geburten ertolgten jeweils 179, 182 bzw. 183 Tage nach beobachteter Kopulation. Während der 3 bis 7 Tage anhaltenden Zyklusphase maximaler Genitalschwellung kopulierten Weibchen 2- bis 8mal täglich mit dem gruppenführenden Männchen innerhalb von 48 bis 72 Stunden. Sexualzyklen varıierten in ihrer Länge zwischen 25 und 42 Tagen. Vier Monate post partum wurden erstmalıg bei säugenden Müttern Menstruationsblutungen entdeckt. Mindestens weitere 4 Monate lang konzipier- ten die Weibchen trotz regelmäßiger Brunstschwellungen und Kopulationen während der Laktation nicht. Bei einer durchschhittlichen Zwischentragezeit von 19 Monaten und einer mittleren Laktations- dauer von 16 Monaten gaben die Weibchen in der Regel nur während der ersten Hälfte der Tragzeit Milch. Ein junges Weibchen wurde mit 312 Jahren geschlechtsreit. Die Jugendentwicklung mehrerer Individuen wurde verfolgt. Zwischen dem 7. und 8. Lebensmo- nat verlieren junge Drills die letzten unpigmentierten Bereiche im Gesicht und weisen die schwarze Maske der Juvenilen und Adulten auf. Nur bis zu diesem Zeitpunkt stellt Muttermilch den Hauptteil der Nahrung dar. Mit zunehmender Umstellung auf feste Nahrung gehen die Verselbständigung von der Mutter und die soziale Integration in die Hierarchie der Erwachsenen einher, die mit 16 Monaten abgeschlossen sind. Imponierversuche, Hüteverhalten und Kopulationsversuche jungadulter Männchen sowie unge- hemmte aggressive Übergriffe junger Weibchen mit sozialer Expansionstendenz führen als tiergarten- biologisch bedeutsame Prozesse zur Labilität im Sozialgefüge. Die Herausnahme junger Männchen und Weibchen aus dem Elternverband sowie deren Vergesellschaftung mit neuen Partnern substitu- ieren als zuchttechnische Maßnahmen unter Zoobedingungen die mögliche Emigration dieser Alters- klasse aus einem freilebenden Sozialverband, verhindern hemmungslose Beißereien mit dem gruppen- führenden Mann und stabilisieren die parentale Zuchtgruppe. Literatur BöErR, M. (1983): Several Examinations on the Reproductive Status of Lowland Gorillas (Gorilla g. gorilla) at Hannover Zoo. Zoo Biology 2, 267-280. CARMANn, M. (1979): The gestation and rearing periods of the mandrill (Mandrillus sphinx) at London Zoo. Int. Zoo Yearb. (London) 19, 159-160. DrrrricH, L. (1984): Jahresbericht des Zoologischen Gartens der Landeshauptstadt Hannover für das Jahr 1984. Hemmingen: Sponholtz-Druckerei. GARTLAN, J. S. (1970): Preliminary Notes on the Ecology and Behavior of the Drill, Mandhrillus leucophaeus Ritgen, 1824. In: Old World Monkeys. Ed. by J. R. and P. H. Narıer. New York: Academic Press. pp. 445480. GARTLAN, ]J. S.; STRUHSAKER, T. T. (1972): Polyspecific associations and niche separation of rain- forest anthropoids in Cameroon, West Africa. J. Zool. 168, 221-266. HaDpipIan, J.; BERNSTEIN, J. S. (1979): Female Reproductive Cycles and Birth Data from an Old World Monkey Colony. Primates 20, 429-442. HoofF, J. A. R. A. M. van (1967): The Facial Display of the Catarrhine Monkeys and Apes. In: Primate Fthology. Ed. by D. Morrıs. London: Weıdenfeld and Nicolson. pp. 7-68. Jouventin, P. (1975): Observations sur la Socio-ecologie du Mandhrill. Terre et la Vie (Paris) 29, 493-532. LITTLEwWoOD, A.; SMITH, ]J. (1979): Breeding and handrearing mandrills at Portland Zoo. Int. Zoo Yearb. 19, 161-165. MELLEn, J. D.; LiTTLEwooD, A. P.; Barrow, B. C.; STEvEns, V. J. (1981): Individual and Social Behavior in a Captive Troop of Mandrills. Primates 22, 206-220. Rörı, M. (1985): Untersuchungen zum Sozialverhalten von Mandrill- (Papio sphinx) und Drill- Fortpflanzung und Verhalten des Drill im Zoo Hannover 281 (Papio leucophaeus)Gruppen in den Zoologischen Gärten Berlin, Hannover und Hamburg. Dipl.- Arbeit, FB Biologie, Freie Univ. Berlin. VOELKER, J.; Sachs, R. (1977): Affen als natürliche und experimentelle Endwirte afrikanischer Lungenegel (Paragonimus africanus, P. uterobilateralis). Tropenmed. Parasit. 28, 137-144. VOELKER, J. (1985): Schriftl. Mitt. an den Verfasser vom 11. 12. 1985. ZUCKERMAN, $.; PARKES, A. S. (1932): The menstrual cycle of primates. Proc. Zool. Soc. Lond. 1932, 139-191. Anschrift des Verfassers: Dr. MıcHAEL BÖöER, Zoologischer Garten der Landeshauptstadt Hannover, Adenauerallee 3, D-3000 Hannover 1, FRG A comparative cytogenetic study on the mitotic and meiotic chromosomes in hamster species of the genus Phodopus (Rodentia, Cricetinae) By T. Haar, H. Weıs, and M. ScHMID Department of Human Genetics, University of Würzburg, FRG Receipt of Ms. 9. 9. 1986 Abstract The mitotie and meiotic chromosomes of the two hamster species Phodopus sungorus and P. roborovsku were examined with the aid of various banding techniques. P. sungorus has a diploid chromosome number 2n = 28. P. roborovskii ıs 2n = 34. Almost complete homoeologies can be demonstrated for all euchromatic segments. The differences in the karyotypes of the two species can be traced to eight centric fusions, three pericentric inversions, one telomeric fusion, as well as to changes in the content and arrangement of the constitutive heterochromatıin. With a high degree of probability, the karyotypes of both species have developed from a common ancestor with 2n = 40 chromosomes. Both species have remarkably large, heteromorphic sex chromosomes. In male meiosis, the euchromatic arm of the X chromosome not paired with the Y chromosome exhibits highly delayed condensation throughout prophase. Introduction The genus Phodopus (Rodentia, Cricetinae) comprises two species of Asıatic hamsters. The habıtat of P. sungorus extends from Ischim (Ural) to the west throughout the steppes of western Sıbiria up into the regions of Mongolia, Djungarıa and Altaı to the southeast. P. roborovskii inhabits the northern and northeastern part of the Gobi desert, as well as the provinces Shansı, Nan-Shang and Zaidan (ArGYRoPuLo 1933). The karyotype of the Djungarıan hamster (P. sungorus) ıs well characterized with varıous banding techniques (THust 1974; PoGosIanz 1975; BIGGER and SavAGeE 1976; Das and SAavAGE 1978; GAMPERL et al. 1978). P. sungorus ıs being successfully used for studies on cancer, mutagenesis and chromosomal nondisjunction (PoGosIanz 1975; HANSMANN 1984). The reasons for the prominence of the Djungarıan hamster in so many areas of research are that these anımals are hardy and easy to keep and that they reproduce rapıdly with quickly developing offspring. In contrast to this, no detailed cytogenetic analyses have been made with P. roborovsku. The present study demonstrates the homoeologies found between the chromosomal banding patterns of P. sungorus and P. roborovskii, and the most probable common U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5205-0281 $ 02.50/0 Z. Säugetierkunde 52 (1987) 281-290 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 282 T. Haaf, H. Weis and M. Schmid ancestral karyotype of these two species is proposed. The possible mechanisms involved in the chromosome phylogeny of P. sungorus and P. roborovskü are discussed. Material and methods Two adult male and female specimens each of P. sungorus sungorus and P. roborovskü were used for this study. Mitotic chromosome preparations were obtained from fibroblast cultures initiated from biopsies of lung and peritoneum according to the usual techniques (WoLr 1974). Male meiotic chromosomes were prepared according to the method of MEREDITH (1969). For conventional karyotyping the preparations were stained with Giemsa dye (5min in a 5% Giemsa solution, pH 6.9) after preidentifying the chromosomes by Q-banding (CAspErsson et al. 1970). Chromosomal R-banding patterns were obtained using the technique of SEHESTED (1974). The chromosomes were numbered ın accordance with the G-banded standard karyotype of P. sungorus established by Pocosıanz (1975). Staining of the constitutive heterochromatin (C-banding) was done according to the method of SuMNER (1972), and silver (Ag) staining of the nucleolus organizer regions (NORSs) according to the method of GooDPASTURE and BLoom (1975). Preidentification of the chromosomes with quinacrine mustard (Q-banding) was necessary because not all chromosomes could be identified with certainty in Giemsa-stained, C-banded or silver-stained preparations. The fluorescence microscopy was performed using a Zeiss fluorescence microscope equipped with an 50W/AC lamp and a filter combination 450-490/FT 510/LP 520. Length measurements of the chromosomes were made on photographs of metaphases. The film negatives were projected at a final magnification of 30,000x and the chromosome lengths were determined with a map measurer. The calculation of the nondisjunction (ND) frequency was made exactly as described by LonG (1978). Results Both conventional staining with Giemsa dye as well as fluorescent banding patterns with quinacrıne mustard of P. sungorus and P. roborovskü are demonstrated in figure 1. The karyotype of P. sungorus (Fig. la, lc) ıs composed of 2n = 28 chromosomes with a fundamental number (FN) of 51 chromosome arms. With the exception of the acrocentric chromosomes 12, 13 and Y, all chromosomes are submetacentric to metacentric. No secondary constriction could be detected in any chromosome. P. roborovskıi (Fig. 1b, 1d) has a diploıd chromosome number of 2n = 34, the fundamental number is 59. Autosome pairs 12, 14 to 16 as well as the Y chromosome are acrocentric, all others are submetacen- tric to metacentric. Chromosomes 12 and 14 possess secondary constrictions located close to the centromeres. After stainıng with the AT-specific fluorescent dye quinacrine mustard (Fig. Ic, 1d), the chromosomes exhibit distinct bandıng patterns (Q-bands) enabling the unequivocal identification of all chromosomes of the two hamster species. No brilliant quinacrine fluorescence characteristic for AT-rich heterochromatic regions could be demonstrated. The C-banded karyotypes of P. roborovskü and P. sungorus, preidentified by their Q- bands, are depicted in figure 2a and 2b respectively. In both species, constitutive hetero- chromatın ıs localızed in the centromeric regions of most autosomes. The autosomes 1, 2 and 4 of P. roborovskü and the autosomes 1, 5 and 12 of P. sungorus are characterized by unusually sparse centromeric C-bands. Interstitial C-bands are evident in the autosomes 3, 5 and 7 of P. roborovsküi and in the autosomes 1, 2, 4, 6 and 7 of P. sungorus. As the comparison of R-banding patterns (Fig. 4) shows, the interstitial heterochromatic seg- ments in the autosomes 3 and 7 of P. roborovskü are homoeologous to the interstitial C- bands in the autosomes 2 and 6 of P. sungorus. The Y chromosome and one arm of the X chromosome are completely heterochromatic in both species. Large amounts of constitu- tive heterochromatin are localized in the pericentromeric regions of autosomes 11 and 13, as well as to both sides of the secondary constrictions ın autosomes 12 and 14 of P. roborovskü (Fig. 2a). The X chromosome of P. sungorus comprises 9.9 % of the haploıd female karyotype, whereas in P. roborovsküi it is only 6.7 %. Chromosomes of Phodopus 283 The nucleolus organızer regions (NORs) of the two hamster species can be specifically demonstrated by using the ammoniacal silver staining (Fig. 3). All metaphases were prestained with quinacrine mustard to identity the NOR-carrying chromosomes. In P. roborovskü, the NORSs are concentrated in the long arms of chromosomes 12 and 14 (Fig. 3a), which exhibit distinct secondary constrictions in conventionally stained preparations RS 1 Q NN = Ar % RR patterns obtained after staining with quinacrine mustard \ IN ® N AMD IR N in TTS S Fig. 1. Karyotypes of male P. sungorns (left) and P. roborovskü (right). (a, b) Conventional staining with Giemsa dye. (c, d) Fluorescent banding 284 T. Haaf, H. Weis and M. Schmid T: _ N Fıg. 2. C-banded karyotypes of P. roborovskü (a) and P. sungorus (b). Note that the Y chromosome and one arm of the X chromosome are completely heterochromatic in both species. All chromosomes were preidentified by Q-banding (Fig. 1b). The NORs of P. sungorus are localized on the short arm telomeres of chromosomes 5, 7, 12 and 13 (Fig. 3b). Secondary constrictions were not observed in this species (Fig. 1a). This is in agreement with the results of BIGGER and Savace (1976). With the exception of the NORs in chromosome 12 of both hamster species, there is no homoeology between the chromosomal sites of nucleolus organizer regions. One can hypothesize, that the common ancestor of P. roborovskü and P. sungorus had NORs not only in terminal positions, but also in interstitial positions. In the karyotypic evolution of Phodopus, the interstitial NORs could have been shifted to terminal positions by small pericentric inversions. On the other hand, terminal NORs could have been lost due to Chromosomes of Phodopus 285 Fıg. 3. Silver (Ag)-stained karyotypes of P. roborovsküi (a) and P. sungorus (b). The arrows indicate the silver precipitation at the nucleolus organizer regions (NORSs). All chromosomes were preidentified by their Q-bands deficiences of NOR-associated heterochromatin or translocated onto other telomeric regions. Figure 4 compares the R-banding patterns of P. sungorus (S) with those of P. roborovskii (R). The presumed chromosome homoeologies are demonstrated in the table. Because P. sungorus has lower number of chromosomes than P. roborovsküi, the chromo- some complement of this species is used as a basıs for the alignment of homoeologous chromosome arms and segments. Sıx chromosomes of P. sungorus and P. roborovskü show a completely identical banding pattern: S3 = R2, S7 = R8, S8 = R9, S9 = R10, S13 = R16 and SY = RY. Four chromosomes differ only in their amount of constitutive hetero- chromatin: S10 = R11, S11 = R13, S12 = Ri12 and SX = RX. There is always more 286 T. Haaf, H. Weis and M. Schmid ! a ‚»== Er I z { Sy, > SI se (R) and P. sungorus (S). The chromosomes are arranged to align homoeologous d between two black dashes are homoeologous. The double arrows in a NV | j ! ! chromosomes R4, R5 and R6 indicate the sites between which an inversion occurred e8 R N N = 2 3 | SIR Ss RERR N SD Fig. 4. A comparison of R-banding patterns between P. roborovsku chromosome segments. All chromosome arms or segments situate pericentromeric heterochromatin localized in the autosomes of P. roborovsku, whereas the heterochromatin in the short arm of SX is distinctly larger than in RX (Fig. 2). The biarmed chromosomes S2, S4, S5, S6 and R1, R3, R7 arose by a series of seven independent centric fusions, which occurred in acrocentric chromosomes of the ancestral karyotype. Chromosomes of Phodopus 287 The biarmed chromosomes R4, R5, R6 de- Table veloped by pericentric inversions in acrocen- tric chromosomes. A comparison of the R- banding patterns of Phodopus indicates a very complex derivation for the chromosome S1. Two originally acrocentric chromosomes P. sungorns P. roborowskii (corresponding to R6 and R14) initially formed a single biarmed chromosome. A subsequent telomeric fusion involving the short arm of this chromosome (R14) and the short arm of a further acrocentric chromosome (correspond- ing to Rip) then gave rise to $1. Interestingly enough, a large interstitial C-band is still iden- tifiable in the long arm of chromosome $1 at exactly the site where the centric fusion should have occurred. The structure and behaviour of the sex chromosomes of Phodopus during male meiosis is demonstrated in figure 5. Using the @-bandıner technique, the X 7 and Y chromosomes can be very distinctly localized in the various stages of meiotic prophase and diakinesis, due to their exceptionally large heterochromatin content (Fig. 2). The strongly p = short arm; q = long arm; inv = stained sex vesicle, which consists of the con- pericentric inversion densed X and Y chromosomes, appears during late pachytene (Fig. 5a) and unfolds again at diplotene stage. In both hamster species, the sex vesicle consists of a dark (heterochroma- tic) and a light (euchromatıic) region: the light region, which corresponds to the long arm of the X chromosome, consists of a rolled-up thread, which has the appearance of a taıl in some of the later pachytene stages. During meiotic prophase and diakinesis, the pairing heterochromatic arms of the X and Y are highly condensed, whereas the non-pairing euchromatic arm of the X chromosome exhibits a dramatic delay of chromatin condensa- tion. Usually, the unpaired arm of the X chromosome appears to be two to three times longer than the arm pairing with the Y chromosome. At diakinesis, the sex chromosomes have regained their normal morphology and exhibit a characteristic end-to-end association (Fig. 5b-e). The XY bivalent can be distinguished from the rod- to ring-like autosomal bivalents by its heteropycnotic appearance. C-banding enabled us to interpret the exact arrangement of the X and Y chromosomes in the bivalent configuration: the hetero- chromatic arm of the X chromosome is terminally associated with the distal part of the long arm of the Y chromosome. The end-to-end arrangement of the X and Y at diakinesis is interpreted to be the result of a partial synapsis followed by a single crossing over and subsequent terminalization (SOLARI 1974). The large autosomes usually exhibit two to three chiasmata, whereas the medium-sized ones have only one or two. An analysıs of 300 meiotic cells in diakinesis and metaphase II showed a nondisjunction (ND) frequency of 2.2% for the species P. sungorus. This value is approximately within the expected range (Long 1978). Chromosomal homoeologies between P. sungorus and P. roborovskii Discussion The karyotypic comparison between P. sungorus and P. roborovsküi confirms that Robert- sonian fusions and pericentric inversions are the most common mechanisms of structural 288 T. Haaf, H. Weis and M. Schmid Fig. 5. C-banded meiotic chromosomes of P. sungorus (a-d) and P. roborovsku (e). (a) Late pachytene stage with prominent sex vesicle (SV). (b) Diakinesis. (c) Autosomal bivalents and XY bivalent of one diakinetic plate. (d, e) Cut-outs of sex bivalents from selected diakineses of P. sungorus (d) and P. roborovs- kü (e). The heterochromatic Y chromosome always as- sociates end-to-end with the terminal segment of the heterochromatic arm of the X chromosome. Note the lower degree of chromatin conden- sation in the unpaired arm of the X chromosome chromosome changes in the course of karyotypic evolution (CApAnnA et al. 1976; FREDGA 1977; VIEGAS-PEQUIGNOT et al. 1983). The chromosome arms present ın the karyotype of the common ancestor of Phodopus have been conserved as units in all chromosomes. The results of thıs study indicate that the karyotypes of P. sungorus and P. roborovskü evolved from an ancestral karyotype with a diploid complement of 2n = 40 comprising 7 metacentric or submetacentric chromosomes (including the X) and 14 acrocentrics (includ- ing the Y). After divergence of the two species, more of the ancestral chromosomes were conserved in the P. roborovskii line than in the P. sungorus line. The karyotype of P. sungorus is more complex than that of P. roborovskü. In the course of evolution, Robertsonian fusions seem to have caused a reduction of the chromosome number in both species. The different numbers of chromosomes in P. roborovskül (2n = 34) and P. sungorus (2n = 28) can be explained by the fact that some chromosome arms of P. sungorus correspond to discrete chromosomes in P. roborovskü. The long arm of chromosomes 1, 4 and 5 of P. sungorus is formed by the chromosomes 6 + 14, 4 and 5 of P. roborovskii. The long arm of chromosome $1 additionally contains a small chromosome of P. roborovsku (R14). The short arm of chromosome 5 of P. sungorus corresponds to chromosome 15 of P. roborovskü. In addition to the structural chromosome rearrangements, marked differences were found in the amount of constitutive heterochromatin as well as in the localization of nucleolus organizer regions between P. sungorus and P. roborovskü. However, this fact does not interfere with the results obtained with R-banding patterns and the presumed Chromosomes of Phodopus 289 chromosome homoebologies. Changes in the position and quantity of constitutively hetero- chromatic regions are well documented among closely related species in varıous mamma- lian genera (PATHAk et al. 1973; PATTON and SHERwOooD 1982; RoBınson et al. 1983). C- band heterochromatin is composed to a large extent of highly repetitive DNA sequences (satellite DNAs) which are not transcribed into RNA for protein synthesis. Because heterochromatin varıations do not produce an obvious phenotypic effect, it is not surprising that these highly repetitive DNAs can evolve very rapidiy Cum and Yas- MINEH 1971). The chromosomal distribution of nucleolus organizer regions is also known to differ significantly both within and among species (SCHWARZACHER and WACHTLER 1983). In several hamster species of the family Cricetidae, the sex chromosomes are of exceptional size. The X chromosome comprises about 10% of the genome and the Y ıs always remarkably longer than the smallest autosomes. The X chromosome of P. sungorus makes up 9.9% of the haploid female karyotype, the X chromosome of P. roborovskü makes up 6.7 %. For comparison, the X chromosome of Mesocricetus auratus constitutes 10.2% and that of M. newtoni 7.4% of the haploid genome (VoıcuLescu 1974). In general, the mammalıan X chromosome comprises only approximately 5% of the haploid genome and the Y chromosome is one of the smallest chromosomes of the complement (OHno 1967). In the evolution of Mammalıa, the euchromatic part of the X chromosome has become highly conserved and many identical X-linked genes are mapped on the X chromosomes in varıous mammalıan species. When the sıze of the mammalıan X chromo- some exceeds 5% of the haploıd complement, the additional chromatın ıs usually genetically inert constitutive heterochromatın (PATHAK 1983). Therefore, the large X and Y chromosomes of the Cricetidae may be caused by the addition of C-band heterochroma- tin to the original types of sex chromosomes during karyotypic evolution (PATHAK and Stock 1974). The additional heterochromatin of the X and Y chromosomes leads to a characteristic behaviour of the sex bivalent. This study presented a detailed comparison of the mitotic and meiotic chromosomes of two closely related hamster species. Karyotypic analysıs permitted the reconstruction of the various structural chromosome rearrangements that have occurred during speciation of Phodopus. The results demonstrate that chromosomal evolution in closely related species can proceed in a very conservative manner, keeping entire chromosome arms as units. Acknowledgements This study was supported by the Universitätsbund Würzburg. We thank Prof. Dr. I. Hausmann, University of Göttingen, for providing some of the hamsters used in this study. Zusammenfassung Vergleichende zytogenetische Untersuchungen an den Hamstern Phodopus sungorus und P. roborovskii (Rodentia, Cricetinae) Die mitotischen und meiotischen Chromosomen der beiden Hamster-Spezies Phodopus sungorus und P. roborovskü wurden mit Hilfe verschiedener Bänderungsmethoden untersucht. P. sungorus weist eine diploide Chromosomenzahl 2n = 28 auf, P. roborovsküi 2n = 34. Die vergleichende zytogeneti- sche Untersuchung zeigt sehr deutlich, daß die Karyotypen-Evolution bei nahe verwandten Spezies ın einer konservierenden Art und Weise abläuft, wobei die Chromosomenarme des ancestralen Karyo- typs als komplette Einheiten erhalten bleiben. Für alle euchromatischen Segmente können fast vollständige Homoeologien nachgewiesen werden. Die Unterschiede in den Karyotypen von P. sungorus und P. roborovsku können auf acht unabhängige zentrische Fusionen, drei perizentrische Inversionen, eine fragliche Telomer-Fusion, sowie auf Veränderungen in Gehalt und Anordnung des konstitutiven Heterochromatins zurückgeführt werden. Die Karyotypen beider Spezies haben sich mit einer hohen Wahrscheinlichkeit von einem gemeinsamen Vorfahren mit 2n = 40 Chromosomen entwickelt. Dabei ist der ancestrale Zustand der Chromosomen bei P. roborovsku eher erhalten geblieben als bei P. sungorus. Auffallend sind in beiden Spezies die großen, heteromorphen Geschlechtschromosomen. Der 290 T. Haaf, H. Weis and M. Schmid kurze Arm des X-Chromosoms und nahezu das gesamte Y-Chromosom bestehen aus konstitutivem Heterochromatin. In der männlichen Meiose zeigt der nicht mit dem Y-Chromosom gepaarte euchromatische Arm des X-Chromosoms eine stark verzögerte Kondensierung während der gesamten Prophase. References ARGYROPULO, A. J. (1933): Die Gattungen und Arten der Hamster (Cricetinae, Murray, 1866) der Paläarktık. Z. Säugetierkunde 8, 129-149. BiGGER, T. R. L.; Savage, J. R. K. (1976). Location of nucleolar organizer regions on the chromosomes of the Syrian hamster (Mesocricetus auratus) and the Djungarian hamster (Phodopus sungorus). Cytogenet. Cell Genet. 16, 495-504. CAPANNA, E.; GROPP, A.; WINKInG, H.; Noack, G.; CivITELLı, M.-V. (1976): Robertsonian metacentrics in the mouse. Chromosoma 58, 341-353. CASPERSSON, 1.; ZECH, L.; JOHAnsson, C.; MoDesT, E. J. (1970): Identification of human chromosomes by DNA-binding fluorescent agents. Chromosoma 30, 215-227. Das, R. K.; Savage, J. R. K. (1978): Chromosome replication patterns in the Djungarian hamster (Phodopus sungorus). Chromosoma 67, 165-176. FREDGA, K. (1977): Chromosomal changes in vertebrate evolution. Proc. R. Soc. Lond. B. 199, 377-397. GAMPERL, R.; VISTORIN, G.; ROSENKRANZ, W. (1978): Comparison of chromosome banding patterns in five members of Cricetinae with comments on possible relationship. Caryologia (Firenze) 31, 343-353. GOODPASTURE, C.; BLOoMm, S. E. (1975): Visualization of nucleolus organizer regions in mammalıan chromosomes using silver staining. Chromosoma 53, 37-50. HansMANN, 1. (1984): Aspects of nondisjunction. In: Mutations in Man. Ed. by G. Obe. Berlin, Heidelberg, New York: Springer-Verlag. 147-155. Long, $. E. (1978): Chiasma counts and non-disjunction frequencies in a normal ram and in rams carrying the Massey I Robertsonian translocation. J. Reprod. Fert. 53, 353-356. MEREDITH, R. (1969): A simple method for preparing meiotic chromosomes from mammalıan testis. Chromosoma 26, 254-258. Onno, S$. (1967): Sex-chromosomes and sex-linked genes. Berlin, Heidelberg, New York: Springer. PATHAR, $.; Hsu, T. C.; ARRIGHI, F. E. (1973): Chromosomes of Peromyscus (Rodentia, Cricetidae). IV. The role of heterochromatin in karyotypic evolution. Cytogenet. Cell Genet. 12, 315-326. PATHAR, $.; Stock, A. D. (1974): The X chromosomes of mammals: karyological homology as revealed by banding techniques. Genetics 78, 703-714. PATHAR, $. (1983): The behaviour of X chromosomes during mitosis and meiosis. In: Cytogenetics of the Mammalian X Chromosome, Part A: Basic Mechanisms of X Chromosome Behaviour. Ed. by A. A. Sandberg. New York: Alan R. Liss. 67-106. PATTOoN, J. L.; SHERWOOD, W. (1982): Genome evolution in pocket gophers (genus Thomomys). 1. Heterochromatin variation and speciation potential. Chromosoma 85, 149-162. Pocosıanz, H. E. (1975): Djungarıan hamster — a suitable tool for cancer research and cytogenetic studies. J. Natl. Canc. Inst. 54, 659-664. Rogınson, T. ]J.; ELDER, F. F. B.; CHapman, J. A. (1983): Evolution of chromosomal variation in cottontails, genus Sylvzlagus (Mammalıa: Lagomorpha): $. aguaticus, S. floridanus, and S. trans- itionalis. Cytogenet. Cell Genet. 35, 216-222. SCHWARZACHER, H. G.; WACHTLER, F. (1983): Nucleolus organizer regions and nucleoli. Hum. Genet. 63, 89-99. SEHESTED, J. (1974): A simple method for R banding of human chromosomes, showing a pH- dependent connection between R and G bands. Hum. Genet. 21, 55-58. SOLARI, A. ]J. (1974): The behaviour of the XY pair in mammals. Int. Rev. Cytol. 38, 273-317. SUMNER, A. T. (1972): A simple technique for demonstrating centromeric heterochromatin. Expl. Cell Res. 75, 304-306. Trust, R. (1974): G-banding and late replication of the Djungarian dwarf hamster chromosomes. Expl. Path. 9, 153-156. VIEGAS-PEQUIGNOT, E.; DUTRILLAUxX, B.; PROD’HoMME, M.; PETTER, F. (1983): Chromosomal phylogeny of Muridae: a study of 10 genera. Cytogenet. Cell Genet. 35, 269-278. VoıcuLgscu, I. (1974): A comparative study of the chromosome banding patterns of Mesocricetus newtoni and Mesocricetus auratus. Z. Säugetierkunde 39, 211-219. WOoLr, U. (1974): Cell culture from tissue explants. In: Methods in human cytogenetics. Ed. by H.G. Schwarzacher and U. Wolf. Berlin, Heidelberg, New York: Springer-Verlag. 39-58. Yunıs, J. J.; YAsMInEH, W. G. (1971): Heterochromatıin, satellite DNA, and cell function. Science 174, 1200-1209. Authors’ address: T. Haar, H. Weıs and M. ScHhMID, Department of Human Genetics, University of Würzburg, Koellikerstraße 2, D-8700 Würzburg, FRG Alter und Linsengewicht von Meriones shawi grandis (Cabrera, 1907) Von R. RÖDEL Eingang des Ms. 29.9.1986 Abstract Age and eye-lens weight in Meriones shawi grandis (Cabrera, 1907) Studied was the relationship between eye-lens weight and age in 103 Meriones of known age. The correlation, r=0.9937, proved to be very close. With y=a + b Inx (eye-lens weight in mg=y and days=x) the regression line was calculated: y=-68,88 + 28,16 Inx Accordingly the age of 1045 (wild) Meriones — caught outdoors — was determined; the oldest male trapped turned out to be 620 and the oldest female 590 days old. Einleitung Von 1979 bis 1981 wurden in Marokko Untersuchungen zur Populationsdynamik von Meriones shawi grandis (Cabrera, 1907) vorgenommen (RÖDer 1982). Hierfür wurde nach zuverlässigen Methoden zur Altersbestimmung gesucht. In Betracht kam unter anderem die Altersschätzung nach dem Gewicht der Augenlinsen. Ihre Zuverlässigkeit hing davon ab, ob ausreichend enge Zusammenhänge zwischen Alter und Linsengewicht bestehen. An gezüchteten Rennmäusen der Art Meriones shawı, deren Alter bekannt war, wurde dieser Zusammenhang untersucht. Material und Methode Im Winter 1979/80 wurden in der weiteren Umgebung von Marrakesch mit Lebendfallen M. shawı gefangen und ım Laboratorıum gehalten. Von 103 Tieren der Tochter- und Enkelgenerationen, die in Gefangenschaft geboren waren, deren Geburtsdatum also bekannt war, konnten ab Januar 1980 die Linsen gewogen werden: das jüngste Tier war 21, das älteste 752 Tage alt. Nicht berücksichtigt wurden Tiere, die durch Giftversuche, Unfälle (beim Füttern und Säubern), Krankheiten, Kannibalısmus und natürlichen Tod gestorben waren. Die Linsen wurden unmittelbar nach der Präparation feucht gewogen (FRIEnD 1967). Protokolliert wurde das Mittel beider Linsen. Es wurde darauf geachtet, daß die Zeit vom Herausschneiden der Augen bis zum Ablesen der Gewichte nie länger als sechs Minuten dauerte. Erkennbar beschädigte Linsen wurden nicht gewogen. Ergebnisse Die Abbildung zeigt die Linsengewichte von 60 Weibchen und 43 Männchen, wobei das Alter in Tagen auf der Abszisse logarıthmisch dargestellt ist. Die Berechnung der Regressionsgeraden (KrEyszıG 1979) hierzu nach y=a + b Inx ergab (bei n= 103): y=-68,88 + 28,16 Inx U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5205-0291 $ 02.50/0 Z. Säugetierkunde 52 (1987) 291-294 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 292 R. Rödel 100 >0 Tage 10 50 100 500 1000 Alter und Linsengewichte von Meriones shawi aus Marokko bei im Labor aufgezogenen Tieren der Tochter- und Enkelgeneration ım Freiland gefangener Tiere. Das Alter ist logarithmisch abgetragen. Die empirischen Werte (Punkte) weichen nur wenig von ihrer Regressionsgeraden ab, die zusätzlich eingetragen wurde Die graphische Darstellung zeigt eine gute Übereinstimmung von errechneten und gemes- senen Werten. Die hohe Korrelation ergibt sich aus: r=0,9937 (Korrelationskoeffizient) r?= 0,9875 (Bestimmtheitsmaß) Um eventuelle geschlechtsspezifische Unterschiede zu berücksichtigen, wurden die Regressionsgeraden für Männchen und Weibchen getrennt gerechnet. Die Gleichung für die Weibchen lautet: (bei n=60): y=-68,31 + 28,06 Inx (r=0,9946) Die Gleichung für die Männchen lautet (bei n=43): y=-70,29 + 28,44 Inx (r=0,9918) Die Gleichungen unterscheiden sich so wenig, daß dies für die Praxis ohne Bedeutung ist. Somit konnte mit einem hohen Maß an Zuverlässigkeit die gemeinsame Regressionsge- rade zur Schätzung des Alters verwendet werden. Danach wurde für 1045 im Freiland gefangene Meriones ein Höchstalter bei Männchen von ca. 620 und bei Weibchen von ca. 590 Tagen ermittelt. Diskussion Unter den Veränderungen eines Wirbeltieres im Laufe seiner Lebenszeit gibt es eine Reihe von Merkmalen, die der Altersschätzung dienen können. Hierzu zählen Form, Tracht, Größe, Gewicht, Maße bestimmter Körperteile, Verknöcherungen und die Veränderungen an den Zähnen (Durchbruch, Wechsel, Ablagerungen von Dentin und Zement und Abnutzung). Alter und Linsengewicht von Meriones shawi grandıs 295 Ein enger Zusammenhang besteht offenbar bei Wirbeltieren zwischen dem Alter und dem Linsengewicht. So stellte bereits P. SMITH 1883 (zit. bei Hockwın et al. 1963) fest, daß die Linsen unterschiedlich alter Menschen verschiedene Absolutgewichte aufweisen und um so schwerer sind, je älter der betreffende Mensch ist. Das wurde seither für einige Wirbeltiere nachgewiesen (FRIEnD 1967; Hockwin et al. 1971). Am bekanntesten gewor- den ist die Untersuchung von LorD (1959) über das amerikanische Kaninchen Sylvilagus floridanus. Die Linsengewichtsmethode wird seit etwa 20 Jahren immer mehr anerkannt und verhilft bei sorgfältiger Anwendung im Vergleich zu herkömmlichen Methoden zu sehr präzisen Aussagen. Sie muß jedoch für jede Tierart neu überprüft werden und ist nicht immer anwendbar (BERRY und TRUsLOVE 1968). Linsentrockengewichte hat schon LEsnE (1979) bei M. shawi untersucht. Von ca. 30 Tieren, deren Geburtsdatum bekannt war, wurden die Augäpfel herausoperiert, in Formalın (10%) fixiert und nach Frankreich geschickt (LEsne, mündl. Mitt.). Nach 20 Tagen Konservierungszeit wurden die Linsen bei der INRA (Institut de la Recherche Agronomique), Jouy-en Josas bzw. Colmar herauspräpariert getrocknet und gewogen. Für das Alter in Tagen (x) und das Linsengewicht in mg (y) wurde eine Regressionsglei- chung nach y=bInx + a errechnet. Dabei konnte für b=32,10, a=-87,84 und für r=0,98 ermittelt werden (LEsnE 1979). Da mit unterschiedlichen Methoden gearbeitet wurde, lassen sich die beiden Ergebnisse nicht ohne weiteres miteinander vergleichen. Die Korrelation ist mit r=0,98 ähnlich gut. Der unbestreitbare und immer wieder hervorgehobene Vorteil der Trockengewichtsme- thode liegt einmal darin, daß nur die Augen, nicht aber die Tiere ins Labor transportiert werden müssen. Zum anderen stellt die Trockengewichtsmethode geringere Anforderun- gen an den Sezierenden, da er nur den Augapfel aus der Augenhöhle entfernen muß. Die Entnahme der konservierten Linse ıst dann nicht weiter schwierig; die Linse kann dabei kaum beschädigt werden. Diese Methode hat jedoch den Nachteil, daß langwierig konserviert und getrocknet werden muß, und sie mehr technischen Aufwand erfordert. Außerdem bergen die Trock- nungs- und Konservierungsprozesse zusätzliche Fehlerquellen (Pucek und Lowe 1975). Die zusätzlichen Arbeitsgänge erhöhen die Zahl der Verwechslungsmöglichkeiten. Weiter- hin liegen die Ergebnisse erst frühestens nach drei Wochen vor. Manche halten die Trockengewichtsmethode für genauer (FRIEND 1967). Die Unterschiede sind ohne Bedeu- tung für die Praxis. Zur Überprüfung meiner Methode habe ich in Marrakesch wiederholt das Alter einzelner Labortiere, deren wahres Alter ich zunächst nicht kannte, nach dem Linsenge- wicht geschätzt. Bis zum Alter von 100 Tagen war die Ditferenz zum wahren Alter in der Regel nicht größer als 2 Tage. In den höheren Altersklassen streuten die Ergebnisse mehr (101 bis 200 Tage, maxımal 4 Tage). Ein Hinweis auf die Zuverlässigkeit auch bei freilebenden Tieren ergibt sich aus der Analyse der Monatsfänge von 1979 bis 1981. Die Reproduktionszeiten der Meriones weisen deutliche Spitzen im Winterhalbjahr und Täler im Sommerhalbjahr bzw. Herbst auf. Wenn die Linsengewichtsmethode zuverlässig ist, müßten sich diese Tendenzen auch in den Alterspyramiden und Säulendiagrammen nachweisen lassen. Dies ıst der Fall (RöDeı 1982). Somit kann festgestellt werden, daß die ermittelte Kurve ein zuverlässiger Maßstab zur Altersbestimmung - selbst für hohe Altersklassen - ıst. Zusammenfassung Die Beziehung zwischen Alter und Linsengewicht wurde bei 103 Meriones shawi bekannten Alters untersucht. Die Korrelation war mit r=0,9937 sehr hoch. 294 R. Rödel Die Berechnung der Regressionsgeraden nach y=a + b Inx ergab (Linsengewicht in mg=y und Tage=x): 7 y=-68,8802 + 28,16 Inx Hiernach wurde das Alter von 1045 ım Freiland gefangenen M. shawi bestimmt. Dabei war das ermittelte Höchstalter bei Männchen 620 und bei Weibchen 590 Tage. Literatur BERRY, R. J.; TrusLovE, G. M. (1968): Age and eye lens weight in the house mouse. J. Zool., London, 155, 247-252. CABRERA, A. (1932): Los Mamitferos de Marruecos. Trab. Mus. Nac. Cienc. Nat. (Zool.), Madrid, 57, 1-360. FRIEND, M. (1967): Some obvservations regarding eyelens weight as a criterion of age in anımals. New York Fish Game ]. 14, 91-121. Hockwin, O.; SCHMUTTER, J.; MÜLLER, H. K. (1963): Untersuchungen über Gewicht und Volumen verschieden alter Rinderlinsen. Graefes Arch. Ophthalmol. 166, 136-151. Hockwin, O.; BECHTEL-EHRING, U.; LICHT, W.; Rast, F. (1971): Concerning the estimation of the age of guinea pigs, rabbits and chickens by means of determining their lens weight. Ophthalmol. Res. 2, 77-85. KreyszıG, E. (1979): Statistische Methoden und ihre Anwendungen. Göttingen: Vandenhoeck und Ruprecht. LeEsne, L. (1979): La Merione de Shaw du Maroc. Compte rendu de la reunion tenue ä Marrakech le 10 Avril 1979 sur la M£rıone de Shaw. 7.-Annexe 1, 14. LorD, J. R., Jr. (1959): The lens as an indicator of age in cottontail rabbits. J. Wildlife Manage. 23, 358-360. PuceEr, Z.; Lowe, P. W. (1975): 3. Age criteria in small mammals. In: Small mammals: their productivity and population dynamics. Ed. by GoLLEY, F. B.; PETRUSEwICZ, K.; Ryszkowsk1, L. Cambridge: Cambridge Univ. Press. Röper, R. (1982): Untersuchungen zur Populationsdynamik von Meriones shawi grandis (Cabrera, 1907), Rodentia, Cricetidae ın Marokko. Diss. Univ. Bonn. Anschrift des Verfassers: Dr. ROGER Röder, GTZ-Office c/o Embassy of the Federal Republic of Germany, 8, Hassan Sabry St., Zamalek, Kairo, Agypten Verhaltensunterschiede zwischen Wild- und Hausmeerschweinchen Von ADELHEID STAHNKE Lehrstuhl für Verhaltensphysiologie an der Fakultät für Biologie der Universität Bielefeld, Arbeitsgruppe Säugetiersozialsysteme Eingang des Ms. 9. 9. 1986 Abstract Behavioral comparison of wild and domestic cavıes Compared the behavior of captive wild cavies (Cavia aperea) and domestic guinea pigs (Cavıa aperea _ f. porcellus). Animals were kept in small groups together with individuals of only their own species, and their total behavior was observed. The wild cavies stayed more attentive to disturbances, which was most obviously indicated by the vocalization pattern “chirp””, which they used very frequently and in a more sophisticated manner than the domestic guinea pigs. The wild cavies showed more agonıstic behavior patterns towards conspecifics than domestic guinea pigs, and they did not have as many amicable social relationships. The vocalization pattern “purr”, which was very prominant in the alpha-males of the domestic guinea pig, was seldomly heard in the wild cavies. This behavior pattern is supposed to function as an important stabilizing pattern used mainly by the alpha-male. Obviously U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5205-0294 $ 02.50/0 Z. Säugetierkunde 52 (1987) 294-307 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Verhaltensunterschiede zwischen Wild- und Hausmeerschweinchen 295 the described differences in behavioral organization are connected with different capabilities of both forms to adjust to the captivity situation. Einleitung Nach neueren Untersuchungen begann die Domestikation des Meerschweinchens vor mindestens 3000, vielleicht vor über 6000 Jahren im südamerikanischen Andengebiet. Dies wird durch Funde aus dem heutigen Peru belegt (MAacNEISsH 1977; HERRE und RÖHRS 1977). Die Abstammung vom Wildmeerschweinchen, Cavia aperea, ist von Hückınc- Haus (1961) zweitelsfrei belegt. HÜCKINGHAUS weist zudem besondere anatomische Ähnlichkeiten mit der Wildmeerschweinchenform, die im heutigen Peru vorkommt, nach. Noch heute werden in Peru, Bolivien und Ecuador Hausmeerschweinchen als Fleischliefe- ranten in vermutlich ähnlicher Weise wie seit einigen tausend Jahren entweder freilaufend oder in Gruben in großer Dichte gehalten (WEIR 1974; HENDRICHS mdl.). Die ersten Hausmeerschweinchen kamen ım 16. Jahrhundert nach Europa. Heute ist das Hausmeerschweinchen eines der wichtigsten Labortiere (WAGNER und MANNING 1976). Es ist ethologisch gründlich untersucht (Kıng 1956; KunkeEL und Kunker 1964; Roop 1972; Young 1969), um nur einige der ersten wichtigen Arbeiten zu nennen (s. auch die Bibliographien bei WEIR 1972; HARPER 1976; STAHNKE 1983). Über die Ethologie von Wildmeerschweinchen ist wenig bekannt. Die einzige ausführ- liche vergleichende Studie wurde von RooD (1972) durchgeführt, der auch einige wenige Verhaltensdaten aus dem Freiland sammeln konnte. WEIR und Roop (1970) untersuchten das Fortpflanzungsverhalten der Wildmeerschweinchen und BEAUCHAMmP et al. (1979) die olfaktorische Kommunikation. Die hier vorgelegte Studie soll die Kenntnisse über das Verhalten von Wildmeer- schweinchen vertiefen. Es wurden dabeı vor allem Unterschiede gegenüber dem Verhalten von Hausmeerschweinchen herausgearbeitet. Da es sich um Gehegeuntersuchungen han- delt, müssen mögliche Auswirkungen der Haltung auf das beobachtete Verhalten beson- ders berücksichtigt werden. Material und Methode Herkunft der Tiere Wildmeerschweinchen, Cavıa aperea: Die ersten Tiere erhielten wir 1982 aus den USA. Es handelte sich um Nachzuchten 1974 in Argentinien in der Nähe von Buenos Aires gefangener Tiere. Die schwersten Tiere wiegen etwa 700g. Die Tiere sind dunkelgrau-braun. Hausmeerschweinchen, Cavia aperea f. porcellus: Die Hausmeerschweinchen stammen aus der seit 1975 bestehenden Zucht der Verhaltensforschung der Universität Bielefeld. Die Tiere sind kurzhaarıig und ein- bis mehrfarbig. Nichtträchtige Weibchen wiegen etwa 800g, die schwersten Männchen über 10008. Haltung Gehege: Wild- und Hausmeerschweinchen wurden in getrennten Gruppen in gleich aufgebauten Gehegen untergebracht. Innengehege waren 2-3 m? groß, Außengehege 15-25 m?. Die Gehege waren mit Asten, Kisten oder Steinplatten eingerichtet. Die Außengehege konnten im Winter gegen die Witterung geschützt werden. Fütterung: Die Tiere wurden täglich zur gleichen Zeit gefüttert. Sie erhielten frisches Wasser, beı Bedarf mit Vitaminzusätzen, und Heu, Gras, Laub, Apfel, Mohrrüben, Futterrüben sowie Mais, Haferflocken oder kommerzielles Preßfutter. Aufbau der Gruppen: Die Tiere wurden über Monate oder Jahre in kleinen Gruppen zusammengehal- ten. Die meisten Gruppen in den Innengehegen setzten sich aus einem adulten Männchen und 2—4 geschlechtsreifen Weibchen zusammen. Ein Teil der Jungtiere wurde mit der Geschlechtsreife herausgenommen. Zudem gab es Gruppen, die nur aus Männchen bzw. Weibchen bestanden. Die Gruppen in den Außengehegen wuchsen auf etwa 20 Tiere an. 296 Adelheid Stahnke In einem Zeitraum von 3 Jahren standen ca. 150 individuell bekannte Wildmeerschweinchen zur Verfügung. Während 8 Jahren wurden einige hundert Hausmeerschweinchen beobachtet. Datenerhebung Täglich wurden alle Gruppen kontrolliert und Besonderheiten notiert. Zu einer festgesetzten Tages- zeit wurden Verhaltensprotokolle erstellt. Während 1-2 Stunden wurden alle auftretenden Verhal- tensweisen aller Tiere einer Gruppe fortlaufend in Kürzeln notiert. Dazu wurde die Zeit im Minutenraster aufgenommen. In den Protokollen sind Aufenthaltsort der Tiere, die räumliche Ausrichtung zueinander, die zeitliche Abfolge einzelner Verhaltensweisen, der Ablauf sozialer Interaktionen sowie Reaktionen auf das Geschehen außerhalb des Geheges festgehalten. Zum Ethogramm: Die Bezeichnungen von KunkEL und Kunker (1964) und von Roop (1972) für die einzelnen Verhaltensweisen wurden weitgehend übernommen, jedoch wurde das Ethogramm stellen- weise erweitert und differenziert (ausführliche Angaben für das Hausmeerschweinchen in STAHNKE 1983). Ergebnisse Zum äußeren Erscheinungsbild der Wildmeerschweinchen Die Wildmeerschweinchen wiegen fast % weniger als die Hausmeerschweinchen. Sie sind schlanker und graziler gebaut, haben dünnere Beine und kleinere Füße. Der Körper wirkt langgestreckt, besonders im Lendenbereich, Trächtigkeit ist schwerer sichtbar. Der Kopf ist schmal und langgestreckt, die Schnauze länger und gerader, die Ohren sind kleiner und stehen nach hinten. Die Bewegungen der Wildmeerschweinchen sind ım Vergleich zu den Hausmeer- schweinchen rascher, heftiger und auch kraftvoller. Die Tiere wirken beweglicher. Sie heben sich häufig an den Wänden oder freistehend auf die Hinterbeine, beides zeigen Hausmeerschweinchen selten. Die Wildmeerschweinchen springen leicht 50-60 cm hoch, Hausmeerschweinchen höchstens 30-40 cm. Bei den Wildmeerschweinchen gibt es häufiger Anzeichen hoher Erregung als bei den Hausmeerschweinchen. Besonders ranghohe Tiere laufen oft - ohne erkennbaren Anlaß - mit gesträubtem Fell herum. Beide Formen vokalisieren viel, jedoch werden die einzelnen Laute teilweise unterschiedlich eingesetzt, und die Klangfarbe unterscheidet sich bei einigen Lauten. Das soziale Verhalten Das Purren Eine häufige Vokalisation der Hausmeerschweinchen in sozialem Zusammenhang ist das „Purren“, ein tiefer, langgezogener, stark vibrierender Laut. Da dieser Laut besonders von adulten Männchen geäußert wird und auch in das Sexual- und Imponierverhalten gegenüber Weibchen eingebaut ist, wird er von vielen Autoren dem Sexualverhalten zugeordnet (KunkEL und KunkeEL 1964; RooD 1972; SACHSER 1983). Eigene Untersu- chungen ergaben zusätzliche weitere Funktionen, vor allem ın der männlichen alpha- Rangstellung und -Rolle (STAHNkE 1983). Verglichen wurde die Häufigkeit des Purrens von adulten Männchen in Gruppen mit einem adulten Männchen und 3 adulten nichtöstrischen Weibchen (Wildmeerschweinchen: 6 Gruppen; Hausmeerschweinchen: 36 Gruppen). Alle Hausmeerschweinchenmännchen purrten wesentlich häufiger als die Wildmeerschweinchenmännchen, im Durchschnitt über 25x/h, die Wildmeerschweinchen 1,5%/h. Bei den Hausmeerschweinchen waren Werte von über 50x/h häufig, vor allem, wenn Jungtiere da waren. Werte von unter 10x/h traten nur in unter 5% der Stunden auf. Bei den Wildmeerschweinchen trat in über 50% der Beobachtungsstunden gar kein Purren auf, die höchsten überhaupt gemessenen Werte waren 16x (ohne Jungtiere in der Gruppe) und über 20x (bei Anwesenheit kleiner Jungtiere). Abb. 1. Wildmeerschweinchen, Cavia aperea, ın Gehegehaltung Aapeeamen Abb. 2. Sonagraphische Auf- zeichnung des Purrens eines Hausmeerschweinchens 250 298 Adelheid Stahnke Anzahl Bei den Hausmeerschweinchen Bulgeimk purren alle jungen Männchen lange /Std. vor der Geschlechtsreife, viele be- 30 reits in der ersten Lebenswoche. Sie purren ın früherem Alter und häufi- ger, wenn es kein alpha-Männchen in der Gruppe gibt (STAHNKE 1983). Bei den Wildmeerschweinchen wur- 20 de Purren bei Jungtieren nicht beob- achtet. Es trat frühestens bei über 2 Monate alten Männchen auf, aber auch dann nur, wenn kein älteres E Männchen in der Gruppe lebte. Un- terlegene Männchen purrten auch mit 3 oder 4 Monaten nicht. Dies weist auf zwei Unterschiede hin: Das Purren der Wildmeerschwein- chen scheint stärker an den Zustand Haus - Wild- der Geschlechtsreife gebunden, und meersehweinchen es scheint fester an den Zustand der Anzahl sozialen Dominanz gekoppelt zu N =10 N=4 der sein. | | Tiere Das Purren unterscheidet sich auch in Lautstärke und Dauer: Bei den Hausmeerschweinchen ist der Laut gut hörbar. Er dauert etwa 1-3 Sekunden und wird mehrfach, 1-2 Minuten lang, wiederholt. Die Wildmeerschweinchen purren tiefer, sehr viel leiser und immer nur kurz, auch selten wiederholt. Das Purren steht mit einer charakteristischen Bewegungsweise in Zusammenhang: Das Tier geht betont langsam (,‚Imponiergang““), oder es schaukelt den Hinterkörper seitlich hin und her, wobei es die Hinterbeine abwechselnd hebt (,,Treteln‘“). Zugleich können die Haare aufgestellt werden. Das Tier umkreist den Käfig, oder es geht seitlich an einem Artgenossen vorbei, oder es folgt einem Artgenossen nach, oder es umkreist ein anderes Tier mit Körperkontakt, wobei es ihm an der Flanke und unter dem Kopf langsam mit der Schnauze entlangfährt. Der eigene Körper kann seitlich zum Artgenossen gekrümmt sein („Seitwärtsstellen‘“). Dazu können auch andere Elemente des Sexualverhaltens auftreten. Diese Bewegungsabläufe sind bei den Hausmeerschweinchen fast immer von Purren begleitet, bei den Wildmeerschweinchen traten sie häufig auch ohne das Purren auf. Die Ausführung der Bewegungen unterscheidet sich deutlich: Bei den Wildmeerschweinchen sind der Imponiergang und das Treteln wesentlich langsamer, fast zeitlupenartig, und das Verhalten erscheint sehr intensiv. Die Haltung ist sehr flach, die Tiere berühren mit dem Körper fast den Boden, während die Hausmeerschweinchen dabei auf gestreckten Beinen, eher höher als sonst, stehen. Abb. 3. Durchschnittliche Häufigkeit des Purrens bei Wild- und Hausmeerschweinchen Rangbeziehungen und soziale Verträglichkeit Bei der Haltung in Gruppen gibt es bei Wildmeerschweinchen eine stärkere Unverträglich- keit einzelner Individuen als beim Hausmeerschweinchen. Einander fremde Tiere zusam- menzusetzen, ıst nur in Einzelfällen möglich. Auch lange zusammengehaltene Tiere können plötzlich unverträglich werden. Dominante Individuen reagieren stärker auf Auseinandersetzungen anderer Tiere. Adulte Männchen suchten mehrfach Weibchen aut, Verhaltensunterschiede zwischen Wild- und Hausmeerschweinchen 299 die kämpften oder sich heftig androhten, schoben sich dazwischen und „balzten“ die Weibchen an oder drohten seitwärts, worauf die Weibchen auswichen. Ein adultes Weibchen (nicht das alpha-Weibchen!) in einer Gruppe ohne adultes Männchen äußerte das gleiche Verhalten - ohne Balzen — gegenüber 2 Monate alten Männchen, die miteinan- der kämpften. Bei den Hausmeerschweinchen wurde bisher ein Eingreifen in dieser Form nicht gesehen. Soziale Beziehungen zwischen Männchen Adulte männliche Hausmeerschweinchen lassen sich langfristig zusammen halten, ohne daß ernste Verletzungen auftreten, wenn keine Weibchen in der Gruppe leben. Nach anfänglichen Kämpfen bilden sich Rang-, andeutungsweise auch gegengeschlechtliche Rollenbeziehungen aus. In zweigeschlechtlichen Gruppen können unterlegene Männchen häufig gejagt und gebissen werden, jedoch waren die Verletzungen nicht lebensgefährlich (STAHNKE 1983; STAHNKE und HENDRICHSs 1986). Nach den Untersuchungen von SACHSER und HENDRICHs (1982) und SACHSER (1983) nehmen Auseinandersetzungen adulter Männ- chen in zweigeschlechtlichen Gruppen ab, wenn die Gruppengröße über eine bestimmte Anzahl erwachsener Individuen anwächst. Dann bilden sıch in der Gruppe Unterstruktu- ren, wodurch sich einige ranghohe Männchen seltener begegnen. Adulte Männchen der Wildmeerschweinchen konnten wir bisher nicht länger zusam- men halten. Heranwachsende Männchen werden mit der Geschlechtsreife plötzlich vom erwachsenen Männchen der Gruppe heftig gejagt und gebissen. 3-4 Monate alte Männchen bekämpfen sich heftig. Unterlegene Tiere sterben leicht, auch ohne größere Verletzungen. Soziale Beziehungen zwischen Weibchen Adulte weibliche Hausmeerschweinchen zeigen wenig, meist leichte Agonistik gegenein- ander. Das volle Kampfverhalten der Männchen kann auftreten, ist aber extrem selten. Auseinandersetzungen können etwas häufiger sein, wenn kein Männchen ın der Gruppe lebt oder wenn eines der Weibchen Jungtiere hat. Art und Ausmaß agonistischen Verhaltens zwischen adulten Wildmeerschweinchen- weibchen sind anders als bei den Hausmeerschweinchenweibchen. Agonistische Verhal- tensweisen sind wesentlich häufiger, mehr heftige Verhaltensweisen treten auf. Der Rang der Weibchen zeigt sich unmittelbar in ihrem Verhalten. Das ranghöchste Weibchen sucht andere Weibchen auf, droht, beißt oder jagt. Wo immer es erscheint, weichen die anderen Weibchen aus. Art und Intensität von Angriffen sind aber nicht unbedingt korreliert mit der relativen Rangstellung beider Tiere. Es scheinen zusätzlich individuelle Merkmale von Einfluß zu sein. Bei den Wildmeerschweinchen züchteten in der Regel nur einige der ranghohen Weibchen einer Gruppe. Junge Weibchen können, wie Hausmeerschweinchen, bereits im Alter von 3-4 Monaten werfen. Dies trat jedoch nur auf, wenn keine älteren Weibchen in der Gruppe lebten. Viele Weibchen waren noch mit 5-6 Monaten nicht trächtig. In den größeren Gehegen, in denen zeitweise 15-20 Weibchen lebten, züchteten regelmäßig nur 4-6 Tiere. Auch sind ältere Weibchen zusammen mit jüngeren Männchen schwer zum Züchten zu bringen. Demgegenüber warfen bei den Hausmeerschweinchen 80-90 % der Weibchen nach der Geschlechtsreife regelmäßig. Ähnliche Fortpflanzungsdaten für Haus- meerschweinchen erhielt SACHSER (mdl.). Soziale Beziehungen zwischen Männchen und Weibchen Das Zusammenhalten von Männchen und Weibchen ist bei den Hausmeerschweinchen in den meisten Fällen unproblematisch. Beliebige Individuen können zusammengewöhnt werden. Zwar kann es bleibende Unterschiede in der Qualität der Beziehungen geben 300 Adelheid Stahnke (STAHNKE 1983; STAHNKE und HEnDRIcHS 1986), jedoch führen soziale Integrations- schwierigkeiten nicht soweit, daß eine Haltung nicht möglich ist. Bei den Wildmeerschweinchen lassen sich lediglich wenige Monate alte Männchen und Weibchen problemlos zusammengewöhnen. Sehr schwer gewöhnen sich ausgewachsene Tiere aneinander. In zwei Fällen starben adulte Weibchen innerhalb einiger Stunden, nachdem ein fremdes Männchen zugesetzt worden war, ohne daß dieses sich den Weib- chen angenähert hatte. In einem anderen Fall ging das anfängliche nicht-agonistische Annähern des Männchens nach heftigem Abwehren und Ausweichen des Weibchens allmählich ın Jagen und Beißen über. Auch die in einer Gruppe ansässigen Weibchen wurden beim heftigen Jagen fremder Tiere beobachtet. In über Monate bestehenden Gruppen kann das Männchen plötzlich beginnen, eines der Weibchen zu jagen und zu beißen. Nach unseren Beobachtungen bildet sich, wenn zwischen zwei Tieren erst einmal harte Agonistik auftritt, keine verträgliche Beziehung mehr aus. Ein adultes Männchen hat zu den einzelnen Weibchen Beziehungen verschiedener Qualität. Nur gegenüber wenigen (bestimmten) Weibchen äußert es agonistisches Verhal- ten. Freundliches Verhalten, Sexualverhalten und Imponierverhalten richtet es auf die verschiedenen Weibchen nicht gleich häufig. Zwei der Männchen zeigten diese Verhaltens- weisen über Monate fast ausschließlich gegenüber dem ranghöchsten Weibchen der Gruppe. Dieses Weibchen suchten sie regelmäßig auf oder folgten ihm nach. Besonders auffällig war, daß bei fast jeder Begegnung dann der Imponiergang oder das Treteln auftraten. Im Verhalten eines alpha-Männchens gegenüber dem ranghöchsten Weibchen ist in den Imponier- bzw. Sexualverhaltensabläufen immer wieder eine leichte Agonistik erkennbar, und, wenn überhaupt, weicht schließlich das Weibchen aus, nicht das Männ- chen. Die Wildmeerschweinchenmännchen markierten die Weibchen (durch Überspringen mit Harnen bzw. Absetzen von Analdrüsensekret, ‚Flanken‘“) häufiger als die Hausmeer- schweinchenmännchen. Fast ausschließlich wurden die ranghohen Weibchen markiert. Die Männchen markierten vorwiegend unmittelbar nach einer leichten Auseinanderset- zung mit dem Weibchen oder auch anschließend an Sexual- oder Imponierverhalten. Schon für Hausmeerschweinchen wurde ein Zusammenhang zwischen dem Flanken und der Überlegenheitsdemonstration vermutet (STAHNKE 1983). Bei den Wildmeerschweinchen erscheint dies noch ausgeprägter. Ranghohe Wildmeerschweinchenweibchen haben eine stärkere soziale Position als Hausmeerschweinchenweibchen. Dies drückt sich schon in ihrem Verhalten zueinander aus, es zeigt sich deutlich ım auffälligen Verhalten des ranghöchsten Weibchens, und es ist auch am Verhalten gegenüber Männchen erkennbar. Adulte Weibchen sind, anders als beim Hausmeerschweinchen, über halbwüchsige Männ- chen dominant, das Männchen wird z.B. gejagt. Sofern ein junges Männchen in dieser Situation überhaupt überlebt, beginnt es schließlich, den Angriffen standzuhalten und nicht mehr auszuweichen. Es sucht dann gerade das ranghöchste Weibchen immer wieder auf, richtet Sexual- oder Imponierverhalten und auch leichte Agonistik gegen das Weib- chen und markiert es häufig. Soziales Verhalten adulter Tiere gegenüber Jungtieren Adulte Hausmeerschweinchenmännchen nehmen zu Jungtieren sehr häufig und intensiv Kontakt auf, sie werden zudem regelmäßig von den Jungtieren aufgesucht, und die Interaktionen verlaufen freundlich. Häufige Verhaltensweisen der Männchen sind Nach- folgen, Beriechen, Purren und der dazugehörige oben beschriebene Verhaltenskomplex. In diesen Begegnungen tritt oft Spielverhalten auf. Ganz ähnliche Abläufe wurden bei den Wildmeerschweinchen beobachtet. Auch die Wildmeerschweinchenmännchen folgen den Jungtieren, treteln, umkreisen sie langsam, purren dabei manchmal. Dieser gesamte Verhaltenskomplex ist häufiger, wenn Jungtiere ın Verhaltensunterschiede zwischen Wild- und Hausmeerschweinchen 301 der Gruppe leben, und er wird auch dann überwiegend gegen diese gerichtet. Aus unseren Beobachtungen läßt sich Roops (1972) Vermutung, die Domestikation der Meerschwein- chen hätte dazu geführt, Sexualverhalten weniger differenziert auch gegen nichtöstrische Weibchen und gegen nichtgeschlechtsreife Jungtiere zu richten, nicht bestätigen. Boxen oder Verjagen von Jungtieren durch adulte Männchen kommt bei Hausmeer- schweinchen so gut wie nicht vor. Bei den Wildmeerschweinchen werden gelegentlich einzelne Jungtiere heftig gejagt und gebissen. Die Versorgung von Jungtieren durch die Mutter verläuft bei Wild- und Hausmeer- schweinchen ähnlich und soll hier nicht beschrieben werden. Bei beiden Formen wurde einige Male das Saugen bei einem anderen Weibchen beobachtet, jedoch unterscheiden beide Formen ım Regelfall zwischen den eigenen und fremden Jungtieren und stoßen fremde Jungtiere weg. Ein Unterschied besteht im Verhalten der Weibchen gegenüber den Jungtieren anderer Weibchen. Hausmeerschweinchenweibchen haben wenig Interaktionen mit fremden Jung- tieren. Kräftigeres Boxen oder Beißen trat nur wenige Male auf, wenn ein Jungtier ein Weibchen intensiv bedrängte. Wildmeerschweinchenweibchen boxen, jagen oder beißen Jungtiere anderer Weibchen öfter, sie suchen sie auch auf und verjagen sıe dann. Aktivität Die Wach- und Schlafphasen von ' bis 1 Stunde Dauer verteilen sich bei den Wild- und Hausmeerschweinchen ähnlich, wobei tagsüber längere Schlafphasen, in den Abendstun- den längere Wachphasen auftreten, wie auch die Untersuchungen von BÜTTNER und WOLLnIk (1982) an isoliert gehaltenen Hausmeerschweinchen und von GOUBOT und BÜTTNER (in Vorb.) an isoliert gehaltenen Wildmeerschweinchen ergaben. Grundsätzlich verschieden ıst jedoch die Aktivität der adulten Tiere während der Wachphasen: Hausmeerschweinchen verbringen die meiste Zeit fressend, wobei sie „‚gra- send“ langsam herumgehen. Andere Aktivitäten, wie Herumgehen, Schnuppern z.B., dauern nur kurz, regelmäßig und über mehrere Minuten treten sie nur beim alpha- Männchen auf. Die Wildmeerschweinchen dagegen verbringen einen Teil einer Wachphase mit raschem Herumlaufen, Schnuppern, häufigem Hochrecken. Das Fressen wird immer wieder minutenlang unterbrochen. Es gibt reine Lokomotion von bis zu 20 Minuten Dauer, ohne eingebautes Erkundungsverhalten und ohne soziale Interaktionen. Verhalten gegenüber Ereignissen außerhalb des Geheges Die Wildmeerschweinchen bleiben beı gleicher Behandlung aufmerksamer und störbarer als Hausmeerschweinchen. Sie unterbrechen bei einer Störung viel eher ihr Verhalten, bei einer leichten Beunruhigung verharren sie, bei stärkerer Störung ziehen sie sich zurück. Bei Panik springen sie auch oder drücken sich reglos ins Heu oder stellen sich an einer Wand senkrecht hoch und sind dann leicht zu greifen. Diese Reaktionen zeigen Hausmeer- schweinchen nicht. Ein Laut, der bei Störungen ın unterschiedlicher Weise auftritt, ıst das „Chirpen“, das auch ın sozialem Kontext eingesetzt wird. Wildmeerschweinchen reagieren auf die Fütterung anders als Hausmeerschweinchen. Einzelne Hausmeerschweinchen quieken laut, sobald sie das Futter riechen oder Futterge- räusche hören, bei den Wildmeerschweinchen wurde dies bisher nicht gehört. Dieses Quieken gleicht einem Kontaktlaut, der bei beiden Formen besonders von Jungtieren gegenüber ıhrer Mutter geäußert wird, der aber bei beiden Formen auch bei adulten Tieren auftritt. Die Wildmeerschweinchen nehmen wesentlich deutlicher als die Hausmeerschweinchen Kontakt zu Tieren in Nachbargehegen auf: Die Tiere schnuppern häufig in den Käfigritzen und an den Wänden hoch, sie schnuppern auch frei auf den Hinterbeinen stehend in 302 Adelheid Stahnke Richtung der Nachbarkäfige. Die meisten adulten Männchen zeigten gegen ein bestimmtes Nachbarmännchen an einer Ecke ihres Käfigs regelmäßig und minutenlang Kampfverhal- ten, Haaresträuben, Zähnewetzen, Breitseitdrohen und Scharrmarkieren. Dieses Verhalten trat bei den Hausmeerschweinchen nicht auf. Das Chirpen Dieser Laut wurde ın zwei verschiedenen Situationen registriert, wobei das Verhalten des Tieres hohe Erregung und leichte Unsicherheit anzeigt und auf einen starken Konflikt hinweist. Zum einen trat der Laut als Reaktion auf ein ungewohntes Objekt oder ein ungewohntes Geräusch auf - z.B. auf die nicht gewohnte Anwesenheit des Beobachters -, zum anderen bei besonderen sozialen Ereignissen, wie Rangauseinandersetzungen oder dem Östrus eines Weibchens. Er tritt nie unmittelbar bei Einsetzen des Ereignisses auf, sondern oft erst Minuten später. In der ersten oben genannten Situation verharren die Tiere lange reglos, ehe sie chirpen. Erst danach bewegen sie sich wieder. Das Chirpen gehört zu den lautesten Vokalisationen der Meerschweinchen (STAHNKE 1983). In rascher, rhythmischer Folge werden die Laute ausgestoßen. Beim Hausmeer- schweinchen sieht man jedesmal die Bewegung des ganzen Körpers, bedingt durch die bei der Lautgebung eingesetzte Kraft. Beim Wildmeerschweinchen ist dies kaum sichtbar. Das Chirpen kann bei beiden Formen 20 Minuten andauern, maximal werden etwa 100 Silben in einer Minute ausgestoßen, über 20 Minuten wurden 800 Silben gezählt. KHz KHz 16 FA { 12 3 \ \ | \ ar \ i \ \ ı g 4 1 % pp 7 1 Een 35 NV “Chirpen” Isec "Chirpen” 0,5 segiı Hausmeerschweinchen Ha usmeerschweinchen KHz as 16 ) ' Fe 2 | | | 8 | h 1} ' ‘ 2 "Chirpen” 1sec Wildmeerschweinchen "Chirpen" Wildmeerschweinchen Abb. 4. Sonagraphische Aufzeichnung des Chirpens von Hausmeerschweinchen und Wildmeer- Verhaltensunterschiede zwischen Wild- und Hausmeerschweinchen 303 Das Chirpen der Wildmeerschweinchen ist höher, es klingt schärfer und weniger voll als das der Hausmeerschweinchen. Als Übergang zum Chirpen gibt es einen sehr tiefen, heiseren Gurrlaut, der in Chirpen umschlagen kann. Diese Vorform trıtt beim Wildmeer- schweinchen häufiger auf, auch ist sie zwischendurch immer wieder andeutungsweise zu hören. Das Chirpen der Wildmeerschweinchen ist flexibler. Von leisen Andeutungen, über wenige Silben, bis hin zur Lautäußerung in voller Intensität treten viele Übergänge auf. Auch können die Tiere immer wieder kurz einsetzen und nach einer Pause erneut chirpen. Beim Hausmeerschweinchen sind die leichten Andeutungen oder kurzes Chirpen seltener. Die Tiere chirpen meist lang andauernd und in voller Intensität. Gegenüber ungewohnten Außenreizen wurde Chirpen bei den Hausmeerschweinchen im Laufe von 10 Jahren lediglich zweimal gehört, beide Male gegenüber dem Beobachter. Wildmeerschweinchen, die an die Anwesenheit eines Beobachters nicht gewöhnt sind, chirpen in den ersten Tagen bei jeder Beobachtung. Erst wenn der Laut nicht mehr auftritt, beginnen die Tiere, sich im Gehege freier zu bewegen. Das Chirpen, bedingt durch soziale, kontliktreiche Ereignisse, trat bei den Hausmeerschweinchen in über 1000 Beobachtungs- stunden 11x in einem Raum mit 10 Tiergruppen auf. Bei den Wildmeerschweinchen wird es unter vergleichbaren Bedingungen fast ın jeder Beobachtungsstunde gehört. Übersicht über die Ergebnisse Stellt man die wichtigsten Ergebnisse noch einmal zusammen (s. Tabelle), so ergeben sich die auffälligsten Unterschiede ım Verhalten der Wild- und Hausmeerschweinchen in den Möglichkeiten des Aufbaus sozialer Beziehungen und in den Reaktionen auf Störungen. Insgesamt lassen sich die Hausmeerschweinchen als sozial verträglicher kennzeichnen. In allen hier beschriebenen sozialen Konstellationen traten zwischen den geschlechtsreifen Tieren weniger, vor allem auch leichtere Auseinandersetzungen auf. Dies erstreckt sich teilweise selbst auf das Verhalten gegenüber Jungtieren. Allerdings waren bei den Wild- meerschweinchen in einigen Fällen die Beziehungen der Gruppenmitglieder durchaus als freundlich zu bezeichnen, wenn auch nie ganz frei von agonistischen Elementen. Die Unterschiede im Purren deuten dessen stärkere Festlegung bei den Wildmeerschweinchen auf die alpha-Position an. Andererseits purren die Hausmeerschweinchen häufiger und demonstrieren also häufiger ihren diesbezüglichen Zustand und die soziale Position. Die bleibende größere Scheu und Störbarkeit der Wildmeerschweinchen ist ein Anzei- chen für andere Aufmerksamkeitsstrukturen. Die Tiere beachten auch nach langer Einge- wöhnung alle Ereignisse und Störungen stärker, sie unterbrechen viel eher ihr momentanes Verhalten. Sie reagieren auch eher auf das Geschehen in anderen Gehegen, und regelmäßig waren intensive Kontaktaufnahmen, sogar Interaktionen, zu den Tieren - oft bestimmten Individuen - der Nachbargehege zu beobachten. Ein Ausdruck der anderen Verrechnungs- möglichkeiten von Konflikten und Störungen ist das im Vergleich zum Hausmeerschwein- chen viel häufigere und flexibler auftretende Chirpen. Diskussion Die deutlichen Unterschiede im Verhalten zwischen Wild- und Hausmeerschweinchen betreffen einerseits die Erscheinung einzelner Verhaltensweisen, vor allem aber Art und Häufigkeit ihres Auftretens. In ihrer Bedeutung für die Organisationsmöglichkeiten der Tiere erscheinen die Unterschiede im zweiten Bereich besonders gravierend. Da sich unsere Beobachtungen ausschließlich auf in Gehegen gehaltene Tiere beziehen und vergleichbare Daten aus dem Freiland nicht vorliegen (vgl. Roop 1972), sind Aussagen über die Ursachen des unterschiedlichen Verhaltens der beiden Formen schwie- rig. Vor allem das Möglichkeitsfeld der Wildform, deren Verhalten unter unserer Haltung 304 Adelheid Stahnke Tabelle Zusammenfassung wichtiger Unterschiede im Verhalten von Wild- und Hausmeerschweinchen Wildmeerschweinchen Hausmeerschweinchen Soziale Beziehungen adulter Tiere Adulte Männchen sind gegeneinander unver- träglich Adulte Weibchen sind gegeneinander verträg- lich; agonistisches Verhalten ist häufig und hef- tig; die Beziehungen sind individuell, unter- schiedlich freundlich Adulte Männchen und Weibchen sind nur be- dingt verträglich; die Beziehungen sind indivi- duell, es gibt freundliche und durch viel Agoni- stik gekennzeichnete Beziehungen; fremde Tiere gewöhnen sich schwer aneinander Die Tiere nehmen Kontakt zu den Tieren in Nachbargehegen auf; ranghohe Tiere interagie- ren mit bestimmten Individuen in Nachbarge- Adulte Männchen sind unter bestimmten - si- tuationsabhängigen —- Bedingungen gegenein- ander bedingt verträglich Adulte Weibchen sind gegeneinander verträg- lich; agonistisches - besonders heftigeres — Verhalten ist sehr selten Adulte Männchen und Weibchen sind verträg- lich; die Beziehungen sind individuell; fremde Tiere gewöhnen sich rasch aneinander Die Tiere nehmen kaum Kontakt zu den Tieren in Nachbargehegen auf; Interaktionen mit Tie- ren in Nachbargehegen treten nicht auf hegen Purren Wenig, leise, kurz Regelmäßig, laut, langdauernd Nur alpha-Tiere purren Auch unterlegene Tiere purren Junge Männchen purren bereits in der 1.-3. Lebenswoche Junge Männchen purren nicht vor der Ge- schlechtsreife Treteln und Imponiergang fast immer gleich- zeitig mit Purren Treteln und Imponiergang selten gleichzeitig mit Purren Reaktionen auf Störungen und Ereignisse außerhalb des Käfigs Bei längerfristiger Haltung vergleichsweise we- niger schreckhaft und störbar Bleiben scheuer und schreckhafter, aufmerksa- mer gegen Störungen, unterbrechen eher ıhr Verhalten, bei Erschrecken flexiblere Reak- tionen Chirpen Sehr häufig, sehr flexibel, Sehr selten, fast nur in voller Ausprägung immer auf ungewohnte Störungen hin, häufig in sozialen Konfliktsituationen wesentlich eingeschränkter erschien als das der Hausmeerschweinchen, läßt sich aus diesen Beobachtungen nicht voll erfassen. Aus der langjährigen Kenntnis der Tiere entstand der möglicherweise subjektive Eindruck, daß die Hausmeerschweinchen in unseren verschie- denen Haltungsbedingungen größere Anteile ihres Verhaltenspotentials hervorbrachten als die Wildmeerschweinchen. Dies geht nicht nur aus den relativ größeren Häufigkeiten für die meisten Verhaltensweisen hervor, sondern auch aus den funktionalen sozialen Gefü- gen, welche sie aufbauten und in die sie sich langfristig ohne größere Belastungsanzeichen integrieren konnten, während den Wildmeerschweinchen dies nur sehr begrenzt und mit großem Aufwand möglich war. Dadurch sind die gemessenen Unterschiede in den einzelnen Verhaltensweisen schwer interpretierbar. Es läßt sich z. B. nicht immer entschei- den, ob eine Verhaltensweise, die bei den Wildmeerschweinchen kaum auftrat, weniger im Möglichkeitsbereich der Wildform liegt, oder ob ihr Auftreten nur ın dieser Haltung gehemmt war. Dann müßte der Unterschied nicht in den Antriebsvoraussetzungen für Verhaltensunterschiede zwischen Wild- und Hausmeerschweinchen 305 diese spezifische Verhaltensweise liegen, sondern könnte dyirch Strukturen bedingt sein, die stärker die Gesamtorganisation der Tiere betreffen. Andere Autoren, die Haustierverhalten an anderen Säugetierarten untersucht haben, nennen als Merkmale, welche sich im Laufe eines Jahrtausende dauernden Haustierstandes herausgebildet haben, Unterschiede gegenüber der Wildform in der Gesamtaktivität und dem Bewegungsdrang, der Aufmerksamkeit und „Fluchtbereitschaft“, der sozialen Inte- grationsfähigkeit und Aggressivität von Haustieren gegenüber der Wildform (HERRE und RöHRrs 1973, 1974; HEMMER 1983). Nach HERRE (1981) sollte zumindest teilweise die Möglichkeit in Betracht kommen, daß es sich bei diesen Unterschieden nicht immer lediglich um Austfallserscheinungen bei der Haustierform handeln muß, sondern daß es sich durchaus auch um Neuanpassungen an die anderen Lebens- und Umweltbedingungen handeln kann. Unsere Daten für die Haus- und Wildmeerschweinchen lassen sich ähnlı- chen Eigenschaften und Merkmalen zuordnen, wie sie eben für andere Arten angeführt wurden. Aus diesen Daten geht eindeutig hervor, daß die Wildmeerschweinchen in anderer Weise mit unseren Gehegebedingungen zurechtkamen als die Hausmeerschwein- chen. Am Beispiel der beiden Verhaltensweisen „Chirpen‘“ und ‚Purren“ läßt sich dies besonders eindrucksvoll aufzeigen: Das Chirpen ist ein Verhalten, welches sich unter anderem dem Aufmerksamkeitsbereich zuordnen läßt. Es tritt bei hoher Erregung und Spannung auf, und es ermöglicht, diese Konfliktsituationen zu bewältigen. Die Wildmeer- schweinchen reagierten stärker und auch differenzierter auf Störungen sowie auf unklare soziale Situationen. Nicht nur chirpten sie wesentlich häufiger als die Hausmeerschwein- chen, sie verwendeten den Laut auch modifizierter in Klangfarbe, Lautstärke und Dauer. Vergleichsweise äußerten die Hausmeerschweinchen das Chirpen selten, nur in extremen Zuständen, die aber weniger häufig auftraten, und fast immer in voller Intensität mit großem Kraftaufwand. Die Wildmeerschweinchen reagierten also mehr und intensiver auf äußere Ereignisse, und sıe hatten mehr Möglichkeiten, diese abgestuft je nach Situation zu verarbeiten. Für die Hausmeerschweinchen ist nicht geklärt, ob sie entsprechende Vorkommnisse teilweise weniger genau registrierten oder ob sie — aktıve - Möglichkeiten haben, diese Wahrneh- mungen anders, unter geringerer Erregung und Spannung, zu verarbeiten. Trotz der vielfältigen Verarbeitungsmöglichkeiten für hohe Erregungs- und Spannungszustände waren die Wildmeerschweinchen wegen ihrer anderen Aufmerksamkeitsstrukturen doch ın unserer Gehegehaltung stärkeren Belastungen ausgesetzt als die Hausmeerschweinchen. Die andere Verhaltensweise, das Purren, steht in seinen Auswirkungen dem Chirpen entgegen. Diese Lautäußerung trat bei den Hausmeerschweinchen viel häufiger und deutlicher auf. Bei den Hausmeerschweinchen gehört das Purren — neben seiner Funktion im sexuellen Bereich - zu den Mechanismen, soziale Prozesse freundlich, unter Vermei- dung von Agonistik, ablaufen zu lassen, mithin die soziale Verträglichkeit der Tiere zu fördern und positive soziale Beziehungen aufzubauen oder zu erhalten. Hausmeer- schweinchen sind in der Lage, differenzierte, auch individuelle, soziale Beziehungen aufzubauen, und sie können sogar in relatıv enger Haltung komplexe soziale Organısa- tionsmuster bilden (vgl. SACHSER 1983; SACHSER und HENDRICHS 1982; STAHNKE 1983; STAHNKE und HENDRICHS 1986). Bei den Wildmeerschweinchen waren die wesentlich geringere soziale Verträglichkeit und Integrationsfähigkeit auffällig. Eine direkte Bezie- hung zwischen den Schwierigkeiten, sozialpositive Beziehungen, ın denen vergleichsweise wenig Agonistik auftritt, aufzubauen und der geringen Häufigkeit des Purrens besteht vermutlich nicht. Jedoch können die Unterschiede in der Häufigkeit des Purrens bei den Wild- und den Hausmeerschweinchen als Anzeichen für die anderen sozialen Möglichkei- ten der Tiere, ın der Gehegesituation mit Artgenossen zurechtzukommen, bewertet werden. Während die Hausmeerschweinchen beständig über das Purren zugleich den Status der Dominanz als auch die funktionale soziale Struktur „mitteilen“ (im kommuni- 306 Adelheid Stahnke kativen Sinne den Gruppenmitgliedern „‚mitteilen‘‘), trat beides bei den Wildmeerschwein- chen weniger auf, obgleich hier das Purren noch stärker an den alpha-Status gekoppelt ist. Die Wildmeerschweinchen konnten unter diesen Bedingungen viel schwerer Möglichkei- ten finden, sich sozial verträglich zu verhalten und langfristig reibungsarme Beziehungen aufzubauen. Vielmehr setzten sie oft heftigste Agonistik ein, die alle Beteiligten stark belastete und bis zu Todesfällen führte. Die Situation des Unterlegenseins zu bewältigen, gelang den Wildmeerschweinchen weniger. Und dominante Individuen zeigten den ihnen unterlegenen Tieren auch bei geklärten Rangverhältnissen immer wieder ihre Überlegen- heit durch Angrıtfe. Die Wildmeerschweinchen verhielten sich also ım Aufmerksamkeitsbereich differen- zierter und reichhaltiger, die Hausmeerschweinchen hinsichtlich Aufbau und Aufrechter- haltung positiver sozialer Beziehungsmuster. Beides dürfte bewirkt haben, daß den Wildmeerschweinchen das Zurechtkommen in der Gehegesituation schwerer fiel. Die Hausmeerschweinchen waren unter diesen Bedingungen in der Lage, Organisationsmuster zu finden, die ein ftunktionales, einigermaßen reibungsarmes Zusammenleben ermöglich- ten. Sie konnten ihre sozialen Beziehungen auf einem ım Vergleich mit den Wildmeer- schweinchen niedrigeren Erregungsniveau regeln. Die Wildmeerschweinchen dagegen befanden sich oft in hohen Erregungszuständen, die sie anscheinend oft gar nicht oder aber nur über heftige Verhaltensäußerungen abbauen konnten. Die eingangs gestellte Frage, wie flexibel das Verhalten der Wildmeerschweinchen sein kann, wie sehr die Ausprägungen sıtuationsabhängig sind und in welchem Maße sich das Verhalten von Wild- und Haus- meerschweinchen unter anderen Lebensbedingungen möglicherweise doch ähnlich ausneh- men kann, bleibt weiteren Untersuchungen vorbehalten. Danksagung Vor allem danke ich Herrn Prof. Dr. H. HENDRICHSs, der diese Untersuchung ermöglichte und sie durch Diskussionen und Kritiken wesentlich förderte. Bei Herrn Dr. M. StüwE möchte ich mich besonders herzlich für die Beschaffung der ersten Wildmeerschweinchen bedanken. Frau C. SCHÜTTE und Herrn H. Krös verdanke ich wertvolle Hinweise und anregende Diskussionen über das Verhalten der Wildmeerschweinchen, desgleichen Frau Y. THYEn, Herrn A. HAEMIscH, Herrn Dr. N. SACHSER und Herrn M. ScHMIDT über das Verhalten der Hausmeerschweinchen. Herrn J. BÖHNER danke ich für seine Hilfe bei der Anfertigung der Sonagramme. Zusammenfassung Das Verhalten von Wildmeerschweinchen (Cavia aperea) und Hausmeerschweinchen (Cavıa aperea 1. porcellus) unter Gehegehaltung wurde verglichen. Die auffälligsten Unterschiede betrafen die Reak- tion auf Störungen und den Aufbau sozialer Beziehungen. Die Wildmeerschweinchen waren bleibend aufmerksamer und störbarer gegenüber dem Geschehen in der Umgebung des Geheges und auch gegenüber ungewohnten Ereignissen im eigenen Gehege. Die Lautäußerung „Chirpen“, die auf Störungen hin auftritt, äußerten sie viel häufiger und vielfältiger als die Hausmeerschweinchen. Bei den Wildmeerschweinchen trat, auch in lange aneinandergewöhnten Gruppen, mehr und härtere Agonistik auf als bei den Hausmeerschweinchen. Freundliche soziale Beziehungen waren seltener, das gemeinsame Halten einander fremder Individuen vielfach nicht möglich. Als ein Indikator für die Fähigkeit zur größeren sozialen Verträglichkeit der Hausmeerschweinchen wurde das häufige Einset- zen der Lautäußerung ‚„Purren‘ der alpha-Männchen gewertet. Diese Verhaltensweise wirkt sozial stabilisierend. Das Purren trat bei den Wildmeerschweinchen selten auf. Die gefundenen Unterschiede im Verhalten der Wild- und Hausmeerschweinchen dürften zusammenhängen mit deren verschiede- ner Fähigkeit, in der Gehegesituation zurechtzukommen. Literatur BEAUCHAMP, G. K.; Crıss, B. R.; WELLINGTON, J. L. (1979): Chemical communication ın Cavia: Responses of wild (C. aperea), domestic (C. porcellus) and F, males to urine. Anım. Behav. 27, 1066-1072. BÜTTNER, D.; Wornık, F. (1982): Untersuchungen zur Kurzzeitperiodik beim Meerschweinchen (Cavia aperea f. porcellus). 1. Bewegungsaktivität unter Licht-Dunkelwechsel, Dauerdunkel und Dauerlicht. Z. Säugetierkunde 47, 321-324. Verhaltensunterschiede zwischen Wild- und Hausmeerschweinchen 307 HARPER, L. V. (1976): Behavior. In: The biology of the guinea pig. Ed. by J. E. WAGNER; P. ]. MannınG. New York, San Francisco, London: Academic Press. 31-51. HEMMER, H. (1983): Domestikation — Verarmung der Merkwelt. Braunschweig, Wiesbaden: Vieweg. HERRE, W. (1981): Domestikation — Ein experimenteller Beitrag zur Stammesgeschichte. Naturwiss. Rdsch. 34, 456-463. HERRE, W.; Röhrs, M. (1973): Haustiere — zoologisch gesehen. Stuttgart: Gustav Fischer. — — (1974): Das Verhalten der Haustiere. In: Grzimeks Tierleben, Verhaltensforschung. Hrsg. von K. Immelmann. Zürich: Kindler. — — (1977): Zoological considerations on the origin of farming and domestication. In: Origins of agriculture. Ed. by C. A. Reep. The Hague, Paris: Mouton Publishers. 245-279. HÜcKINGHAUs, F. (1961): Zur Nomenklatur und Abstammung des Hausmeerschweinchens. Z. Säugetierkunde 26, 108-111. Kıng, J. A. (1956): Social relations of the domestic guinea pig living under semi-natural conditions. Ecology 37, 221-228. KunkeL, P.; Kunker, 1. (1964): Beiträge zur Ethologie des Hausmeerschweinchens. Z. Tierpsychol. 21, 602-641. MACNEISH, R. S. (1977): The beginning of agriculture in Central Peru. In: Origins of agriculture. Ed. by €. A. Reep. The Hague, Paris: Mouton Publishers. 753-801. Roop, ]J. P. (1972): Ecological and behavioural comparisons of three genera of argentina cavies. An. Behav. Mon. 5, 1-83. Roop, J. P.; WEIR, B. J. (1970): Reproduction in female wild guinea pigs. J. Reprod. Fert. 23, 393409. SACHSER, N. (1983): Soziale Beziehungen, räumliche Organisation und Verteilung agonistischer Interaktionen in einer Gruppe von Hausmeerschweinchen (Cavia aperea ft. porcellus). Z. Säuge- tierkunde 48, 100-109. SACHSER, N.; HENDRICHs, H. (1982): A longitudinal study on the social structure and its dynamics in a group of guinea pigs (Cavıa aperea f. porcellus). Säugetierkdl. Mitt. 30, 227-240. STAHNKE, A. (1983): Zur Sozialisation männlicher Hausmeerschweinchen: Der Einfluß ranghoher adulter Männchen. Dissertation, Universität Bielefeld. STAHNKE, A.; HENDRICHS, H. (1986): Social varıability in male guinea pigs: different dominance quality resulting from early social experience. Zool. Beiträge 29, 413—435. WAGNER, J. E.; ManniIng, P. J. (Eds.) (1976): The biology of the guinea pig. New York, San Francisco, London: Academic Press. WEIR, B. J. (1972): Special bibliography on guinea-pig behaviour. Guinea Pig News Letter 5, 18-32. — DR Notes on the origin of the domestic guinea pig. Symp. zool. Soc. Lond. 34, 437446. Young, W. C. (1969): Psychobiology of sexual behavior ın the guinea pıg. Advan. Study Behav. 2, 1-10: Anschrift der Verfasserin: Dr. ADELHEID STAHNKE, Schwärzlocher Str. 64, D-7400 Tübingen Food consumption and seasonal variation in the diet of the aardwolf Proteles cristatus in southern Africa By P.R. K. RicHARDSoN MeGregor Museum, Kimberley, South Africa and Animal Ecology Research Group, Department of Zoology, University of Oxford, England Receipt of Ms. 9. 9. 1986 Abstract Investigated were the food and foraging activity of twelve free ranging adult aardwolves Proteles cristatus and their cubs. For most of the year aardwolves are nocturnal and feed almost exclusively on the harvester termite Trinervitermes trinervoides (Sjostedt). During winter however, it is usually too cold for termites to forage at night, so the aardwolf becomes more diurnal and supplements its diet by feeding on another harvester termite Hodotermes mossambicus (Hagen), which is mainly active by day U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5205-0307 $ 02.50/0 Z. Säugetierkunde 52 (1987) 307-325 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 308 P. R. K. Richardson and during the winter. Nevertheless, the aardwolf still suffers a food shortage during winter. This is evident from the weight loss (up to 20% of body weight) experienced by all animals during this period. It is argued that the aardwolf’s distribution across Africa is determined by the availability of those Trinervitermes species which are active throughout most of the year. Introduction The aardwolf Proteles cristatus is one of the most highly specialised carnivores in existence (Kruuk and Sands 1972). Not only does it feed almost exclusively on termites but, throughout its African distribution, it feeds primarily on a few species of the genus Trinervitermes. The particular species changes as one moves across Africa, with T. trinervoides being the main species selected in South Africa (CooPER and SKINNER 1979). Two other predominant species have been found in the aardwolf’s diet in Botswana/ Zimbabwe (SMITHERS 1983) and in East Africa (Kruuk and Sans 1972). These three species appear to be ecological equivalents and are characterised by a few very important features. These are: (1) that they are each the only local species of termite which regularly harvests/forages for dry grass in dense concentrations and completely exposed on the soil surface; (11) they are active for most of the year (Kruuk and Sans 1972); and (iii) like all other nasute termites (Nasutitermitinae), they posses a highly effective means of chemical defence. The soldiers have well developed frontal glands from which they can squirt very sticky and noxious terpene-based secretions (PRESTWICH 1983; BRAEKMAN et al. 1984). These secretions appear to be highly successful in deterring attack from ants (LONGHURST et al. 1978, 1979), which are believed world-wide to be the major predators of termites (EISNER et al. 1976; DELIGNE et al. 1981), as well as most myrmecophagous (ant- and termite-eating) mammals (Lugın and MONTGOMERY 1981; REDFORD and DorEA 1984). It appears therefore that the chemical defence of these termites is so successful, that during their development the harvesters were able to forego the protection of the mud galleries used by most other termites and forage quite exposed on the soil surface (KRUUK and Sanps 1972). RicHARDSON (1985) has therefore suggested that the ability of the aardwolf to tolerate these terpene secretions must be its most significant adaptation, as this has made these termites readily available to the aardwolf and virtually without competition from any other mammal (Kınspon 1971; Kruuk and Sanns 1972; MELToON 1976). Having this ability, the aardwolf has developed a number of more obvious physical adaptations like a very broad tongue and palate, large salıvary glands and well developed senses of smell and hearing (FLOwER 1869), which have enabled it to become extremely adept at feeding almost exclusively on this one food source. KRUUK and Sans (1972) have suggested that it is this extreme specialisation of the aardwolf which is responsible for its being absent from West and Central Africa. Harvesting species of Trinervitermes are present ın these more wooded regions, but they appear to be less suitable for aardwolf predation, being less common and not regular foragers throughout the year (Sans 1961; OHracu 1979). This hypothesis therefore implies a very strong dependence by the aardwolf on having a harvesting Trinervitermes species as a regular food source throughout most of the year. The aim of the present paper is to test this hypothesis by investigating the aardwolf’s dependence on T. trinervoides as a food source in the northern Cape of South Africa, and to relate its seasonal availability to aardwolf foraging activity and body weight during the year. Materials and methods The study was carried out on Benfontein (28°50’S, 24°50’E), an 11,300 hectare (ha) game farm near Kimberley, South Africa (Fig. 1). This is a semi-arid region which experiences large temperature fluctuations on both a daily and a seasonal basis. The winter months (May to August) are typically dry Food consumption and seasonal variation in the diet of the aardwolf 309 a ° SS 2450 BOTSWANA 4 % Kimberl| imloarloy KEN wi SOUTH AFRICA ” 7 Nememe * Study area N „a Secondary road — — Farm boundary Pan (non-Pperennial water) Contour interval 100 feet 3800 } y Fr 28°50 N — : 7 # 2, ° ( (®) 900, 0 OS ee ° [6) BENFONTEIN Study area () x „N gn >= % 13= ——— u 7 > 1 0) 1 2 3 4 S) KM Fig. 1. The farm Benfontein showing the main study area and the more important geographical features with no cloud cover, so the days are frequently warm (regularly over 20°C), while at night the temperature may fall below zero. The mean annual rainfall is 437 mm (Weather Bureau, Pretoria). The main study area, of 2015 ha, lies in the eastern half of Benfontein and follows a gradual incline from a pan basın in the north-west to an area of Kalaharı thornveld, which is the only area with trees (Acacıa erioloba), in the south-east. The slopes in between are dominated by various perennial grasses and Karoo shrubs, particularly Pentzia spp. Due to periodic waterlogging of the soil, mounds of T. trinervoides are completely absent from the pan basın. They are present throughout the remaining parts of the study area, and are larger and more abundant in areas with the densest grass cover. All observations were made from a four-wheel-drive vehicle. Aardwolves were followed at very slow speed, at a distance of 15-30 m. Visual contact was maintained by keeping an anımal in the headlights of the vehicle, or following ıt with the beam of a spotlamp mounted above the driver’s window. All observations were recorded on a pocket dictaphone. The aardwolves were undisturbed 310 P. R. K. Richardson by the spotlamp or by the sound of the human voice. Binoculars (10x40) were used for observing them at a distance. The resident aardwolves of five different territories were followed between July 1981 and September 1984. Individual recognition was aıded by fitting eleven of the adults and seven of the cubs either with light-weight collars made from sterkolite (Sterkolite, Johannesburg) or, during the last year of the study, with radio-collars (seven adults and one cub). All collars had small coloured reflectors riveted to them to facılitate recognition of the bearer. Uncollared aardwolves were identified by natural earmarks and theır stripe patterns. In order to fit the collars, aardwolves were immobilised, using either blowdarts or a dart gun (RıcHArDson 1983). Various body measurements were taken from the recumbent anımals. Nomination of a focal anımal was done on a rotational basis in order to obtain data on each pair’s territory every year. However, in order to get more complete data on breeding and feeding activities, observations were concentrated on two pairs of aardwolves, sometimes to the detriment of obtaining data on the other pairs. Aardwolves were usually followed from the time they left the den, until an hour or two after midnight. If one had already left the den, or was lost while being followed, then the study area was searched until it, or another aardwolf, was found. During winter, because aardwolves were usually active for only a few hours in the late afternoon and early evening, it was frequently possible to follow them until they retired. However, over the rest of the year aardwolves were usually active all night. On these occasions, observations were usually terminated when an aardwolf lay down, or had covered most of its territory. During monthly “all- night studies”, an aardwolf was observed for its entire period of activity. Mapping The method used to plot the movements of aardwolves was based entirely on the use of 420 colour- coded stakes, which were placed throughout the five aardwolf territories. Stakes were located at most dens and middens and as landmarks ın areas in between. Observations of aardwolves were made whenever one passed near a stake, changed direction, or when there was a noticeable change in social or feeding behaviour. For each observation the aardwolf’s distance and bearing to the nearest marker was recorded. Distances of over 30 m were estimated to the nearest 10 m. These data were used to calculate the coordinates of the aardwolf’s position. From September 1982 to July 1984 all data were computerised and analysed according to a grıid matrıx of 100x100 m (1 ha) cells. Feeding Two categories of feeding data were collected: food type, and a relative measure of feeding intensity. The food type was almost invariably one or other of two species of termites — Trinervitermes trinervoides or Hodotermes mossambicus. These were easily ıdentified by driving to where an aardwolf had just feed and shining a hand-held spotlamp at the ground. Trinervitermes are small, orange-coloured termites that feed in very dense concentrations, whereas Hodotermes are very much larger, nearly black in colour and forage in a more widely spaced fashion. Feeding intensity was recorded on a scale of one to three. “One” was recorded ıf an aardwolf had fed on termites once, or only a few times since the last observation. “Two” was recorded if the aardwolf was feeding quite frequently, but still spent >50 % of the time walking. “Three” was recorded if an aardwolf spent more time feeding than walking. The feeding data were analysed assuming a uniform speed of movement during each observation period (see later) and were allocated to each hectare in proportion to the time spent there. Thus the relative feeding value per hectare was derived by multiplying the feeding intensity (0, 1, 2 or 3) by the time spent in that hectare. The estimation of the feeding intensity was clearly subjective, but the results were highly correlated (see later) with those of the detailed feeding analysis recorded during the all-nıght studies. On these nights, in addition to recording the above data, each observation of feeding or walking was timed (with a stopwatch) during alternate 15 min periods. This was achieved with relative ease because the average feeding bout lasted over 20 sec (RıcHARDSoN 1985), so the spells of feeding and walking fell into quite discrete periods and were easily recorded. It was assumed that the feeding behaviour during the 15 min periods of detailed observations was, taken over the whole night, the same as during the “off” periods. These data then provided information on the total time spent feeding per night. Faecal analysis Fresh aardwolf faeces were collected approximately every two weeks during the last two years of the study. They were collected from middens in three different territories. An attempt was always made to collect a whole defaecation and the most recent one present. They were collected in a paper bag, labelled and later sun-dried. A total of 136 defaecations was collected. Each sample was first oven-dried at 80°C for 24 hours, then gently crushed with a pestle and Food consumption and seasonal variation in the diet of the aardwolf Jı mortar till the contents could spread evenly. A rough estimate was then made of the proportion, by volume, of Trinervitermes to Hodotermes present. These termites are easily identified, as the heads of both soldiers and workers persist in the faeces and are very different in terms of size, shape and colour. During the all-nıght studies an attempt was made to collect every defaecation of the animal being followed, as well as the first defaecation of the following night. These faeces were processed in the same way as above. In addition, an estimate of the total number of termites in one complete set of faeces was made in order to calculate the total number eaten per night. In doing this count, the first defaecation of that night was excluded and replaced by the first of the following night. The analysis was done by taking small samples from each defaecation and counting the total number of soldier’s heads — which are more easily ıdentified than the worker’s heads and may also survive better (Lusın A MONTGOMERY 1981). These samples were recorded by weight as a proportion of the whole efaecation. Climatological data The Kimberley airport is about 5 km west of Benfontein, so the daily maximum and minimum temperatures were taken from its weather station. During the last two years of the study the ambient temperature in the field was also recorded. To do this a thermometer on a foam rubber base was placed on the back of the truck and readings were taken periodically through the night. Results Foraging behaviour The data on the foraging behaviour of aardwolves were collected while following 12 adults and 18 cubs. The times spent following these anımals and the distances covered are summarised in Table 1. Table 1 Distance and time spent following 30 aardwolves between September 1982 and July 1984 Name Territory! 1 Distance km Achilles Seven Winds William Tell Nausicaa Errol Flynn Muses Zebra Robin Hood Maid Marion Telemachus’ Cyclops Galatea Adults Total (n = 12) 382.96 Cubs Total (n = 18) 51.82 HsceHmsmesstmesHs ! Territories are named after the initials of the resident male. - * Time does not include rest periods. -— * Sub-adult which did not leave his natal territory. Aardwolves are almost exclusively solıtary foragers (Kruuk and Sans 1972; RiCHARD- son 1985), and feed by walking around their territories and stopping to lick up termites whenever they locate them. Thus their average foraging speed ıs directly related to the supply of termites available (Table 2). The table shows that they walk at an average speed of 2.3 km/h when they are not feeding, at 1.0 km/h when they are feeding at maximum intensity, and at regular intermediate speeds for feeding levels in between. This therefore justifies the methods which assume a uniform speed of movement during each observation. 312 P. R. K. Richardson Table 2 Characteristics of foraging aardwolves while feeding at different intensities Relative Feeding Intensity Distance km 1 Speed km/h 346.9 137.5 122.4 DI 7339 Most of the detailed observations of feeding behaviour were recorded during the all- night studies. The results of the 13 nights that were successfully completed (one with no feeding), are summarised in Table 3. In this table the original values (every alternate 15 min) are given for the number of termite patches utilised, so they should be doubled to get the values per night. The correlations (values per ha) between the relative feeding value and the actual time spent feeding are also presented. On all nıights the correlation was highly significant (p<0.01). Table 3 Foraging data for the all-night studies No. of Food Relative Feeding Value vs Termite Total Feeding Seconds/ha Patches R Seven Winds Seven Winds Seven Winds Seven Winds Seven Winds Achilles Achilles Apollo Achilles Achilles Seven Winds Seven Winds Seven Winds 1 r = Spearman rank correlation coefficient. The regression between the total values of these same two measures of feeding per night, is presented in Fig. 2. Here again there is a highly significant correlation (p<0.001) between the relative and actual values of feeding. These results indicate that the relative feeding data are reliable and can therefore be used to determine feeding values throughout the year. In Fig. 3 the total amount of feeding on Trinervitermes is plotted against the maxımum temperature for that day. The data used here are the relative values of feeding only for nights that aardwolves fed exclusively on Trinervitermes. This is because Hodotermes may feed during the day when it is too cold for them (or Trinervitermes) at night, and therefore their foraging is determined by a different temperature regime from that of Trinervitermes. X1000 VALUE FEEDING RELATIVE 70 60 50 40 30 20 10 X 1000 RELATIVE FEEDING VALUE Food consumption and seasonal varıation in the diet of the aardwolf 313 80 % Seven Winds Pr et . we O Achilles —% E 70 Oo . DE %$ + Apollo er > .” ic 60 222 Ba . . Sei (0) .. e .” .. .” .-. es 50 5 et .. - .” Be Bo ee Be + eo. . 40 > .- zoo ° oa .” % 2a Pig . _ * a .” $ +. 30 Be .” P2 .. .. Ed .” .” . er er Pr 20 .. zen ze .. Ba 2 . oO 9 Be a 10 ° e=- .- Ar .. De Ü) a [0] Bez (0) 1000 2000 3000 4000 5000 6000 7000 8000 TOTAL FEEDING SECONDS Fıg. 2. Regression of the total relative feeding values per individual aardwolf against the total time it spent feeding during the all-nıght studies (r=0.97, df=13, p<0.001; y= -293 + 0.118x; 95 % confidence limits are indicated) 1 2 3 x l ) X ) X I I l l l l 1 B N I 1 j l I | l I x ) ) ) l | 1 ' l j 1 x | I I I l | l l I I N I j I I h I j l | | I I I l 1 l I I 1 | I I I l I I I l I l I l x | I I I l — N l I l | X N | I 1 | 1 I N I l j | | N I I I | l | ) ) ) ) l 1 N N l I l I N | l I I I I | | I | I I l 1 1 | I | i 1 1 I 1 l i I l l | I | | j ı I N j ) l I | | ) ) je ! | | | | l l l N x x l ) ) ) | I m ıX l | | j I L A [ j N j I I m I ) j | 1 | | X x m 1 | l | ] I I I I l | I I Xı ) 1! N m ) l | I ae N W I 1 | l I N 1 j u j j 16 18 20 22 24 26 28 30 32 34 36 38 DAILY’ MAXIMUM TEMPERATURE ° Fig. 3. Total relative feeding value per individual aardwolf plotted against the daily maximum temperature. Data are only from nights when aardwolves fed almost exclusively on Trinervitermes and when one anımal was followed for its entire activity period. Threshold temperatures (1, 2 and 3), which determine termite activity, and the mean relative feeding values (A, B and C), are indicated (@) 40 (relative feeding value / time ) RATE FEEDING 160_ 314 P. R. K. Richardson There are four distinct levels of teeding, and these appear to be determined by the effect of temperature on Trinervitermes. Below 18°C there is no feeding. From 18-24 °C there is a low level of feeding. Above 24°C there is a very much greater level of feeding, which drops again as the temperature rises above 34°C. These results suggest that when the daily maximum is less than 18°C, it gets too cold for the termites to forage in the evening. Above this temperature termites can forage for a few hours in the early evening, but it soon cools down and they return to their nests. A daily maximum of approximately 24°C seems to be the turning point at which it no longer gets too cold for termite foraging. They can therefore forage all night, or at least until midnight. Thus there is a sudden increase in the termites available to aardwolves. As the temperature rises even higher, so termite activity declines again (NEL 1968; RıcHARDSoN 1985), and with it the feeding of aardwolves. x x o Original sunset temperatures 8 10 12 14 16 18 20 22 24 26 28 TEMPERATURE AT SUNSET Ze Fig. 4. Curve illustrating the dependence of aardwolf feeding rate (total relative feeding value per night/active foraging period) on the temperature at sunset. Data are only from nıghts when aardwolves fed almost exclusively on Trinervitermes and when one anımal was followed for its entire activity period In order to show that the pattern illustrated in Fig. 3 is a result of termite activity, and not simply a function of the time aardwolves spend feeding, the feeding rate is plotted against temperature at sunset in Fig. 4. The feeding rate is the total relative feeding value per night, divided by the time the aardwolves spent foraging. It therefore excludes any rest periods. Sunset temperature is used because this is the approximate time when the termites become active, and therefore should be an more accurate measure than the daily maximum. On most warm days the sunset temperature was not recorded, so these missing data were estimated using a regression of temperature at sunset against maximum daytime tempera- ture. This ıs reasonable because these temperatures are very highly correlated (r=0.97, df=17, p<0.001). Ya BE x x Sunset temperatures derived from regression 30 Food consumption and seasonal variation in the diet of the aardwolf 315 This figure shows similar trends to Fig. 3, although it also suggests that termites become increasingly active as the temperature rises above 13°C. This is presumably because there are more nests becoming active, so aardwolves have more chance of finding foraging termites. | The time, in relation to sunset, at which aardwolves were first observed to start feeding on Trinervitermes ıs plotted in Fig. 5. In this figure the only data used are for those days when aardwolves were followed from at least an hour before sunset. The earliest an aardwolf was seen to start feeding on Trinervitermes was 51 min before sunset, but it is clear that most termites become active ın the last 15 min before sunset. If cold is the major factor limiting the nocturnal foraging of the Trinervitermes during winter, then one would expect that once they have commenced | foraging, these termites should continue for as long as possible to make the expedition energetically worth- while. This was tested by comparing the times (rela- 7 tive to sunset) and temperatures at which aardwolves retired to their dens during winter, for nights when they had and had not fed on Trinervitermes (Table 4). ? Although the mean temperature at which aardwolves retired for the night was 1.3°C lower when they had not fed on Trinervitermes, this difference was not significant. However, if they had fed the mean time of retiring was nearly two hours later, and this differ- ence was highly significant (p<0.001). The tlexibility in aardwolf foraging behaviour dur- ing winter is further ıllustrated in Table 5, which summarises the data on total foraging periods (includ- ing rest periods) and the mean percentages actıvity for ee nen summer and winter. During the summer months ne, eedins on Inzer aardwolves foraged tor significantly longer periods „termes during daylight, during the (p<0.001) than in winter, but the percentage activity winter months over the total foraging period was significantly lower (p<0.05). This was presumably because during winter, wıth the termites being active for a very limited period each night, aardwolves made maximum use of the opportunity to feed and therefore did not rest during the night. They did sometimes rest during the afternoon if Sunset FREQUENCY 60774577307 157 0 15 30 TIME (min) RELATIVE TO SUNSET Table 4 Time and temperature at which aardwolves retired to their dens after they had, or had not fed on Trinervitermes during winter Mean Time Std Dev Range Mean T°C Std Dev Range After Sunset min Trinervitermes 162.4 84.0 37-348 n 1 a) foraging Trinervitermes 65.5 49.0 0-160 £ 2.98 1.5-12.0 12 not foraging Mann-Whitney U test — Feed vs Not Feed: Time: W = 727, nl = 12, p < 0.001 Temp: W = 648, nl =12,p >01 316 P. R. K. Richardson Table 5 Characteristics of foraging activity during summer and winter Data excludes nıghts which were incomplete Season Foraging Period (h) Overall % Activity Mean Range Mean Range Summer 6.65-10.0 90.0 84.6-100 Winter! 3.50 1.10-5.92 94.4 323-100 ' Excluding night when Achilles walked 24 km. Mann-Whitney U test - Summer vs Winter: Duration of actıve period: W = 538, nl = 9, p < 0.001 Mean % activity per night: W = 742, nl = 9, p < 0.05 they had become active earlier but found no Hodotermes foragıng. They usually rested till about half an hour before sunset, then recommenced foraging. There was however one exceptional nıght during June 1983, when a male, Achilles, walked 24 km ın 10 hours and remained actıve until the temperature had dropped to zero — four hours after he had last fed on termites (Table 3). On this occasion he was involved in early mating season activities, and spent the whole nıght scent-marking his own and a neighbouring territory (RICHARDSON 1985). Because of this, the data for this day are not included in the above statistical analyses. The general pattern of foraging activity throughout the year is ıllustrated in Fig. 6. During summer aardwolves omeralllı became active about an hour after sunset and retired TIME OF DAY — HOURS 13 214 157 2167 17 187 197207721772223722472.01551022E 03 0 AO 50T, INN TEN INN LILLM TE NN HEHE |'| | Ba lei A — = at BE || Em T BE sn | ||| |||] RG RE EEE || III 1112111000000 DOM Sunset Sunrise Activity Period LI LITT Leave den Return to den n; Leaving den INN UND Range n, Returning to den Fig. 6. Aardwolf activity periods during the year relative to sunset and sunrise. Data include only those observations when aardwolves were seen to emerge from, or retire to, their dens Food consumption and seasonal variation in the diet of the aardwolf 317 before dawn. The winter pattern appears more flexible, with aardwolves retiring to their dens on some nights at times earlier than they had left the den on others. All these variations can be explained in terms of temperature and the activity of termites. The total relative values of Hodotermes eaten per night during winter are illustrated in Fig. 7. There are a number of conclusions one can draw from this figure: 1. Hodotermes were sometimes active on days when it was too cold for Trinervitermes that night (daily maximum temperature <18°C). 2. Within the temperature range that Flodotermes were observed to be active (daily maximum = 13-28 °C), they were still more frequently inactive than actıve. By contrast, Trinervitermes were always active ın the evening if the daily maximum was over 18°C. 3. Although the mean feeding value for 1984 was greater than that for 1983, this difference was not significant (Mann-Whitney U test; W=124, n1=9, p>0.1), nor was the frequency of days on which Hodotermes were active (y’=0.23, df=1, p>0.1). The general conclusion one can draw from this figure, is that although Fodotermes does sometimes provide an alternative food source for aardwolves during winter, it is not very reliable as such, nor does it provide the same amount of food per night as does Trinervitermes during summer. 1983 x1000 1984 nı=9 n2=15 w = — < > x © z [=} w w m r Ww > < | w cc (6) 10 20 30 40 0 10 20 30 40 DAILY MAXIMUM TEMPERATURE °C Fıg. 7. Total relative values of Hodotermes consumed on days and nights during winter 1983 and 1984. Data include nights when aardwolves also fed on Trinervitermes, but are only for the days when one anımal was followed for its entire activity period. The number of days on which Hodotermes were actıve (nl) and inactıve (n2) are indicated Faecal analysis The mean percentages of Trinervitermes and Hodotermes present in the 136 aardwolf defaecations collected from middens are ıllustrated in Fig. 8. In all the samples, Triner- vitermes and Hodotermes together accounted for almost the entire number of food items present, wıth other species of termites or insects comprising only traces. It ıs clear that, apart from during the winter months, Trinervitermes was practically the only item in the diet. This point is ıllustrated in Table 6, which shows that Trinervitermes comprised over 318 P. R. K. Richardson DIET OF PERCENTAGE MEAN Fig. 8. Percentage occurrence of Trinervitermes and Hodotermes in 136 aardwolf defaecations collected over two years 99% (by volume) of the diet in 75 % of the faeces collected and was totally absent in only 1.4% of them. It should be noted when interpreting Fig. 8, that it represents only the mean percen- tages of species present in each defaecation, and therefore does not take into account the volume of the faeces, which was gen- Table 6 erally much smaller during winter. Thus the overall proportion of Hodotermes in the annual energy budget is given an inflated value, although such a value may reflect its Defaecations % volume of faeces containing Total importance in winter when other Irinervitermes s E N nen Sonnen foods are less available. In testing ET THREE REGEN TE RES whether there was a significant ın- Number 2 20 102 60 136 crease of Hodotermes in the diet be- Percentage I 2 x tween 1983 and 1984, the data for June, July and August for the two years were used, comparing the fre- quency of faeces in which Triner- vitermes or Hodotermes comprised the majority of the defaecation (Table 6). On this basıs, there was a significant increase (p<0,001) of Hodotermes in the diet during 1984. Composition of 136 aardwolf defaecations collected from middens during 1983 and 1984 ' Traces of other species present. Food consumption Estimates of the total amount of Trinervitermes eaten by aardwolves per night and per year, are based entirely on the total relative feeding values and daily maximum temperature (Fig. 3). Because the results for the all night-studies showed that there were four distinct levels of feeding according to maximum temperature (I=<18°C, 1834°C), the mean values for each temperature range were taken as representative values for the whole year. These values were then used to calculate the total Trinervitermes consumption per year, using the data on the daily maximum temperatures for the last two years of the study (September 1982-August 1984). The frequency distribution of the four temperature ranges per month are illustrated in Fig. 9. The average monthly consumption of an adult aardwolf (Fig. 10A) was calculated by multiplying the frequency of these four temperature ranges per month by their respective relative feeding values (Fig. 3) then summing them. This figure (B) also illustrates the means of the original feeding values, recorded on nıghts when the entire activity period of an aardwolf was monitored. The mean values per month (B) and the values derived from the temperature distribution (A) are highly correlated (r=0.94, df=7, p<0.001) and their absolute values are also very similar, although the latter have lower peaks. Therefore these latter values (A) have been assumed to give a reasonable estimation of the Trinervitermes consumption of an adult aardwolf per year. Ihe mean consumption of Hodotermes per month is illustrated in Fig. 10B. It is noticeable that although the maximum values during winter may be very high, because of the relatively low frequency at which Hodotermes were active (Fig. 7), the mean values consumed per month were always lower than those ot Trinervitermes. Thus although Hodotermes may, on some days, provide an alternative food source to Trinervitermes, on average ıt still provides less, and possibly helps little to alleviate the food shortage the aardwolf experiences during winter. The estimation of the total number of Trinervitermes consumed per year is derived from the data above and the total number of termites counted in a day’s faeces. The total dry weight of the complete day’s faeces that was used for counting termites, was 853 g. Of this 10.8 g was sampled from five separate defaecations and 1,297 soldiers were counted. This gave a total of 99,400 soldiers and, using a mean soldier/worker ratio of 0.32 (RICHARDSON 320 P. R. K. Richardson 60 50 40 30 X1000 20 VALUE [= a 2 > zZ — — > 7} [o) z o FEEDING 20 30 RELATIVE 40 1 50 60 70 B UT Hodotermes Fig. 10. Mean monthly consumption of Trinervitermes by an adult aardwolf during the year. A: data calculated from the temperature frequency distribution (Fig. 9). B: Mean values (with ranges) from nights when aardwolves were followed for the entire activity period. There were no data for October or November, and the September night was incomplete (only until midnight). Feeding on Hodotermes in winter is also indicated 1985), a grand total of 309,600 termites for the night. This figure may appear re- markably high, however, it compares well with estimates from two other sources: (1) The total numbers of Triner- vitermes ın two aardwolf stomachs taken from indi- viduals killed by motor vehic- les were 249,000 and 270,000 (RıiCHARDSON 1985). (11) On the night the faeces was col- lected the aardwolf (Apollo) fed on 396x2=792 termite patches (Table 4). This would give an average of 309,600/ 792=375 termites per patch. As on another occasion, over 500 termites were collected from two patches at which aardwolves had already fed, and over 4000 from some un- disturbed patches (RICHARD- son 1985), this result is clear- ly quite possible. A major assumption in the above calculation is that the total defaecations during a night (excluding the first of day 1 and taking the first of day 2) are all derived from the same night’s feeding. It was not possible to prove this, but providing there are no great fluctuations in temperature, the amount of termites eaten on consecutive nights will be fairly constant. Similarly, the production of faeces should remain constant, so any 24 hour accumulation of faeces should give a faırly accurate estimate of food eaten during that period. The faeces sampled above were collected on a night when the total relative feeding value was 41,600. As the total consumption per year is 14,100,000 (Fig. 10), this gives an estimate of: WEIGHT BODY 10 Food consumption and seasonal variation in the diet of the aardwolf 924 14,100,00 x 309,600 11.600 = 104.9 = 105 million termites per year The average standing crop of Trinervitermes per territory is approximately 160 million termites (RICHARDSON 1985), and therefore each adult pair (without the cubs) consumes about 1.4 times the standing crop per year. Most termites, in fact most invertebrates, have a production/biomass ratio (P/B) of about three (WooD and Sanns 1978). If this is also true for 7. trinervoides, then it means that aardwolf predation alone could account for more than half the total production per year before even considering other energy losses, such as the production of alates, respiration and other mortality factors. This point illustrates the importance of these two species to each other and is discussed in another paper on aardwolt territoriality (RICHARDSON in prep.). Body weight The dependence of aardwolves on Trinervitermes as their main food source is ıllustrated by their decline in body weight during winter (Fig. 11). Although there are insufficient data for statistical analysis, there is a clear trend amongst both adults and cubs to lose weight during winter. The males were slow to gain weight after winter, but this ıs probably because they spent a great deal of time guarding cubs at the den while the female was away foraging (RICHARDSON 1985). These results do not take into account the weight of stomach contents. However, because most of the aardwolves were darted after they had been feeding for a few hours (they are most easily darted while they are feeding), there was probably little difference in the masses of their stomach contents, and certainly not sufficient to produce the 2.2 kg difference which was recorded tor one male between May and September 1984. Adults WINTER CUBS AT THE DEN J F M A M J J A S (0) N D Fig. 11. Change in body weight of aardwolf adults and cubs during the year and in relation to periods ot stress. Sample sizes and ranges are indicated 322 P. R. K. Richardson Cub survival The survival of cubs born during the three years of the study is also regarded as being indicative of the feeding conditions. As cubs were first seen after they were about eight weeks old (RicHArDson 1985), these data are taken to represent the numbers of cubs born — even although some may already have died. Cubs first start feeding on termites when they are about three and a half months old, but continue to suckle to a maximum age of five months (RıcHArpson 1985). The total number of cubs that survived for 12 months are presented in Table 7. All cubs survived for at least five months but those that died, disappeared during the winter. Significantly more cubs died during 1984 (p<0.001, n=6) than did in the previous two years (n=1). This last year was towards the end of what has been the worst drought in the past 60 years in South Africa (Lunnp 1983), and during this period there was a significant die-off of Triner- Table 7 vitermes mounds in certain parts of the study Survival of cubs up to 12 months of age area (RICHARDSON 1985). These results show that winter must be a time of food stress for aardwolves. The adults Survived de appear to cope because, like many other mam- mals (Hanks 1981; Kruuk and ParısH 1983), 1981-3 10 they are able to utilise their fat reserves. By the 1983/4 5 ; end of summer aardwolves have considerable Fisher exact probability test: p < 0.001 amounts of thick, yellow fat in the abdomen (ERASMUS pers. comm.; pers. observation). The cubs, however, may not be able to build up such large reserves of fat and may be more dependent than the adults on having food during the winter. In this respect it is interesting to note that there was a higher incidence of Hodotermes in the faeces of aardwolves during 1984 than there was in 1983, which may reflect a relative decrease in the availability of Trinervitermes during 1984. The combined effects of this, and of the winter, may have been the cause of the high incidence of cub mortality during this year. Discussion The above results confirm the importance of Trinervitermes (and secondarily Hodotermes) in the diet of the aardwolf as found by Kruuk and Sans (1972) in East Africa, and CoOoPER and SKINNER (1979), BOTHMA and NEL (1983) and SMITHERS (1983) in southern Afrıca. However, in southern Africa none of these authors collected faeces on a regular monthly basis, and consequently did not notice the pronounced seasonal change which clearly relates to the cold temperatures prevalent during winter. Before pursuing this aspect, it is necessary first to consider an important physical difference between these termites, viz. pigmentation. In order to test their tolerances of ultra-violet radiation, Hewıtt et al. (1972) conducted an experiment whereby they exposed workers of T. trinervoides and Hodotermes to two hours of direct sunlight. The experimental containers were maintained at a temperature of 20-25 °C during exposure and afterwards the termites were provided wıth ample drinking water. All the Trinervitermes, which are poorly pigmented, died within 24 hours of exposure. By contrast, only 21% of the darkly pigmented Hodotermes died. This was to be expected, because Hodotermes are known to be most active during the daylıght hours of winter and late autumn (NEL and HEwITT 1969). This therefore explains why Trinervitermes is never actıve by day (except occasion- ally in the late afternoon), and thus why Hodotermes features so prominently in the aardwolf’s diet during winter. Temperature is clearly the most important single factor influencing the availability of Food consumption and seasonal variation in the diet of the aardwolf 323 Trinervitermes during the year. Being unable to tolerate direct sunlight 7. trinervoides is almost exclusively a nocturnal forager. On nıghts when the temperature at sunset is <18 °C virtually no foraging takes place, and therefore, throughout most of winter (May-August), Trinervitermes ıs seldom available to aardwolves. This places considerable stress on these anımals. This was evident from the weight loss experienced by all aardwolves during winter, and which probably caused the death of most of the cubs that died in their first year. These last two features suggest a great dependance by the aardwolf on Trinervitermes and also an apparent inability to feed successfully on alternative sources of food. Although Hodotermes was an important substitute for Trinervitermes during winter, it was an irregular forager and therefore could not entirely make up the deficit. Furthermore, the aardwolt appears to be remarkably inefficient at feeding on insects other than harvester termites. During periods of low termite abundance aardwolves were occasionally seen to snap at locusts or antlions, but were never seen to eat substantial amounts of these insects. Thus the aardwolf appears to be so specialised at feeding on termites that it does not really regard other insects as food. This may be related to the extra energy expenditure required to catch more mobile insects, but ıs nevertheless in marked contrast to the bat-eared fox Otocyon megalotis, which feeds extensively on Hodotermes but is also most successtul at catching various insects and arachnids, and digging up beetle larvae (NEL 1978; SMITHERS 1983; personal observation). Some explanation for this situation can possibly be found in the extreme specialisation of the aardwolf’s tongue, which ıs very large, flat and spatulate. The peculiarity of this tongue is reflected in its use. When feeding a captive aardwolf, Spineıry (1970, p. 82) stated that “It seems to have great difficulty in feeding on what one would regard as ‘normal’ tood because it uses its tongue, which is coated with sticky salıva, to pick up its food. The difficulties appear to be so great that it soon stops feeding altogether.”’ SPINELLY overcame this problem by feeding the aardwolf small balls of mince held in the palm of her hand. Thus it would appear that the aardwolf is highly efficient at licking up small insects off the soil surface, but thıs is at the expense of being able to handle larger items ot food. As mentioned earlier, Kruuk and Sanps (1972) have postulated that ıt ıs ıts extreme specialisation for certain types of Trinervitermes that is responsible for the aardwolf being absent from Central and West Africa. In East Africa the western boundary of the aardwolf’s distribution (SMITHERS 1983) approximately follows that of 7. bettonianus (Sanps 1965), which is the aardwolf’s main prey species in that region (Kruuk and SanDs 1972). In the savanna regions of West Africa, T. geminatus is a common harvester termite, so ıt ıs not immediately obvious why the aardwolf cannot survive there. The answer appears to lie in the foraging behaviour of the termites, and how well their harvesting habıt (storing grass underground in nests) is developed. All harvester termites prefer to collect mature or dry grass rather than green grass, which is toxic to some species (SanDs 1961; BoTHA and HewıtT 1978; OHıaGu 1979). Therefore, as green grass ıs most abundant during the wet season, this is when the true harvesters are least active. This ıs well documented for Hodotermes in southern Africa (NEL and Hewırr 1969), and for T. geminatus in West Afrıca (Sanns 1961; OH1aGu 1979). In northern Nigeria T. geminatus ıs largely inactive for three months during the wet season (July-September) and then again for four months during the cold dry season (Novem- ber-February) (Sanps 1961). In southern Nigeria the wet season lasts very much longer, and virtually no harvesting takes place from May to October (OH1Acu 1979). Presumably the absence of harvester termites for sıx months of the year would place intolerable stress on the aardwolf (particularly the cubs), and thus prevent it from living in West Africa. Although T. bettonianus and its ecological equivalents in southern Africa do harvest grass, they do so to a far lesser extent than 7. geminatus and rely more heavily on browsing grass throughout the year (Kruuk and Sanns 1972). Because of this they cannot afford to remain inactive for extended periods, so they forage throughout the year except when 324 P. R. K. Richardson weather conditions inhibit ıt. They therefore provide a highly reliable food source - to such an extent that ıt has permitted the aardwolf to specialise almost exclusively on these few species. Acknowledgements I would like to thank De Beer’s Consolidated Mines for access to their game farm, Benfontein; for the use of computer facilities, and for technical help from varıous members of staff. This project received financial support from the following organısations or funds: The British Council, The Cape Department of Nature and Environmental Conservation, The Endangered Wildlife Trust, Johannes- burg Consolidated Investment Company, Kango Wolf Power Tools, Wolf Garden Machinery Manufacturers, Oliver Ashe Overseas Scholarship Fund and Overseas Research Student Awards. Mrs. D. Fock translated the summary and Drs. M. Cox, P. H. HEwITT, J. D. SKInNEr, and O. A. E. Rasa commented on earlıer drafts of the manuscript. Zusammenfassung Nahrungsbedarf und saisonale Variabilität in der Ernährung des Erdwolfes Proteles cristatus in Südafrika In Südafrika lebt der Erwolf Proteles crıstatus fast ausschließlich von 2 Arten der an der Oberfläche ihre Nahrung findenden Grasschneidertermiten Trinervitermes trinervoides und Hodotermes mossambicus. Den größten Teil des Jahres ist der Erdwolf nachts aktiv und ernährt sich von Trinervitermes, die nur schwach pigmentiert ıst und kein direktes Sonnenlicht verträgt. Wenn die Temperatur im Winter bei Sonnenuntergang <18°C beträgt ist Trinervitermes nur eine spärliche Nahrungsquelle für den Erdwolf. Hodotermes ist dunkel pigmentiert und im Spätherbst und Winter hauptsächlich tagsüber aktıv. Auch der Erdwolf wird ım Winter mehr zu einem Tagtier und ergänzt seine Nahrung durch Aufnahme von Hodotermes. Trotzdem leidet der Erdwolf noch unter Nahrungsmangel. Das beweist seine Gesichtsabnahme (bis zu 20 % des Körpergewichts), die bei allen Tieren im Winter auftritt. Nahrungsmangel ist sehr wahrscheinlich die Todesursache bei den meisten Jungen, die im 1. Lebensjahr eingehen. In ganz Afrika lebt der Erdwolf vorwiegend von einigen Arten der Trinervitermes. Es wird daher angenommen, dafs das Vorkommen des Erdwolfs vom auftreten von Trinervitermes abhängt. Diese bestimmt die Verbreitung des Erdwoltes in Afrika. References Acocks, J. P. H. (1975): Veld types of South Africa. (2nd Edn). Mem. Bot. Surv. sth. Afr. No. 40. BoTHa, T. C.; Hewrrr, P. H. (1978): Influence of diets containing green material on laboratory colonies of Hodotermes mossambicus (Hagen). Phytophylactica 10, 93-97. BoTHMa, J. Du P.; Ner, J. A. J. (1980): Winter food and foraging behaviour of the Aardwolf Proteles cristatus in the Namib-Naukluft Park. Madoqua 12, 141-145. BRAEKMAN, J. C.; DaLozE, D.; DuponT, A.; PASTEELS, J. M.; Josens, G. (1984): Diterpene composition of defense secretion of four West African Trinervitermes soldiers. J. Chem. Ecol. 10, 1363-1370. CoOPER, R. L.; SKINNER, J. D. (1979): Importance of termites in the diet of the aardwolf Proteles cristatus in South Africa. S. Afr. J. Zool. 14, 5-8. DELIGNE, J.; QUENNEDY, A.; Brum, M. B. (1981): The enemies and defence mechanisms of termites. In: The socıal insects. Ed. by H. R. Hermann. New York: Academic Press, Vol. II, 2-76. EisnEr, T.; KrısTon, I.; ANESHANSLEY, D. J. (1976): Defensive behaviour of a termite (Nasutitermes exitiosus). Behav. Ecol. Sociobiol. 1, 83-125. FLowEr, W. H. (1869): On the anatomy of Proteles cristatus (Sparrman). Proc. Zool. Soc. Lond. 474—4%. Hanks, J. (1981): Characterisation of population condition. In: Dynamics of large anımal popula- tions. Ed. by C. W. Fowrer and T. D. SmitH. New York: J. Wiley and Sons. 47-73. Hewitt, P. H.; Ner, J. J. C.; Schorman, I. (1972): The solar and ultraviolet radıiation tolerances of several termite species. J. ent. Soc. sth. Afr. 35, 119-121. Kıncoon, J. (1971): East African Mammals. London: Academic Press, Vol. 1. Kruur, H.; Parish, T. (1983): Seasonal and local differences in the weight of European badgers (Meles meles L.) in relation to food supply. Z. Säugetierkunde 48, 45-50. Kruuk, H.; Sans, W. A. (1972): The aardwolf (Proteles cristatus Sparrman, 1783) as a predator of termites. E. Afr. Wildl. J. 10, 211-227. LoNGHUuRsT, C.; Jonunson, R. A.; Woon, T. G. (1978): Predation by Megaponera foetens (Fabr.) (Hymemoptera: Formicidae) on termites in the Nigerian southern Guinea savanna. Oecologia 32, 101-107. Food consumption and seasonal variation in the diet of the aardwolf 325 LONGHURST, C.; JOHNSON, R. A.; Woop, T. G. (1979): Foraging, recruitment and predation by Decamonium uelense (Sanstchi) (Formicidae: Myrmicinae) on termites in southern Guinea savanna, Nigeria. Oecologia 38, 83-91. Lugın, Y. D.; MONTGOoMERY, G. G. (1981): Defenses of Nasutitermes termites (Isoptera, Termitidae) against Tamandua Anteaters (Edentata, Myrmecophagidae). Biotropica 13, 66-76. Lunp, B. G. A. (1983): How bad is the present drought? S. Afr. J. Scı. 79, 440-441. MeErrton, D. A. (1976): The biology of the aardvark (Tubulidenta-Orycteropodidae). Mamm. Rev. 6, 75-88. NEL, J. A. J. (1978): Notes on the food and foraging behavior of the bat-eared fox Otocyon megalotıs. Bull. Carneg. Mus. nat. Hist. 6, 132-137. NEL, J. J. €. (1968): Die invloed van temperatuur en relatiewe humiditeit op die gewigsverlies en oorlewing van die grasdraertermiete Hodotermes mossambicus (Hagen) en Trinervitermes triner- voides (Sjostedt). S. Afr. Tydskr. Landbouwet. 8, 151-155. Ner, J. J. C.; HewItT, P. H. (1969): A study of the food eaten by a field population of the harvester termite, Hodotermes mossambicus (Hagen) and its relation to population density. ]J. ent. Soc. sth. Afr. 32, 123-131. OHnacut, C. E. (1979): A quantitive study of seasonal foraging by the grass harvesting termite, Trinervitermes geminatus (Wasmann), (Isoptera, Nasutitermitinae) in southern Guinea savanna, Mokwa, Nigeria, Oecologia 40, 179-188. PRESTWICH, G. D. (1983): The chemical defenses of termites. Sci. Am. 249, 68-75. REDFORD, K. H.; Dora, J. G. (1984): The nutritional value of invertebrates with emphasis on ants and termites as food for mammals. J. Zool., Lond., 203, 385-395. RICHARDSoNn, P. R. K. (1983): An improved darting system for immobilizing smaller mammals in the wild. S. Afr. J. Wildl. Res. 13, 51-54. RıcHARDson, P. R. K. (1985): The social behaviour and ecology of the aardwolt, Proteles cristatus (Sparrman, 1783) in relation to its food resources. Unpubl. D. Phil. thesis, Univ. Oxford. SanDs, W. A. (1961): Foraging behaviour and feeding habıts in five species of Trinervitermes in West Africa. Ent. exp. appl. 4, 277-288. Sands, W. A. (1965): A revision of the termite subfamily Nasutitermitinae (Isoptera: Termitidae) from the Ethiopian region. Bull. Br. Mus. nat. Hist. Entomol. Suppl. 4, 1-172. SMITHERS, R. H. N. (1983): The mammals of the southern African subregion. Pretoria: Univ. Pretoria. SPINELLY, N. (1970): Hand-rearing an aardwolf, Proteles cristatus. Int. Zoo Yb. 10, 82-83. Woop, T. G.; Sanns, W. A. (1978): The role of termites in ecosystems. In: Production ecology of ants and termites. Ed. by M. V. Brıan. Cambridge: Cambridge Univ. Press, 245-292. Author’s address: P. R. K. RıcHArpson, D. Phil, Mammal Research Institute, University of Pretoria, Pretoria 0002, South Africa BIISENBESPIRIEETLUING EN Harroıv, D. C. S.: The mammals of Nigeria. Oxford: Clarendon Press 1987. 402 S., zahlreiche Abb. und Tab., 10 Fototafeln, 112 Verbreitungskarten. Leinen £ 75.00. ISBN 0-19-857565-3 Der Autor, der viele Jahre im Sudan und in Nigeria verbracht hat, legt hier eine sorgfältige und inhaltsreiche Säugetierfauna vor, die er zugleich tiergeographisch-ökologisch analysiert. Auf 220 Seiten werden die 247 Arten Nigerias in systematischer Folge unter folgenden Gesichtspunkten behandelt: Allgemeine Verbreitung, Fundorte in Nigeria (dazu Karten und ein Verzeichnis der Fundorte mit ihren Koordinaten), Häufigkeit und Verteilung („Status“), Merkmale, Okologie, Fortpflanzung und taxonomische Bemerkungen. Als Maße finden sich überwiegend nur kursorische Angaben zu den üblichen Körpermaßen, Gewicht und Schädellänge, teils als Mittel- und Extrem- werte, oft auch nur „typische Werte“ oder nur Extrema. Mit wenigen Ausnahmen beschränken sich die Beschreibungen auf äußere Kennzeichen. Zahn- und Schädelmerkmale werden nur ganz aus- nahmsweise angegeben. Die innerartliche Variabilität wird nicht behandelt. Dafür gibt es Bestim- mungsschlüssel zu den Familien, Gattungen und Arten, die im allgemeinen recht brauchbar erschei- 326 Buchbesprechungen nen. Die Angaben zur Verbreitung sind besonders wertvoll, weil jeder Fundort mit Quelle und Beleg aufgeführt, kartiert und aufgrund der Koordinatenangaben auch leicht auffindbar ist. In den allgemei- nen Kapiteln werden die Säugetiere der verschiedenen Vegetationszonen nach unterschiedlichen Gesichtspunkten klassifiziert, etwa im Regenwald nach Körpergewicht, Ernährungsweise, Vertikal- vorkommen, oder nach allgemeinem Verbreitungstyp. Die Fauna wird mit der anderer afrikanischer Länder verglichen und ist gemessen an der Landestläche reichhaltig. Endemismus ist im Regenwald- Anteil am höchsten. Die Klimageschichte in den letzten 40 000 Jahren und ihr möglicher Einfluß auf die Differenzierung der Säugetierfauna wird ebenso behandelt wie Faunengrenzen, die Biomasse der Säuger in den verschiedenen Vegetationszonen und die Einpassung der Fortpflanzung in den Jahreslauf. Ein weiteres Kapitel ist dem Einfluß des Menschen gewidmet, dessen Bevölkerungszahl zur Zeit jährlich um 3,3 % (etwa 2,3 Millionen) wächst. Die Rolle der Säuger als Nahrungsquelle, Jagdwild, Schädlinge, Überträger von Krankheiten werden behandelt, Nationalparks und Wildreser- vate in einer Tabelle unter Angabe ihrer Fläche zusammengefaßt und auf einer Karte dargestellt. Im Anhang finden sich u. a. ein Glossar, eine Begründung für die Entfernung von Arten von der Liste nigerianischer Säugetiere durch RosEvEAR (1953) und eine Tabelle der nigerianischen Arten mit Angabe ihres Vorkommens in 16 westafrikanischen Ländern. Da die wichtigsten Ordnungen, die Fledertiere, Nager und Raubtiere durch RosEvEAR (1965, 1969, 1975) ausführlicher taxonomisch behandelt wurden, die Spitzmäuse durch HUTTERER und Harroıo (1983), ıst die Beschränkung von Morphologie und Systematik ım vorliegenden Band kein Nachteil. Etwas mehr Sorgfalt hätte man sich in den zu den höheren Taxa führenden Abschnitten gewünscht. Die Merkmale beziehen sich offenbar nur auf die in Nigeria nachgewiesenen Arten (Stacheln bei Igeln, zusammengedrückter Schwanz bei Otterspitzmäusen, Zahnformel von Lagomor- phen und Musteliden). Warum dann aber 28-44 Zähne bei den Spitzmäusen? Von solchen kleinen Schwächen abgesehen ist das Buch jedoch dank der übersichtlichen Gliede- rung, der gleichbleibenden Anordnung in den Artenkapiteln, der straffen Darstellung, in Druck und Qualität der Abbildungen als afrıkanische Landesfauna mit Betonung von Okologie und Faunenge- schichte vorbildlich. J. NIETHAMMER, Bonn OLsen, S$. J.: Origins of the Domestic Dog. The Fossil Record. Tucson: Univ. Arizona Press 1985. 118 pp., 39 fıgs. US $ 19.95. ISBN 0-8165-0909-3 Among all domesticated mammals dogs play a special role because of their unique relationship to man. Especially the diversity of modern races is an outstanding phenomenon as it shows an immense intraspecific varıability in morphological as well as in physiological features, which has not been reached to that extent by breeding of other species. These attributes can, to a certain degree, serve as a hint to the manifold human needs although dogs are mainly important as pets in modern civilizations. The origin of dogs as well as the time and the place of their domestication have already in the past puzzled scientists for a long time, and there still remain some unsolved problems to date as well. The book in question deals with these problems from an archaeozoological point of view as it describes the fossil record of dogs world wide. It is written by $. J. OLsEn, an expert in this field who studied most of the ancient finds himself and with help of his son J. W. Orsen. The diverse chapters are entitled as follows: Fossil ancestry of Canis; tamıng and domestication of early wolves; early prehistoric dogs in North America; small pleistocene wolves of China as ancestral dogs; prehistoric dogs in mainland East Asıa; prehistoric dogs ın Europe and the Near East; canıds of uncertain taxonomic status; multivariate analysıs of sample fossils. All in all the book is well written and especially well illustrated showing a great amount of photographs of ancient skulls and bony fragments. Reinvestigations of finds and critical evaluations concerning the question on wolf versus dog remains are made accordingly. The wolf ancestry of dogs is again recorded excluding an origin from other canid species. Several newer finds especially those from China and Asıa are included. In general earliest dog remains were dated between approximately 10,000 B.P. and 8,000 B.P. at different sites in China, Kamtschatka, North America, Near East and Northern Europe. In 1984 the oldest domestic dog find in the world has been reported from the site of Palegawra cave in northeastern Iraq. It was dated to 12,000 B.P. However, this is based on a single small mandibular fragment and seems doubtful until additional finds of comparable time will assure this. The importance of dogs in early human communities and cultures are stressed in the book, and the dingo problem ıs discussed as well. There are only some crıtical remarks to be made. E.g. the author tends to originate dogs from especially small sized wolf subspecies although there is necessarily no proof for this. Decrease in size of ancient domestic forms compared with their wild living progenitors also occurs in several other mammalıan species and may likewise be of consequence within a few generations of selective breeding. On the other hand, some ancıent skulls or skull fragments are assumed to be remains of tamed wolves rather than of dogs. For the reviewer this statement also seems to be doubtful because tamed or captive wolves can hardly be differentiated from wild living individuals on bony materials. These general remarks may show certain limits of archaeozoological research. However, from the book we learn that about 10,000 years ago domestic dogs were already Buchbesprechungen 327 spred throughout several regions of the world. Unfortunately, the initial phase of the domestication process and the creation of dogs as well as the duration of wolf to dog change still remain open questions. D. Kruska, Kiel Guraya, $. S.: Biology of Ovarian Follicles in Mammals. Berlin, Heidelberg, New York, Tokyo: Springer 1985. 320 pp., 76 figs. DM 198,-. ISBN 3-540-15022-6 Die Vielseitigkeit des Ovars als Bildungsstätte der weiblichen Keimzellen und als Bildungsstätte von Hormonen ist weitgehend an die Follikel gebunden. Das Verständnis der komplexen physiologischen Vorgänge, deren dauernder Wechsel und ihr Fortdauern bis in die postnatale Lebensphase setzt eine Integration der Ergebnisse vieler Spezialdisziplinen voraus, die nur selten erreicht wird. Das vorlie- gende Buch ist der Versuch zu einer derartigen Synthese. Es gliedert sich in Kapitel über 1. Primärfol- likel, Stromazellen; 2. Follikelwachstum, Oocyte, Zona pellucida, Theca und Granulosa, Follikel- liquor; 3. Eireifung; 4. Ovulation; 5. Luteinisierung, Steroidgenese während der präovulatorischen und ovulatorischen Phase; 6. Follikelatresie. Auf einem Gebiet, das derartige Beachtung wie dieses findet und praktische Bedeutung für Human- und Veterinär-Medizin, Züchtungsbiologie, Genetik und allgemeine Biologie hat, ist zu erwarten, daß nur ein Zwischenbericht über das bisher Erreichte vorgelegt werden kann, denn vieles ist im Fluß. Ein solcher Bericht ist aber natürlich von größtem Wert für alle, die mit Reproduktions- problemen befaßt sind. Die Probleme werden auf verschiedenen Ebenen behandelt (biochemisch, histochemisch, ultra- strukturell-Jichtmikroskopisch). Der Text ist klar und übersichtlich, die Abbildungen sind vorzüg- lich. Befunde an verschiedenen Säugerarten werden berücksichtigt, doch neigt der Autor zu generali- sierter Betrachtung und übergeht phylogenetisch-taxonomische Gesichtspunkte. Als Informations- quelle über die genannten Aspekte ist das Buch jedoch zu empfehlen. Besonders verwiesen sei auf das sehr umfangreiche Schrifttumsverzeichnis von 30 Seiten. D. STARcK, Frankfurt/M. BOECKH, J.; PFANNENSTIEL, H.-D. (Hrsg.): Zoologie 1985. Bilanz und Perspektiven. Stuttgart: G. Fischer 1986. 56 S. DM 14,-. ISBN 3-437-30526-3 In der vorliegenden Schrift wird versucht, nach dem Gespräch einiger Zoologen in Bad Wildbad 1974 (herausgegeben von H. Markr) Bilanz zu ziehen und den heutigen Standort des Faches Zoologie erneut zu bestimmen. Die Gespräche auf einem neuen Treffen in Berlin (Februar 1985) werden in fünf Beiträgen zusammengefaßt. H. ALTNER berichtet über die Situation aus wissenschaftspolitischer Sicht. R. WEHNER legt einen Vergleich der Zoologie in der Bundesrepublik mit der Situation im Ausland, allerdings beschränkt auf USA, England und Schweiz, vor. RATHMAYER behandelt die Auswirkungen der Gespräche von 1974 mit der heutigen Lage. D. von Horst stellt in nüchterner, sachlicher und kritischer Weise zusammen, was aus der Sicht zoologischer Forschung zu der Debatte um die Tierschutzgesetzgebung zu sagen ist. Bilanz und Aussichten der verschiedenen Teildisziplinen werden ausführlich von den beiden Herausgebern untersucht. Offensichtlich war das Hauptthema der Tagung, das kontrovers diskutiert wurde, die Frage, ob die Zoologie als übergreifendes Fach, definiert durch ihren Forschungsgegenstand, das Tier, wie es Mark 1974 hervorhob, erhalten bleiben soll und kann oder ob, erzwungen durch methodische Spezialisierung von Teildisziplinen, eine Abspaltung von mehr und mehr Teilgebieten bis zur Auflösung der Zoologie als wissenschaftliche Disziplin erforderlich, nötig oder wünschenswert wäre. WEHNER hat die Situation in den genannten Ländern beschrieben und auf die absolute Dominanz der experimentellen Teilfächer und ihre Abwanderung (eigene Zeitschriften, Kongresse), aber auch die Verarmung der zoologischen Kongresse in den angelsächsischen Ländern und in der Schweiz hingewiesen. Dabei fällt auf, daß? umfangreiches Material nur aus diesen drei Ländern vorgelegt, die Situation in der Bundesrepublik aber nur sehr summarisch dargestellt wird. WEHNER und auch ALTNER übergehen in ihren Berichten die Tatsache, daf neben den methodisch spezialisierten experimentellen ‚„‚Biosciences“ ein bedeutender Komplex im Rahmen der Zoologie übrig bleibt, der sich der vergleichenden Methode und der Erfassung komplexer Zusammenhänge auf allen Ebenen bedient (Systematik, vergl. Morphologie, Phylogenie) und sich zugleich eine Vielzahl technischer Verfahren zunutze macht. Die Ergebnisse sind gerade in den verflossenen 10 Jahren beachtlich und demonstrieren die Verankerung der experimentellen Teildisziplinen in einer zentralen ‚Wissenschaft vom Tier‘ (z.B. im deutschen Sprachbereich 1983 und 1986 zwei gut besuchte Internationale Tagungen über „Functional Morphology in Vertebrates‘“‘). Die Gefahr, daß eine totale Ablösung der modernen Zweigfächer vom Stamm zu einem neuen Reduktionismus führen muß, wird offenbar von einigen Fachkollegen nicht gesehen, wenn auch nicht verschwiegen werden soll, daß die Herausgeber in ihrem speziellen Report über die Teilgebiete einen sachlichen Standpunkt vertreten. Die Forderung nach einer einheitlichen Wissenschaftssprache „Englisch“ ist heute, auch in Deutschland, bereits weitgehend realisiert, wenn auch die sprachliche Ausbildung während des 328 Buchbesprechungen Studiums in Deutschland noch einige Wünsche offen läßt. Einführung eines Pflicht-Unterrichtes in zwei Fremdsprachen während des Studiums naturwissenschaftlicher Fächer wäre erstrebenswert. Objektive Argumente für Englisch als Wissenschaftssprache stehen reichlich zur Verfügung. Emotio- nale, polemisch wirkende Gesichtspunkte (WEHNER S. 16: „Wurde doch der englische Stil von dem Bemühen um Klarheit geprägt — der deutsche von dem Bemühen, unbeanstandet durch die Zensur zu kommen.‘“‘) dienen nicht der Verständigung. Der Konsens wird durch derartige unsachliche Verallge- meinerungen keineswegs gefördert. D. STARcK, Frankfurt/M. Heser, R.; STROMBERG, M. W.: Anatomy and Embryology of the Laboratory Rat. Wörthsee: Biomed-Verlag B. Hebel 1986. 271 pp. DM 146,-. ISBN 3-9801234-0-5 Es handelt sich um ein neu bearbeitetes Ergänzungswerk des 1976 erschienenen Buches „Anatomy of the Laboratory Rat“ (Baltimore: Williams and Wilkins Co.), das ın 13 Kapiteln die deskriptive Anatomie der Labor-Ratte behandelt und um einen Abschnitt über Embryologie ergänzt ist. Die Gliederung nach Organsystemen bringt eine ausführliche und genaue Beschreibung der makroskopi- schen Anatomie, vielfach ergänzt durch Einbeziehung mikroskopisch-anatomischer Übersichten. Jedem Kapitel ist ein umfangreiches Literaturverzeichnis beigefügt. Der Text wird durch zahlreiche, z. T. neue Abbildungen von ausgezeichneter Qualität ergänzt. Neues Schrifttum bis 1986 ist eingear- beitet. Besonders wertvoll ist die umfangreiche Bearbeitung des Nervensystems. Im embryologischen Teil werden Eireifung, Ovulation, Befruchtung, Furchung und Implantation besprochen. Es folgt ein Atlas von 29 Schnittbildern (Mikrophotos) von 17 Stadien von der Furchung bis zum 22. Tag (42 mm KRlänge). Der zugehörige Text faßt für jedes Stadium den Reifezustand nach Art einer Normentafel zusammen. Leider ist dıe Bebilderung der einzelnen Kapitel sehr ungleich. So vermißt Ref. ausrei- chende Illustration der Makromorphologie des Gehirns und bildliche Darstellung der ım Text besprochenen zentralnervösen Leitungsbahnen. Auch dürfte es für den Benutzer nützlich sein, wenn die Photos der frühen Embryonalphasen (Abb. N 4-7) durch erläuternde Situationsskizzen ergänzt wären. Das Buch wird als Nachschlagwerk, vor allem für den Experimentalforscher zweifellos von Nutzen sein, setzt allerdings Vertrautheit mit den Grundfakten der Morphologie und Ontogenie der Säugetiere voraus. D. STARcK, Frankfurt/M. GoODALL, Jane: The Chimpanzees of Gombe. Patterns and Behavior. Cambridge, Mass., London: The Belknap press of Harvard University Press 1986. 673 pp. US $ 29.95. ISBN 0-674-11649-6 Frau Goopaut legt mit vorliegendem Werk einen umfassenden und detaillierten Bericht über ihre bahnbrechenden Beobachtungen an einer freilebenden Schimpansengruppe im Gombe Wald Tansa- nıas vor. Diese Forschungen erstrecken sich nunmehr auf 26 Jahre und sind damit wohl die bisher am längsten währende, kontinuierliche Forschung an einer Population. Der sensationelle Erfolg der Autorin, die Akzeptanz und Gewöhnung der Schimpansen an die Anwesenheit der Beobachterin, setzte jahrelange, intensive und geduldige Arbeit unter extremen Bedingungen voraus. Das große Engagement, ihre Geduld, Einfühlungsgabe und Erfahrung haben es ermöglicht, eine weit gespannte Synthese vorzulegen, die von jedem Interessenten mit Gewinn gelesen wird. Die einleitenden Kapitel berichten in knapper Form über die Geschichte der Forschung über Menschenaften, speziell über Pan troglodytes, über den Anfang der eigenen Forschungsarbeit und über die Probleme der Kontaktfindung zwischen Pan und Mensch. Die eigenen Erfahrungen und Beobachtungen werden in 19 Kapiteln in sehr detaillierter Form vorgelegt, erläutert durch zahllose instruktive Photos (darunter 8 Farbtafeln). Im einzelnen werden folgende Teilprobleme behandelt: Individualität bei Wildtieren, Sozialstruktur, Kommunikation, Wandel der Gruppe in der Folge der Jahre, Nahrung und Jagd, Aggression, Dominanzverhalten, Sexualverhalten und dessen Entwicklung, Territorialität, Werkzeuggebrauch und kognitive Leistung. Zahlreiche Tabellen und Schemata ergänzen den Text. Das Buch enthält ein umfassendes Schrift- tumsverzeichnis, Namen- und Sachregister. Bedauerlich ist, daß der historische Bericht einige neuere Forschungen nur unzureichend berücksichtigt (KORTLANDT, SuGIyaMa). Die Karte über Schimpan- senhabitate und Verbreitung dürfte durch KorTLAnDTs Karte (1969-1983) ergänzungsfähig sein. . Als Dokumentation von GooDauıs Forschungen ist das Werk von großem Wert. Es wird über den Kreis der Spezialforscher hinaus auch bei einem weiten Publikum Interesse finden. Die Ausstat- tung ıst von bester Qualität, der Preis erstaunlich gering. D. STARcK, Frankfurt/M. Neu in der Reihe "Biologie und Evolution interdisziplinär": Psychozoikum | Evolution und Mechanismus der menschlichen Erkenntnisfähigkeit. Von Prof. Dr. Erhard Oeser, Institut für Philosophie, Wien. 1987. 208 Seiten mit 40 Zeichnungen. Gebunden DM 48,-- ISBN 3-489-63234-6 Unter dem Begriff "Psychozoikum" faßt Erhard Oeser seine mit Konrad Lorenz und Rupert Riedl entwickelten Ideen zu einem Zwei-Stufen-Modell der Evolutionären Er- kenntnistheorie zusammen. Dieses Modell besagt, daß selbst die einzigartigen Er- kenntnisleistungen des Menschen, die unsere heutige wissenschaftlich-technische Zi- vilisation hervorgebracht haben, ausgegangen sind von den Mechanismen des Infor- mationsgewinns, deren Grundlage allen Lebewesen gemeinsam ist. Selbst-Organisation Ordnungsprozesse in sozialen Systemen aus ganzheitlicher Sicht.Von Prof. Dr. Gil- bert Probst, Universität Genf, vormals Institut für Betriebswirtschaft der Hochschule St.Gallen für Wirtschaft und Sozialwissenschaften. 1987. 208 Seiten mit 10 Abbil- dungen. Gebunden DM 39,80 ISBN 3-489-63334-2 Man weiß seit langem, daß sich Menschengemeinschaften auch ohne Vorschriften, Organigramme oder Organisationsspezialisten strukturieren. Dazu haben die Biologen und die Physik in den letzten Jahren wesentliche Erkenntnisse gewonnen. Hier wird Selbst-Organisation als Konzept für die Entstehung und Entwicklung von Ordnung in sozialen Systemen beschrieben. Dies führt zu einem vollkommen neuartigen Um- gang mit sozialen Systemen. Die Aufgaben, Fähigkeiten und Vorgehensweisen des Organisierens auf materieller und sinngebender Ebene werden in diesem Buch darge- legt. . PAUL PAREY -» BERLIN und HAMBURG Erscheinungsweise und Bezugspreis 1987: 6 Hefte bilden einen Band. Jahresabonnement Inland: 298,- DM zuzüglich 13,80 DM Versandkosten, Gesamtpreis 311,80 DM einschließlich 7% Mehrwert- steuer. Jahresabonnement Ausland: 298,- DM zuzüglich 24,- DM Versandkosten. Das Abonne- ment wird zum Jahresanfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, Spitalerstraße 12, D- 2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesellschaft für Säugetier- kunde“ erhalten die Zeitschrift unberechnet im Rahmen des Mitgliedsbeitrages. Z. Säugetierkunde 52 (1987) 5, 265-328 Lehrbuch der Schafkrankheiten Von Prof. Dr. med. vet. Heinrich Behrens, ehem. Direktor des Tiergesundheitsamtes der Landwirtschaftskammer Hannover, apl. Professor an der Tierärztlichen Hochschule Han- nover. 3., überarbeitete und erweiterte Auflage. 1987. 312 Seiten mit 108 Abbildungen, davon 20 farbig auf 4 Tafeln. Kartoniert DM 98,- Das bereits in seinen beiden Vorauflagen als Standardwerk bewährte Lehrbuch liegt nun in der dritten, überarbeiteten und erweiterten Auflage vor. Der Verfasser, ein seit langem als Hochschullehrer, Wissenschaftler und praktischer Berater weithin bekannter Fach- mann, gibt eine umfassende Darstellung der Schafkrankheiten, ihrer Erkennung und Be- handlung sowie aller prophylaktischen Maßnahmen zur tierärztlichen Betreuung von Schafen in allen Haltungsformen. Dazu gehört in immer noch zunehmendem Umfang die Koppelschafhaltung, insbesondere in Form der Hobbyhaltung, die eine stärkere Inan- spruchnahme der praktizierenden Tierärzte und der tierärztlichen Untersuchungsämter mit sich bringt. Die Neuauflage enthält zusätzlich 4 Schwarzweiß- sowie 8 farbige Abbildungen auf 4 Tafeln. Es wurden 28 Krankheiten neu aufgenommen und, insbesondere für die Hobbyschaf- haltung, die Kälteeinwirkung auf neugeborene Lämmer sowie die mutterlose Lämmerauf- zucht und Lämmeraufzucht durch fremde Mutterschafe berücksichtigt. Die im Anhang zusammengefaßten Kapitel über klinische Normalwerte, Blutmeßgrößen und Altersbestim- mung stellen gleichfalls eine Erweiterung dar. Die Literatur wurde neu ausgewählt. Auf dem neuesten Stand steht das Lehrbuch der Schafkrankheiten nun wieder für einen großen In- teressentenkreis in Studium, Praxis und Beratung zur Verfügung. Aus Buchbesprechungen zur 2. Auflage: „Auch die 2. Auflage des Lehrbuches stellt den gegenwärtigen Stand des Wissens in sehr gestraffter und übersichtlicher, aber dennoch gründlicher Weise dar und ist für den Tierarzt und ebenso für den Studenten der Tiermedizin, der sich mit den Krankheiten des Schafes und ihrer Bekämpfung beschäftigt, unentbehrlich. Insgesamt ein ausgezeichnetes und von dem Verlag hervorragend ausgestattetes Lehrbuch, dem eine breite Anwendung zu wünschen ist.“ Deutsche Tierärztliche Wochenschrift „Das nicht allzu groß angelegte Lehrbuch gibt eine umfassende Darstellung der wichtigsten in unserem Raum auftretenden Infektions-, Mangel- und Stoffwechselkrankheiten sowie vor allem der parasitären Probleme. Auch Vergiftungen und Erbkrankheiten sind ent- sprechend berücksichtigt. Wertvoll sind Hinweise über die Behandlung dieser Tierart, vor allem Anweisungen, die für den heute oft unerfahrenen kleineren Schafhalter von beson- derer Bedeutung sind. Auch der Tierarzt erfährt manches, was ihm bei der Behandlung einzelner Hobbytiere, die dem Besitzer oft sehr ans Herzgewachsen sind, nützlich sein wird. So ist der Erwerb dieser Neuauflage allen, die sich mit Schafen befassen, vor allem aber dem Tierarzt, der zur Behandlung und zur Betreuung von Schafbeständen zugezogen wird, bestens zu empfehlen.“ PAJL Banz Berlin und Hamburg Zuchthygiene Biol 52 (6), 329-384, Dezember 1987 ISSN 0044-3468 C21274F = ZEITSCHRIFT FÜR SAUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Organ der Deutschen Gesellschaft für Säugetierkunde Owen, R. D.; Qumsiyeh, M. B.: The subspecies problem in the Trident leafnosed bat, Asellia tridens: homomor- phism in widely separated populations. — Die Unterarten-Problematik bei der Dreizack-Blattnase, Asellia tridens: Homomorphismus in räumlich weit getrennten Populationen 329 Gerell, R.: Distribution of Myotis mystacinus and Myotis brandtii (Chiroptera) in Sweden. — Verbreitung von Myotis mystacinus und Myotis brandtii (Chiroptera) in Schweden 338 Wauters, L. A.; Dhondt, A. A.: Activity budget and foraging behaviour of the Red squirrel (Sciurus vulgaris Linnaeus, 1758) in a coniferous habitat. — Aktivität und Verhalten bei der Nahrungssuche von Eichhörnchen (Sciurus vulgaris Linnaeus, 1758) im Nadelwaldhabitat 341 Ylönen, H.; Viitala, J.: Social organization and habitat use of introduced populations of the Vole Clethrionomys rufocanus (Sund) in Central Finland. — Soziale Organisation und Habitatpräferenz von Clethrionomys rufocanus — Populationen in Mittelfinnland 354 Ventura, J.; Gosälbez, J.: Reproductive biology of Arvicola sapidus (Rodentia, Arvicolidae) in the Ebro delta (Spain). — Zur Fortpflanzungsbiologie von Arvicola sapidus (Rodentia, Arvicolidae) im Ebro-Delta in Spanien 364 Stadler, S. G.; Hendrichs, H.: Zum Lautrepertoire eines ursprünglichen Cerviden, Muntiacus reevesi micrurus (Sclater, 1875). - On the sound repertoire of the Formosan muntjac, Muntiacus reevesi micrurus (Sclater, 1875) 372 Bekanntmachung 382 Buchbesprechung 384 Verlag Paul Parey Hamburg und Berlin V HERAUSGEBER/EDITORS P. J. H. van BreEE, Amsterdam - W. FIEDLER, Wien — H. Frick, München — W. HERRE, Kiel - H.-G. Krös, Berlin - H.-J. KuHnn, 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/EDTLORTAT OBEIGE 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 Mikroverfilmung oder der Vervielfältigung auf anderen Wegen, bleiben, auch bei nur auszugswei- ser Verwertung vorbehalten. Das Vervielfältigen dieser Zeitschrift ıst auch im Einzelfall grund- sätzlich verboten. Die Herstellung einer Kopie eines einzelnen Beitrages oder von Teilen eines Beitrages ist auch im Einzelfall nur in den Grenzen der gesetzlichen Bestimmungen des Urheber- rechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 ın der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmungen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Vervielfältiger zu kennzeichnen. Copyright-masthead-statement (valid for users in the USA): The appearance of the code at the bottom of the 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. Mit einer Beilage des Verlages Paul Parey. Fortsetzung 3. Umschlagseite © 1987 Paul Parey. Verlag: Paul Parey, Hamburg und Berlin. Anschriften: Spitalerstr. 12, D-2000 Hamburg 1; Lindenstr. 44-47, D-1000 Berlin 61. — Printed in Germany by Westholsteinische Verlagsdruckerei Boyens & Co., D-2240 Heide/Holst. The subspecies problem in the Trident leaf-nosed bat, Asellia tridens: homomorphism in widely separated populations By R. D. Owen and M. B. QumsivEH The Museum and Department of Biological Sciences, Texas Tech University, Lubbock, Texas Receipt of Ms. 16. 12. 1986 Abstract Examined univariate and multivariate morphometric variation in populations of a desert bat, Aselha tridens, from most of its range ın Africa and Asıa. Morphometric and color variation in this species suggest that two distinct taxonomic groups exist. One of these represents at least two widely disjunct populations with a large intervening region occupied by the other morphologically distinct group. The existence of widely separated homomorphic populations may prove to be more common than previously believed in mammalian species with large ranges, and may cause difficulty in recognizing subspecies based on classical taxonomice criteria. Introduction The trident leaf-nosed bat, Asellia tridens (E. Geoffroy St.-Hilaire, 1813), is known from arıd and semiarid regions from Pakistan and Afghanistan through Arabia to Morocco and Mauritania, and as far south as Somalia and Gambia (CorgEr and Hırı 1980). The infraspecific systematics of this species has been complicated by descriptions of several subspecies that were based on one or a few specimens, and by the generally patchy distribution of this species in desert valleys and oases, and near human habitation. Understanding morphologic variation in this species is important for several reasons. First, Asellia ıs adapted to xeric conditions, but avoids the extremely arıd and usually lifeless extremes ın many parts of the Sahara and Arabian deserts. This results ın disjunct populations, which may adapt to the specialized conditions of local habitats. Second, several subspecific designations have been based primarily on pelage color and small differences in size; thus, an understanding of how color relates to morphometric variation (and subspecific designations) is of critical importance in establishing the taxonomy of this or any other color-varıable species. Third, size varıation in this species is interesting because larger specimens are reported to be found in peripheral portions of its range (e.g., Afghanistan, southwestern Arabia, and southern Sudan), with smaller individuals occur- ring ın the central (and major) portion of the range of the species (Kock 1969). Similar patterns of size variation occur in other North African desert mammals (OsBoRN and Hermv 1980; QumsıvEH 1985), and it thus is of interest to document this quantitatively across the range of a widely distributed species like A. tridens. Such documentation allows evaluation of the relation of morphologic variation to recognized zoogeographic regions. Previous studies of variation in A. tridens either did not utilize statistical methods or, at most, employed only univariate statistics (HARRISON 1957, 1964; Kock 1969; DEBLASE 1980). In none of these studies was sexual dimorphism considered, or controlled for, in the accompanyıng analyses of geographic variation. In order to comprehensively assess morphometric variation in A. tridens, we undertook a series of multivariate analyses that allowed us to determine the extent of overall variation between sexes and among geo- graphic populations. We also evaluated the usefulness of pelage color to describe geo- graphic groupings within A. tridens. U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5206-0329 $ 02.50/0 Z. Säugetierkunde 52 (1987) 329-337 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 330 R. D. Owen and M. B. Qumsiyeh Materials and methods We recorded three external and 18 cranıal measurements from 153 specimens from across most of the range of A. tridens (Fig. 1, Appendix I). In addition to the measurements listed in Appendix II, we recorded dorsal and ventral pelage color on these and additional specimens (Appendix I), by Y Zr ' 1.1000 km nnd — . ie „eo ._ Pe ae A I? kt a Fig. 1. Sample localities for Asellia tridens. Numbers refer to samples and groups used in this study. Two localities in Pakistan and Afghanistan are not shown here (outside boundaries) and are indicated by numerals with arrows comparing specimens directly with the color standards of Rıngway (1912), updated by SmITHE (1974, 1975). The data inıtially were subjected to a Principal Components Analysıs (PCA) of the standard- ized character correlation matrix. PCA is an exploratory method involving no a priori assumptions concerning groups within the data set. The procedure allows ordination of individuals on a mınımum number of axes accounting for amaxımum amount of the data varıance, in order to determine visually the groups that may occur within the overall sample. The significance of such groups then may be verified by appropriate statistical tests. The most appropriate test for this is a two-way (sex versus locality) Multivarıate Analysıs of Variance (MANOVA). This test recognizes and utilizes inter- character correlations in contrast to the analogous univarıate statistics, which cannot test for multivariate differences (Wırric et al. 1986). Because a significant sex-by-locality interaction effect was found in the two-way MANOVA, one-way MANOVA and Analyses of Variance (ANOVAs) were run to elucidate the manner in which sexual dimorphism differs between geographic groups. MANOVA utilizes only those specimens that have no missing data, and thus only 147 specimens were usable in multivarıate analyses. We used PCA again to search for subgroups within the two geographic groupings that were determined by the first analyses. Because sexual dimorphism had been demonstrated, each sex was treated separately in these analyses. Where a subgroup was found, MANOVA again was used to test its significance. After ascertaining statistical significance of groups within the data, we used Canonical Discrimi- nant Analysis (CDA) to define the differences among groups. CDA is a multivariate procedure that produces a discriminant function which maximizes the effects of those characters which differ most among groups. Finally, we were concerned with the problem of several phenotypically intermediate specimens. We used classificatory Discriminant Function Analysıs (DFA) to determine the closest affinıties of these specimens. Morphometric variation in Asellia tridens 331 Results The principal Components Analysıs of all specimens indicated two primary clusters, as defined by the projections of specimens onto Components I and II, which account for 38.4 and 12.0 percent of the total variation, respectively (Fig. 2). No other component or combination of components was found that showed clustering of specimens. The first cluster (group A) included specimens from Chad, Egypt, Algeria, Tunisia, People’s Democratic Republic of Yemen, and Sudan (populations 0, 1, 4, 5, 8, and 9, respectively). The second cluster (group B) included specimens from Morocco, southwestern Asıa, Palestine, and Mauritanıa (populations 2, 3, 6, and 7, respectively). We applied subsequent tests of geographic varıation based on these two groups. The two-way (group versus sex) MANOVA of these two phenotypic groups showed very high significance for both group and sex (p < .001), and also high significance for group-by-sex interaction (.01 > p > .001, see Table). Univarıate ANOVAs indicated significant between-group differences in 19 of the 21 characters, between-sex differences ın nıne characters, and a significant interaction effect in seven characters. The one-way MANOVAs indicated highly significant sexual dımorphism in both groups A and B (Table). Of the 21 characters tested, the one-way ANOVAs indicated that 12 are significantly dimorphic in group A, whereas four are dımorphic in group B. Because of the sexual dimorphism and the geographic grouping, we considered four groups for further analyses (A males, A females, B males, and B females). Within these Il Fıg. 2. Plot of Asellia tridens specimens on principal components I and II, accounting for 38.4 and 12.0 percent of the data variance, respectively. Numbers for specimens refer to localıty groups as ın Fig. 1. Arrows indicate four specimens of concern, that were determined by discriminant function analysıs to be most sımilar to other specimens from their respective localıty groupings 382 R. D. Owen and M. B. Qumsiyeh groups, additional PCAs suggested clusters only in group A females, in which the two Yemeni specimens differ in components I and II from those specimens from the remainder of group A localıties. A MANOVA indicates that the difference is highly significant (p < 0.005), and the univariate ANOVAs show eight of the 21 characters to differ significantly. Group means and standard errors for 21 characters in Asellia tridens Also shown are F-values and significance levels? for 2-way MANOVAP and ANOVAs, and the same statictics for I-way tests between sexes of each group separately. Characters are listed and described in Appendix II Group A° Group B Two-way tests Sexual Dimorphism Variable Males Females Males Females Group Sex Interaction Group A Group B MANOVA 5g,29%+% 3,17#8* 2,58% 2,548* 2,9288 FAL 49.89 41.92*** 0,50 6.92** 4,50* 2.93 +0.24 MTC3 35.70 (01.20.13*** 0,57. 5.682 370770 +0.24 | : MTC4 29.50 79 27.02*** 0.08 0.69 GLS 85 45.24*** ##* 1,70 CBL 72% 2,36 BBC IOC MASTB ZYGB RBCA RBMAX PEW NASB PALL | PPLL UTRL | UPMLM BRCD SOME eG: 0.27 MANDL 12.19 11.97 12.40 12.20 27.99*** MANDTR 7831 7. 69 10.69** CORH Sy se. 5.10* | 2*,0.05>P>0.01; **, 0.01>P>0.001; ***, 0.001>P.-?® Using Wilks’ criterion for calculation of exact F.-° Chad, Egypt, Algeria, Tunisia, Yemen, and Sudan (populations 0, 1, 4, 5, 8, and 9). N =45 88,45 2 2.-“ Morocco, SW Asia, Palestine, and Mauritanıa (populations, 2, 3, 6, and | ).N=49 88,14 29. Morphometric variation in Aselha tridens 333 The Canonical Discriminant Analysis of sexual dimorphism within the two geo- graphical groups shows that there is considerable overlap of the sexes on canonical variate I in both groups (more so in group B). The CDA of geographic group differences, in contrast, indicates that, for either sex, a linear function may be constructed that entirely separates the two groups, the separation being more distinct in males. The same function for the females also successfully separates the two Yemeni specimens from the remaining group A females. Classificatory Discriminant Function Analysis for morphometrically intermediate indi- viduals was performed separately for each sex. In all cases this analysis indicated that the individuals in question are associated with the other specimens from their respective populations. Discussion Morphometric variation in Asellia tridens Harrıson (1957) observed that forearm measurements were larger and ears relatively smaller for specimens from Iraq and Yemen (which he reterred to A. t. murraiana) than for those from Oman, Hofuf Oasis in Saudi Arabia, Sudan, and Egypt (which he referred to A.t. tridens). Although he included both males and females and provided no statistical tests of these differences, HARRISON (1957) concluded that his data show that this species may migrate from Iraq to Yemen in the winter months. GAISLER et al. (1972) and QumsıyEH (1985) found a slight difference between populations in northern and southern Egypt, but allocated all Egyptian material to the nominate subspecies. Kock (1969) suggested that size difference (especially forearm length) may be due to climatic (primarily temperature) differences. All earlier studies listed above of A. trıdens were limited to cursory examination of few specimens with at best univariate statistics. More importantly, these studies involved pooling ot males and females into single samples, resulting in confusion of variatıion due to sexual differences with that due to geographic differences. Our results, based on specimens from most of the range of A. tridens, suggest that two distinct clusters occur (groups A and B), which are distinguished by morphomerric analysıs. Of special interest is the highly significant sex-by-group interaction effect in the MANOVA (Table). In univarıate ANOVA, a difference of either direction or degree may cause an interaction effect among treatments; ı.e., a different group may be larger, or the same group may be larger to greater or lesser extent. In the multivarıiate case, a third possibility (not associated with the other two) exists that may also cause an interaction effect. This is a difference in intercorrelation among characters. In the example of A. tridens, both group A and group B are highly dimorphic sexually, but the expression of dimorphism is quite different in the two groups. Group A is dimorphic for 12 of the 21 characters examined (males larger ın all 12), whereas group B is dimorphic for only four (males larger in three, females in one). The Canonical Discriminant Analysis indicates that males are more geographically variable than are females. However, this sex-related difference in varıability is not expressed as a greater or lesser degree of overall dimorphism in one or the other geographic group (the MANOVA F-values for sexual dimorphism are essentially equal, see Table). Rather, the greater male plasticity appears to reflect a higher correlation among characters in group A than in group B (allowing a greater number of significant between-sex ANOVAs without a concomitant increase in MANOVA significance level). Thus, the species’s morphology may be more severely constrained in the environment occupied by group A bats than in those of group B. Group A males are simply larger than females ın 334 R. D. Owen and M. B. Qumsiyeh most characters, whereas group B bats exhibit a more complex suite of between-sex differences; many of these are not significant by a univariate test, but together combine for significant multivariate difference. Systematic value of color in Asellia tridens In many parts of its range, A. tridens ıs dimorphic in color with red individuals occurring among the “normal” gray or pale-colored individuals (HArrıson 1957, 1964; ATTALLAH 1978). Harrıson (1957) stated that the red phase is not found in A. t. murraiana. However, we observed dichromatism in series of specimens from Assa, Morocco; Niger; Benni Abbes, Algeria; Mauritania; Afghanistan; and Iran. These data agree with the finding of Kock (1969) that presence or absence of the red phase is not useful as a systematic character. Most specimens of the morphometrically defined group A are pale in color. A typical specimen has a dorsal coloration of smoke gray with bases even a paler smoke gray (no. 44 in SMITHE’s [1975] colors), and ventrally is pale pınkish buff. A specimen from Dandara temple (type localıty of A. t. tridens) has a whitish ventrum and dorsal fur with glaucous tips and pearl gray bases. Specimens from Kassala, Sudan, also are pale in color with ventral hairs and bases of dorsal hairs pearl gray, tips glaucous, and membranes pale in color. Although specimens from group A are varıable in color, all exhibit overall pale coloration similar to that described for A. t. tridens and A. t. pallida. Bases of hairs in specimens from Iran (group B) range from smoke gray to drab gray (colors of SMITHE 1974, 1975) and the membranes are darker in color than in Egyptian specimens. The two specimens examined from Pakistan are even darker than those from Iran. Color in southwestern Asıan specimens ıs noticeably darker than in those from Saharan populations discussed above as group A, suggesting that there are two groups distinguished by color, with the mountain form (group B) being darker. It should be noted, however, that a series from Palestine includes pale, dark, and intermediate specimens, and that specimens from Yemen (which cluster with group A) have dark coloration overall (ventrum and bases of dorsal hairs being drab in color with tips olive brown). This suggests that color represents a local adaptation, and does not correspond completely with the morphometric groupings, thus limiting its value for systematic studies in this species. Relationship of color and morphometric variation to the named subspecies of Asellia tridens Geographic varıation in mammals tradıtionally has been understood in light of discrete groupings termed subspecies (LiDICKER 1962; Wırson and Brown 1953). Here we discuss, and relate the results of our multivariate analyses to, the named subspecies of Asellia tridens. A. t. tridens (E. GEorFrroY St.-HıLaıre, 1813). The type locality of the nominate subspecies is from the Valley of the Kings and Dandara, Egypt (E. GEoFFRoY St.-HILAIRE 1818). All our specimens from Egypt (populations numbered 1) including those from the type locality are small in size and pale in coloration, and are included in the morphologi- cally defined group A. We recommend that the name A. t. tridens be applied to all populations represented in group A. A. t. diluta Andersen, 1918. The type specimen is from EI Golea, Algerian Sahara. ANDERSEN (1918) referred other specimens from Biskra to this taxon, which he disting- uished as being larger than A. t. tridens, with the pelage “conspicuously” pale. All our specimens referable to A. t. diluta are from Algeria (populations numbered 4). These Morphometric variation in Aselha tridens 995 clearly are ıdentifiable wıth group A, both morphometrically and by color, and we recommend that A. t. diluta be considered a synonym of A. t. tridens. A. t. pallida Laurent, 1937. Described from Oued Tatta, Antı Atlas Mountains, Morocco, this subspecies is presumably a much paler form than A. t. diluta, with color reaching white ventrally and the membranes being pale. BRossET and CAUBERE (1960) referred specimens from Figuig (Morocco), Biskra, and El Golea (Algeria) to A. t. diluta. These specimens were darker than pallida as described by LAURENT (1937), and BRossET and CAUBERE (1960) concluded that the specimens examined by LAurEnT (1937, 1942) either were bleached or represented an evolutionarily localized population. Our specimens from Morocco are from a locality only a few kilometers from the type locality of A. t. pallida, but are clearly referable to group B both morphometrically and by color (most specimens being much darker than those examined from Algeria). Because we have not examined the type material of A. t. pallida, we cannot determine the nomenclatorial status of populations currently referred to this name, or of our material from Morocco and Mauritanıa. A. t. murraiana Anderson, 1881. Described from Karachı, Sınd (now Pakistan), as a subspecies darker and larger, but with relatively smaller ears, than A. t. tridens. Specimens examined from Pakistan, Afghanistan, Iran, and Iraq (referred by all previous authors to A. t. murraiana) were identifiable with group B. This name would constitute the earlıest available for populations included in group B. A. t. italosomalica DeBeaux, 1931. This subspecies, from Somalıa, was distinguished by its small size (DEBEAux 1931). We examined no specimens of this taxon and cannot comment on Its status. To summarize, within the populations of Asellia tridens examined, we recognized two groups based on multivariate analyses of morphologic variation. One group (A), to which the name A. t. tridens ıs applicable, has a wide range throughout the Sahara and Arabian deserts. The second group (B) includes at least two disjunct populations (separated geographically by group A). The southwestern Asıan populations (3 and 6) clearly are referable to A. t. murraiana. Populations from northwest Africa (2 and 7, indistinguishable in our analyses from A. t. murraiana) pose a systematic problem. Most authors would recognize geographically disjunct populations of a species that have “their own evolutionary tendencies” as distinct subspecies (LIDICKER 1962). The northwestern Afrıcan population of group B may fit this definition of a subspecies, but these specimens are morphologically indistinguishable from those of populations of A. t. murralana. This homomorphism is not limited to overall size or even to a simple shape- related suite of characters. Rather, it includes similarity of geographic patterns of inter- character correlations across sexes. Thus, it is difficult to postulate from these data that the two populations are evolutionarily distinct entities. This evolutionary and nomenclatorıal problem may perhaps be resolved by additional data, such at mitochondrial DNA or protein electrophoretic mobilities, which should further elucidate the evolutionary his- tories of these populations. Our study of this desert-adapted bat demonstrates the value of multivarıate mor- phometric analysis of specimens from the entire distribution of such a species. We believe that other studies utilizing such analyses on widely distributed species will document that the phenomenon of homomorphism in disjunct populations is more common than previously suspected in natural populations of mammals. Appendix I Specimens examined. The locality is given first followed by (in parentheses) the museum abbreviations (see Appendix II) and number of specimens examined: Morocco: Agadir Prov., Assa Khetara (USNM 39 38). 336 R. D. Owen and M. B. Qumsiyeh Algeria: El Golea (BMNH 1 ?,FMNH1 6,1 2), Bechar (ROM 1 2), Beni Abbes (CM 1 9,9 88; SMNS 3 22). Mauritania: Sahel Saranna, 3 km s Aleq (USNM 2dd 1 2). Tunisia: 4 km s Redeyf (ELC 322, 4 88), 12 km n Tozour (ELC 8 dd, 8 22). Chad: Ouniang Kabir (USNM 322, 18). Egypt: Red Sea Gov., Bir Abraq (FMNM 1 2, USNM 1 2), Qena Gov., Qena (FMNH 12), Qena Gov., Dandara Temple (SMNS 1 ?, CM 1156, 11 22), Qena Gov., Luxor (CM 3 22, 11 dd), Western Desert Gov., Sıwa (FMNH 1 2), Sınaı Gov., Sinai, (USNM 22P), Giza Gov., Saqgara (SMNS 19), Sudan: Kassala (USNM 1 8). Yemen: Lahej (USNM 2 22). Palestine: Coastal Prov, Tel Aviv-Jaffa (TAU 2 22,1 8), Jordan Valley Prov., Yarmouk, near Jordan Bridge (TAU 1 2), Jordan Valley Prov., Wadı Ashak, Chama el Malah, near Mehola (TAU 2 38). Iraq: Baghdad Liwa, Baghdad (UCONN 1 %, FMNH 2 29,2 88). Iran: Lurestan, Dehloran (FMNH 1 9, 3 34); Shuh or Susa (ROM 2 22). Afghanıstan: Dilaram (FMNH 1 8,1 2). Pakistan: Baluchistan, Panjgur (FMNH 2 ?P). Appendix II List of abbreviations Museums: BMNH - British Museum of Natural History; CM - Carnegie Museum of Natural History, Pittsburgh; ELC - E.L. Cockrum Collection, Tucson, Arızona; FMNH - Field Museum of Natural History, Chicago; ROM - Royal Ontario Museum, Toronto, Ontario; SMNS - Staatliches Museum für Naturkunde, Stuttgart, Germany; TAU - Tel Aviv University, Zoology Dept. Museum; USNM - National Museum of Natural History, Washington, D.C.; UCONN - University of Connecticut Museum of Natural History, Storrs. Measurements: FAL - forearm length; MTC3 - third metacarpal length; MTC4 - fourth metacarpal length; GLS - greatest length of the skull; CBL - condylobasal length (from occipital condyles to anterior alveolar margin of canıne); BBC - breadth of braincase (taken at base of zygomatic arch); IOC - interorbital constriction; MASTB - mastoıd breadth; ZYGB - zygomatic breadth (maximum breadth of zygomatic arches); RBCA - rostral breadth at canine alveolars; RBMAX - rostral breadth (maximum breadth taken on outside margins of maxilla); PEW - palatal emargination width (maximum); NASB - nasal breadth (maximum); PALL - palatal length (from posterior margin of palate to ventral tip of foramen magnum); PPLL - post-palatal length (from posterior margin of palate to ventral tip of foramen magnum); UTRL - upper toothrow length from canıne to last molar; UPMLM - upper toothrow length from premolar to last molar; BRCD - braincase depth (taken externally but excluding sagittal crest,; MANDL - mandible length; MANDTR - mandibular toothrow length from anterior margin of canıne to last molar; CORH - coronoid height (maximum height of coronoid process of mandible). Acknowledgements We are grateful to the curators of the museums who kindly allowed access to their mammal collections and/or loaned us specimens for this study. These museums are listed in Appendix II. Thanks are due also to E. L. CockRum for loaning us his private collection of Asellia from Tunisia. This manuscript was reviewed by $. A. MıLLEr, M. R. Wıric, J. K. JONES, Jr., C. JOnEs, and R. J. BAKER, whose efforts we appreciate. We thank ROBERT HUBER for translating the abstract. This project was funded in part by the University of Connecticut, Texas Tech University, Carnegie Museum of Natural History, and the U.S. Naval Medical Research Unit No. 3. Zusammenfassung Die Unterarten-Problematik bei der Dreizack-Blattnase, Asellia tridens: Homomorphie in räumlich weit getrennten Populationen Morphometrische Unterschiede in verschiedenen Populationen der wüstenbewohnenden Fledermaus A. tridens wurden vom überwiegenden Teil des Verbreitungsgebietes in Asien und Afrika mittels uni- und multivariater Statistik untersucht. Diese Daten und Unterschiede in der Färbung legen die Existenz zweier verschiedener taxonomischer Gruppen nahe. Das Vorhandensein homomorpher, jedoch räumlich weit getrennter Populationen scheint üblicher zu sein als früher angenommen wurde. Diese Tatsache erschwert die Erkennung von Subspecies anhand klassischer taxonomischer Kriterien. Morphometric varıation in Aselha trıdens 957, Literature ANDERSEN, K. (1918): Diagnoses of new bats of the families Rhinolophidae and Megadermatidae. Ann. Mag. Nat. Hist. Ser. 9, 2, 374-384. ANDERSON, J. (1881): Catalogue omammalia in che Indian Museum, Calcutta. Part I. Calcutta: Indian Museum Publ. ATALLAH, S. 1. (1978): The mammals of the eastern Mediterranean region: their ecology, systematics and zoogeographical relationships. Säugetierkde. Mitt. 26, 1-50. BrosSET, A.; CAUBERE, B. (1960): La colonie d’Asellia tridens de l’Oasıs de Fıguig (Chiropteres). Mammalia 24, 222-227. CoRBET, G. B.; Hırı, J. E. (1980): A world list of mammalian species. Ithaca, New York: Cornell Univ. Press. DeBeAux, ©. (1931): Spedizione del Bar’one Raimondo Franchetti in Dancalia-Mammiferi. Ann. Mus. Civ. Stor. Nat. Genova 55, 183-217. DeBLase, A. F. (1980): The bats of Iran: systematics, distribution, ecology. Fieldiana Zool. (new ser.) 4, 1424. GAISLER, J.; MADKOUR, G.; PELIKAN, J. (1972): On the bats (Chiroptera) of Egypt. Acta. Sc. Nat. Brno 6, 140. : GEOFFROY ST.-HILAIRE, E. (1813): Sur un genre de chauve-souris, sous le nom de Rhinolophes. Ann. Mus. Hist. Nat. Parıs 20, 254-266. — (1818): Description de l’Egypte. Historie naturelle. Description des mammiferes qui se trouvent en Egypte, 2, 99-135. HaARrRIsonN, D. L. (1957): Some systematic notes on the trident bats (Asellia tridens E. Geoffroy) of Arabia. Mammalia 21, 1-8. — (1964): The mammals of Arabia, vol. I: Introduction, Chiroptera, Insectivora, Primates. London: Ernst Ben. Kock, D. (1969): Die Fledermaus-Fauna des Sudan. Abh. Senckenb. Natur. Ges. 521, 1-238. LAURENT, P. (1937): Une forme nouvelle du genre Asellia: A. tridens pallida subsp. nov. Mammalıa 1, 111-116. — (1942): La chauve-souris trident, d’Etienne Geoffroy-Saint-Hilaire (Asellia tridens E. Geoffroy 1813), description complementaire. Bull. Mus. Nat. Hist. natl. (Paris), Ser. 2, 14, 4348. LiDICkErR, W. Z., Jr. (1962): The nature of subspecific boundaries in a desert rodent and its implications for subspecies taxonomy. Syst. Zool. 11, 160-171. OsBorRn, D. J.; Hermy, I. (1980): The contemporary land mammals of Egypt (including Sinai). Fieldiana Zoology 5, 1-579. QuMSIYEH, M. B. (1985): The bats of Egypt. Spec. Publ. Mus., Texas Tech Univ. 23, 1-102. Rıpcway, R. (1912): Color standards and color nomenclature. Washington, D.C. SMITHE, F. B. (1974): Naturalist’s color guide. Supplement. New York: Amer. Mus. Nat. Hist. — (1975): Naturalist’s color guide. New York: Amer. Mus. Nat. Hist. 86 color swatches. Wırrıc, M. R.; Owen, R. D.; CoLBErT, R. L. (1986): Assessment of morphometric varıation in natural populations: the inadequacy of the univariate approach. Syst. Zool. 35, 195-203. Wırson, E. O.; BRown, W.L., Jr. (1953): The subspecies concept and its taxonomic application. Syst. Zool. 2, 97-111. Authors’ addresses: Dr. ROBERT D. Owen, The Museum, Texas Tech University, Lubbock, Texas 79409, USA; Dr. Mazın B. QumsiyEH, Department of Biochemistry, Texas Tech University Health Sciences Center, Lubbock, Texas 79430, USA Distribution of Myotis mystacinus and Myotis brandtü (Chiroptera) in Sweden By R. GERELL Department of Animal Ecology, University of Lund, Lund Receipt of Ms. 4.12.1986 Abstract Studied the distribution of the two sibling species of whiskered bats Myotis mystacinus and M. brandtii in Sweden. The northern range of M. mystacinus reaches about the latitude of 62°N while that of M. brandtii extends to 64°N. The two species are sympatric in South and Central Sweden. M. brandtii seems to be the more frequent species in Sweden. Introduction When a species is splitted into two species one has to rebuild the distribution picture of the original species. Lack of knowledge, however, often brings the handbook writer to preserve the old distribution of the original species. The new species, on the other hand, is handled as a rare species. So ıt has been done with the whiskered bats, Myotis mystacinus (Kuhl, 1819) and M. brandtii (Eversmann, 1845). The distribution of M. mystacinus and M. brandtii in Sweden has been dealt with by BaAGOE (1973), who examined Swedish museum records, respresenting 17 different localities (Fig. 1). Material and methods Within the scope of a project on bats supported by World Wildlife Fund, dead bats were collected through the public. Up to now, about 300 bats have been received of which 45 specimens have been identified as whiskered bats. All records obtained are referred as summer records. Another two specimens, one of each species, were nettrapped at the same localıty in Skäne (southernmost part of Sweden). Results The sex and age distribution of the collected specimens (Table) don’t differ statistically when comparing the two species (x =1.75, d.f.=3, P> 0.50). The male majority among the juveniles can not be explained. Table Sex and age distribution of the records of Myotis mystacinus and M. brandtii Species Females Ad. Juv. M. mystacınus M. brandtii 6 6 U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5206-0338 $ 02.50/0 Z. Säugetierkunde 52 (1987) 338-341 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Distribution of Myotis mystacinus and Myotis brandti in Sweden 339 The new records obtained (see Fig.) strengthen the distributional differences of the two species indicated by the earlier records examined by BaaGoE (1973). Hence, the northern range of M. mystacinus reaches about the latitude of 62°N while that of M. brandtii extends to 64>N. The two species are sympatric in South and Central Sweden. Distribution of Myotis mystacinus (triangles) and M. brandtiüi (circles). Filled symbols denote records obtained in this study, open ones refer to BAAG®E (1973) 340 R. Gerell The records of M. brandtu ıs ın majority (63 %) in the collection as it was among the earlier records examined (70%). The distribution map gives the impression that the two species are evenly distributed but when relating the number of records to the encounter rate (dependent on human population density), both species are probably more common north of the latitude of 60>N. Most records of both species (80%) were obtained in dwelling houses, which is to expect with regard to the way of collecting. Discussion The two sıbling species are sympatric ın Central Europe (GAUCKLER and Kraus 1970; HanäAk 1971; RUPRECHT 1974) but separated in greater part of the Soviet Union (STRELKOV 1983). In Sweden, M. brandtü ıs found farther north (64°N) compared to M. mystacinus (62°N). The same distribution pattern ıs found in Finland (LEHMANN 1983). This agrees with the statement by STRELKOV (1983) that M. brandtii is a typical boreal forest species. M. mystacinus seems to be the more common species in Central Europe while M. brandtii is more frequent in Sweden and Finland (LEHMANN 1983), based on the number of localities obtained. In southern Norway, M. mystacinus predominates while in Denmark, only M. mystacinus is recorded, except from the island of Bornholm where both species occur (BAAGDE 1983). No evident differences in the habitat selection of the two sıbling species have yet been found (GAUCKLER and Kraus 1970; OHLENDORF 1983). Both species have been trapped by mistnets at the same locality in Skäne indicating that the two species can coexist. RygerG (1947) found that M. mystacinus (s.1.) was more common in the northern part of Sweden than in the southern one. Censuses by means of bat detectors has confirmed that (GERELL 1981). M. mystacinus (s.s.) ıs rare ın Zealand while it is rather common on the island of Bornholm. BAAGBE (1984) has explained the difference of the population density of the species in Denmark as due to competition with Pipistrellus pipistrellus. P. pıpistrellus is absent on the island of Bornholm but very common in Zealand. This hypothesis fıts also the distribution pattern of the three species in Sweden. P. pipistrellus forms the biggest nursery colonies in houses in South Sweden (up to 500 individuals) while M. mystacınus/ brandtii makes the biggest ones in North Sweden. Furthermore, P. pipistrellus is absent on the island of Gotland while M. mystacinsus (Ahlen, 1983) and M. brandti are common species. Acknowledgements Many thanks to Dr. H. J. BaaGoe for his kind help with checking my species identifications. The study was supported by grants from World Wildlife Fund, Sweden. Zusammenfassung Verbreitung von Myotis mystacinus und Myotis brandtü (Chiroptera) in Schweden Die Zone sympatrischen Vorkommens der beiden Bartfledermausarten Myotis mystacinus und M. brandtii umfaßt Süd- und Mittelschweden. Die Verbreitung von M. brandtü erstreckt sich bis zur Latitüde von 64°N, während die von M. mystacinus nur 62°N erreicht. M. brandtü erscheint die häufigste der beiden Arten zu sein. References AHLEN, I. (1983): The bat fauna of some isolated islands in Scandinavia. Oikos 41, 352-358. BAAGBE, H. J. (1973): Taxonomy of two sibling species of bats in Scandinavia Myotis mystacinus and Myotis brandtii (Chiroptera). Vidensk. Meddr dansk naturh. Foren. 136, 191-216. — (1984): Bornholms flagermus. Fjaelstaunijn 8, 80-89. Distribution of Myotis mystacinus and Myotis brandtii in Sweden 341 GAUCKLER, A.; Kraus, M. (1970): Kennzeichen und Verbreitung von Myotis brandtü (Eversmann, 1845). Z. Säugetierkunde 35, 113-124. GERELL, R. (1981): Bat conservation in Sweden. Myotis 18-19, 11-15. HaAnAR, V. (1971): Myotis brandtü (Eversmann, 1845) (Vespertilionidae, Chiroptera) in der Tsche- choslowakei. Vestn. Cs. spol. zool. 35, 175-185. LEHMANN, R. (1984): Myotıs mystacinus (Kuhl, 1819) and Myotis brandtii (Eversmann, 1845) in Finland. Myotis 21-22, 96-101. ÖOHLENDORF, B. (1983): Die Große Bartfledermaus, Myotis brandtii (Eversmann, 1845), ein fester Bestandteil der Harzer Fauna. Nyctalus (N.F.) 1, 577-584. RUPRECHT, A. L. (1974): The occurrence of Myotis brandtii (Eversmann, 1845) in Poland. Acta Theriol. 19, 81-90. RyBErg, O. (1947): Studies on bats and bat parasites. Stockholm: Bokförlaget Svensk Natur. STRELKOV, P. P. (1983): Myotis mystacinus and Myotis brandtii in the USSR and interrelations of these species. Part 2. (In russian, wıth english summary). Zool. Zh. 62, 259-270. Author’s address: Dr. Rune GERELL, Ecology Building, S-223 62 Lund, Sweden Activity budget and foraging behaviour of the Red squirrel (Sciurus vulgaris Linnaeus, 1758) in a coniferous habitat By L. A. WAUTERS and A. A. DHONDT Department of Biology, University of Antwerp, Wilrijk, Belgium Receipt of Ms. 26. 9. 1986 Abstract The daily and seasonal actıvity-pattern, habitat use and foodchoice of free-ranging red squirrels (Scinrus vulgaris Linnaeus, 1758) were studied over a one year period using radio-telemetry. The daily activity-rhythm changed over the year. In winter squirrels were only active for a few hours, emerging from the drey with fırst light. Their activity was concentrated towards foraging, feeding mainly on high-energy seeds of pines. The large amount of time spent in the nest reduced the energy-costs of thermoregulation in cold weather. In spring the active period progressively expanded throughout the afternoon. This increase in the time spent active per day was probably caused by changes in environmental factors, such as increasing daylength and temperature, and a decreasing abundance of primary food-resources. From May to October the activity-pattern of the whole population was bimodal. Most likely endogenous factors, related to the depletion of the stomach content, form the basıs of this bimodal pattern. Relating external factors to the length of the active phase, it was shown that daylength and temperature played an important role, being able to explain 93 % of the variation in the length of the active period throughout the year. Undoubtedly feeding-requirements are also of importance. Introduction Like most other tree-squirrels the European red squirrel (Sciurus vulgaris Linnaeus, 1758) is a dıurnal anımal, which offers the possibility to study its activity-pattern and behaviour through field observations. Some general information is available about activity and food- preference of red squirrels from direct field observations (SHORTEN 1962; PULLIAINEN 1973; ZwAHLEN 1975; TITTENSOR 1977; WAUTERS 1984), or from stomach-contents analysıs of shot animals (DEGn 1974; TITTENSOR 1977; GRONWALL 1982). These studies show that, in general, red squirrels have a long and bimodal pattern of activity in summer, a short unımodal pattern in winter, and an intermediate pattern ın spring and autumn. Their foraging-behaviour is concentrated on high-energy food resources, especially tree- U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5206-0341 $ 02.50/0 Z. Säugetierkunde 52 (1987) 341-353 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 342 L. A. Wauters and A. A. Dhondt seeds (of coniferous and deciduous trees). It is shifted towards alternative food-items such as fungi, buds, shoot, flowers and even anımal matter when seeds are scarce or no longer available (see also MOoLLER 1983). It is assumed that the 24-hours activity rhythm exhibited by the red squirrel has an endogenous basis, as ıs the case in most other mammals (AscHorr 1963). However still little is known about the influences of environmental factors on the rhythm and the pattern of actıvity. Tonkın (1983) examined the varıation ın the actıvity-pattern and foraging behaviour throughout the year and its relation to environmental factors in a deciduous wood, using radıo-tracking. This paper presents a comparable study where it is tried to quantify the activity-budget of radıotagged squirrels in a coniferous habitat, and where the influences of environmental factors and behaviour on the patterns observed are examined. Study area A study area of 40.5ha was chosen in a large coniferous wood (212ha) at Herenthout, Province of Antwerpen, Belgium. It was dominated by Scots pine, Pinus sylvestris and Corsican pine, Pıinus nigra var. calabrıca (72%), with some larch, Larix decidua (7%) and a few mixed plots of oak, Quercus robur and beech, Fagus sylvatica (15%). A detailed vegetation-study of the ditferent plots had been carried out previously (WAUTERs 1984). Material and methods Field work Between May 1985 and May 1986, twenty- eight different squirrels, 14 males and 14 females, were tagged with radio-transmitters (squirrel transmitter from Biotrack, U.K.). Each squirrel carrıed a transmitter for 1 to 5 months. In any one month eight to nine squirrels were monitored. Two types of radiotracking techniques were Ts 1.e. radiosurveillance and point-fixes. 1. radiosurveillance: a tagged squirrel was located and then followed and observed for about one hour. 2. point-fixes: the radıo-tagged squirrel was located and monitored for 5 minutes. For both methods the following information was recorded every minute: a. Foraging-place: i.e. the tree-species on which the squirrel was observed or the ground. b. Height and substrate: a distinction was made between 4 height classes (0-5 m, 5-10 m, 10-15m, >15 m) and 3 substrate-classes (stem, large branches, small branches). c. Behaviour: behaviour was assigned to one of the following categories 1. Foraging, including searching, handling and recovering stored food, 2. Food-storage Ber iour, 3. Travelling, not related to foraging, 4. Interactions with other squirrels, 5. Grooming, 6. Resting outside the drey, 7. Other behaviour, 8. Not active, when sleeping or resting in the nest. d. Remarks: additional comments about the observed behaviour when necessary, especially for categories 1, 4 and 7. e. Position: all locations were mapped using X, Y coordinates. In addition to the radio-tracking data casual observations were made of other squirrels. The information recorded was the same as that noted for tagged anımals. For every observation day some characteristics of the weather were noted: 1. Proportion cloud- cover (0/8 to 8/8 scale), II. rainfall (none, slight, heavy, snow), III. wind-direction and wind-force, IV. Temperature (at 12.00 hours). Data analysis Although data were recorded on a one minute basis, the time-unit was further subdivided when necessary. Therefore data could be treated as monitored continuously. Foraging place, substrate and height The time spent on each tree-species or on the ground was expressed as a percentage of the total hours of active behaviour observed in each month. Data on the substrate used and the height-classes were treated the same way. It must be mentioned that the latter were not recorded for all "observations. Activity budget and foraging behaviour of the Red squirrel 343 Activity The daily activity pattern of each month was deduced from all data collected that month using both radio-tracking methods. For each hour of the day all the time spent active was summed as was the time spent in the drey. The amount of activity per hour was expressed as a percentage, using the following formula: time spent active x 100 percentage activity/hour (A) = ———— time spent active + time in drey The mean number of ‚hours of activity per day for each month was then calculated the following way: Ai time active per day JE TR with n = number of hours activity recorded. i=1 As regards the activity-budged of the squirrel, the time allocated to each of the behavioural categories was expressed as a percentage of the total time spent active in each month. Foraging-behaviour Since foraging was quantatively the most important type of behaviour it was examined more closely. The percentage of time spent searching for and handling different food-items was estimated on a monthly basis, so that changes in the diet could be detected. Results Activity rhythm The activity rhythm varıed considerably over the year (fig.1). From December to February activity started somewhat before sunrise (fig.3), with a peak in the first hours after sunrise. The number of hours spent active per day during this period was less than five (table 1). The percentage of daylength when squirrels were actıve was lowest in February (48,4 %, table 1). In early spring (March-Aprıil) squirrels still left their dreys at first light (fig. 2), but now activity lasted much longer. There was still only a morning peak, but the active period progressively expanded in the afternoon (fig. 1). In May the activity-rhythm showed a bımodal pattern, with an intense peak starting soon after sunrise, and a less pronounced and shorter one in late afternoon (fig. 1). In this month the number of hours of activity per day (10h06’) was the highest of the whole year (table1). From May to September the pattern stayed bimodal, but the percentage of afternoon activity was variable (fig. 1). The largest amount of afternoon activity was found in July (48%) when peaks were also less pronounced and there was little synchrony between start or oftfset of activity for different squirrels. In October and November afternoon activity became less important (fig. 1). Neverthe- les November was the month with the highest proportion ot available daylight hours spent active (72%) and the only one ın which activity lasted till after sunset (fig. 2). Although it has been shown that the activity-rhythm often has an endogenous basıs tor mammals (AscHoFF 1963), environmental factors were proven to be of great importance in squirrels (PULLIAINEN 1973; ZWAHLEN 1975; Paurs 1979; Tonkın 1983). Therefore we related the onset of activity to sunrise and offset of activity to sunset (fig. 2). We found that during the period when daylength was shortest (October to April), squirrels became active soon after first light, thus somewhat before sunrise. In the long summer months there was more than one hour difference between daybreak and the onset of activity (fig.2). The relation between the end of the active period and sunset was less pronounced. Only in autumn and in May activity may have been ended through nightfall (fig. 2). Variation in daylength and temperature contributed both to the variation in the length of the active period (table 1). The multiple correlation coetficient was highly significant (r = ae SZES5Z5Z55 A sa Bes, RE = 0 PEN = [>=] [eeJa ee 100 = == Has Der MN! N | | | v EHEN nissen I ill ! [ m —— m m m m nom Io %& ACTIUITY | | se I — AI Hill | | | = | il] ia il | Il I Il Il Il ll N | er [0,0) an N 86 M 86 09.90 4.00 8.00 12.00 16.00 HOURS Fıg. 1. Activity rhythm over a whole year. The times of sunrise and sunset are indicated by the curves. Radiotracking data were obtained for those hours marked as vertical blocks Activity budget and foraging behaviour of the Red squirrel 345 Table 1 Daylength, mean monthly temperature, mean time spent active per day per squirrel, percentage of daylength spent active, time spent active in the afternoon per day, and afternoon activity as a percentage of total daily activity Mean hours Mean monthly Mean monthly Number of hours active per day between sunri- max.t°(’C) min.t° (°C) se and sunset Month total afternoon % 1985 May June July Aug. Sept. Oct. Nov. Dec. 1986 Jan. Febr. March April May 52% AFTER SUNSET BEFORE 52 10% 152 15% 10% 52 AFTER SUNRISE BEFORE 52 102 J J A Ss 0 N D J F n A n Fig. 2. Start and offset of activity related to sunrise and sunset. The time interval between start of activity and sunrise and between end of activity and sunset is expressed as a percentage of daylength for each month 346 L. A. Wauters and A. A. Dhondt 0.966; F = 63.43; df = 2, 9; p < 0.01). Squirrels had a longer activity period when days are longer (partial correlation coefficient r = 0.958; p < 0.01) or when temperature is higher (r = 0.855; p < 0.01). Activity budget The major proportion of active time was spent foraging. It stayed fairly constant over the year, varyıng between 63% and 78% of total activity (table 2). There was however a difference in the proportion foraging time between the spring-summer (April-September) and the autumn-winter period (October-March), showing a tendency towards relatively more time spent foraging in the latter (table 3). In contrast other types of behaviour became more important in the spring-summer period. Travelling, which always accounted for more than 10% of the activity (table2) was observed most in summer (table 3). Movements seemed nearly always to be direct, whether they were between feeding-sites or towards dreys. To test that the proportion of time available to squirrels for searching and handling food, was influenced by the proportion travelling time, a rank correlation was calculated between the monthly percentages foraging time and the monthly percentages travelling time. A statistically significant negative correlation was found (Spearman rank correlation; r = -0.736; n = 12; p < 0.02). Squirrels groomed themselves throughout the year (table 2). However this behaviour was also observed more often in the spring summer period (table 3). Table 2 Activity budget of radiotagged squirrels. Percentage behaviour classes comprising total monthly observed activity Month June "July "Aue. 2 Sep ©er. . Nove Dec Janusnkeb: April May 1985 1985 1986 Sample size 17h36 22h33 24h37 16h25 27h17 25h35 22h00 22h48 22h43 19h35 31h35 Radiotagged 10 8 8 8 14 9 squirrels Behaviour % of total time spent active Foraging 272.5) 7.4.02 47.8:.08 77:.7007581 Seed-hoarding 37720 12.50 1:0:8 140 Travelling 1014. 322154541256 Interactions ; 053; &L8 Grooming : i 2,608 1 2.1 Resting 21001 2:5 Mating chases _ - Others Dr 3r5 Table 3 Mean percentage of different behaviour classes in summer and winter, calculated from table 2. A Mann-Whitney U-test was performed to compare the proportion of time spent on each behaviour class between the summer (April-September) and winter (October-March) period Behaviour % of total activity % of total activity U-value of Mann- in summer (n = 6) in winter (n = 6) Whitney U-test Foraging 67.7 (+ 3.9) 75.5 (+ 2.1) P = 0.002) I Travelling 17732 @:21%9) 13#82(1°2) Grooming 4.4 (+ 1.3) 2.4 (+ 0.3) Resting 5.0 (+ 0.6) 3.01 €..072) Gi@ne@ nm m 1 OrAWe Activity budget and foraging behaviour of the Red squirrel 347 Grooming occurred regularly during the day, taking less than a minute up to a maximum of about 3 minutes. The longer grooming-sessions were performed mainly after emerging from the drey. Dozing occurred also significantly more in summer than in winter (table3). It was seen most frequently from May till August (table2), during the months when the absolute time spent active was highest (table 1). While dozing, which took from a few to about 20 minutes, squirrels were seen using different postures according to varying weather conditions: In extreme conditions (heavy, cold winds, or very hot weather) they held the tail folded over the back, while in less severe weather they often let it hang down. Some types of behaviour were only seen during certain seasons. This was the case for seed caching behaviour in autumn (table 2). Mating-chases only take place in the reproduc- tive period. In 1986 females came into oestrus late January (unpublished data) and mating- chases were observed ın January and early February (table2). The second mating period started in May, when again males were seen chasing females. Interactions between squirrels not related to reproductive behaviour never seem to be numerous, although they occur regularly. The fact that they are only short, from less than a minute to 3 minutes, is partly responsible for the low percentages of total activity they comprise (table2). Expressed in absolute time, interactions only account for 35’’ per day ın February to a maximum of 4'37'' per day in June. The varıation in the amount of aggressive interactions between months is rather small and these data do not allow to distinguish different periods when squirrels are more or less aggressive, although other data suggest these differences exist (WAUTERS and DHONDT 1985; WAUTERS 1986). Habitat use Since the study-area consists mainly of Scots and Corsican pine, these two tree-species were most used by squirrels (fig. 3). Furthermore oak was used regularly during certain seasons. Although larch and beech were also present in most parts of the wood, squirrels 100 90 TREES 80 70 60 50 40 30 20 AFORAGING ON GROUND MONTH Fig. 3. Seasonal variation in habitat use of radiotracked squirrels. Percentage of total time Spent active on different treespecies and on the ground. Scots pine [_], Corsican pine ®) , oak DJ, other tree- species [IM 348 L. A. Wauters and A. A. Dhondt were only occasıonally observed foraging on them (fig.3). The time spent ın oak-trees showed extreme seasonal variation, related to the temporal availability of different food- items. From May to early July, which ıs the period between two successive cone crops, pines were used less compared to the rest of the year (fig. 3). The pines had shed their seeds by late April-early May, and it was not until the second half of June that new cones were taken. Differences in use between both dominant pine-species can be explained by the higher abundance of the Scots pine and by the fact that the Corsican pine had a much lower seed-crop that year (unpublished data). The differences were most extreme in June and July because the cones of the Corsican pine took longer to grow than those of the other species, and therefore were used only about a month later. Squirrels foraged a considerable amount of time on the ground in autumn, in April and early summer (fig. 3). Looking at the different parts of the tree used by squirrels (fig. 4), we see that foraging was mainly observed on small branches, as we would expect since most food ıs found there. Large branches, especially those of oak, were used somewhat more in June and July, related to food-searching behaviour. The trunk of the tree is mainly used for travelling to and from the ground and therefore climbed on more often when squirrels spent more time on the ground (fig. 4). 100 90 80 70 60 Dany —a N 4 DE HETIVE EINE SON DIFFERENT SUBSTRATES Sys | N ug N a a a a einen. 5 MONTH Fig. 4. Foraging substrate of squirrels in a coniferous wood. Percentage of active time spent on the ground and on different parts of the tree. Ground M, trunk IM ‚ large branches I] , small branches I Foraging behaviour and food-choice Foraging behaviour was further analysed, expressing the time used for searching and handling different food-items as a proportion of total time spent foraging (table 4). Early June 1985 the diet was still diverse with 5 food-items comprising more than 4 % each. About 20% of the time was spent in oak trees searching and eating caterpillars (11.1 %) and investigating dead branches for insect and fungi living under the bark (7.3 %). Squirrels were still regularly found foraging on the ground (18.2 %) where they searched for the remaining seeds in fallen pine-cones, or dug up their last food-reserves. The most important food was already found in Scots pine, with male flowers eaten early June (10.4%) and already starting on the new green cones of the 1985 seed-crop the second half of the month (40.4 %). From July onwards till April of the next year pine seeds became the most important food-resource (table4), with for Scots and Corsican pine together max- Activity budget and foraging behaviour of the Red squirrel 349 Table 4 Seasonal variation in feeding-behaviour of radiotagged squirrels. Searching and handling food- items expressed as monthly percentages of observed foraging time | Month June July Aug. . Oct. Nov. Dec. Jan. Feb. March April May 1985 1985 1986 Sample size 11h03 15h05 17h39 20h47 19h34 17h20 17h47 17h04 24h45 13h23 22h53 Feeding-behaviour % of total foraging-time Ground ; \ : DZ De e 3 2 10), OR Searching ‚OMEES 331425162 EB EE Sr) Fallen and cached Ä Ä . a en : r S 10) 22 Pine cones Fallen and cached acorns Berries Fungı Insects Seed-hoarding 3.3 Ka ES 7 o 1.4 0.7 30902 1.7 N — Corsican pine Searching Pine cones Buds and shoots Others Scots pine 56.7 64.3 Searching HINEARDHTNSG Pine cones k Ä 3 , 1 } RE Buds and shoots a | _ _ - = Male flowers 2 — _ Others Larch Searching Larch cones Others Spruce Searching Spruce cones Buds and shoots Oak Searching Acorns (+ cached ones) Fungi and insects living Under the bark Catterpillars Flowers Others Others om» ON& O2 [0% N FD Fe ao | Mor 0090| w SI nei Son DNS Sean are Lurr in imum percentages in January (91.1 %) and February (93.5 %). Some varıation was brought into this homogenous but energy-rich diet with other, so called secundary, food-items. In July squirrels were eating berries of an abundant shrub-species, Vaccınium myrtillus (10.1%), while mainly in September (3.9%) and during the following months they searched the ground for fungı. From September onwards acorns became an important part of the squirrels’ diet (table 4). Squirrels intensively foraged in the oak trees along the lanes during autumn, with searching and handling acorns comprising 17.6 % of total foraging time in September 1985, and still 12.6 % in November. When autumn progressed and more and more acorns were 350 L. A. Wanters and A. A. Dhondt shed, the ratio of searching time on handling time increased (0.14 in September, 0.35 in November, see table4). By the end of November squirrels shifted their foraging behaviour, again concentrating mainly on coniferous trees. During the same autumn period the squirrels spent a lot of time on the ground (table 4). Pine cones and acorns were bitten off and brought down where they were hidden in food- caches. This activity was performed mainly in September (4.9%) and October (4.8 %). Rather a lot of the cached food was recovered soon after ıt was hidden. Therefore searching for and handling stored food comprised 8.5 % and 11% of total foraging behaviour in October and November respectively (table 4). During the winter squirrels concentrated their feeding-efforts on pine-seeds, until late March (table 4). From Aprıl onwards the diversity of the diet increased. Warm April days caused the bursting of cones and a large proportion of the seeds shed. Moreover cone density had decreased strongly by now, especially in Corsican pine where the 1985 cone- crop was poor. This resulted in a marked increase in the ratio of searching to handling time in Scots pine (0.10 in March, 0.24 ın Aprıl and 0.26 in May) as well as in Corsican pine (0.11 ın March, 0.34 ın April and 1.45 in May)! Mainly buds of coniferous trees (12.8 %) were used ın April, while in May the young shoots and male flowers became more important (12.2 %). Squirrels found however the most suitable food-resource in flowering oak-trees. The oak flowers, abundant but only available for a short period of time, were the main food- item in this month, their handling time comprising 22.4 % of foraging time (table 4). By the end of the month flowers withered but now a new temporary food-resource became increasingly available, the emerging caterpillars. Up to 11.9% foraging time was spent searching for the insect-larvae and eating them. Discussion The actıvity-rhythm that was observed resembled the results obtained for red squirrels in different parts of its distribution area (OGNEV 1940 in the Soviet Union; SAINT-GIRONS 1966 ın France; DEGNn 1974 ın Denmark; PuRROY and Rey 1974 ın Spain; TonkIn 1983 in Great-Britain). The bimodal pattern in summer, the unimodal activity-peak in winter and an intermediate pattern in spring and autumn was found in most cases. Nevertheless there seems to be some geographical varıation and possibly varıation related to the type of habiıtat. In the West of France (SAINT-GIRONS 1966) squirrels stayed active throughout the day in winter, while in deciduous woods in England (Tonkın 1983) the early afternoon drop in activity during the summer months, found ın this study, was not present. Mild weather in winter and a low abundance of energy-rich food in summer are environmental factors that might have caused this increase in the time spent active. Combining our data on the activity budget, habitat use and changes in foraging behaviour allows us to explain the seasonal varıation observed in the activity pattern. Winter From December to February squirrels emerge from the drey before sunrise with all individuals showing an activity-peak the first daylight hours. Total active period is short and about 18-19 hours per day is spent in the drey. A very high proportion of activity is spent foraging for energy-rich food resources. Pine seeds, which were very abundant this year, comprise more than 90% of the diet, allowing the squirrels to have a high rate of food-intake. Therefore the energy-demands can be fulfilled in only a few hours of foraging. Staying active in cold weather causes severe heat loss and hence increases the energy-costs to maintain the body-temperature. Therefore hardly any time ıs spent on Activity budget and foraging behaviour of the Red squirrel 35 other activities, such as grooming and resting, and squirrels reenter the drey around noon. PuLLıaınen (1973) showed that in Northern Scandinavia dreys are very important for thermoregulation. The temperature in a drey while a squirrel is in it ıs about 10 to 20 >C above air-temperature. Since related species (GoLIGHTLY and OHMART 1978) are heterothermic, showing a decrease in body temperature while resting in a nest, we can assume that in cold conditions, resting in the drey is an important behavioural response resulting in a decrease of the energy-costs of thermoregulation. Spring When spring progresses, the seed-crop of the coniferous trees gradually gets depleted, while warmer weather in late April causes the cones to burst open and shed their seeds. This decrease in the abundance of primary food-items forces the squirrels to spent more time travelling between good feeding-sites, and furthermore to increase the diversity of their diet. Buds, shoots, flowers and even insects become important food-items. The change in habitat use, with squirrels foraging more often on the ground looking for scattered food-stores, also increases travelling time. This shift in diet and actıvity-pattern causes a decrease ın the rate of energy-intake and hence forces the squirrels to spend more hours active per day. Therefore activity gradually expands throughout the afternoon. It is also shown here that changes in habitat use are closely linked with the presence of food resources that are locally abundant but only used for a short time. Squirrels are concentrating their feeding efforts on oak flowers in May, shifting towards caterpillars in late May and June. This low-energy diet in May seems to be the major cause for the maximum in absolute time spent foraging (7h20’) which was observed in this month. Summer One of the aspects of the squirrels’ actıvity-rhythm that remains a puzzle is the pattern becoming biphasic and staying that way in summer. Although this pattern is the one most commonly observed ın anımals (AscHorr 1966) ıts mechanism is not yet understood. Concerning its function in squirrels ıt very likely has something to do with thermoregula- tion. The summer resting phase in the drey occurs during the warmest period of the day, when staying active might bring the squirrels into overheating problems. Another possibil- ity is that after the morning feeding period, the contents of the stomach reach capacity, and a break in activity ıs therefore necessary (ToNKIn 1983). Differences in physiological condition related to the stomach content may also be able to explain why in winter (after a long period without any food ıintake and a high cost of thermoregulation) activity starts wıth daybreak, while in summer (only a short nıght-rest without food-ıntake and a low cost of thermoregulation) it often takes an hour or more before squirrels emerge from the drey. This might be an important element of the endogenous basis of the activity-pattern. The long actıve period (about 8 to 10 hours) is, apart from daylength, related to the activity-budget and food-choice. In summer squirrels spend an important amount of time resting outside the drey during which they take postures that seem to have a temperature- regulating function. Hence it ıs likely that dozing is a behavioural response to avoid overheating during the actıve period. Also grooming comprises more than 5% of the activity-budget. The increasing grooming activity is most likely related to the higher amount of skin-parasites the squirrels suffer from, compared to the winter-period. In early summer squirrels still need to travel a lot to find good feeding-places, or between scattered food-resources, leaving less time available for searching and handling the food. In June for example, they travel some 108 minutes per day against only 35 in January. Furthermore squirrels take, mainly in June and early July, still an important 552 L. A. Wanuters and A. A. Dhondt amount of food-items wıth a low energy-content, such as the remaining seeds in fallen pine-cones, berries and insects. Therefore the rate of energy-intake is rather low, forcing the squirrels to forage for a longer time period. In July afternoon activity was most important, and the peaks were not very pronounced. This could be explained by the fact that at that time nearly all the radıo-tagged squirrels were adult females, which had their litter born early July (WAUTERs and Duonpr 1985). During the first week of lactation they spent a lot of time nursing the young and often emerged from the drey several hours after sunrise. Not only the onset of activity was delayed, but also the synchronicity was lost, since different females left the drey at different times. Autumn In autumn the afternoon activity peak becomes less important, nevertheless the absolute time spent active stays around 6h30’. During this time of the year high-energy food is most abundant, which is retlected in the decrease of travelling time and the increase of feeding time. Squirrels temporarily shift their food-choice and concentrate on the seeds of deciduous trees, that are only available for a short period of time. In 1983 beech was regularly visited, harvesting the large beechmast crop while oak, which had a very poor acorn crop that year, was disregarded (WAUTERsS 1984; WAUTERS and DHonDrT 1986). In the autumn of 1985, with a crop failure in beech but an abundant crop for oak, they concentrated on the latter. Feeding only on energy-rich food-items the rate on energy- intake becomes very high. Furthermore the air temperature is not yet that low that squirrels would have high costs of thermoregulation, nor do they seem to have to perform behaviour to avoid extreme heat loss. It is also at this time of the year that their body- weight is highest (unpublished data). Another response to the high abundance of temporary primary food-resources is food- caching behaviour, which has an important effect on food-availability later on. Hiding acorns decreases the intensity of interspecific competition with animals harvesting the seeds as soon as they have fallen to the ground. The caching of cones provides the squirrels with energy-rich food in the late spring period between two successive seed-crops. Stored in the damp upper-layer of the forest ground the cones remain closed and seed-shedding is delayed. The data presented here on the proportion of each food-item ın the diet, should be combined with quantitive data about the rate of energy-intake to enable us to calculate the daily energy-ıintake for each month. Lab-experiments should be done, measuring the energy-costs of metabolism under different conditions allowing to calculate the energy-balance of the squirrel, and hence understand better the influences of environmental factors and certain types of behaviour on the squirrels’ condition. Acknowledgements This study was financed by a Grant obtained from the Institute for the development of Scientific Research in Industry and Agriculture (I.W.O.N.L.) ın Brussels. Zusammenfassung Aktivität und Verhalten bei der Nahrungssuche von Eichhörnchen (Sciurus vulgaris Linnaeus, 1758) im Nadelwaldhabitat Der tägliche und jahreszeitliche Aktivitätsrhythmus, die Habitatnutzung und die Nahrungswahl von freilebenden Eichhörnchen (Sciurus vulgaris) wurden mit Hilfe von Radiotelemetrie in einem Nadel- wald untersucht. Der tägliche Aktivitätsrhythmus änderte sich während des Jahres. Im Winter waren die Eichhörnchen nur wenige Stunden aktiv. Sie suchten am frühen Vormittag vor allem energiereiche Kiefersamen. Die lange Zeit, die sie im Nest bleiben, reduziert die Kosten der Thermoregulation. Im Frühling verlängere sich die Aktivitätsperiode allmählich, und die Tiere wurden auch nachmittags Activity budget and foraging behaviour of the Red squirrel 338 aktiv. Dieses wurde wahrscheinlich durch Milieufaktoren verursacht, wıe wachsende Tageslänge, zunehmende Temperatur und abnehmende Mengen von Baumsamen. Der Aktivitätsrhythmus der Population wurde bimodal. Für die monatlichen Mittelwerte erklären Tageslänge und Temperatur 93% der Variation der Aktivitätsdauer. References ÄSCHOFF, J. (1963): Comparative physiology, diurnal rhythms. Ann. Rev. Physiol. 25, 581-600. ASCHOFF, J. (1966): Circadian activity pattern with two peaks. Ecology 47, 657-662. Decn, H. J. (1974): Feeding activity in the red squirrel (Sciurus vulgaris) in Denmark. J. Zool., London, 174, 516-520. GOLIGTHLY, R. T. Jr.; OHMART, R. D. (1978): Heterothermy in free ranging Abert’s squirrels (Scinrus aberti). Ecology 59, 897-909. GRONWALL, ©. (1982): Aspects of the food ecology of the red squirrel (Sciurus vulgaris L.). Ph. D. Thesis University of Stockholm. Sweden. Motıter, H. (1983): Foods and foraging behaviour of Red (Sciurus vulgarıs) and Grey (Sciurus carolinensis) squirrels. Mammal Review 13, 81-98. Ocnev, $. I. (1940): Animals of the U.S.S.R. and Adjacent Countries. 4. Rodents. Moscow, Leningrad. An SSSR-Israel Programme for Scientific Translations 1966, Jerusalem. Paurs, R. W. (1979): Body temperature dynamics of the Red squirrel (Tamiasciurus hudsonicus). Can. J. Zool. 57, 1349-1354. PULLIAINEN, E. (1973): Winter ecology of the Red squirrel (Sciurus vulgaris L.) ın northeastern Lapland. Ann. Zool. Fennici 10, 487494. PuURRoY, F. ]J.; Rey, J. M. (1974): Estudio ecologico y sistematico de la ardılla (Sciurus vulgaris) en navarra distribution, densidad de poblaciones, alımentacion actıvidad dıana y anual. Buletin de la Estacion Central de Ecologia 3, 71-82. SAINT-GIRONS, M.-C. (1966): Le rhythme circadien de l’activit€ chez les mammiferes holarctiques. Mem. Mus. Nat. Hist. Naturelle, Paris. Serie A. 40, 101-187. SHORTEN, M. (1962): Squirrels: Their biology and control. Bull. Ministry Agrıc., Fisheries and Food. London: HMSO. 184. TITTENsSoR, A. M. (1977): Red squirrel. In: The Handbook of British Mammals. Ed. by G. B. Corbet and H. N. Southern. Oxford: Blackwell Scientific Publications. 2nd ed., 153-164. Tonkın, J. M. (1983): Activity patterns of the red squirrel (Sciurus vulgaris). Mammal Review 13, 99-111. WAUTERS, L. (1984): Inleidende studie tot de populatie-ecologie van de Europese Rode eekhoorn (Scinrus vulgaris L.). Licentiaatsverhandeling. Universitaire Instelling Antwerpen. WAUTERS, L. (1986): Population dynamics of red squirrel (Sciurus vulgaris) populations in different habitats. Lutra 29, 346. WAUTERS, L.; DHONDT, A. A. (1985): Population dynamics and social behaviour of red squirrel populations in different habıtats. Proc. XVII. Congr. Internat. Union Game Biologists 311-318. WAUTERS, L.; DHONDT, A. A. (1986): Dichtheid en home ranges van een populatie eekhoorns Sciurus vulgarıs L., 1758 in Belgie. Lutra 29, 243-260. ZWAHLEN, R. (1975): Die lokomotorische Aktivität des Eichhörnchens (Sciurus vulgarıs). Oecologia 1857269316. Authors’ address: Luc WAUTERs and Anpr£E A. DHOoNDT, Departement Biologie, Universitaire Instelling Antwerpen, Universiteitsplein 1, B-2610 Wilrijk, Belgie Social organization and habitat use of introduced populations of the Vole Clethrionomys rufocanus (Sund) in Central Finland By H. YLÖönen and J. VırTaLA Department of Biology, University of Jyvaskylä, Finland Receipt of Ms. 5.9. 1986 Abstract Grey-sided voles (Clethrionomys rufocanus) from northern Finland were introduced on two islands of the Lake Konnevesi in 1983. The islands were emptied of local rodents before the introduction took lace. n One of the populations died out due to predation by stoats (Mustela ereminea) already during the first summer. The new population introduced in autumn either didn’t survive over the next winter. The island was recolonized by the bank vole during next summer. The other population on an island of 1.7 hectares survived from summer 1983 until 1986 and reached a peak density of 69 individuals in August 1985. The immigrant bank voles were not able to establish a population during the presence of grey- sided voles. The voles preferred moist oligo-mesotrophic habitat with hummocs of Ledum palustre together with Vaccinium myrtillus and V. uliginosum. This habitat is characteristic to northern Finland. In summer the breeding females were mutually amicable but territorial. Mature males behaved mutually aggressive but were not strictly territorial. Contrary to the north Finnish populations the introduced voles aggregated on most suitable overwintering habitats. In spring there were lots of signs of foraging on those areas. Dispersal of the voles to neighbouring islands was also observed. Characteristics of the social organization and habitat preference of the introduced populations are compared with other populations. The social organization of this population was quite flexible, too. Introduction The socıal organızation is studied from two points of view: 1. has it something to do with cycling of some small rodent populations or 2. is the social system a stabilizing factor in the population during all phases of a possible cycle (Vırrara et al. 1986). The ways of studying these problems are either experimental manıpulation of populations, long term survey of natural populations, or both. In 1982 we started an enclosure study on social organızation of bank vole (Clethrio- nomys glareolus Schreber) in Konnevesi Central Finland. The study has been going on since then and the first results have been reported by YLöneEn and VırraLa (1985) and YLÖNEN et al. (1986). The work was inspired by the study on social organizations of subarctic populations of the voles Clethrionomys rufocanus (Sund) and Microtus agrestıs (L.) by Vırrara (1977) in Kilpisjärvi, Finnish Lappland where distinct differences have been found concerning social organization and social population regulation between these species that adapted to different habitats (c.f. VırraLa and HOFFMEYER 1985; VIITALA et al. 1986). This forced us to try to introduce populations of grey-sided voles in Central Finland as control populations for the local bank vole populations in a long-term study on social organızation. The introduction could give information on influence of latitude and different environ- mental conditions on forming of social organızation and on possible changes in competi- U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5206-0354 $ 02.50/0 Z. Säugetierkunde 52 (1987) 354-363 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Grey-sided vole in Central Finland 335 tive situation between rodent species in new environment (HENTTONEN et al. 1977; Hansson 1974). It could also clarıfy factors determining geographical distribution of Clethrionomys rufocanus, which just in the lowland of Fennoscandia and Sovjet Union is rather curious (HENTTONEN and VıITaLa 1982; STENSETH 1985). Study area Konnevesi Research Station is situated in northern Central Finland (62°15’N, 26°26’E) and it lies about 250 km from the nearest northern and eastern ranges of C. rufocanus (Fig. 1). After two unsuccesful introduction trials in 1981 (c.f. VırraLa et al. 1986) and 1982 study populations could be founded on two islands of the Lake Konnevesi in summer 1983. Both islands Siimarinsaarı (1.3 ha) and Iso-Korppi (1.7ha) show the same main habıtats (Fig. 2). The middle parts of the islands are higher and more oligotrophic than the brims. Both grow forest with either Picea abies and Betula pendula or Pinus silvestris as main trees. On Iso-Korppi Populus tremula is also common. The field layer is built up mainly by Vaccinium myrtillus or V. vitis-ıdaea. Parts of the field layer are almost free of vascular vegetation because of strong shading effect of spruce forest. Fig. 1. Study area with the islands with introduced populations of the vole Clethrionomys rufocanus and the map of the distribution of the species in Finland. Also the location of the study site in Konnevesi, Central Finland ıs given. 1 = Sıimari-Island, 2 = Iso-Korppi The brims of the islands are oligo-mesotrophic, partly paludified heath forest with the same tree species as the middle parts. The field layer ıs dominated by knechigh brush of Ledum palustre, Vaccinium uliginosum and V. myrtillus. The ground is a mosaic of brushy hummocs and depressions free of vascular plants between them. This vegetation type appears normally in northern Finland. Siimarinsaarı is a bit smaller and higher and that is why the vegetation belt dominated by Ledum- Myrtillus brushes on the brims is smaller than on Iso-Korppi. Siimarinsaarı at the head of a bay has the minimum distance of 120 meters to the mainland. Iso-Korppi lies 220 meters from the nearest small island and more than 1.5 km from the mainland. 356 H. Ylönen and J. Viitala DR FH 2 EeIi3 Le Fig. 2. Vegetation of the study islands. 1 = Ledum-Myrtillus bruhes, 2 = Fresh spruce and decidous forest with mainly Vaccinium myrtillus field layer, 3 = dry pine forest with Vaccinium vitis-idaea Material and methods The small mammal communities on the islands before the freesetting of grey-sided voles consisted of Clethrionomys glareolus, Micromys minutus, Sorex aranens and S$. minutus. These species were captured by snaptrapping before introduction of C. rufocanus or later during live trapping. The islands were covered by a grid of trap stations with ten meter intervals. On Iso Korppi the traps on each trap line were set at 20 meter intervals. On Sıimari island and Iso Korppi 96 and 76 lıve traps of Ugglan Special type (Hansson 1967) respectively were set. On Iso Korppi each trap station had a trap for half of the total trappıng time. The trapping periods lasted five days and the traps were checked ten times each period. The islands were trapped once a month from July to October in 1983 and from May to October in 1984. In 1985 we were able to trap Iso Korppi in June, August and October. Following observations were made every time an anımal was captured: trap locality, time, species, individual identity, sex, sexual status, body weight, infestation by ticks, scars and possible other signs of aggression and behaviour of the anımal(s) in the trap. In October 1985 and June 1986 some snap and live trappıng was done on the neighboring islands to determine dispersal of C. rufocanus. The existence of C. rufocanus can be seen the basıs of typical feeding places, too, ı.e. heaps of sticks of blueberry stems under natural covers and hollows — often also in live traps. To determine possible changes in food preference of the grey-sided vole between its native and introduction area all possible information on foraging habits were noticed during the study periode. In spring 1986 after the snow melt winter foraging remains were collected around every trap station. Results Population fluctuations We released three males and three females of C. rufocanus on Siimarı island at the end of June 1983. Altogether 19 individuals were captured 72 times until the beginning of September. Then the whole population was killed apparently by a single stoat, Mustela erminea. Reintroduction after the removal of the stoat was unsuccesful and the island was recolonized by the native ©. glareolus. Already in the beginning of June 1984 24 male and 27 female bank voles were captured on the island concentrating on brims. Grey-sided vole in Central Finland 357 Four females and three males were released on Iso-Korppi island in June 1983. During that summer 36 individuals were marked. The density was at highest in September: seven females and thirteen males (Fig. 3). Only one female and three males survived through the winter. The population recovered well in 1984, however, and after good winter survival it grew to 67 individuals (39 ind./ha) in summer 1985. In August twelve females and four males from the population were introduced on another, a little bit larger island to save the population after the possible winter collapse on Iso-Korppi. In May 1986 only a single male was found and the population of C. rufocanus was lost. 1983 1984 Fig. 3. Population fluctuations of C. rufocanus on Iso-Korppi between 1983 and 1986. Introductions or removals of voles are shown with arrows. Also the amount of immigrant bank voles is given. Solid line = females, broken line = males Late summer every year individuals of C. glareolus were immigrating the islands populated by C. rufocanus. They were unable to occupy the most favored brim areas. None of them were able to survive through the winter. Habitat preference Both introduced anımals and their offspring were captured almost without exception on the Ledum-Myrtillus belt of the brims and on other moist habitats with rich undergrowth (Fig. 4). The rather dry inner parts of the islands covered by pine and spruce forest with sparse field layer were avoided by C. rufocanus. Only the few ımmigrant bank voles, C. glareolus, were forced to use these areas. In the absence of,C. rufocanus in 1984 on Siimarı Island ©. glareolus also preferred the moist Ledum-Myrtillus areas (Fig. 4). There were no significant differences ın habitat use of C. rufocanus between seasons and years (Table 1). Social organization The number of breeding females did not exceed five on Iso-Korppi during the two first study years (Fig.5). The number of mature males varıed between two and four. As in previous studies (c.f. review by VırTaLa and HOoFFMEYER 1985) females were mutually tolerant but mature and submature males were highly aggressive towards each other. After high winter survival ten females were mature already in mid of June 1985. Only one trap was visited by two different breeding females even though home ranges of these 358 H. Ylönen and J. Vütala Table 1 Percentual use of the trapstations of different habitats during the study years 1983-1985 on Iso-Korppi N shows the amount of captures in that year. All differences are significant with p < 0.001 Habitat Ledum-Myrt. Myrtillus Vıitis-ıdea 52 42 47 Fig. 4. Movements of the small population of C. rufo- canus on Siimari-Island in July/August 1983 (a) and the frequence of trap visits by C. glareolus ın August 1984 (b) after the recoloni- zation of the island by the species. The figure shows the similar habitat use of the two species without compe- tıtion Jun 1985 Aug 1985 Fig. 5. Home ranges of mature females and movements of mature males of C. rufocanus on Iso-Korppi in June and August 1985 (Z >[ger ospe 995) Tergey saypudapy-wnpo7 ya ur sum J9JuIm ay2 107 P91e3s133e uonerndod yı ey opnppuo9 3M susıs ZuLIIUIMIPAO Jo uonepnuns ayı yaım 1942930, ], 'Osje spewrue sınyeur Jo J9urpIone Sıpsadsenur jo 9se3195p pue spenprarpur [e19A9s Aq sden a[3urs Jo asn Zursesssur ue Moys sıynsaı Surdden Sy], "Sursgummısao Jo sußıs 0u= 4° s9JoA ayı Aq 191uUIM Surunp poJ Arısqfıq pur -9nIgq zo sureurı yonur A19A = | 3I9UM soLIOZIYIE9 ınoy ur p>ypissepp pur uones dem A1aa9 punoAe pauru11zI9p SEM SULISIULMIIAO Jo suSis Ay], 'TeI01 ay2 JO 9, 98 sem s[eNPIATPUT 9J0UI 10 0M1 Aq pasn sden ur soımıdes Jo saqumu ayy, 'UMOUYS sı (suumjo9 papeys a[fduıs) sden asayı panısta JAeY YIIyM STenpIArpur Jo Jaquınu ayı pue (suumpo9 papeys sso19) sıısta 0MI uey 310UI yıım uoneıs dem ysea jo Ayuanba.y Zurdden ayj, put Mous ay2 Ao1fe ısnl 9867] Surids ur sudıs JurısgurmıaAo Jo vonnqrnstp syı pur ‘sg6] ur Surdden uumane ıse] sy Surnp uonepndod ayı jo Asuanbayy zısıa den 942 pur soJoA Papıs-A913 Seulsf ayı Jo SUIUIWAOM| ‘9 '31.] sBEL Idas s86L Idag | EI dj] -— = l . [| N 360 H. Ylönen and J. Viitala Table 2 The distribution and amount of food remains collected by the grey-sided vole in different habitats The detected storages in the 10x 10 m area around the trap station form a single sample. Dry weight in grams (W) of the stored plant material found in May 1986 is given. The group “others” consists of annual shoots and leaves of Betula pendula, Picea abies, Pinus silvestris and Populus tremula, which occured ocassionally in the samples. The amount of trap stations in each habitat is given in table 1 Habitat and the Vaccınium Vaccınıum Ledum “others” amount of samples myrtillus vitis-idaea palustre W % W % W % W % Ledum-Myrt. 120330 50.7 a il 33.24 14.0 10.04 4.2 (24) Myrtillus 26.38 43.4 316910525 0.520009 oh 8 (9 Vitis-idaea 11.12 27.9 23.38 58.6 5.39, 13,5 (6) females covered bigareas. Thusthey were stricetly territorial. In August there were eight breeding females behaving in the same way (c.f. Fig. 5). Also the number of mature males in June was twıce as high as in previous summers. This resulted ın very much aggression between them. Fresh wounds were noticed in every trapping day. In August 1985 there were only two mature males left. Males moved on larger areas than females. Territorial behaviour begun to de- crease towards autumn. The number of trap stations visited by several different individuals whether mature or not was increasing. In October distinct aggrega- tions were recognized (Fig. 6). Fıg. 7. Dispersal of grey-sided voles to the neighbouring islands during the late summer 1985. The graphs show the amount of voles and shrews captured in 100 trap nights. The arrows show the dispersal of grey-sided voles BIC. &r Aläle S.Sp. Overwintering The survival of the population was very low in the first winter and there were very little signs of winter foraging. Next winter 1984/85 the survival was about 90 percent for both sexes (s. Fig.3). Already the autumn trapping gave the impression of aggregation forma- tion. The signs of winter foraging cumulated also on the southern but especially northern brim parts of the island (Fig. 6). The big population going to overwinter in autumn 1985 seemed to form distinct aggregations in late September. Even though the population died out probably in late winter - early spring, the foraging signs concentrated on the sites of the aggregations. Grey-sided vole in Central Finland 361 The preferred food, collected in heaps typical for C. rufocanus in winter, were Vaccıninm myrtillus, but also in great amount V. vıtis-idaea. Ledum palustre and annual shoots of local spruce, pine and deciduous trees were also found in the heaps (Table 2). Dispersal Dispersal of C. rufocanus was evidenced by snap trapping on two nearby islands a small one 70m from Iso-Korppi and a little bigger one 200m from Iso-Korppi. On four more islands 400 to 600m away (Fıg.7) Microtus agrestis and C. glareolus only were captured. Small heaps of blueberry stems and branches of willows out by voles were found on three of these islands probably indicating unsuccessful attempts of C. rufocanus to colonize these islands. Submature bank voles were immigrating Iso-Korppi every year in late summer. They did not survive through winter, however. Not a single individual of M. agrestis was ever observed on the islands populated by C. rufocanus even though field voles are numerous on the other islands of the lake Konnevesı. Discussion The attempt to introduce C. rufocanus on central Finnish mainland enclosure was unsuccessful (Vırrara et al. 1986). Also the fırst introduction attempts on islands failed because of poor habitats or predation by Mustelids. According to OKSANEN and OKSANEN (1981) one stoat or weasel is enough to destroy the whole population on such an island as we used in our study. Also on Iso-Korppi a stoat was observed in August 1984. It caused a decline of the population but not its total destruction. It ıs possible that the crash during winter 1985/86 have been due to predation. There were lots of Mustelids in Konnevesi area during the crash after two high density years of small rodents. The decline of the population must have happened during late winter or early spring. That was indicated by the big heaps of blueberry stems (c.f. Table 2) with also fresh cuttings used as winter food. The single male captured in May 1986 was in very poor condition. The reasons of the crash are out of scope of the present paper, however. The populations of small rodents on rather small islands must often be founded by immigrants allmost every year (c.f. Pokkı 1981). The reasons of extinctions may be predation, hard climatical conditions and shortage of food. The population colonized foremost the brims of the islands with rich growth of Zedum palustre and Vaccinium myrtillus. These areas with rocks and hummocs provide food and shelter. Nesting holes are important, because according to HENTTONEN and VIITALA (1982) grey-sided vole is not able to dig very well. Dry mossy areas preferred by it ın northern Finland (KareLA 1957; KaLeLa et al. 1971) do not exist on these islands. Also bank voles preferred the same brim areas on Siimari-Island 1984 in the absence of C. rufocanus. The immigrant bank voles on Iso-Korppi were captured with only one exception in the dryer middle parts of the island. Presence of C. rufocanus obviously forced it to use poor habitats where it did not have the possibility to overwinter. In northern Finland C. rufocanus is stronger in interspecific competition of these two species (HENTTONEN et al. 1977). Hansson (1974), however, assumes that in areas, where C. rufocanus is on the southern border of its range C. glareolus could be in competitive advantage because it is very numerous on those habitats expectedly preferred by C. rufocanus. In the present island study this could not be the case because the few individuals immigrating the island were not strong enough to survive over the winter. The immigra- tion of the submature bank voles took place ın so late summer that the immigrants did not breed anymore - probably also due to interspecific inhibition (KAARSALO and WALLGREN 362 H. Ylönen and J. Viitala 1985). The recolonization of Siimari-Island by C. glareolus after the disappearence of grey- sided voles was very successtull. The immigrant bank voles entering the island must have swum at least the distance of 200 metres from the nearest neighbouring island. There are some unpublished observations on swimming ability of C. rufocanus during lemming migrations (HENTTONEN unpubl.; KAIKUSALO unpubl.; VırTara unpubl.). During the present study they had evidently swum at least 200 metres. Some signs of blueberry storages were found also on two islands about 500 metres north from Iso-Korppi. No direct evidence by trapping exist, however. Interesting ist the total absence of Microtus agrestis on the islands colonized by grey- sided voles (c.f. VırTaLa 1977; HENTTONEN et al. 1977) although the field vole was the most numerous rodent species on the neighbouring islands. Usually M. agrestis is much a more efficient colonizer than any of the Clethrionomys species (VırTaLa 1977; PokkI 1981). The social structure of the breeding colonies of the introduced C. rufocanus colonies were the very same observed in other wild Clethrionomys populations (for review s. e.g. VırraLA and HOFFMEYER 1985; VIITaLa et al. 1986) ı.e. strict territoriality of mature females and overlapping of much larger home ranges of mature males. Contrary to the observations of VırTaLA (1977) and HENTTONEN and VIITALA (1983) in northern Finland the island population of C. rufocanus aggregated during winter in Middle-Finland. The population on Iso-Korppi behaved similarly as the mainland popula- tion of ©. glareolus native to the area (YLÖNEN and VırraLa 1985). The flexibility of the overwintering strategy of C. rufocanus is suggested already by VırTaLaA and YLÖNEN (1985). Heat saving by aggregating seems to be especially advantageous on the islands where the snow cover is eaven thinner than on the mainland of Central Finland. This kind of behaviour may, however, increase small Mustelid predation, and cause population crashes described also in the present study. The small introduced C. rufocanus populations on the islands or on mainland in Central Finland had the dangers of founder populations (MCARTHUR and WıLson 1967; DIAMOND 1984). They may die because of environmental or endogenous reasons or by accident. All our study population died out during the five years study. Some new populations - still unknown for us -— may have been founded by dispersers. According to Pokkı (1981) only one pregnant female is enough to found a new island population. Acknowledgements We thank Asko KaıkusaLo and HEIKKI HENTTONEN for providing us with most of the grey-sided voles introduced on the islands. The study has been supported financially by the State Board for Science, Emil Aaltonen foundation, Finnish Cultural Foundation and by Jyväskylä University. Zusammenfassung Soziale Organisation und Habitatpräferenz von Clethrionomys rufocanus-Populationen in Mittelfinnland Graurötelmäuse, Clethrionomys rufocanus (Sund), aus Nordfinnland wurden auf zwei Inseln des Sees Konnevesi (Mittelfinnland) im Sommer 1983 eingebürgert. Vor der Einbürgerung wurden die heimischen Nager der Inseln mit Schlagfallen abgefangen. Eine der Populationen wurde schon während des ersten Sommers durch ein Hermelin (Mustela erminea) zerstört. Die im Herbst neu eingebürgerte Population überlebte den folgenden Winter auch nicht. Die Insel wurde im nächsten Sommer durch Rötelmäuse (Clethrionomys glareolus) sehr schnell wieder kolonisiert. Die andere neu gegründete Inselpopulation konnte sich vom Sommer 1983 bis Frühjahr 1986 erhalten und erreichte im Spätsommer 1985 die höchste Dichte von 69 Individuen auf einer Fläche von 1,7 ha. Die auf die Insel immigrierenden Rötelmäuse hatten keine Möglichkeit, während der Anwesenheit der Graurötelmaus eine Population zu stabilisieren. C. rufocanus bevorzugte feuchte Habitate, die ein Vegetationspolster von Vaccınium myrtillus, Grey-sided vole in Central Finland 363 Ledum palustre und Vaccinium uliginosum aufwiesen. Diese Habitate sind charakteristisch für Nordfinnland. Während des Sommers lebten die reproduktiven Weibchen territorial. Geschlechtsreife Männchen waren während der Reproduktionsperiode untereinander sehr aggressiv. Im Gegensatz zu nordfinnischen Populationen verbrachten die eingebürgerten Graurötelmäuse den Winter im sozialen Verband in den Vorzugshabitaten. Anzeichen der gemeinsamen Überwinte- rung waren nach der Schneeschmelze reichlich vorhanden. Abwanderungen von Graurötelmäusen auf Nachbarinseln wurden auch nachgewiesen. Die soziale Organisation sowie die Bevorzugung bestimmter Habitate durch die eingebürgerte Population wird mit anderen Clethrionomys-Populationen verglichen. Die eingebürgerten Graurötel- mäuse zeigten, wie schon für andere Clethrionomys-Arten nachgewiesen, eine flexible soziale Organisation. References Diıamono, J. M. (1984): “Normal” extinctions of isolated populations. In: Extinctions. Ed. by M.H. Nitecki. Chicago: Chicago Univ. Press. 192-246. Hansson, L. (1967): Index line catches as a basis of population studies on small mammals. Oikos 18, 261-276. Hansson, L. (1974): Nya utbrednings uppgifter för grasiding, med synpunkter pä eventuella kon- kurrens med skogssork. Fauna Flora 69, 91-94. HENTTONEN, H.; KAIKUSALO, A.; TAsT, J.; VIITALA, J. (1977): Interspecific competition between small rodents in subarctic and boreal ecosystems. Oikos 29, 581-590. HENTTONEn, H.; VIITALA, ]J. (1982): Clethrionomys rufocanus (Sundevall, 1846) - Graurötelmaus. In: Handbuch der Säugetiere Europas. Bd. 2/1. Ed. by J. Niethammer and Krapp. Wiesbaden: Akademische Verlagsgesellschaft. 147-164. HENTTONEN, H.; VIITALA, J. (1983): Aspects on the social behaviour of cyclic microtines in winter. Oulanka Reports 4, 79-81 (in Finnish, with English summary). KaarRsALo, K.; WALLGREN, H. (1985): Inhibition of pregnancy in Clethrionomys rutilus females exposed to conspecific males and to males of C. rufocanus. Abstract No. 0333. In: Abstracts of papers and posters. Fourth Intern. Theriol. Congr., Edmonton, Alberta, Canada, 13-20 August 1986. KAterLa, ©. (1957): Regulation of reproduction rate in subarctic populations of the vole Cleth- rionomys rufocanus (Sund). Ann. Acad. Sci. Fenn. A IV 34, 1-60. KALELA, O.; Koponen, T.; Yrı-PıETILÄ, M. (1971): Übersicht über das Vorkommen von Klein- säugern auf verschiedenen Wald- und Moortypen Nordfinnlands. Ann. Acad. Sci. Fenn. A IV 185, 1-13. MCARTHUR, R. H.; Wırson, E. ©. (1967): The theory of island biogeography. Princeton, New Jersey: Princeton Univ. Press. OKSANEN, L.; OKSANEN, T. (1981): Lemmings (Lemmus lemmus) and grey-sided voles (Cleth- rionomys rufocanus) in interaction with their resources and predators on Finnmarksvidda, north- ern Norway. Rep. Kevo Subarctic Res. Stat. 17, 7-31. Pokk1, J. (1981): Distribution, demography and dispersal of field vole, Microtus agrestis (L.), in the Tvärminne archipelago, Finland. Acta Zool. Fennica 164, 148. STENSETH, N. C. (1985): Geographic distribution of Clethrionomys species. Ann. Zool. Fennici 22, 215-219. VIrTaLa, J. (1977): Social organization in cyclic subarctic populations of the voles Clethrionomys rufocanus (Sund) and Microtus agrestis (L.). Ann. Zool. Fennici 14, 53-93. VIITALA, J.; HOFFMEYER, 1. (1985): Social organization in Clethrionomys compared with Microtus and Apodemus: social odours, chemistry and biological effects. Ann. Zool. Fennici 22, 359-371. VIITALA, J.; YLÖNEn, H. (1985): Variability of social overwintering strategies in the Genus Cleth- rionomys. Aquilo Ser. Zool. 24, 107-111. VIITALA, J.; BONDRUP-NIELSEN, S.; BuJaLskA, G. (1986): Social structures in Clethrionomys in relation to resource abundance. Holarctic Ecol. (in press). VIrTALa, J.; KoJoLa, T.; YLöNnEn, H. (1986): Voles killed by ticks - an unsuccessful attempt to introduce north Finnish Clethrionomys rufocanus into an enclosure in central Finland. Ann. Ent. Fennicı 52, 32-35. YLÖNEn, H.; VırTara, J. (1985): Social organization of an enclosed winter population of the bank vole Clethrionomys glareolus. Ann. Zool. Fennicı 22, 353-358. YLönen, H.; KoJoLa, T.; VırTaLa, J. (1987): Changing spacing behaviour and demography in an enclosed population of Clethrionomys glareolus (Schreb). Holarctic Ecol. (submitted). Authors’ addresses: Hannu YLÖNEN and Jussı VırTaLa, Department of Biology, University of Jyväskylä, Yliopistonkatu 9, SF-40100 Jyväskylä and Konnevesi Research Sta- tıon, SF-44300 Konnevesi, Finland Reproductive biology of Arvicola sapidus (Rodentia, Arvicolidae) in the Ebro delta (Spain) By J. VENTURA and ]. GosSÄLBEZ Catedra de Vertebrados, Universidad de Barcelona, Spain Receipt of Ms. 16. 10. 1986 Abstract Studied the reproductive cycle of the southwestern water vole, Arvicola sapidus in the Ebro delta (Tarragona, Spain). 145 specimens (79 8d, 66 ??) captured between 1981 and 1984 were analysed. To evaluate the maturity and sexual activity in males the cytological analysis of the testicular content, minor and major diameter of the testicle and length of the seminal vasicle were considered. In the females, the state of the vulva, development and vascularisation of uterus and ovarıies, and embryos and placental scars were registered. The presence of spermatozoids is constant in adult individuals throughout the year. The seminal vesicles and the testicles vary considerably in size throughout the year. The minimum is reached in December and January, and the maximum in June. Between March and October all adult females are sexually active. The sexual activity is lowest in December-January with values around 35445 %. The presence of pregnant females is limited to the February-October period. The average litter size is 3.31 # 1.20, n=19. The continued sexual activity of some females during autumn and winter, the constant presence of spermatozoids in males throughout the whole year as well as the capture of juvenile individuals in January, suggest the possibility of reproductive activity in November and December. Extrinsic factors may regulate to a certain degree the increase or decrease of sexual activity in different years. Introduction There is little information so far on the different aspects of the bionomic status of Arvicola sapidus. This species inhabits all of the Iberian Peninsula and southern, central and the north west tip of France (MıLLErR 1912; HEIM DE BaLsac and GuisLaın 1955). The information concerning the characteristics of the biological cycle of this rodent in all of ıts area of distribution is very scarce. Apart from the results obtained by DELoST (1968) about the reproductive characteristics of a population near the Tarn river (France) during one whole annual cycle, the rest of studies are less detailed. Such studies deal with descriptions on the sexual state of specimens obtained during non periodical catches (GosALBEZ 1976; ZaBALA 1983) or those dealing with references included within general treatises on mammals or rodents (DiDIER and Rope 1935, 1939; SAINT-GIRONS 1973; LE Louarn and SAINT-GIRONS 1977; REICHSTEIN 1982; BauDoın 1984). The purpose of this paper is to determine the reproductive characteristics and par- ticularıties of Arvicola sapidus ın north east Iberia. This study, the first one concerning the subject in the Iberian Peninsula, was carried out in the Ebro delta (Tarragona). Material and methods The specimens analysed have been trapped each month from January 1983 to December 1984, at the banks of the canal network surrounding the Encanyissada lagoon in the Ebro delta (Tarragona, Spain). In addition, specimens were collected in the same area in January 1980, March, April and December 1981, and between January and March 1982. In all, 79 males and 66 females were studied. U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5206-0364 $ 02.50/0 Z. Säugetierkunde 52 (1987) 364-371 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Reproductive biology of Arvicola sapidus in the Ebro delta 365 From each specimen the body weight was taken, as well as the following measurements: head and body length, length of tail, hind foot length and ear length. In order to determine the stage of maturity and sexual activity in males, the following characteris- tics and parameters were considered: position of testicles (abdominal or scrotal), testicular cell content (GoSÄLBEZ et al. 1979), minor and major diameters of the testicle and length of the seminal vesicle (GoSÄLBEZ and Sans-CoMmA 1976). In the testicular squashes the following categories were distinguished: immature: specimens lacking spermatozoids and spermatids in the testicle. submature: specimens with few spermatozoids and spermatids in the testicle. mature: specimens with complete reproductive capacity determined by a large amount of sper- matozoids in the testicle. In the females, the characteristics under consideration were: status of the vulva (open or closed), presence of mammary tissue, stage of development and irrigation of uterus and ovaries, and presence and number of embryos and placental scars. The categories established are: immature: specimens with closed vulva, whose uterus and ovaries are not much developped nor vascularised and lack placental scars. Submature: specimens with open or closed vulva and whose uterus and ovaries are slightly developped and never show placental scars. Inactive mature: specimens with closed vulve and uterus and ovaries developped, though scarcely vascularised. Placental scars may be present. Active mature: specimens with open vulva and with completely developped and vascularısed uterus and ovaries. There may be placental scars. Results Reproduction in males Fig. 1 reveals that all individuals weighing less than 150g are immature. Individuals between 150 and 176g are at different stages of maturity, while all specimens weighing more than 176g are sexually mature. Some specimens captured in November are adults accounting to their size but their testicular squashes reveal a very small number of spermatozoids. The testicular lengths of these individuals ranged between 7.0 and 8.9mm and the lengths of the seminal vesicles were 5.0-9.5 mm (fig. 2). The length at which the individuals show reproductive capacity at the population level are 12.5 mm for the testicle and 9 mm for the vesicle. That is, even though these specimens have reached adult dimensions morphologically and metrically, the small sıze of both organs suggests that the spermatogenesis has been altered, possibly due to an attenuation in the rhythm of gonad development because of environmental changes. If this were the case, those individuals born in the second half of summer would reach maturity later than those born in the spring, given the same bodily development. Even though the adult specimens always present spermatozoids in the testicle, this organ and the seminal vesicle undergo important changes in size throughout the year (fig. 3). In December-January both parameters reach minimum values in the adults. In February, a progressive size increase commences, reaching the maximum in June. From this month on a decrease in size of both organs can be discerned. The product of the two testicular diameters and the length of the seminal vesicle during the February-October period are clearly correlated (fig. 4). Below 65 mm? for the testicular diameter product and 5.5 mm of seminal vesicle length only immature individuals appear. Submature individuals are limited to the 65-105 mm? range of testicular diameter product and 6.0-8.0 mm range for the length of the vesicle. Above upper lıimits all the specimens can be considered mature. Reproduction in females From March to October, 100 % of the mature females studied were sexually active (fig. 5). In November, this percentage decreased notably and the minimum values, around 35-45 %, were reached in December and January. suaummads aınyeur :(@) ‘sprozoreunads Ma} yıım suswmdads »zıs-ıpnpe :(-o-) ‘susunmads aınzeurgns :(e) !susunsads 9ınzewun :(O) “(wur ur °7]) aJonsaı ayy Jo yı3ua] ayy pue “(wu ur TA) 2PPIsOA feurwss 9yı Fo yI3ua] ay2 pue sofeur Jo snyeis Jenxas ayl u99MI9g dıysuonejsy 'erppp 01 ayı wor snpıdvos vponaay ‘7 "37 Q N (©) S v f f W Vv W J f J. Ventura and J. Gosalbez 366 suaunsads Jınyeur :(@) ‘sprozoreuuisds MaJ yıım suswmads azıs -jnpe :(-o-) ‘susumads sanyeurgns :(@) ‘susummads aınzewun :(0) ‘ıe9A 9y3 InoydnoAayy sajeur Jo snyeIs [enxas pue (3 ur) ıydıam Apoq U99MI9q dıysuonepy 'eIypp 01 ayı wo1F supıdos vrosaay ‘I "317 Q N (0) S v r r W V W 3 r 0S Sl 001 szl oOsL SıLl 003 STT 0S2 GLT 00€ ILHOI3M 367 Reproductive biology of Arvicola sapidus in the Ebro delta suaumads Jınzew :(@) ‘susunsads Jınyewans :(®) ‘suswmads sınzewun :(0) "(wu ur “J) >PIs>A jeunuss ayı Jo ı3ua] ayı pur („ww ur ‘] @ x @) s1992ueIp zejnonsaı ayı Jo 19npo1d ayı US9MI9q UONEPA1O9 ayı pur sofeul 70 SMIENS [enxas ayp usamaaq dıysuonepoy '"erppp oIgy Sy WON} snpıdvs vjonauy 'p "BJ 0SZ 007 ost 001 0S suaumoads Jınyeu ur (T,L) Sppnsaı Jyı Jo pue (TA) >PPIs9A Jeuruss ay2 Fo yısuay aderaae ayı jo uonersea AJyguow "erppp O1gA ayı woAy snpıdvs vponauy 'g 1] Q N (6) S Vv r f N V W J f 368 J. Ventura and J. Gosalbez % 100 90 80 70 60 50 40 30 20 10 J F M A M J J A S oO N D Fıg. 5. Arvicola sapidus from the Ebro delta. Variation throughout the year of the percentage of active mature females referred to the total mature females (upper line), and percentage of pregnant females over the total of active mature females (bottom line). (I): inactive mature; (A): active mature; (P): pregnant WEIGHT 250 200 150 100 50 Fıg. 6. Arvicola sapidus from the Ebro delta. Relationship between body weight (in g) and sexual status of females throughout the year. (O): immature specimens; (©): submature specimens; (@): active mature specimens; (-@): inactive mature specimens; (4): pregnant specimens Reproductive biology of Arvicola sapidus in the Ebro delta 369 Pregnant females appear between February Table and October. However, the small sample size (n= 19) does not allow us to discern the repro- ductive peaks. It can only be said in this re- spect that the maximum value appeared in March, in which 80% of the active mature females captured were pregnant. Litter size in Arvicola sapidus from the Ebro delta from embryo counts in pregnant females Number of embryos All specimens weighing less than 130g are u ie immature (fig.6). Between this weight and February 3 3 170g, there is a wide band of specimens that Nu 2 : show different stages of maturity and activity ns > but never are pregnant. All the females with June 1 body weights above 170g can be considered July 1 sexually mature. The lightest pregnant female August ! ! September 2 weighed 183g. BR 3 It is important to note that the presence of specimens weighing less than 130g in De- Total 19 cember and January may be evidence repro- ductive activity in November. Two of these specimens were captured towards the end of January (28. 1. 1982) and weighed respec- tively 115 and 128g. Such individuals suggest the possibility of reproduction in December. This phenomenon is irregular and may be determined to some degree by the occurrence of relatively milder autumns. Litter sıze was 2-5 embryos (see table). The average number of embryos per female was 3.31 # 1.20 (n=19). No annual pattern in litter size could be determined. There seems to exist an increase of the average from February to April. Likewise no such preference of embryos would be detected. The distribution frequencies were: OR-2L (n=3); IR-2L (n=3); IR-IL (n=2); 2R-1L (n=2); 2R-2L (n=2); 3R-2L (n=2); 4R-IL (n=2); OR-3L (n=1); 2R-OL (n=1); 2R-3L (n=1). When the gestating females are grouped into three weight intervals: 180-199, 200-219g and more than 219g the following average litter sizes appear respectively: 2.0 (n=3); 3.87 (n=8); 3.25 (n=8). Litter sıze increases wıth body weight only in the two firsts intervals and the difference between the second and third interval is not significative. Discussion The literature concerning the reproductive cycle of Arvicola sapidus is to the date scarce and occasional. Thus, it ıs not possible to compare the results obtained here with those of other populations, except for those of DELosT (1968). This author mentions the presence of sexually active males throughout the whole year at the Tarn rıver (France). These results coincide with those observed in the Ebro delta. The decrease in sıze of the testicles does not seem to suppress the spermatogenic capacity. However the specimens from the summer- autumn litters take longer to acquire sexual maturity due to the decreased testicular development. The varıatıon in sıze of testicle and seminal vesicle throughout the year coincides with the seasonal varıation in sexual activity reported by DErosrt (1968) for males. The period of maximum activity ıs established by this author between April and October. The minimum values appear in December-January, and in February the activity is almost totally reestablished. The reproductive activity of the females is maintained at maximum levels between February and October. Even though activity is less during the remaining months, this 370 J. Ventura and J. Gosalbez decrease never includes the whole population as approximately 35 % of the mature females continue to be active. Pregnant females were found between February and October, during a longer period than between Aprıl and September reported by DeLosrt (1968). These differences may be due to environmental and climatic factors as suggested by PerLıkAn (1972) for Arvicola terrestris. Even though no pregnant females appear between November and January, the constant presence of mature males throughout the year, the continued activity of some mature females during this period and the capture of juvenile individuals in December and January suggest the possibility of reproductive activity in November and December. Anyway, it is significantly lower that observed during the rest of the year. The litter sizes found were 2-5 embryos. These values fall within the range established by Divier and Rope (1935) and DeLosrt (1968) in french populations, although the maximum values obtained by both authors (7 and 8 embryos respectively) are higher than that observed in the Ebro delta. The most frequent litters were those consisting of 2 and 3 embryos and after, those formed by 5 embryos. These results are also lower than the modal value (6 embryos) recorded by Diver and Rope (1935). Even though the reproductive peaks have not been established, the high values in March agree with the observations of many authors for Arvicola terrestris (VAN WIJNGAARDEN 1954; PELIKAn 1972; WIELAND 1973) that it is precisely at the beginning of the reproduc- tive period when the sexual activity is at its highest. At the population level, A. sapıdus of the Ebro delta shows a general period of high sexual activity between February and October. The decrease in November-December may be conditioned by intrinsic factors of the females, because spermatozoids are present in all mature males throughout the year. Therefore despite the apparent reproductive capacity of the males, the non-receptivity of a high percentage of temales determine meager chances for reproduction. Additionally to be kept into account is the size reduction of the testicles and seminal vesicles during these months. Even though the possession of spermatozoids enables the animal to be considered sexually active, the efficience of these cells for reproduction is yet to be verified. Possibly the seminal vesicles have a direct influence on the viability of the spermatozoids, since the substances contained in the secreted liquid are important for the nutrition and protection of the reproductive cells (Guyron 1967; Houssay 1969), while also facilitating fecundation (HOGARTH 1978). Therefore even if spermatozoids are produced, an animal with an insufficient seminal vesicle may not be apt for reproduction. Despite these factors, the existence of a certain reproductive activity in December suggests the influence of extrinsic (environmental) factors which can alter to a certain degree this activity in different years. Acknowledgements The authors wish to express their gratitude to Dr. M*. J. LöPEZ-FUSTER, G. GÖTZENS, $. Ruız and A. EscoLA, members of the Mastozoology team of the Department of Vertebrates for their colaboration in the fieldwork. Likerwise we want to acknowledge ICONA, and very specially the assistance of the guard J. MARTT during our stays in the Ebro delta. Zusammenfassung Zur Fortpflanzungsbiologie von Arvicola sapidus (Rodentia, Arvicolidae) im Ebro-Delta in Spanien An 79 Männchen und 66 Weibchen von Arvicola sapidus, die in den Jahren 1981-1984 im Ebro-Delta in Spanien gefangen worden waren, wurde das Fortpflanzungsverhalten und seine Abhängigkeit von Gewicht und Jahreszeit untersucht. Dazu wurden die folgenden Parameter bewertet: Bei den Männchen das mikroskopische Bild des Hodens, sein größter und kleinster Durchmesser und die Länge der Vesicula seminalia; bei den Weibchen der Zustand der Vulva, Größe und Durchblutung der Reproductive biology of Arvicola sapidus in the Ebro delta 37a Uteri und Ovarien, das Auftreten und die Anzahl von Embryonen und das Auftreten vom Maculae cyaneae. 2 Bei den adulten Männchen findet man ganzjährig Spermatozoen, obwohl die Größe von Hoden und Vesikeldrüsen ım Jahreslauf stark schwanken. Am kleinsten sind sie von Dezember bis Januar, am größten im Juni. Von März bis Oktober sind alle geschlechtsreifen Weibchen auch geschlechtstätig. Am geringsten ist ihr Anteil im Dezember und Januar. Trächtige Weibchen kamen nur von Februar bis Oktober vor. Die durchschnittliche Embryonenzahl war 3,31. Der Nachweis von Spermien bei den Männchen und geschlechtstätigen Weibchen auch im Winter sowie von Jungtieren im Januar deutet darauf hin, daß Fortpflanzung gelegentlich auch im Winter vorkommt. References BAuponn, C. (1984): Le campagnol amphibie, Arvicola sapidus. In: Atlas des mammiferes sauvages de France. Societe Frangaise pour l’Etude et la Protection des Mammiferes. Paris: Fayard Editeurs. Derost, P. (1968): Etude comparative de la reproduction chez quelques Rongeurs sauvages non hibernants dans les differents regions de France. Ent. Chize, | Physiol. 23-50. Dipier, R.; RoDe, P. (1935): Les mammiferes de France. Parıs: Societe Nationale d’Acclimatation de France. — — (1939): Les micromammiferes de la faune francgaise. I. Campagnols. Mammalıa 3, 110-121. GosALBEZ, J. (1976): Contribuciön al conocimiento de los roedores del nordeste de la Peninsula Iberica y su interes biolögico. Tesıs Doctoral. Universidad de Barcelona. GOSÄLBEZ, ]J.; SANS-CoMA, V. (1976): Sobre el topillo rojo, Clethrionomys glareolus Schreber, 1780, en la region del Montseny (Cataluna, Espana). Säugetierkdl. Mitt. 24, 77-79. GOSÄLBEZ, ].; LÖPEZ-FUSTER, M. ].; DuRFORT, M. (1979): Ein neues Färbungsverfahren für Hodenzellen von Kleinsäugetieren. Säugetierkdl. Mitt. 27, 303-305. GuyTon, A. C. (1967): Tratado de Fisiologia Medica. Madrid: Ed. Interamericana. HEIM DE BaLsac, H.; GussLaın, R. (1955): Evolution et speciation des campagnols du genre Arvicola en territoire francais. Mammalıa 19, 367-390. HOoGARTH, J. P. (1978): Biology of reproduction. Glasgow: Blackie. Houssay, B. A. (1969): Fisiologia Humana. Madrid: El Ateneo. LE LovArn, H.; SAINT-GIRoNs, M.-C. (1977): Les rongeurs de France. Faunistique et biologie. Paris: Institut National de la Recherche Agronomique. MILLER, G. S. (1912): Catalogue of the Mammals of Western Europe. London. PELIKAN, J. (1972): Arvicola terrestris (L.), indexes of reproduction ın Czechoslovakıa. Acta Scı. Nat. Acad. Sci. Brno. 11, 3-50. REICHSTEIN, H. (1982): Gattung Arvicola Lacepede, 1799. In: Handbuch der Säugetiere Europas. Hrsg. von J. NIETHAMMER und F. Krapp. Wiesbaden: Akademische Verlagsgesellschaft. SAINT-GIRONS, M.-C. (1973): Les mammiferes de France et du Benelux. Parıs: Doin Editeurs. WIELAND, H. (1973): Beitrag zur Biologie und zum Massenwechsel der Großen Wühlmaus (Arvicola terrestris L.). Zool. Jb. Syst. 100, 351-428. WIJNGAARDEN, A. van (1954): Biologie en bestrijding von de Woelrat, Arvicola terrestris terrestris (L.) ın Nederland. Diss. Eindhoven. ZABALA, J. (1983): Primeracita de Arvicola sapidus Miller, 1908 (Mammalia, Rodentia) para la provincia de Guipüzcoa. Munibe 35, 109-114. Author’s address: Dr. JoaQuIm GOSÄLBEZ and JACINT VENTURA, Cätedra de Vertebrados, Depar- tamento de Zoologia, Facultad de Biologia, Universidad de Barcelona, Av. Diagonal 645, E-08028 Barcelona Zum Lautrepertoire eines ursprünglichen Cerviden, Muntiacus reevesi micrurus (Sclater, 1875) Von $S. G. STADLER und H. HENDRICHS Lehrstuhl für Verhaltensphysiologie, Fakultät für Biologie, Universität Bielefeld Eingang des Ms. 22.9. 1986 Abstract On the sound repertoire of the Formosan muntjac, Muntiacus reevesi micrurus (Sclater, 1875) Studied was the sound repertoire of Reeves’ muntjac to specify acoustical qualities, situations of occurence, and to discuss underlying internal conditions. From May 1983 through December 1984 17 anımals of the Formosan muntjac (Muntiacus reevesi micrurus) living in four groups in large outdoor enclosures were observed. 200 h of observation have been evaluated. 50 % of the observations were conducted at close distance to gain tape recordings for sonagraphical analyses. Sıx different sounds have been recorded, three of which are documented sonagraphically. These sounds include: Yappıng, Barking, Fawn’s-Scream, Whistle, Footstamp, Teethgrinding. Yapping and Barking occured ın situations when an anımal faced a potential danger or, less frequently, a dominant muntjac. The Fawn’s-Scream was uttered by fawns when frightened or alone and searching for the mother. The Whistle is subdivided into three subtypes: The Short-Whistle of dominant males or adult females with young is considered an expression of self-assertiveness, the Long-Whistle of subdominants an expression of uneasiness. The Soft-Whistle is a vocalization of fawns uttered in ambivalent contexts. The Footstamp is considered a warning sound. The internal state can be described as highly aroused. Teethgrinding is part of the threat display of adult bucks. It was also heard from anımals in conflict. The concept of displacement activity is discussed for this particular situation. Einleitung Für das Verständnis sozialer Prozesse sind Kenntnisse der akustischen Ausdrucksmöglich- keiten einer Art hilfreich. Lautäußerungen gelten als mögliche Indikatoren der atfektiven Tönung eines kommunikativen Aktes (MorToNn 1977; JÜRGENS 1979; SCHERER 1985). Hinweise auf innere Zustände wie „Erregung“ oder „Sicherheit“ der Tiere (vgl. HEND- RICHS 1978) sind wichtig für die Analyse der Regulation sozialer Prozesse (BiscHor 1972, 1985; BaLpwın und BaLpwin 1977). Eine umfangreiche Übersicht über Lautäußerungen bei Huftieren gibt Kıry (1972). Für Cerviden liegt eine Reihe von Arbeiten zu verschiedenen Aspekten der akustischen Kommunikation vor (z.B. EspmArk 1971; GRADL-GRAMS 1983; HıRTH und McCur- LOUGH 1977; RiCHARDSoN et al. 1983). Lautäußerungen bei Muntjaks wurden vor allem von YAHNER (1980), aber auch von CHarLin (1971) und BARRETTE (1977a) dargestellt. In dieser Arbeit werden Lautäußerungen des Chinesischen Muntjaks (Muntiacus ree- vesi) beschrieben und sonagraphisch dokumentiert; außerdem werden Zuordnungen quali- tativer Art zum sozialen Kontext ihres Auftretens gegeben. Aufbauend auf dieser Grund- lage wird versucht, einige Aspekte der im Tier vorliegenden Bedingungen für die Produk- tion der einzelnen Laute zu beschreiben. Diese Bedingungen, die über das beobachtbare Verhalten erschlossen werden müssen, können in Form hypothetischer Variablen auf der Ebene des inneren Zustandes dargestellt werden (vgl. z.B. BiscHor 1985). U.S. Copyright Clearance Center Code Statement: 0044-3468/87/5206-0372 $ 02.50/0 Z. Säugetierkunde 52 (1987) 372-382 © 1987 Verlag Paul Parey, Hamburg und Berlin ISSN 0044-3468 Zum Lautrepertoire von Muntiacus reevesi 373 Material und Methode Seit 1974 werden am Lehrstuhl für Verhaltensphysiologie der Universität Bielefeld mehrere Gruppen der Formosa-Unterart des Chinesischen Muntjaks, Muntiacus reevesi micrurus (zur Systematik s. WHITEHEAD 1972), in mehreren Grofßsgehegen (900-5000 m?) gehalten. Für diese Untersuchung wurden von Mai 1983 bis Dezember 1984 regelmäßig 11 Tiere in 3 Gruppen (I: 1,1 + 0,2 jüngere Tiere; II: 1,1 + 2,0; III: 1,1 + 0,1) sowie unregelmäßig 6 Tiere in einer weiteren Gruppe (IV: 2,2 + 2,0) beobachtet. Es wurden ca. 200 Beobachtungsstunden ausgewertet. Davon entfallen ca. 50 % auf Stallbeobach- tungen, ın denen die Tiere der Gruppen I, II und III auf Distanzen zwischen einem bis zu fünf Metern beobachtet werden konnten. Protokolliert wurde fortlaufend in ein Ein-Minuten-Raster; es wurden nach der ad libitum-Methode (Artmann 1974) ca. 60 verschiedene Verhaltensweisen aufgenommen und ausgewertet. Tonbandaufnahmen wurden (sowohl im Freien als auch in den Stallgebäuden) mit einem Tonbandgerät 4400 der Firma UHER und einem Richtmikrophon der Firma ac bei einer Aufnahmegeschwindigkeit von 19 cm/sec gemacht. Eine klangspektrographische Analyse erfolgte auf einem Sonagraphen Vibralyzer 7030 A der kay Elemetrics Corp. mit Filtereinstellung ‚‚wide‘“ (300 Hz) bzw. „narrow“ (80 Hz). Ergebnisse Es wurden 6 verschiedene Lautäußerungen in insgesamt 8 Ausprägungen unterschieden: Kläffen ua) Bellen (24) Kitzschrei (13) Fiepen kurz oXe) (297) 22 (20) lang (669) leise (12) Hufstampfen (18) Zähneknacken (49) *, absolute Häufigkeit des Auftretens während des gesamten Beobachtungszeitraumes. Kläffen Lautbeschreibung (vgl. Abb. 1): Der einzelne Laut ist zwischen 0,08-0,15 sec lang, wird meistens 3- bis 5mal ım Abstand von ca. 0,1-0,3 sec geäußert und ist im Bereich zwischen 500-2500 Hz relativ geräuschhaft mit Energiemaxima bei etwa 600 Hz, 1200 Hz und 2000 Hz. Das Frequenzmaxımum liegt ım Bereich um 3500 Hz. Kläffen trat auf, wenn ein Tier überrascht wurde, z.B. von einem Beobachter. Es wurde während des Flüchtens geäußert. Danach sicherte das Tier ın aufrechter Haltung und mit gespitzten Ohren in Richtung der Störquelle und ging entweder zum Bellen über oder verstummte. Kläffen wurde bei adulten ?? und jungen dd beobachtet. Bellen Lautbeschreibung (vgl. Abb. 2): Der einzelne Laut ist zwischen 0,35-0,80 sec lang und wurde in der Regel im Abstand von ca. 4 bis 6 sec geäußert. Eine Lautserie konnte bis zu 20 mın andauern; dabei wurden in der Regel die Abstände zwischen den Lauten mit zunehmender Zeit etwas länger (bis 10 sec); die letzten Laute einer längeren Lautserie wurden oft leiser und kürzer. Der Laut hat seine Hauptenergie zwischen 800-1200 Hz, eine Oberschwingung im Bereich um ca. 2000 Hz ist deutlich, oberhalb dieses Wertes bleibt der Laut leise und geräuschhaft mit einem Frequenzmaximum bei ca. 4500 Hz. 374 S. G. Stadler und H. Hendhrichs Neben diesem Grundmuster traten starke interindividuelle Unterschiede auf. So zeigte ein ca. 2 Jahre altes 8 einen relatıv hohen Anteil tonaler Anteile im Bereich von ca. 1100 Hz (hier auch Grundfrequenz und Hauptenergie) mit einer Oberschwingung bei ca. 2500 Hz. Ein junges ? (2,0) sec lang. Sein zeitlicher Verlauf ist sehr varıabel. Die Grundfrequenz, die auch die Hauptenergie trägt, liegt zwischen 1100-1500 Hz, schwach ausgeprägte Oberschwingun- gen (bis max. 4) können bis zu einer Maximalfrequenz von ca. 6000 Hz auftreten. Große Schwankungen in der Lautstärke treten auf, lassen sich aber aus aufnahmetechnischen Gründen nicht quantitativ angeben. Neben diesem Grundmuster wurde beı einem subadulten d folgender Laut im gleichen Verhaltenskontext registriert: Der Laut ist 0,15-0,20 sec lang. Seine zeitliche Reihung ist 376 $. G. Stadler und H. Hendrichs ebenso vom Verhalten des Interaktionspartners abhängig. Grundfrequenz und Hauptener- gie liegen zwischen 1000-2000 Hz (in abfallender Frequenz), leise Oberschwingungen mit einem Frequenzmaximum bei ca. 4000 Hz können auftreten. Das Lange Fiepen wurde ım allgemeinen von ?2 oder von jungen dd auf die vorher geschilderten Annäherungen der adulten 84 geäußert. Normalerweise wichen die unterle- genen Tiere, wenn sie lang fiepten, aus (492mal = 65 %). Sie konnten aber auch am Ort bleiben (183mal = 24 %) (vor allem während ihrer Ruhephasen), zeigten dann aber andere Verhaltensweisen der „Unterwerfung“. Das Lange Fiepen wurde auch von einem adulten und dominanten d geäußert (19mal), ın vergleichbarer Situation. Leises Fiepen Lautbeschreibung (vgl. Abb. 3c): Der einzelne Laut ist 0,07-0,10 sec lang und wurde meistens ebenfalls in Interaktionen als Reaktion auf das Verhalten eines anderen Tieres geäußert, darum läßt sich auch hier keine exakte Angabe für das zeitliche Muster des Auftretens geben. Die Grundfrequenz und Hauptenergie liegt zwischen 2000-2200 Hz, eine sehr schwach ausgeprägte Oberschwingung im Bereich um 4500 Hz kann auftreten, sonst wirkt der Laut relativ rein. Das Leise Fiepen ist eine Lautäußerung der Kitze. Es wurde während der Aktivitätspe- rıoden vom Jungtier geäußert, besonders dann, wenn ein anderes Tier (auch die Mutter, speziell das adulte d) sich annäherte und an Körper oder Kopf des Jungtieres schnupperte bzw. das Jungtier mit den Voraugendrüsen markierte. Das Leise Fiepen trat besonders in den ersten 8-12 Lebenswochen auf. Fiepen kHz oa b R 8 4 BG nun 5 kHz a b Fiepen 8 8 ık [A 4 L} % er Rn Bir nn nn un et re (Se EN NIT NT N (0) 05 (0) 05 0) 05 sec. 0) 05 1.0 15 sec. FILTER "wide” FILTER "wide" Abb. 3 (links). Sonagramme Fiepen. a= Kurzes Fiepen eines adulten d (Wide-band-Filter); b = Kurzes Fiepen eines adulten ? (Wide-band-Filter); c= Leises Fiepen eines sechs Wochen alten Bockkitzes (Wide-band-Filter). -— Abb. 4 (rechts). Sonagramme Fiepen. a= Langes Fiepen eines adulten ? (Wide-band-Filter); b = Langes Fiepen eines subadulten d (Wide-band-Filter); c = Langes Fiepen eines adulten ? (Wide-band-Filter) Hufstampfen Lautbeschreibung: Da die Lautstruktur des Hufstampfens von der Beschaffenheit des Untergrundes abhängt, läßt sich kein eindeutiges Bild des Lautes zeichnen. Typisch ist das sehr kurze und dumpfe Geräusch, das in jedem Fall erzeugt wird. Das Hufstampfen wurde Zum Lautrepertoire von Muntiacus reevesi IL meist im Abstand von 1-4 sec in Serien von 2- bis zu 10mal hintereinander, und zwar überwiegend von ?? oder von jungen dd, gezeigt. Bei den ?? lösten oft unklare Situationen ın der Nähe ihres Jungtieres Hufstampfen aus, häufig wenn adulte 2? sich mit ihrem Jungtier im Stallgebäude aufhielten und dort gestört wurden. In einer solchen Situation zeigte ein 26 Tage altes Jungtier bereits Hufstampfen. Ein junges d, das aus einem großen Gehege vorübergehend in ein sehr kleines (30 m?) umgesetzt wurde, reagierte auf die Annäherung eines Menschen mit Umherlaufen, Sichern, Zähneknacken und Hufstampfen. Die Haare waren dabei, beson- ders im Bereich der weißen Schwanzunterseite, gesträubt. Hufstampfen trat auch in „rein“ sozialen Situationen auf: Als das ? aus Gruppe II nach 5 Tagen Abtrennung wieder mit den 3 dd zusammengelassen wurde, gingen alle 4 Tiere unruhig umher, beschnupperten sich gegenseitig, sicherten ungewöhnlich lange und stampften ca. 1 min (im Drei-Sekun- den-Abstand) lang mit den Hufen. Das adulte, dominante 8 ‚beendete‘ diese Situation, als es begann, das ? zu treiben. Zähneknacken Lautbeschreibung: Das Zähneknacken ist ein klickender Laut, der 2- bis 10mal jeweils im Abstand von etwa 1 sec geäußert wird. Wie der Laut erzeugt wird, ist gegenwärtig nicht endgültig geklärt. BARRETTE (1977c) und LEvEr (1977) geben an, daß der Laut durch Reibebewegungen der Molaren erzeugt wird. Immerhin war der Laut bei einem unserer dd, das beide Canıni abgebrochen hatte (vgl. Woopvine 1984), leiser als bei den übrigen 8d. Außerdem wurde es nur bei dd beobachtet, die wenigstens 2 Jahre alt waren. Dies deutet auf eine Beteiligung der Canını an der Erzeugung des Lautes hın. Das Zähneknacken trat vor allem während der „Drohkämpfe‘“ (>90 %) auf, welche die adulten dd an den Gehegezäunen, an denen sıe optischen, akustischen und olfaktori- schen Kontakt hatten, ausführten. Dabei wurde die von Dugost (1971) als „‚Imposition“ bezeichnete Verhaltensweise gezeigt, die im allgemeinen als Drohgeste interpretiert wird (Dusost 1971; BARRETTE 1977a). Eine solche Situation trat auch innerhalb einer Gruppe - zwischen dem Vater und seinem geschlechtsreifen Sohn - auf. Hier knackte jedoch nur der dominante Vater mit den Zähnen. Außerdem wurde das Zähneknacken beim Sexualverhalten beobachtet, und zwar, wenn das ? nicht wie gewöhnlich mit Ausweichverhalten auf die Annäherung des & reagierte, sondern wenn es stehenblieb und eine Demutgebärde (Kopf tief und Ohren gespitzt) zeigte, oder wenn es sich mit Schnauzenstößen in die Seite des d oder einer als „Lift“ (BARRETTE 1977a) bezeichneten Verhaltensweise, bei der das ? das d mit der Schnauzen- spitze anhebt, gegen dessen Annäherung wehrte. In einem Fall reagierte ein jungadultes J, das in ein sehr kleines Gehege umgesetzt worden war, u.a. mit Zähneknacken auf die Annäherung des Menschen auf eine ungewohnt nahe Distanz (<5 m). Eine Fluchtmög- lichkeit war nicht gegeben, und in dem großen Gehege, in dem es vorher gelebt hatte, war es niemals einer solchen Nähe zum Menschen ausgesetzt gewesen. Diskussion Das Kläffen erscheint als ein Untertyp des Bellens und auch motivational eng mit diesem verwandt. Da das Kläffen fest an den Verhaltenskontext der Flucht gebunden ist (vgl. BARRETTE 1977b) und von der Lautstruktur eindeutig spezifiziert werden kann, erscheint es gerechtfertigt, es vom übrigen Bellen abzutrennen. Da zu wenig Beobachtungen zur Reaktion von Artgenossen vorliegen, muß die Frage einer möglichen Warnfunktion offen bleiben. Bellen ist anfänglich mit gesteigerter Aufmerksamkeit und Anspannung verbunden; dies äußert sich auch ın den begleitenden Verhaltensweisen (Sichern, gespannte Körperhal- 378 S. G. Stadler und H. Hendrichs tung). Da Bellen in Situationen auftritt, die sowohl interspezifisch (Raubfeinde) als auch intraspezifisch (Unterlegenheit) als bedrohlich angesehen werden können, nehmen wir als zugrundelegende Bedingung einen Zustand inneren Ungleichgewichts an, der möglicher- weise mit Hilfe der Lautäußerung abgebaut wird. Dies zeigt sich sowohl in den mit dem Nachlassen einer Störung länger werdenden Abständen zwischen den Bellauten der Lautserie als auch ın der dann abnehmenden Lautstärke. Nach mehreren Minuten Bellen wirken die Tiere zudem entspannter, was sich am ruhigen Umhersehen, an der gelockerten Körperhaltung mit gelegentlichem Schnuppern am Boden zeigt. BARRETTE (1977b) nımmt an, daß das Bellen ‚„‚fast ausschließlich als Antwort auf eine potentielle Gefahr“ geäußert wird, schreibt ihm aber nur in einem anderen Zusammenhang (BARRETTE 1977a,b) eine innerartliche Signalfunktion zu. YAHNER (1980) schlägt nach einer ausgedehnten, experimentellen Untersuchung des Bellens vor, diese Lautäußerung eher als ‚‚distress call‘ denn als „alarm call‘ anzusehen. Er führt verschiedene Gründe an, warum seiner Meinung nach das Bellen nicht in erster Linie als Laut mit Warnfunktion angesehen werden sollte. Entsprechend den Argumentationen von HAMILTON (1964), MAYNARD SMITH (1965) und TRıvErs (1971) zur Erklärung der Evolution von Warnlauten, sei die mehr oder weniger solitäre und territorıiale (vgl. HarvınG 1982) Lebensweise der Muntjaks als Hinweis auf das Wirken von Individual-Selektion, nicht auf Verwandten- Selektion zu deuten. Die Ausdehnung einzelner Bellserien über mehrere (bis zu 20) Minuten lasse ebenso eine Warnfunktion unwahrscheinlich erscheinen, da nach SMYTHE (1977) ein einzelner Warnruf, gefolgt von augenblicklicher Flucht, zur Warnung von Artgenossen ausreiche. Das häufige Fehlen von deutlich erkennbaren Reaktionen anderer Muntjaks auf bellende Artgenossen deutet in die gleiche Richtung, mag jedoch in beiden Untersuchungen auch einen Gewöhnungseffekt aufgrund der Gehegehaltung widerspie- geln. EISENBERG und LOcKHART (1972) berichten, daß Axishirsche (Axıs axis) auf Bellen von Muntiacus muntjak ım Freiland nur vereinzelt mit erhöhter Aufmerksamkeit und Sichern reagieren. Auch Rehe (Capreolus capreolus) reagieren nach HEnnıG (1962) häufig nicht oder nur schwach auf das Schrecken eines Artgenossen. Im Gegensatz dazu stehen Aussagen von MÜLLER-UsInG und SCHLOETH (1967) für Reh- (Capreolus capreolus) und Rotwild (Cervus elaphus), und von SCHALLER (1967) und HENDRICHSs (1975) für Axis- (Axis axis), und von SCHALLER (1967) für Barasıngha- (Cervus duvauceli) und Sambarhır- sche (Cervus unicolor). Alle diese Arten reagieren im Freiland stets mit erhöhter Aufmerk- samkeit, Sichern und gegebenenfalls selbst mit Bellen oder gar mit Flucht auf das Bellen von Artgenossen. Die häufig geäußerte Vermutung, das Bellen stehe im Zusammenhang mit der Fort- pflanzungsaktivität (SOPER 1969; CHAPLin 1971; LEVER 1977), wird durch unsere Beob- achtungen nicht gestützt. Bellen trat nicht zusammen mit Sexualverhalten auf. Ähnlich äußert sich BARRETTE (1977b). Möglicherweise können jedoch die insgesamt während der Östrus- oder Setzzeit eines ® erhöhte Aktivität und die höhere allgemeine Erregbarkeit eine leichtere Auslösbarkeit des Bellens herbeiführen und den Eindruck erwecken, das Bellen stünde mit dem Reproduktionsverhalten in engerer Beziehung. Ob die großen individuellen Unterschiede innerhalb der Kategorie Bellen eine Untertei- lung in distinkte Lauttypen rechtfertigen, kann hier aufgrund des geringen Materials nicht entschieden werden. BuBEnık (1965, S. 103) gibt an, daß beim Rotwild (Cervus elaphus) „.... Rauhheit und Lautstärke durch den Frregungszustand bedingt werden“. Da zudem verschiedene Autoren (HınpE und RoweEıı 1962; KırLey 1972; SCHERER 1985) auf den graduellen Charakter von Lautäußerungen hinweisen, soll zunächst an der zusammenfas- senden Einteilung festgehalten werden. Dem Kitzschrei kann wahrscheinlich als innere Bedingung ein Zustand der Unsicher- heit oder Angst bei gleichzeitig hoher allgemeiner Erregung zugeordnet werden. BUBENIK (1965) nennt eine im gleichen Verhaltenskontext beobachtete Lautäußerung des Rehkitzes (Capreolus capreolus) „Angstgeschrei“. Die beobachteten Verhaltenskontexte, ın denen Zum Lautrepertoire von Muntiacus reevesi 379 der Laut auftrat, deuten darauf hin, daß es sich eventuell ontogenetisch um einen Vorläufer des Bellens handeln könnte. Da jedoch auch bei adulten Tieren, die gefangen werden mußten, laute, klagende Schreie beobachtet wurden, kann hier die Frage der ontogeneti- schen Entwicklung des Lautes nicht eindeutig geklärt werden. Eine hohe „Verteidigungsbereitschaft‘“ des Muttertieres wurde ebenfalls bei der Reh- geiß (Capreolus capreolus) beobachtet (BUBENIK 1965). Eigene Beobachtungen am Chinesi- schen Wasserreh (Hydropotes inermis) ergaben, daß die Weibchen sich niemals ihrem schreienden Kitz näherten, solange der Mensch in der Nähe war. Alle für das Kurze Fiepen sowohl der dd als auch der 22 gefundenen Situationen deuten auf zwei grundlegende Komponenten des inneren Zustandes der Tiere hin. 1. sind die Tiere grundsätzlich in einer aktiven Phase und bemüht, sozialen Kontakt herzustellen (vgl. BARRETTE 1977a) und 2. wird der Laut von solchen Individuen geäußert, die zum aktuellen Zeitpunkt mit einem hohen Maß an „Selbstsicherheit“ in die Interaktion hineingehen. Unter Selbstsicherheit wird hier ein Aspekt der psychischen Konstitution eines Individuums verstanden, der es ermöglicht, sein Verhalten in einer gegebenen Situation gemäß den aktuellen Einstellungen des eigenen inneren Zustandes zu organisie- ren. Auf der Verhaltensebene wird diese Fähigkeit als konsequentes Durchführen begon- nener Verhaltensabläufe sichtbar; man könnte von einer Selbstbestimmung des eigenen Verhaltens sprechen. Auch wenn häufig Sexualverhaltensweisen im Zusammenhang mit dem Kurzen Fiepen der Männchen auftraten, so reicht dies nicht aus, den Laut alleın dem sexuellen Kontakt zuzuordnen. Möglicherweise kann diese Komponente steigernd auf die Produktion des Lautes wirken: trotz Gegenwart des &-d fiepte ein subdominantes d wiederholt gegenüber ?? (33mal). Das Lange Fiepen, von YAHNER (1980) ‚„‚whine‘ genannt, wird in gewisser Weise ın einer entgegengesetzten Situation geäußert, insofern es 1. nicht nur bei Aktivität geäußert wurde, sondern häufig auch in Ruhephasen und 2. überwiegend (>95 %) von unterlege- nen Tieren, deren innerer Zustand ın der betrachteten Situation mit „Unbehagen“ zu beschreiben wäre. Unbehagen wird verstanden als ein Maß für die Abhängigkeit des aktuellen inneren Zustandes, und somit der eigenen Verhaltensmöglichkeiten, von Zustand und Verhalten anderer, dominanterer Tiere, in gewisser Weise ein der Selbstsicherheit entgegengesetzter Zustand. Auch wenn ein Tier einen festen Platz in einem sozialen und räumlichen Gefüge und damit eine längerfristige Sicherheitsbasis hat (vgl. HENDRICHS 1978), kann kurzfristig das eigene Verhalten durch das anderer, vor allem überlegener Tiere gestört oder beendet werden. Das Verhalten ist dann fremdbestimmt. Das Unbeha- gen führt zu einer kurzfristigen Belastung, die sich in der Lautäußerung ausdrückt. Es wird vermutet, daß gleichzeitig die Lautäußerung ein Mittel darstellt, mit dessen Hilfe solche Belastungen gemildert werden können. Für das Leise Fiepen der Kitze könnte ein entsprechender Zusammenhang angenom- men werden; es wäre dann möglicherweise ontogenetisch ein Vorläufer vom Langen Fiepen. Das Leise Fiepen wird jedoch auch in Situationen geäußert, in denen Mutter und Kitz allein miteinander sind und freundliche Kontakte haben; daher könnte es auch als Vorläufer des Kurzen Fiepens angesehen werden. Daten für eine genaue zeitliche Analyse des Auftretens der verschiedenen Fieplaute in der Ontogenese liegen nicht vor. Daher kann nicht ausgeschlossen werden, daß eine ausgedehnte Analyse Unterschiede in der Laut- struktur ergäbe, die möglicherweise mit den verschiedenen Verhaltenskontexten korrelie- ren. Findeutigere Befunde zur Lautstruktur wären zu einer Klärung der Ontogenese der Fieplaute notwendig. Dem Leisen Fiepen ist sowohl mit dem Langen Fiepen als auch mit dem Kurzen Fiepen gemeinsam, daß sie zu einer Verringerung aggressiver Tendenzen beim Interaktionspartner führen und dadurch eine räumliche Nähe der Interaktionspartner ermöglichen. Diese Funktion wird im allgemeinen mit „Beschwichtigungsgebärden‘ (TINBERGEN 1959) ver- bunden. Für ein Muntjak-Jungtier erscheint es aufgrund der noch relativ geringen Durch- 380 S. G. Stadler und H. Hendrichs setzungs- und Ausweichfähigkeit nützlich, Mechanismen zur Abwehr aggressiven Verhal- tens des Interaktionspartners zu besitzen. Dem Hufstampfen bei Cerviden wird ım allgemeinen eine Droh- und eine Warnfunk- tion zugeordnet (BUBENIK 1965; MÜLLER-UsInG und SCHLOETH 1967; RICHARDSON et al. 1983). Eine Verhaltensantwort der Kitze, etwa das „Sich-Drücken“ (z.B. BuBENIK 1965 für Capreolus capreolus und Cervus elaphus), haben wir nie beobachtet. Das Hufstampfen trat im Zusammenhang mit potentiellen Feinden (Mensch) auf und wirkte auf andere Tiere, die aufmerksam wurden, sicherten (vgl. RICHARDSoN et al. 1983), und z.’T. ebenfalls mit Hufstampfen begannen. Wenn unterlegene Tiere mit Hufstampfen begannen, nachdem sie von Überlegenen gejagt worden waren, hatte dies manchmal zur Folge, daß die überlegenen Tiere von der Verfolgung abließen und sichernd stehenblieben. Dies könnte für eine Warnfunktion sprechen. BARRETTE (1977a,b) diskutiert einen ähnlichen Zusammenhang für das Bellen. Sicher muß dem Hufstampfen ein Zustand relativ hoher Erregung zugeordnet werden, der durch Reizkonstellationen der äußeren Umwelt bewirkt und mit Hilfe der Verhaltensweise verarbeitet wird. Auch ist offensichtlich eine Kontliktsituation geeignet, das Hufstampfen auszulösen (vgl. Dugosrt 1970). Dem Zähneknacken wird eine Drohfunktion zugeordnet (BuBENIK 1965; MÜLLER- UsınGg und SCHLOETH 1967; BARRETTE 1977a). Dies wird durch unsere Beobachtungen bestätigt. In über 90 % der beobachteten Fälle trat das Zähneknacken zusammen mit eindeutigem Drohverhalten zwischen dd auf. Nach MÜLLER-Usıng und SCHLOETH (1967) kann Zähneknacken beı Cerviden in beiden Geschlechtern auftreten. Wir haben es nur bei dd beobachtet. Interessant ist der Fall des jungen d, das, in ein kleines Gehege gesetzt, bei erzwunge- ner Nähe zum Menschen mit Zähneknacken reagierte. Dabei zeigte sich ein starker Konflikt verbunden mit hoher Erregung (Piloerektion). Da eine Möglichkeit, der Situation durch Flucht zu entkommen, nicht bestand, wurde möglicherweise eine aggressive Kom- ponente im Motivationsgefüge aktiviert. Auch in den geschilderten Situationen, in denen ein d nach längerem, intensivem und vom 2 abgewehrten Sexualverhalten Zähneknacken zeigte, könnte sich eine aggressive Komponente verstärkt haben. Zur Beurteilung dieser Situationen scheint das Übersprungkonzept von TINBERGEN (1940, 1952) hilfreich zu sein. Danach würde es sich um einen Fall der Blockierung oder Vereitelung (‚‚thwarting“) einer stark aktivierten Handlungstendenz, hier des Fluchtverhaltens bzw. des Sexualverhaltens, handeln, wodurch ein anderes Verhalten - die ‚„Alternativbewegung‘“ nach LEYHAUSEN (1952), hier aus dem Bereich des Drohverhaltens - aktıviert wird. TINBERGEN (1952) unterscheidet auf der funktionalen Ebene zwischen einer primären und einer sekundären Funktion von Übersprungbewegungen. Diese Unterscheidung liefert möglicherweise einen Schlüssel zur Erklärung des Zähneknackens der adulten dd. Drohverhalten wird häufig als in einem Kontlikt entstanden gedacht zwischen einer aggressiven und einer Fluchttendenz (z.B. LORENZ 1963). Die beiden oben angeführten Situationen haben eine wichtige Gemeinsamkeit: daß die Ausführung einer Verhaltensten- denz durch weitere Bedingungen verhindert wird. Es ist vorstellbar, daß Zähneknacken zunächst nur als Ventil zum Abbau einer hochaktivierten, aber blockierten Verhaltensten- denz, namentlich des Fluchtverhaltens, gedient hat - primäre Funktion sensu TINBERGEN. Auf dem Wege der Ritualisierung wäre es dann zusätzlich soziales Sıgnal im Drohverhalten geworden — sekundäre Funktion sensu TINBERGEN. Danksagung Herrn Dipl.-Biol. JÖRG BÖHNER gilt unser besonderer Dank für stete Hilfsbereitschaft in akustisch- technischen Dingen und auch für wertvolle Anregungen sowie wiederholtes Lesen und Diskutieren früherer Fassungen des Manuskriptes. Des weiteren danken wir Frau Dipl.-Biol. KArRın VOLGER und Herrn Dipl.-Biol. AnDrEAs HAEMIScH für kritische Durchsicht des Manuskriptes. Zum Lautrepertoire von Muntiacus reevesi 381 Zusammenfassung Von Mai 1983 bis Dezember 1984 wurden 17 Tiere einer kleinen Gehegepopulation des Formosa- Muntjaks beobachtet; ca. 200 Protokoll-Stunden wurden auf die aufgetretenen Lautäußerungen hin ausgewertet. Die Tiere zeigten 6 verschiedene Laute (Kläffen, Bellen, Kitzschrei, Fiepen, Hufstamp- fen, Zähneknacken), von denen das Fiepen in drei Untertypen unterteilt wurde. Die Laute werden beschrieben; Kläffen, Bellen und Fiepen werden sonagraphisch dokumentiert. Kläffen und Bellen traten in Situationen auf, die mit Feindvermeidung bzw. Unterlegenheit gegenüber einem dominanten Sozialpartner beschrieben werden können. Es wird als zugrundeliegen- der Zustand ein inneres Ungleichgewicht angenommen, das möglicherweise mit Hilfe der Lautäuße- rung gemildert werden kann. Eine Warnfunktion des Bellens kann nicht nachgewiesen werden, läßt sich jedoch auch nicht ausschließen. Der Kitzschrei wurde von Kitzen und Jungtieren im Alter bis zu etwa einem Jahr als Ausdruck von Angst geäußert. Ob es als ontogenetische Vorstufe des Bellens angesehen werden kann, erscheint fraglich. Das Fiepen wird unterteilt in drei Laute: Kurzes Fiepen: Es wird von dominanten Männchen oder Kitze führenden Weibchen geäußert. Ihm wird ein innerer Zustand der „Selbstsicherheit“ zugeord- net. Langes Fiepen: Es wird von unterlegenen Tieren geäußert. Dieser Laut ist äußerst variabel. Ihm wird eın innerer Zustand des „Unbehagens‘ zugeordnet. Leises Fiepen: Es wird von Kitzen geäußert und wird als ontogenetische Vorstufe der anderen Fieplaute diskutiert. Dem Hufstampfen wird eine Warnfunktion zugeschrieben. Der innere Zustand ist als hoch erregt zu bezeichnen. Zähneknacken trıtt als Bestandteil des Drohverhaltens adulter Männchen auf, auch in Konflıiktsi- tuationen. Die Anwendbarkeit von Übersprungkonzepten wird diskutiert. Literatur ALTMANN, J. (1974): Observational study of behavior: sampling methods. Behaviour 49, 227-267. BaLpwin, J. D.; BaLpwin, J. I. (1977): The role of learning phenomena in the ontogeny of exploration and play. In: Primate Biosocial Development: Biological, Social and Ecological Determinants. Ed. by $. CHEVALIER-SKOLNIKOFF and F. E. PoIRIER. New York, London: Garland Publishing. 343-406. BARRETTE, C. (1977a): The social behaviour of captive muntjacs, Muntiacus reevesi (Ogilby, 1839). Z. Tierpsychol. 43, 188-213. BARRETTE, C. (1977b): Some aspects of the behaviour of muntjacs in Wilpattu National Park. Mammalia 41, 1-34. BARRETTE, C. (1977c): Fighting behaviour of muntjac and the evolution of antlers. Evolution 31, 169-176. BiscHor, N. (1972): Inzuchtbarrieren in Säugetiersozietäten. Homo 23, 330-351. BiscHor, N. (1985): Das Rätsel Odipus. München, Zürich: Piper. BUBENIK, A. B. (1965): Beitrag zur Geburtskunde und zu den Mutter-Kind-Beziehungen des Reh- (Capreolus capreolus L.) und Rotwildes (Cervus elaphus L.). Z. Säugetierkunde 30, 65-128. CHAPLIN, R. E. (1971): Some observations on the mating, vocal and territorial behaviour of wild muntjac deer. Deer 2, 723-725. Dusost, G. (1970): L’organısation spatial® et sociale de Muntiacus reevesi Ogilby, 1839 en semi- liberte. Mammalıa 34, 331-355. Dusost, G. (1971): Observations Ethologiques sur le Muntjak (Muntiacus muntjak Zimmermann, 1780 et M. reevesi Ogılby, 1839) en captıvite et semi-liberte. Z. Tierpsychol. 28, 347427. EISENBERG, ]. F.; LOCKHART, M. (1972): An ecological reconnaissance of Wilpattu National Park, Ceylon. Smithson. Contrib. Zool. 101, 1-118. EspmARK, Y. (1971): Individual recognition by voice in reindeer mother-young relationship. Field observations and playback experiments. Behaviour 40, 295-301. GRADL-GRAMS, M. (1983): Lautäußerungen und Kontaktverhalten beim Sıkawild (Cervus nıppon nıppon Temminck, 1838) (Teilstudie III). Zool. Anz. (Jena) 210, 101-118. HamıLton, W. D. (1964): The genetical evolution of social behaviour. J. Theor. Biol. 7, 1-52. Harpıng, S. P. (1982): The muntjac project: A brief progress report. Deer 5 (9), 450451. HENDRICHs, H. (1975): The status of the tiger Panthera tigris (Linne, 1758) in the Sundarbans Mangrove Forest (Bay of Bengal). Säugetierkdl. Mitt. 23, 161-199. HENDRICHs, H. (1978): Die soziale Organisation von Säugetierpopulationen. Säugetierkdl. Mitt. 26, 81-116. Hennig, R. (1962): Über einige Verhaltensweisen des Rehwildes in freier Wildbahn. Z. Tierpsychol. 19, 223-229. HinDE, R. A.; RowELL, T. E. (1962): Communication by postures and facial expressions in the rhesus monkey (Macaca mulatta). Proc. Zool. Soc. Lond. 138, 1-21. 382 $. G. Stadler und H. Hendrichs Hırra, D. H.; McCuLLoucH, D. R. (1977): Evolution of alarm signals in ungulates with special reference to white-tailed deer. Amer. Nat. 111, 3142. Jürcens, U. (1979): Vocalization as an emotional indicator: A neuroethological study in the squirrel monkey. Behav. 69, 88-117. Kırzy, M. (1972): The vocalizations of ungulates, their causation and function. Z. Tierpsychol. 31, 171-222. LEveEr, ©. (1977): The Naturalized Animals of the British Isles. London: Hutchinson. LEYHAUSEnN, P. (1952): Theoretische Überlegungen zur Kritik des Begriffes der ‚„‚Übersprungbewe- gung“. In: Antriebe tierischen und menschlichen Verhaltens. Ed. by K. Lorenz and P. Lry- HAUSEN (1973), 4. Aufl. München: Piper. 77-88. LORENZ, K. (1963): Das sogenannte Böse. Wien: Borotha-Schoeler. MAYNARD SMITH, J. (1965): The evolution of alarm calls. Am. Nat. 99, 59-63. Morron, E. $. (1977): On the occurrence and significance of motivation-structural rules in some bird and mammal sounds. Am. Nat. 111, 855-869. MÜLLER-Usıng, D.; SCHLOETH, R. (1967): Das Verhalten der Hirsche. Hdb. d. Zool. Bd. 8, 43. Lief., Beitr. 28, 1-60. RICHARDSON, L. W.; JAcoBson, H. A.; Muncy, R. ]J.; PERKINS, J. C. (1983): Acoustics of white-tailed deer (Odocoileus virginianus). J. Mammalogy 64 (2), 245-252. SCHALLER, G. B. (1967): The deer and the tiger. Chicago: Univ. Chicago press. SCHERER, K. R. (1985): Vocal affect signalling: A comparative approach. Adv. in the Study of Behavior 15, 189-244. SOPER, E. A. (1969): Muntjac. A study of these small elusive Asiatic deer which colonized an English garden. London: Longmans, Green. SMYTHE, N. (1977): The function of mammalian alarm advertising: social signals or pursuit invitation? Am Nat 192 TINBERGEN, N. (1940): Die Übersprungbewegung. Z. Tierpsychol. 4, 140. TINBERGEN, N. (1952): „Derived‘“ activities: their causation, biological significance, origin and emancipation during evolution. Quart. Rev. Biol. 27, 1-32. TINBERGEN, N. (1959): Einige Gedanken über „Beschwichtigungsgebärden“. Z. Tierpsychol. 16, 651-665. TRIVERS, R. L. (1971): The evolution of reciprocal altruism. Quart. Rev. Biol. 46, 35-57. WHITEHEAD, K. G. (1972): Deer of the world. London: Constable. WOooDpving, K. W. (1984): Do muntjac bucks “cast” their canıne teeth? Deer 6 (5), 167. YAHNER, R. H. (1980): Barking in a primitive ungulate, Muntiacus reevesi: Function and adaptive- ness. Am. Nat. 116 (2), 157-177. Anschrift der Verfasser: STEFAN G. STADLER und Prof. Dr. HUBERT HENDRICHS, Arbeitsgruppe Säugetiersozialsysteme, Lehrstuhl für Verhaltensphysiologie, Fakultät für Biologie, Universität Bielefeld, Postfach 8640, D-4800 Bielefeld 1 BEKANNTMACHUNG Protokoll über die Mitgliederversammlung der Deutschen Gesellschaft für Säuge- tierkunde e.V. am 28. September 1987 im Vortragssaal des Institutes für Pflanzen- physiologie der Freien Universität Berlin Der 1. Vorsitzende, Herr KuLzEr, eröffnet die Versammlung um 17.30 Uhr. 1. Die Tagesordnung wird angenommen. 2. Herr Kulzer verliest den Bericht über das abgelaufene Geschäftsjahr 1986: Im Berichtsjahr erschien der 51. Band der „Zeitschrift für Säugetierkunde“ in 6 Heften mit 384 Seiten. Zusätzlich wurde allen Mitgliedern das Sonderheft mit den Kurz- fassungen der Vorträge und Posterbeiträge der Tagung in Stuttgart zugesandt. Auf Einladung der Herren DIETERLEN und NEUGEBAUER fand die 60. Hauptversamm- lung der Gesellschaft vom 28. 9. bis 2. 10. 1986 in Stuttgart statt. Über 200 Tagungs- 10. PR 383 teilnehmer besuchten die 92 wissenschaftlichen Veranstaltungen. Eine ganztägige Exkursion zum Steinheimer Becken und in das Nördlinger Ries schloß die Tagung ab. Durch den Tod verlor die Gesellschaft tolgende Mitglieder und Freunde: Dr. KAarı-HEınz Casper, Liebenwerda Prof. Dr. BERNHARD GRZIMEK, Frankfurt Prof. Dr. WERNER MATTHEs, Halle Dr. Karı MEUNIER, Kiel ULRICH ROEDER, Herne ELFI von ULATOwsKkI, Ost-Berlin Im Jahre 1986 konnten 36 neue Mitglieder geworben werden (bei 29 Abgängen). Damit hat die Gesellschaft 608 Mitglieder. Erstmals wird der Förderpreis verliehen. Herr Kurzer überreicht ihn Herrn Dr. MARTIN $. FISCHER und dankt dem Stifter ebenso wie der Jury. . Herr EvErTS verliest den Kassenbericht für das Geschäftsjahr 1986 und übergibt sein Amt Frau KüHnkrıchH. . Die Kassenprüfer Herr SCHRÖPFER und Herr BERGER haben keinen Anlaß zur Beanstandung gefunden. . Die Anträge zur Entlastung des Schatzmeisters und des Vorstandes werden einstimmig angenommen. . Als Kassenprüfer für das Geschäftsjahr 1987 werden die Herren BOHLKEN und SCHLIEMANN gewählt. . Die Mitgliedsbeiträge betragen unverändert für Vollmitglieder 90,- DM, Studenten 60,- DM, Ehegatten ohne Zeitschriftenbezug 10,- DM. . Die Mitglieder nehmen einstimmig die Einladung der Herren BERGER, SCHRÖPFER und RUEMPLER an, die 62. Hauptversammlung vom 2. 10. bis 6. 10. 1988 in Münster abzuhalten. Als Schwerpunktthemen werden „Stoffwechsel und Thermoregulation“, „Ethologie der Säugetiere“ und „Ökologische Tiergeographie“ sowie eine Diskussionsrunde über das Thema „Das Tier im Paragraphendschungel“ gewählt. Die Einladung von Herrn VoGer, 1989 ın Lausanne zu tagen, wird durch Akklama- tion angenommen. . Die Beihefte sollen künftig als Supplement der „Zeitschrift für Säugetierkunde“ gelten und damit unter Angabe des Bandes und der Jahreszahl zitierfähig werden. Die Herren KuLzEr und VoceL werden in das Herausgebergremium der „Zeitschrift für Säugetierkunde“ gewählt. Für die in den USA erscheinende „Mammalian Species“ werden deutsche Spezialisten gesucht. Die Sitzung endet um 18.30 Uhr. Prof. Dr. E. Kurzer Prof. Dr. U. ScHMIDT Dr. H. FRÄDRICH 1. Vorsitzender Geschäftsführer Schriftführer BIÜIE EBESTEREISENEITNIE, MEanow, R. H.; UERPMANN, H.-P.: Equids in the ancient world. Beihefte zum Tübinger Atlas des Vorderen Orients. Reihe A (Naturwissenschaften) Nr. 19/1. Wiesbaden: Dr. Ludwig Reichert Verlag 1986. 421S., zahlreiche Tab., Ktn., Taf. u. graph. Darst. DM 190,-. ISBN 3-88226-282-6 Der vorliegende Band über die Pferde der Alten Welt ist das Ergebnis eines Symposions, das 1982 in Tübingen unter Beteiligung zahlreicher Fachkollegen aus Europa und Übersee stattfand. Nach Bekundung der Herausgeber „the time was ripe for a small gathering of researchers to discuss the problems...“, um hinsichtlich des vielfach widersprüchlich beurteilten taxonomischen Status der vor- und frühgeschichtlichen Equiden Kleinasiens, des Vorderen und Mittleren Orients und Nord- afrıikas zu einem Konsens zu gelangen. Entsprechend dieser Zielsetzung gilt das Augenmerk der Beiträge in erster Linie der Erörterung osteologischer Kriterien, mit deren Hilfe die bei Ausgrabungen geborgenen Knochen und Zähne einer der Untergattungen der Gattung Equus (echte Pferde, Esel und Halbesel) zugewiesen werden können. Weil aber die Artdiagnosen nicht auf Größe und Form einzelner Skelettelemente, sondern im wesentlichen auf dem äußeren Erscheinungsbild basieren, stellen sich einer sicheren Determinierung der oft nur fragmentarisch vorliegenden, aus dem Skelett- verband isolierten prähistorischen Knochen erhebliche Schwierigkeiten in den Weg. Hinzu kommen eine oft nicht bekannte innerartliche Variabilität, mögliche Artbastardierungen und domestikationsbe- dingte Einflüsse, die manche Bemühungen um eine sichere Zuordnung einzelner Funde zunichte machen. Für die meisten der Funde dürfte der Hinweis von M. A. ZEDER Gültigkeit haben, daß „the identification of the bulk of... equids as either ass or hemionus is problematic...“, und zwar deshalb, weil „the comparative materials which provide the bases for our ıdentifications fail to provide clear-cut features for species discrimination“. Der Band enthält 14 Beiträge, die sich mit der Unterscheidung der rezenten Untergattungen und der Bestimmung vorgeschichtlicher Equidenfunde aus der Türkei, aus Ägypten, Israel, Syrien, Irak und dem Iran befassen. Einer ausführlichen Erörterung und Würdigung der einzelnen Aufsätze und der nicht immer widerspruchsfreien Ansichten muß hier entraten werden. Eine bloße Aufzählung der Beiträge sollte zumindest in Ansätzen erkennen lassen, wo gegenwärtig die Schwerpunkte der Erforschung prähistorischer Equiden im vorderasiatischen Raume liegen. C. P. Grovss: The taxonomy, distribution, and adaptions of recent equids. V. EISENMANN: Comparative osteology of modern and fossil horses, halfasses and asses. V. EISENMANN and S. BECKOUCHE: Identification and discrimination of metapodials from pleistocene and modern Eguus, wild and domestic. J. Zarıns: Equids associated with human burials in the third millennium B. C. Mesopotamia: two complementary facets. J. N. POSTGATE: The equids of Sumer, Again. J. CLUTTON- Brock: Östeology of the equids from Sumer. R. BURLEIGH: Chronology of some early domestic equids in Egypt and Western Asia. P. Ducos: The equid of Tell Muraibit, Syria. H.-P. UERPMAnN: Halafıan equid remains from Shams ed-Din Tannıra in northern Syria. R. H. MEApDow: Some equid remains from Cayönü, southeastern Turkey. $S. Bökönyı: The equids of Umm-Dabaghiyah, Iraq. P. F. TURNBULL: Measurements of Eguus hemionus from Palegawra Cave (Zarzian, Iraq). M. A. ZEDER: The equid remains from Tal-e Malyan, southern Iran. CH. $. CHURCHER: Equid remains from neolithic horizons at Dakhleh Oasis, Western desert of Egypt. Das Kernstück der meisten Arbeiten bilden Maßtabellen. Sie enthalten nicht nur die an prähistori- schen Knochen abgenommenen Maße, sondern auch zahlreiche, bisher nicht publizierte metrische Daten von rezenten Equidenskeletten, die sich in Zoologischen Museen der Alten und Neuen Welt befinden. Ergebnisse metrisch-quantitativer Analysen werden in zahlreichen graphischen Darstellun- gen anschaulich gemacht, nichtmetrische Befunde (Schmelzschlingenmuster) ım Bilde vorgestellt. Literaturverzeichnisse beschließen die einzelnen Beiträge. Wenn auch die Aufsätze unterschiedliche Schwerpunkte setzen, gelegentlich auch das kulturge- schichtliche Moment solcher Untersuchungen betonen, so liefern sie alle letztlich doch Beiträge zur Frage nach der prähistorischen Verbreitung der Equiden in Kleinasien, Nordafrika und dem Vorderen und Mittleren Orient. Wer sıch künftig mit Pferden der Alten Welt zu beschäftigen hat, wird das Erscheinen des vorliegenden Bandes (dem ein zweiter folgen soll) dankbar begrüßen. Das Buch gehört trotz seines sehr speziellen Inhaltes in jede säugetierkundliche Bibliothek. H. REICHSTEIN, Kiel Deutsche Gesellschaft für Säugetierkunde 61. Hauptversammlung in Berlin, 27. September bis 1. Oktober 1987 Kurzfassungen der Vorträge und Posterdemonstrationen. Herausgegeben von Prof. Dr. Heinz-Georg Klös, Dr. Hans Frädrich, Prof. Dr. Carsten Niemitz, alle Berlin. Zusammenstellung: Christel Schmidt, Bonn. 1987. 57 Seiten. 24,5x 16,5 cm. Kartoniert 24,- DM Nach 30 Jahren fand die Hauptver- sammlung der Deutschen Gesellschaft für Säugetierkunde wieder in Berlin statt. Die überaus große Zahl der an- gemeldeten Vorträge und Poster be- wies, daß die nunmehr 750jährige Metropole - sie ist Gründungsort der Gesellschaft - für deren Mitglieder und Freunde an Anziehungskraft nichts eingebüßt hat. Säugetierfor- schung wird in Berlin an verschiede- nen Fachbereichen der Freien Univer- sität, der Technischen Universität sowie im Zoologischen Garten betrie- ben; ein Naturkundliches Museum, das nach der Teilung im Westteil der Stadt lange fehlte, befindet sich nun im Aufbau. Ein Hauptziel der Deutschen Gesell- schaft für Säugetierkunde ist, aufihren Jahrestagungen über Säugetiere arbei- tende Wissenschaftler verschiedenster Fachrichtungen zusammenzuführen, den Gedanken- und Erfahrungsaus- tausch anzuregen, um so Erkenntnisse aus den einzelnen Forschungsgebieten zu integrieren. Kommunikation, Fort- pflanzung und Funktionsmorphologie der Säugetiere waren 1987 die Schwer- punkte, über die berichtet wurde. Sie bieten durch ihre Vielfalt die Grund- lage für einen intensiven Gedanken- und Erfahrungsaustausch, eines der Hauptziele der Gesellschaft. Der Wunsch der Veranstalter, daß Berlin als Stätte des Dialoges den geeigneten Rahmen bilden würde, hat sich voll erfüllt. Zu beziehen durch jede Buchhandlun Verlagsbuchhandlung PA)L Paul Parey Spitalerstraße 12 a.] 2000 Hamburg 1 Erscheinungsweise und Bezugspreis 1987: 6 Hefte bilden einen Band. Jahresabonnement Inland: 298,- DM zuzüglich 13,80 DM Versandkosten, Gesamtpreis 311,80 DM einschließlich 7% Mehrwert- steuer. Jahresabonnement Ausland: 298,- DM zuzüglich 24,— DM Versandkosten. Das Abonne- ment wird zum Jahresanfang berechnet und zur Zahlung fällig. Es verlängert sich stillschweigend, wenn nicht spätestens am 15. November eine Abbestellung im Verlag vorliegt. Die Zeitschrift kann bei jeder Buchhandlung oder bei der Verlagsbuchhandlung Paul Parey, Spitalerstraße 12, D- 2000 Hamburg 1, bestellt werden. Die Mitglieder der „Deutschen Gesellschaft für Säugetier- kunde“ erhalten die Zeitschrift unberechnet ım Rahmen des Mitgliedsbeitrages. Z. Säugetierkunde 52 (1987) 6, 329-384 Lehrbuch der Schafkrankheiten Von Prof. Dr. med. vet. Heinrich Behrens, ehem. Direktor des Tiergesundheitsamtes der Landwirtschaftskammer Hannover, apl. Professor an der Tierärztlichen Hochschule Han- nover. 3., überarbeitete und erweiterte Auflage. 1987. 312 Seiten mit 108 Abbildungen, davon 20 farbig auf 4 Tafeln. Kartoniert DM 98,- Das bereits in seinen beiden Vorauflagen als Standardwerk bewährte Lehrbuch liegt nun in der dritten, überarbeiteten und erweiterten Auflage vor. Der Verfasser, ein seit langem als Hochschullehrer, Wissenschaftler und praktischer Berater weithin bekannter Fach- mann, gibt eine umfassende Darstellung der Schafkrankheiten, ihrer Erkennung und Be- handlung sowie aller prophylaktischen Maßnahmen zur tierärztlichen Betreuung von Schafen in allen Haltungsformen. Dazu gehört in immer noch zunehmendem Umfang die Koppelschafhaltung, insbesondere in Form der Hobbyhaltung, die eine stärkere Inan- spruchnahme der praktizierenden Tierärzte und der tierärztlichen Untersuchungsämter mit sich bringt. Die Neuauflage enthält zusätzlich 4 Schwarzweiß- sowie 8 farbige Abbildungen auf 4 Tafeln. Es wurden 28 Krankheiten neu aufgenommen und, insbesondere für die Hobbyschaf- haltung, die Kälteeinwirkung auf neugeborene Lämmer sowie die mutterlose Lämmerauf- zucht und Lämmeraufzucht durch fremde Mutterschafe berücksichtigt. Die im Anhang zusammengefaßten Kapitel über klinische Normalwerte, Blutmeßgrößen und Altersbestim- mung stellen gleichfalls eine Erweiterung dar. Die Literatur wurde neu ausgewählt. Auf dem neuesten Stand steht das Lehrbuch der Schafkrankheiten nun wieder für einen großen In- teressentenkreis in Studium, Praxis und Beratung zur Verfügung. Aus Buchbesprechungen zur 2. Auflage: „Auch die 2. Auflage des Lehrbuches stellt den gegenwärtigen Stand des Wissens in sehr gestraffter und übersichtlicher, aber dennoch gründlicher Weise dar und ist für den Tierarzt und ebenso für den Studenten der Tiermedizin, der sich mit den Krankheiten des Schafes und ihrer Bekämpfung beschäftigt, unentbehrlich. Insgesamt ein ausgezeichnetes und von dem Verlag hervorragend ausgestattetes Lehrbuch, dem eine breite Anwendung zu wünschen ist.“ Deutsche Tierärztliche Wochenschrift „Das nicht allzu groß angelegte Lehrbuch gibt eine umfassende Darstellung derwichtigsten in unserem Raum auftretenden Infektions-, Mangel- und Stoffwechselkrankheiten sowie vor allem der parasitären Probleme. Auch Vergiftungen und Erbkrankheiten sind ent- sprechend berücksichtigt. Wertvoll sind Hinweise über die Behandlung dieser Tierart, vor allem Anweisungen, die für den heute oft unerfahrenen kleineren Schafhalter von beson- derer Bedeutung sind. Auch der Tierarzt erfährt manches, was ihm bei der Behandlung einzelner Hobbytiere, die dem Besitzer oftsehr ans Herz gewachsen sind, nützlich sein wird. So ist der Erwerb dieser Neuauflage allen, die sich mit Schafen befassen, vor allem aber dem Tierarzt, der zur Behandlung und zur Betreuung von Schafbeständen zugezogen wird, bestens zu empfehlen.“ d Zuchthygiene PAJL RE Berlin und Hamburg INS S31YVYYg1I7 -!DPRARIEO SMENSUNAN _ INSTIIULIUN DNOILTLILSNI NUNOSHLNS Saar 7) = = zZ r 53 on 7 ul u) = \ > = & zn N 2 u N = Sa, j 3 <<. am < ZINN Bi © = = = or = Q\ & = 3 ea & 3 an N & — — ” = _ 4 X w JAN INSTITUTION NOILNLILSNI_NVINOSHLINS_ S31UVYIIT LIBRARIES” SMITHSONIAN r =: N; = En = 2 = = u NuR2 = ei > MH, © 4.2 5 = N 5 x > Fin, 2 > = N \ m > Er SAL? 0 > = NS I> au 2,97 E23 ws s3ıyvygıı LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS Sali IB A S 20 2 < = 58 = 2 = = ZU ERS _ > = y b \z G 29 5 = 5 N = 5 2 7. ) 8 F, AN ö TEEN a oa 4 Pe = = = N 2, E 2,4 ze S N. > = NUN =” = =’ EEE z 77) A < 177) Hi zZ 177) z 'ZE AN INSTITUTION NDLEFILLLSNE U NINDEHLINS = IUVYUg817 LIBRARIES SMITHSONIAN T Bw (dp) > Sa) [797 — ir z m 7 = z MM = 2 77 = = 77 DD; iR ce = ce a = < 7 L72 ER: = = a1 we = x er 5 = es = cn = 21 3 ®) = oO _ oO a o “WE = Se = id z He zz 7E AS s31uvugıı LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWNS s3® | = e) Ef 3 = E Aa E z = Di > = > > fe = = = 274 = = D E | = > zn 00 z 5 = 7 ‚AN INSTITUTION, NOILALILSNI_ NVINOSHLINS S31UVWUBII LIBRARIES SMITHSONIAN INS) | | < N N = « < = 2 Ei N I z z f N NEN: - = ZN > : JENS = 8; N 2 YE R 2 3: = = N = = ur = > z e- > # ns S31Uvaaıı_ „SMITHSONIAN _ INSTITUTION NOILNLLLSNI_NVINOSHLINS Sal | = aM e z = 6 = 5 1 x = E = RS c = m = co 3 m = 1) Im 23 ; > 3 > er = 2% ‚AN_ INSTITUTION NOILNAILSNI s3l4Uvygil LIBRARIES 1.9 = en = ® ä ev = u = — © 4,2 = Eu = > m EN es = 3 E . E =. = 5 = 5 = Dr ZN | m b 7) = n = u = > a INS „>3 ıuvy819 ” INSTITUTION NOILNLILSNI NVINOSHLINS S3ITR Ie= < = Al 4 | = .—. 2mE | = = = = 4 > 3 4% Y 2 U: ug ATS | s) © TE a) 2 2 5 2 u | | 72, = ES E 2 = zy#I Ks A A 5 = Ei AN_ NOLLNLILSNI_NVINOSHLINS SIIUVYHIT LIBRARIES SMITHSONIAN INSTI: | es on > 02) 3 72) > —_E | = Fan = = [077] _ 177) } \ a = 3 Me Z = 3 | |. de = = ge = Bi e = fs) = EN — = 4 ER 8 = ) En o ze o “r =E 7m - — 2 RR 2 SE! I IWsS S3I3W44131 LIBRARIES SMITHSNNIAN INSTITITION NOiNnitichi AHINNMNEHTIAG © NOILNLILSNI NOILNLILSI NOIL INSTITUTION SMITHSONIAN Noll LIB NVINOSHLINS S31IUVYY48I71 LIBRARIES INSTITUTION Nol w in wi pr ke ul = “2 = nz <: px < Fr IN Noa« ac 3 ax S Q oc = | ° & ° un 5 > = = 2 20 4 NOILNLILSNI_NVINOSHLINS 53 I4vvyg En LIBRAÄRI ESS Na INSTSIOTION E > Er oO n_ oO G IE es En = co _ T [ee Eu NN wu 2 = I zo ZN IN = > E 2 2 NS = 2 N 2 = 92) = l LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_ NVINOSHIINS S3 I9V4yg917 e x = z x 2 z 2 $ 3 RW: 3 5 ji > ®) 5 > N N x: A u _.NVINOSHLUIWS S31UVJAI1_LIBRARIES SMITHSONIAN INSTITUTION Te — = - z u 7 1 u Wi e- = x 4 FE DR, = SEX = — & 2 GG Bw < = — < = < / Wh za. e- = = =) x Zee @ ° \ = =) =3 o < ee z Rz = gt BRARI ES _ SMITHSONIAN INSTITUTION _NOILNLILSNI_ S31I4V4911 er = m = a I B a 5 an 5 > SD A > = >. = Be) zu ß DE Ps = I = Er 7) ze ei In A z 5 2 NOILNLILSNI NVINOSHLIWNS S3I4UYY81IT LIBRARIES 0 = o z ge: Q = zZ ‚< zZ E u = x 5 2 3 „uf 4 2 = = ZN EIEZT N ID: 2 = 7’ 2: 2 E > >= > = ER S _< 7 zZ 7) ar - 7 > SMITHSONIAN _ INSTITUTION NOILNLILSNI NVINOSHLINS S31yVyY > " = wu = Rz Lu 27) 7) za 177) = u m. ZI x 22. x B {4 =3 = = = R & £ BE) _ oO _ oO = #2 23 = 3 zZ Be N S313V4811 LIBRARIES SMITHSONIAN INSTITUTION NO. 3 = E = ® = s = - = ; > = » > ap es) = = 'x2 = = > rm. = un Bi 0-5 2 = RE SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIWS S31I4V4917 2 zn 9) z o R = = = = = / 3 z =. Js E: GI Ben = = RT) u LP: Ga N oO =I- CE L ©) i G PT: N = = 2, = = ln = Sa = > | | - NOILNLILSNI _NVINOSHLIWNS S314VYY8g1I71 LIBRARIES INSTITUTION “ 0 | ng 3 a 3 ho v3 & —_ X ee = x ar we = = = 5 e ° zZ u a lt = LIBRARIES SMITHSONIAN aan Aa hrhe ter EEBERRTE TEE TE FEST BEEERTEZ Ben 16h FAN i ib, ” H Ah ” v ar = = = + z. Er > £5 NH oe Anke er Ha