ht » EIERN EHRE re / Ye, N NaslyulaN) eat du he" FASO y BETETE rl it Ba at Para z! DE N 2 a 2 a E Ver pwuizk Kuba ale en Se NARBEN IRA ON NET MR Bullen NN A a”, EI So wu AIR 1 A ‚ DESWEEIT ee SER TEL E BULL EG SDR DEREN TEL. BIEESIDEN DEITRR EN ED LTE an SVEN ar Ang ee VETERLSEEE Aarazeele ne AN weil Se nen al Ad Bee Kr FRE NG Peer ME aAndhr dh Eh N Fataıne Ga NN Ip Ar et ta eat An t. DEU DER EEE EEE Tee ne LS TEE El anal RE „m eine EL) MU Y r lan Ba IN? eines NT EN wi Se he N ann an av dir life ua j DREH ER en NUR 7 A dich har a DREH ER ELSE TR PRLIIDLUE Sao BEZ \ u! PERLE PONTE ELLE LE Y enien DZ Na merkenge a HN, ERZIELTEN Yu ya al)n bil n udn POL SEE TEE Ra eraun c ae 2 ner De Ka he er BESTEN = Buswe Baar UN ne ee unge a ee Le went ia Eee Sein dk Her u dt DaPE Er Dre Bin RN u BIETE v BR % en wernie ee ap IDEE BER TR Yen al Di GREEN Tr a a aaa A nu fein El BEER Te a Nehmen, Zee er ie Beisefmae £h Ey: AN ARTEN 5 uva A un DPRESR) FEESIO FETT Ann or Rah (ei, usa zus ee IS Eee {ErIe SP Pe DU E NAT AUCYO ER DECKEL EEE et Bunyarlat te NEIy Lee) N RE BUTARE DRLIELERDETN BONES Pure} SH BENET l u EN ee VE ODE x WÜN eva ET RT DEN a5, asknigit DAR Dr SS 4 Da u nr en w Matt Sa Bann ee Re NR re KO SORTE EFT DEREN glas DEIEREITEeRT NR 22 Ben Ta EREO NDR Ewa - He ENDEN u ee yed PA Ban BR DEIN ON Ay Dune AR Fulda ai DEE IU RT 2, SSH NEUN DEREN wersye Mu ee 77 nn Su Sy PR TIER LU Ira en N EBTEIRELErTn vi KERN DET EIER DER EN IS RER 2 ANR eg yravnd tn IE Er OLE IE TE A NE PR INN } EL DE RLENN na menu De Aer ar Ua SA, VE PEUL a EE BER Un vida Bei nam win A NN DE Gh tan EIER RL ER ERDE TIT nn N nn lhrnybein BP er = RR be RE SRERNN Kae NER NEAR LTE N Bch eh LETSEER ILL DREI BIT RE N a Es Ner nat yo dt ENDE NET TIN ea Een ER BE EEE EEE era ze WEN “r er PERRRORER ENG Vaniewaln en PAR, Wehe tät we WIRST HARE we ee Sein Var Se MR N DE PER ya Yale aa us aM DDR, WR EIN MEAN yes BESTEHT REN Due kn avelhen eh 1. Mean nt FREE RT Le Fe Base dire Ne Ur RR r z Sen Be ORTEN cr ySles Nun zei lese Vrstinehind Usa len DIDR DEN: Br. et grund N EN an w h e dr Se Dr un UNE an EN FR A eu ” Iran BRERETEN EERTRENENe re vw : ran MN RE rn an tern BE “ DR FERN Et it aan len wur 2% 62; RE NN ch N) RN UNE en in Dar ge RE ee NET] YEEe u irladee un Yang m RER REN Aa nein Ve Te nen ae POPITETSLIE TUT GE ZZ EEE ET ae EAS RP ESP E 7 05 DL EA yet ST a N MuLEDCEe Br DENT Par ee PIE TER ReRr FE EL ABEL ELLE BEER u IN ang CR ke N D 1. x 1 RE N sine DH EN ser 5 Yale KA Van Be VIE N N ar we LTE ART Kae ER AR and vi 7 I AT Ser ke " Pe van Du ei RES] DEE Sn) ee NE TE 327 rue PL NE PEONES EL u Auer er TEN ea hen Yin Verst ME Sy BEUTE SUPER ER" a esräntev. ar tes ae iR REN] Krahıen ansehen \ NET nen len yes RE Ur Genen “N re ehe RR ee a Sauna “ SE ns fes han ante NE TEEN Ne DERWIEETN En 57 IRRE UELI wann AR a erh en ER Bay tee U N URN AL NEN BERICHTETEN Ent ee Zn u ERS lid BOPSRERNESE NER, Dre DErEEETO rer Aenesinr eher: nn annn me Ruin DE RER GT un BEER RREEPRU TR N wein ve EEE .. a Beer ni mungen He - . win .. - . i 4 I ER BUN RYAN: WER Di Bin 109 Nr t TEN Br RR DR u h ı DET | ve nz 3% © i ch Wehe Ni: ZEITSCHRIFT FÜR SÄUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Herausgeber/Editor Deutsche Gesellschaft für Säugetierkunde Schriftleitung/Editorial Office D. Kruska, Kiel - P. Langer, Giessen Wissenschaftlicher Beirat/Advisory Board W. Fiedler, Wien - S. Halle, Jena - G. B. Hartl, Kiel - R. Hutterer, Bonn -H.-G. Klös, Berlin - E. Kulzer, Tübingen - P. Lüps, Bern — W. Maier, Tübingen - O. Anne E. Rasa, Bonn -H. Reichstein, Kiel - M. Röhrs, Hannover - H. Schliemann, Hamburg —- G. Storch, Frankfurt Volume 64 1999 II Inhaltsverzeichnis von Band 64 Wissenschaftliche Originalarbeiten Antinuchi, €. D.; Busch, Cristina: Intrageneric comparisons in urine-concentrating capacity and renal morphology among three species of Akodon from different geographic rainfall regimens. — Intragene- rische Vergleiche zwischen der Fähigkeit, Urin zu konzentrieren und der Morphologie der Niere bei drei Akodon-Arten aus geographischen Regionen mit unterschiedlichen Niederschlagsmustern .... Balmelli, Lorenza; Nentwig, W.; Airoldi, J.-P.: Nahrungspräferenzen der Feldmaus Microtus arvalis in der Agrarlandschaft unter Berücksichtigung der Pflanzeninhaltsstoffe. - Food preferences of the common vole Microtus arvalis in the agricultural landscape with regard to nutritional components of plants .. Barlow, Kate E.; Jones, G.: Roosts, echolocation calls and wing morphology of two phonic types of Pipistrellus pipistrellus. - Quartiere, Echoortungslaute und Flügelmorphologie von zwei akustischen Iypenivon Pipistrellus’pipistrellus........&: -EOen, Be 0.0. 2. 0b. Sm Auer ur Ger Borkowska, Anetta: Genetic and morphological variation among populations of the bank vole Clethrio- nomys glareolus from north-eastern Poland: the seasonal aspect. - Genetische und morphologische Differenzierung zwischen Rötelmauspopulationen (Clethrionomys glareolus) aus dem Nordosten Polens: Saisonale Aspekte... rare Burda, H.; Zima, J.; Scharff, A.; Macholän, M.; Kawalika, M.: The karyotypes of Cryptomys anselli sp. nova and Cryptomys kafuensis sp. nova: new species of the common mole-rat from Zambia (Rodentia, Bathyergidae). - Die Karyotypen von Cryptomys anselli sp. nova and Cryptomys kafuensis sp. nova: neue Arten des Graumull von Sambia (Rodentia, Bathyergidae). . . 2.2. 2.2.2. 2.2.2 sr Camin, S.: Mating behaviour of Ctenomys mendocinus (Rodentia, Ctenomyidae). — Paarungsverhalten von Ctenomys mendocinus/(Rodentia: Gtenomyidae) 222 2.2.2... 22 2 Cassinello, J.: Allosuckling behaviour in Ammotragus. — Fremdsaugen bei Ammotragus. ............ Cervantes, F. A.; Ramirez-Silva, J. P.; Marin, Arıadna; Portales, Gloria L.: Allozyme variation of Cotton- tail rabbits (Sy/vilagus) from Mexico. — Allozym-Variation von Wollschwanz-Kaninchen (Sy/vilagus) aus Mexiko... .. 2 dau 2 a ae een ee Elmeros, M.; Madsen, A. B.: On the reproduction biology of otters (Lutra lutra) from Denmark. — Über die Fortpflanzungsbiologie von Fischottern (Zutra lutra) in Dänemark ................22202.. Garcia, Beatriz A.; Martino, Angela; Chiappero, Marina B.; Gardenal, Cristina N.: Allozyme variation and taxonomic status of Calomys hummelincki (Rodentia, Sigmodontinae). — Allozymvariation und taxonomische Stellung von Calomys hummelincki (Rodentia, Sigmodontinae) ................. Garde, J. M.; Escala, Carmen: Coats and moults of the water vole Arvicola sapidus Miller, 1908 (Roden- tia, Arvicolinae) in southern Navarra (Spain). — Felle und Fellwechsel bei Arvicola sapidus Miller, 1908 (Rodentia, Arvicolinae), aus dem Süden von Navarra (Spanien)... ..........:cc..22cc0n: Gimenez, Mabel D.; Bidau, C. J.; Argüelles, Carina F.; Contreras, J. R.: Chromosomal characterization and relationship between two new species of Ctenomys (Rodentia, Ctenomyidae) from northern Cör- doba province, Argentina. — Chromosomale Charakterisierung und karyologische Beziehungen zwischen zwei neuen Arten von Ctenomys (Rodentia, Ctenomyidae) aus dem Norden der Provinz Cördobäl Argentinien... 1.9 mem IT. mearler 2 Me SE Gonzälez-Esteban, J.; Castien, E.; Gosälbez, J.: Morphological and colour variation in the Pyrenean desman Galemys pyrenaicus (Geoffroy, 1811). - Morphologische und Färbungsvariationen beim Py- renden-Desman'Galemys:pyrenaicus (Geotftoy, 181). 2... 2.2.2022 2.2. Goymann, W.; Leippert, D.; Hofer, H.: Parturition, parental behaviour, and pup development in Indian false vampire bats, Megaderma Iyra. - Geburt und Aspekte der Jungenaufzucht und -entwicklung bei Indischen 'Falschen Vampiren, Megadermallyran2 22.2.2222. 22.0 ee Granjon, L.; Bonnet, Amelie; Hamdine, W.; Volobouev, V.: Reevaluation of the taxonomic status of North African gerbils usually referred to as Gerbillus pyramidum (Gerbillinae, Rodentia): Chromoso- mal and biometrical data. - Neubewertung der taxonomischen Stellung nordafrikanischer Renn- mäuse, die gewöhnlich Gerbillus pyramidum (Gerbillinae, Rodentia) zugeschrieben werden: Chromosomale und/biometnische,Datene 2.0... 2a 2 154 257 285 36 230 363 356 193 30 2 91 321 298 Inhaltsverzeichnis von Band 64 Hale, Molly O.; Fuller, T. K.: Estimating porcupine (Erethizon dorsatum Linnaeus, 1758) density using radiotelemetry and replicated mark-resight techniques. — Dichteschätzungen von Baumstachlern (Erethizon dorsatum Linnaeus, 1758) mittels Radiotelemetrie und Sichtungen markierter Individuen Karczmarski, L.; Cockcroft, V. G.: Daylight behaviour of Humpback dolphins Sousa chinensis in Algoa Bay, South Africa. - Tagesgang im Verhalten von Buckeldelphinen Sousa chinensis in Algoa Bay, Süd- DIE ots ee ee ee Re N Kloskowski, J.: Otter Lutra lutra predation in cyprinid-dominated habitats. - Beuteerwerb des Fischotters RrnstenlutrasneyonW\Veißtischen dommierten Habitaten 2... ....... un. uernneen Kunze, Bärbel; Dieterlen, F.; Traut, W.; Winking, H.: Karyotype relationship among four species of Spiny mice (Acomys, Rodentia). — Karyotypische Verwandtschaftsbeziehungen zwischen vier Arten der Geltung Aecomys (Node) Ser Me ee Pohl, Birgit; Meyer, W.: Eine spezielle Markierungsweise und ihre strukturelle Grundlage beim Krabben- waschbär (Procyon cancrivorus). — A specific type of scent marking and its structural basis in the Cab-eaing COMMITTEE Eee Reinhardt, Ilka; Halle, S.: Time of activity of a female free-ranging Lynx (Lynx !ynx) with young Kittens in Slovenia. — Aktivität eines Luchsweibchens (Zynx !ynx) während der Jungenaufzucht in Slowenien Revilla, E.; Delibes, M.; Travaini, A.; Palomares, F.: Physical and population parameters of Eurasian bad- gers (Meles meles L.) from Mediterranean Spain. — Physische und demographische Parameter des Bitasischen®Dachses/(Melesimelesl.) im mediterranen Spanien... ..........uooueeccecennn. Ruckstuhl, Kathreen E.; Ingold, P.: Aspects of mother-kid behaviour in Alpine chamois, Rupicapra rupi- capra rupicapra. — Aspekte der Mutter-Kindbeziehung bei der Alpengemse (Rupicapra rupicapra FEBÜCDDTU) oo 0 0 reed a N A re Schilling, Nadja; Fischer, M. S.: Kinematic analysis of treadmill locomotion of Tree shrews, Tupaia glis (Scandentia: Tupaiidae). -— Kinematische Analyse der Fortbewegung von Tupaia glis (Scandentia: INpandao)ysaugdemmBautbande a ee Smorkatcheva, Antonina V.: The social organization of the Mandarine vole, Lasiopodomys mandarinus, during the reproductive period. — Soziale Organisation der Mandarin-Wühlmaus (Lasiopodomys Mandanımus)swährend’derReproduktionsperiode .2..2..... 0... .0.0.0.ooneenuernen. Suchentrunk, F.; Polster, Karin; Giacometti, M.; Ratti, P.; Thulin, C.-G.; Ruhle, C.; Vasil&Ev, A. G.; Slotta- Bachmayr, L.: Spatial partitioning of allozyme variability in European mountain hares (Lepus timidus): gene pool divergence across a disjunct distributional range? — Räumliche Verteilung der Allozymvariabilität bei europäischen Schneehasen (ZLepus timidus): Genpool-Divergenz in einem dis- Sözen, M.; Colak, E.; Yigit, N.; Özkurt, S.; Verimli, Reyhan: Contributions to the karyology and tax- onomy of the genus Spalax Güldenstaedt, 1770 (Mammalia: Rodentia) in Turkey. - Zur Karyologie und Taxonomie der Gattung Spalax Güldenstaedt, 1770 (Mammalia: Rodentia) in der Türkei...... Torre, I.; Bosch, M.: Effects of sex and breeding status on habitat selection by feral House mice (Mus musculus) on a small Mediterranean island. — Auswirkungen von Geschlecht und Fortpflanzungsstatus auf Habitatwahl bei freilebenden Hausmäusen (Mus musculus) auf einer kleinen Mittelmeerinsel. .. Weinandy, R.; Gattermann, R.: Parental care and time sharing in the Mongolian gerbil. - Elterliche Jun- genpflege und zeitliche Kooperation bei der Mongolischen Wüstenrennmaus .......... 2.2.2.2... Wissenschaftliche Kurzmitteilungen Anderson, R. P.; Soriano, P. J.: The occurrence and biogeographic significance of the southern Spiny pock- et mouse Heteromys australis in Venezuela. — Verbreitung und biogeographische Bedeutung der Süd- IchenStacheltaschenmaus kleteromys.australisıin Venezuela... nn. Angelici, F. M.; Luiselli, L.; Politano, E.: Distribution and habitat of selected carnivores (Herpestidae, Mustelidae, Viverridae) in the rainforests of southeastern Nigeria. — Verbreitung und Habitat ausge- wählter Carnivora (Herpestidae, Mustelidae, Viverridae) in den Regenwäldern des südöstlichen INS ea N RE ee ee ee le re III 85 19 201 220 12 65 269 76 129 344 308 210 176 169 121 IV Inhaltsverzeichnis von Band 63 Colak, E.; Yıgit, N.; Verimli, Reyhan: On the karyotype of the Long-clawed mole vole, Prometheomys schaposchnikovi Satunin, 1901 (Mammalia: Rodentia), in Turkey. - Über den Karyotypen der Pro- metheusmaus, Prometheomys schaposchnikovi Satunin, 1901 (Mammalia: Rodentia), in der Türkei. . Dyzenchauz, F. J.; Massarini, Alicia, I.: First cytogenetic analysis of the genus Bibimys (Cricetidae, Rodentia). — Erste cytogenetische Analyse der Gattung Bibimys (Cricetidae, Rodentia).......... Fiedler, W.; Alder, H. U., Wohland, Pia: Zwei neue Nachweise der Weißrandfledermaus (Pipistrellus kuhli) für Deutschland. — Two new records of Kuhl’s pipistrelle (Pipistrellus kuhli) in Germany..... Gigirey, A.; Rey, J. M.: Faecal analysis of the edible dormouse (Glis glis) in the northwest Iberian Penin- sula. - Kotanalyse von Siebenschläfern (Glis glis) von der nordwestlichen iberischen Halbinsel..... Kock, D.; Künzel, T.: The maned rat, Lophiomys imhaussii Milne-Edwards, 1867, in Djibouti, NE-Africa (Mammalia: Rodentia: Lophiomyinae). — Die Mähnenratte, Lophiomys imhaussii Milne-Edwards, 1867, in Djibouti, NO-Afrika (Mammalia: Rodentia: Lophiomyinae) ................ceeee... Kreb, Daniölle: Observations on the occurrence of Irrawaddy dolphin, Orcaella brevirostris, in the Maha- kam River, East Kalimantan, Indonesia. -— Beobachtungen zum Vorkommen des Irravaddy Delphins, Orcaella brevirostris in dem Fluß Mahakam, Ostkalimantan, Indonesien... ................... Lod&, Th.: Comparative measurements of terrestrial and aquatic locomotion in Mustela lutreola and M. putorius. — Vergleichende Messungen von terrestrischer und aquatischer Lokomotion bei Mustela lutreola: und M. putorius.. na 21er a RL I SE re Nair, N. Gopukumar; Elangovan, V.; Sripathi, K.; Marimuthu, G.; Subbaraj, R.: Foraging behavior of the Indian short-nosed fruit bat Cynopterus sphinx. — Nahrungssuche beim indischen Kurznasen-Flughund Cynopterus,sphinx, -.......:. 2 B tus aaa ee se a Siciliano, S.; Lailson Brito Jr., J.; de F. Azevedo, A.: Seasonal occurrence of killer whales (Orcinus orca) in waters of Rio de Janeiro, Brazil. — Saisonales Vorkommen von Schwertwalen (Orcinus orca) in den Gewässern vor Rio:de Janeiro, Brasilien. . 2 .....0... 0 acın 2a. 2. 2 0 Siemers, B. M.; Kaipf, Ingrid; Schnitzler, H.-U.: The use of day roosts and foraging grounds by Natterer’s bats (Myotis nattereri Kuhl, 1818) from a colony in southern Germany. — Nutzung von Tagesquartieren und Jagdgebieten durch Fransenfledermäuse (Myotis nattereri Kuhl, 1818) in einer Kolonie in Süd- deutschländ...........4:. 2. au wien er e e Szapkievich, Valeria B.; Cappozzo, H.L.; Crespo, E. A.; Bernabeu, R. O.; Comas, Cecilia; Mudry, Marta D.: Genetic relatedness in two Southern sea lion (Otaria flavescens) rookeries in the south- western Atlantic. — Genetische Verwandtschaft zwischen zwei Populationen von Seelöwen (Otaria flavescens) im südwestlichen Atlantik... 2.2... 20.0 Traut, Ilona M.; Ries, Edith, H.; Donat, Britta; Vareschi, E.: Spacing among Harbour seals (Phoca vituli- na vitulina) on haul-out sites in the Wadden Sea of Niedersachsen. — Individualabstände von See- hunden (Phoca vitulina vitulina) auf Sandbänken im niedersächsischen Wattenmeer. ............ Tschapka, M.; Wilkinson, G. S.: Free-ranging Vampire bats (Desmodus rotundus, Phyllostomidae) survive 15 years in the wild. — Freilebende Vampirfledermäuse (Desmodus rotundus, Phyllostomidae) über- leben 15. Jahre.in.der Natur...2.2.4...0.. 40 1a See h% Sen ee Re Pe 126 59 107 376 371 54 110 187 21 241 246 31: 239. Mitteilungen.der, Gesellschaft 11... ht a4. wire ran Sa 63, 128, 380 Buchbesprechüngen 2 72282: 2 FREE MEERE PER ER RE 192, 256, 319, 383 Printed in Germany © Urban & Fischer Verlag 1999 % INTERNATIONAL JOURNAL. OF MAMMALIAN BIOLOGY Gonzälez-Esteban, J.; Castien, E.; Gosälbez, J.: Morphological and colour variation in the Pyrenean desman Galemys pyrenaicus (Geoffroy, 1811). - Morphologische und Färbungsvariationen beim Pyrenden-Desman Galemys pyrenaicus (Geoffroy, 1811) ....... BE N A RE N RESORT ee Ba: Pohl, Birgit; Meyer, W.: Eine spezielle Markierungsweise und ihre strukturelle Grundlage beim Krabbenwaschbär (Procyon cancrıvorus). - A specific type of scent marking and its structural basis in the Crab-eating raccoon un N Fe N RT REES ER ER ee a Ra en, L.; Cockcroft, V. G.: Daylight behaviour of Humpback dolphins Sousa chinensis in Be Bay, South Africa. " - Tagesgang im Verhalten von Buckeldelphinen Sousa chinensis in Algoa Bay, Südafrika.......u.unnenseaseaeesennenne. Garcia, Beatriz A.; Martino, Angela; Chiappero, Marina B.; Gardenal, Cristina N.: Allozyme variation and taxonomic sta- tus of Calomys hummelincki (Rodentia, Sigmodontinae). - Allozymvariation und taxonomische Stellung von Calomys en Rodenkia, SIgMOdOHEINae).........0..4.0-0 Rennen nenntune nen sunnenrünnennnnnsunnennenen en Esln = Burda, H.; Zima, J.; Scharff, A.; Macholän, M.; Kawalika, M.: The karyotypes of Cryptomys anselli sp. nova and Crypto- mys kafuensis sp. nova: new species of the common mole-rat from Zambia (Rodentia, Bathyergidae). - Die Karyo- typen von Cryptomys anselli sp. nova and Cryptomys kafuensis sp. nova: neue Arten des Graumull von Sambia ee en also ä sea nnn an len w en name dmeh nun pa lnhendnnenund nennen Wissenschaftliche Kurzmitteilungen Traut, Ilona M.; Ries, Edith, H.; Donat, Britta; Vareschi, E.: Spacing among Harbour seals (Phoca vitulina vitulina) on naul-out sites in the Wadden Sea of Niedersachsen. - Individualabstände von Seehunden (Phoca vitulina vitulina) BE anebanken ım hiedersachsischen Wattenmeer ..........0=».0025>0200 20000 Ran nos annanana en innnnensnnunanen nenne E Kreb, Danielle: Observations on the occurrence of Irrawaddy dolphin, Orcaella brevirostris, in the Mahakam River, East Kalimantan, Indonesia. - Beobachtungen zum Vorkommen des Irravaddy Delphins, Orcaella brevirostris in dem Fluß Topp anne nanenasnanenien Dyzenchauz, F. J.; Massarini, Alicia, I.: First cytogenetic a of the genus Bibimys (Cricetidae, zum. - Erste cytogenetische Analyse der Gattung Bibimys (Cricetidae, Rodentia) ........... ne ereren a a Isa neh n enannennennedanenanene se ISSN 0044-3468 - Z. Säugetierkunde - 64(1999)1 - 5. 1-64 - Februar 1999 "URBAN & FISCHER 12 19 30 36 51 re 1999 E; wol ra ar De E ZEITSCHRIFT FÜR Se: SÄUGETIERKUNDE INTERNATIONAL JOURNAL SEE OF MAMMALIAN BIOLOGY Herausgeber/Editor Deutsche Gesellschaft für Säugetierkunde Schriftleitung/Editorial Office D. Kruska, Kiel - P. Langer, Giessen Wissenschaftlicher Beirat/Advisory Board W. Fiedler, Wien - S. Halle, Jena - G. B. Hartl, Kiel - R. Hutterer, Bonn - H.-G. Klös, Berlin - E. Kulzer, Tübingen - P. Lüps, Bern — W. Maier, Tübingen - O. Anne E. Rasa, Bonn -H. Reichstein, Kiel - M. Röhrs, Hannover - H. Schliemann, Hamburg - G. Storch, Frankfurt Deutsche Gesellschaft für Säugetierkunde Altvorsitzende/Living Past Presidents D. Starck, Frankfurt (1957-1961, 1967-1971) - H. Frick, München (1972-1976) -— M. Röhrs, Hannover (1977-1981) — H.-J. Kuhn, Göttingen (1982-1986) - E. Kulzer, Tübingen (1987-1991) — U. Schmidt, Bonn (1992-1996) Amtierender Vorstand/Managing Committee Vorsitzender/President: H. G. Erkert, Tübingen Mitglieder/Board Members: H. Frädrich, Berlin - R. Hutterer, Bonn — D. Kruska, Kiel — Marialuise Kühnrich, Hamburg - R. Schröpfer, Osnabrück - Angela Wöhrmann-Repenning, Kassel Z. Säugetierkunde 64 (1999) 1-11 ZEITSCHRIFT © 1999 Urban & Fischer Verla _ _____ ___ SÄUGETIERKÜUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Morphological and colour variation in the Pyrenean desman Galemys pyrenaicus (Geoffroy, 1811) By J. GONZALEZ-ESTEBAN, E. CASTIEN, and J. GOSALBEZ Soc. Ciencias Aranzadi, Donostia-San Sebastian; Servicio de Conservaciön de la Naturaleza, Gobierno de Navarra, Pamplona; and Departament de Biologia Animal, Universitat de Barcelona, Barcelona, Spain Receipt of Ms. 10. 03. 1998 Acceptance of Ms. 18. 11. 1998 Abstract This study presents information on biometry and colouring of Galemys pyrenaicus. The material stu- died covers the entire distribution area of this species on the Iberian Peninsula. Based on the data ex- amined, no significant geographic differences were found as far as skull size was concerned. However, skull sizes and some of the body measurements taken from animals native to the Pyrenees and in- cluded in the subspecies pyrenaicus, are intermediate between subspecies from Galicia and the Iber- ian Mountains, and included in the subspecies rufulus. These data diminish some of the validity ac- corded to craniometric measures as one criterion to justify the existence of subspecies. Colour allows for a better differentiation although the chorological patterns do not coincide with those established by other authors. We suggest that Galemys pyrenaicus is distributed in totally or partially isolated po- pulations, which could favour the process of morphological differentiation. This emphasises the impor- tance of preserving every population area of this species. Key words: Galemys pyrenaicus, taxonomy, morphology, colour Introduction Since the first description of this species by GEOFFROY (1811) in the last century, and de- scription by GRAELLS (1897), sufficient differences in coloration have been seen between the typical form and the desmans from the central Iberian Peninsula in order to consider the latter as a new variety. Only in studies by six authors (MıLLer 1912; CABRERA 1914; NIETHAMMER 1970; PALMEIRIN and HOFFMANN 1983; JUCKWER 1990) has there been any mention of the variability in form and size seen in the desman Galemys pyrenaicus in its distributional area. GEOFFROY (1811) described the dorsal fur as being dark brown (“brun marron”) in col- our. Years later GrRAELLS (1897) based his new variety, Myogalea rufula, noting that the yellow colour of the hairs on the back renders the animal to appear to have gold reflexes, while in the type specimen these reflexes appear silver. He also reported as a distinguish- ing trait, the bright yellow colour of the paws and nails of the desmans from the Central Mountains. MirLeEr (1912) described the coloration of Galemys pyrenaicus rufulus as not as dk as that of the subspecies of the Pyrenees. This author reports that the form rufulus ap- pears to be clearly differentiated from pyrenaicus owing to its larger size and especially to the greater length and strength of the skull. 0044-3468/99/64/01 — 001 $ 12.00/0 2 J. GONZALES-ESTEBAN, E. CASTIEN, and J. GOSALBEZ CABRERA (1914) did not find any substantial differences in coloration between the two forms and he even claimed that the original description of rufulus was totally useless for recognising the anımal in question. He did, however, maintain the existence of two sub- species attributing the bulk of the differentiation to the size; indicating, as discriminating parameters, the length of the rear foot and the condylobasal length. NIETHAMMER (1970) cautioned that although the form rufulus is larger in size, the mea- surements do overlap. He also pointed out that fur colour varies depending on wear. PALMEIRIN and HOFFMANN (1983) considered that the subspecies are distinct, although morphological differences between them are not great. Lastly JuckweEr (1990), through the examination of body weight and condylobasal length, attempted to reaffirm the idea that the subspecies rufulus is greater in size than the subspecies pyrenaicus. The purpose of thıs study ıs to provide information on the biometry and coloration of the desman, analyse the geographic variation of this species on the Iberian Peninsula and discuss its present taxonomic status. Material and methods The material studied consisted of all together specimens from different collections (Tab. 1). In addi- tion we used biometric data provided by MirLer (1912), n = 6; NIETHAMMER (1970), n = 18; JUCKWER (1990), n = 20 (Tab. 1). It is important to note that material from the Central Mountains was not avail- able. The presumable precariousness of the populations living in these mountains made it inadvisable to collect material there. In order to evaluate the geographic variation of the size, the specimens were divided into 5 groups (Fig. 1). Table 1. Origin and composition of the material. Sources: 1 — Museo Luis Iglesias, 2 — Estaciön Biolö- gica de Dofana, 3 — Museo de Ciencias Naturales, (Alava), 4 -— Collection from PABLO AGUIRRE, 5 — Facultad de Biologia, Universidad de Leön, 6 — Museo Luis Iglesias (Santiago de Compostela), 7 — NIETHAMMER (1970), 8— MILLER (1912), 9 — JucKwER (1990), 10 — own material KLUB EIME Skull Fur Source Province Location Population Occidental 1 La Coruna: - - - sl 1 Lugo: Trascastro-Incio 2IPREIB2 5 - 2 Saa de Incio 29TPH32 8 - 2 San Pedro de 29TPH32 5 - 2 Incio 29TPH32 10 _ 2 San Romän 29TPH54 1 - 6 Cadramön 29TPJ10 1l = 6 Barbeitos 29TPH67 1 - 6 Rio Ulla 1 1 1 Pontevedra: Rio Riobö 1 1 1 Vila de Cruces 29TNH63 — 1 1 Orense: Xunqueira de 29TPG07 1 _ 6 Sierra de 29TPG36 1 - 6 Ramiräs 29TNG77 fl - 6 Riobö 29TPG26 1 - 6 Leön: Matarrosa del Sil 29TQOHO03 3 _ 2 Paramo del Sil 29TOHO04 2 - 2 Peranzanes 29TPH95 2 - 2 Manzaneda 29TQ0G28 2 — 2 Sierra Cabrera 29TPG97 1 - 2 Montrondo 29TOH24 1 1 2 Morphological and colour variation in Galemys pyrenaicus Table 1. Continued Province Population Occidental 2 Asturias: Cantabria: Pyrenees Navarra: Alava: Iberian Burgos: La Rioja: Central Location Pola de Somiedo Arriondas Caleao Turieno Rio Quiviesa Rio Hijar Reinosa Espinama Besande Candemuela Riano Pajares San Emiliano Burguete Quinto Real Aoız Orbara Oroz-Betelu Saigos Akerreta Lanz Huizi Oricain Rıba Itoız Österiz Garzaron Guerendain Santo Domingo de Santo Domingo de Sierra de Cameros Sierra de Cameros Ajamil Peroblasco Aguillar de Rio Vadillos Valdepecillo Sierra de Gredos USTEME 29TOH27 30TUP20 30TUNO08 30TUN67 30TUN67 30TUN96 30TVN06 30TUN57 30TUN45 29TOH46 30TUN35 30TTN76 29TOH46 30TXN36 30TXN26 30TXN34 30TXN45 30TXN35 30TXN25 30TXN15 30TXN16 30TWN87 30TXN14 30TXN15 30TXN34 30TXN25 30TWN96 30TXNO05 30TVM64 30TVM64 30TWM27 30TWM27 30TWM46 30TWM67 30TWMS84 30TWM47 30TWM75 30TTK95 Skull dar r Hr os rHvDreHrpHD» MP DM —DD HH m Hr Bwur | Fur Source SPP, OS DGOND The following measurements were studied: Body: HBL (head + body length), TL (tail length), HFL (hind foot length) W (body weight in grams); cranium: CBL (condyle-basal length), IE-VE (de M° length), BW (skull case width), IOW (interorbital width), ML (mandibular length), CH (coronoid height), BH (skull case height). The CBL was measured following JUCKWER (1990). Since the points used by this author to define this parameter do not coincide with those used by MiıLLEr (1912) and NIETHAMMER (1970) this study did not rely on the LCB measurements given by the latter authors. The relative size was assessed according to GRULICH (1967) for Talpa, taking into account tooth wear. Although all the teeth suffer from deterioration, it is more visible in the incisors, canines, and premolars. Animals with the cusps of teeth intact or only little evidence of wear were considered young. Animals showing abrasion down to the middle of the crowns were considered adults, and those that had a highly deteriorated set of teeth - in some cases the crown was completely lacking - were considered old adults. Where possible, characteristics of the reproductive system, such as size and degree of development were taken into account to supplement the age criteria. 4 J. GONZALES-ESTEBAN, E. CASTIEN, and J. GOSALBEZ Fig. 1. Study area and different population groups. Occl: animals from Galicia and the west of the province of Leön; Occ2: animals from Asturias, Cantabria, and the north of the province of Leön; Pyr: animals from Navarra; Ibe: animals from La Rioja and the south of the province of Burgos; Cen: ani- mals from the province of Avila In order to evaluate the geographic variation of coloration, only those animals were considered showing new fur or fur grown after the moult. Thus, we counted 5 specimens from the Occidentall po- pulation, 9 from Occidental 2, 17 from the Pyrenees, and 5 from the Iberian Mountains. The normality of the distribution of the variables was determined by the Kolmogorov-Smirnov test (Nie et al. 1975). The comparison between mean sampling pairs was carried out according to the method of Scheffe (Nie et al. 1975). The degree of differentiation between the different groups in terms of sex, age, and season of the year that the animal was caught was estimated based on the set of variables chosen through the multivariate analysis of variance (MANOVA) (Nie et al. 1975). The geo- graphic variation of size was evaluated by a step-by-step discriminating factorial analysis choosing the variables which maximise the D* of Mahalanobis between the two closest groups (NIE et al. 1975). Results Size Morphometric and craniometric data of the populations are given in table 2. An analysis of the total variability of the parameters considered resulted in no differences concerning size that may be attributed to sex (T” from Hotellings = 0.794; F = 1.192; p > 0.05), age (T” from Hotellings = 1.914; F = 1.355; p > 0.05), or season (T” from Hotellings = 3.058; F = 1.416; p > 0.05). Individual comparisons between pairs of samples (Tab. 3) indicate that desmans from the Occidental 1 population have smaller skulls. This is the group that has the greatest number of contrasts with significant differences of several measures (18 out of 22com- parisons). As far as body measurements are concerned, the Iberian group has the lowest values. However, from the comparisons it is not possible to establish a clear pattern of geographic variation of these parameters. A preliminary discriminating analysis was carried out based on five craniometric vari- ables (CBL, I-M°, BW, IOW, ML) and on the first four groups considered (OCC1, OCC2, PYR., IBER.). The discriminating function which explains a greater percentage of the variance (66.83%) selects four variables: D = (-0.063)x CBL + (1.091) xI'-M° + (-1.498) x IOW + (2.189) x ML - 54.384. The standardised coefficients of this function Morphological and colour variation in Galemys pyrenaicus 5 Table 2. Morphometric (in mm and grams) and craniometric (in mm) values of the populations stu- died N 00} [a] ra & (©) 6) ur) no) e {oe} zZ en = DIDIO XAUAd NID agı % B > % 00 soo = = = = M S AUAd QDIO agl III0 NED »% s00 s00 x % % x > % x 13H na NID 2 DIO Ad TIDIO agı s00 % % x x % 5: 3 3 x ai z DIDO TIDI0 Aid NaD adgı 3% 3 s00 00 00 > 3 3 > 2% Td9H = DIO agı Ad IDI0 = = x — 5 > = % soo soo HI an Agı DO Ad TIDDO - = s0'0 = 5 x = s00 3 s00 IN [09] > aal YAUAd ZDDIO TDIO - = = = 2 % = s00 s00 s00 Ha N 2 YAd Do Agı 1990 - = soo = 3% 3 = 3 00 s00 MOI 2 dal DIOr ME 1990 = = 3 — % > = 00 s00 00 Mä DIO dgl Ad IDI0 = = % z 3 soo = s00 B s00 eN-;1 agı DO Ad TIDO = — > = 3 x = s00 00 soo Bt:o) NH NH) 441 NY 441 UAd AAd DIO 9ZIS JsoySTy 0) JOUIW 4gl Ad TAU DO IIO TIIO -IIDO -IIIO -III0 -IIIO ‘(S ueyJ sso[ 9zıs afdures) no PaLLIeI Jou Sem JS9L, = — 'SSIUSIOJFIP JuesIpugIs ON = „ 'SSIUSIOYFIP Juesipugis JIe S19yJ USyUM P9J0U SI 10119 Jo Ayıpıgego1d sy] 'uonepndod fenuss :NII :uonepndod uerniag] :ggI ‘uonerndod uesus1kgq :NAd {uonerndod z [eIusPIIO :DIO e :uonejndod [ [eJuspI>IO :TIIO "YPYIS Jo poyyou ayJ 0) ZurpIoase pasAfeue AfgereA Y9e9 10] sopdwes Jo sıred usOMISq suosLiedwo9 fenprarpuf °E OIqeL Morphological and colour variation in Galemys pyrenaicus 7 Frequency (number of individuals) -28 24 20 -16 -1.2 -08 -04 -0.0 04 08 12 16 20 24 28 32 36 40 44 48 52 56 Discriminant Function EN OCCIDENTAL 1 [LJOCCIDENTAL2 BE PYRENEAN BI IBERIAN Fig. 2. Distribution of the sample analysed according to the discriminating function considering five craniometric sizes (CBL, T'-M°, BW, IOW, ML) (ML = 0.86; I'-M° = 0.46; IOW = -0.33; CBL = -0.04) and the correlation coefficients be- tween the variables and the function (ML = 0.89; T'-M’ = 0.54: IOW = 0.01; CBL = 0.57) highlight I'-M° and ML as the variables with the greatest discriminating power. Although their significance reaches a value which is high enough to warrant attention (Lambda de Wilks = 0.623, p< 0.001), it should be noted that only 61.6% of the sample (Fig. 2) is ac- curately classified. According to the data obtained, there is a wide overlap between the groups under comparison. The ordination of the centroid values of the groups situates the Oceidental 1 (-0.61) and Pyrenean (-0.45) populations on the negative side of the func- tion and the Occidental 2 (0.77) and Iberian (1.62) populations on the positive side. Since I-M° and ML have a positive correlation with the function, this ordination highlights the smaller skull size of the animals from Occidental 1 and Pyrenean populations. A second discriminating analysis was carried out, adding the body parameters HFL, HBL, and TL to the variables used in the first analysis. In this case only the animals from the Occidental 2, Iberian, and Pyrenean populations presented complete information of the parameters under consideration. The discriminating function which explains a higher percentage of variance (89.6%) makes a selection of seven variables: D = (-0.81) x CBL + (-1.57)xT-M° + (2.07)xIOW + (-1.41)xML + (0.91)xHFL + (0.04)xTL + (0.05)xHBL + 29.04. The standardized coefficients of this function (ML = -0.56; 1 M° = 0.43; ITOW = 0.45: CBL = - 0.44; HFL = 0.78; TL = 0.23; HBL = 0.29) and the corre- lation coefficients among the variables as well as the function (ML = -0.41; IT’-M° = -0.33; IOW = 0.18; CBL = -0.25; HFL = 0.36; TL = 0.21; HBL = 0.30) point to T-M°, ML, CBL, IOW, and HFL as the parameters with the greatest discriminating power. 8 J. GONZALES-ESTEBAN, E. CASTIEN, and J. GOSALBEZ Frequency (number of individuals) A E 15 _ ———- 12 16 2.0 L]OCCIDENTAL2 BIPYRENEAN BB IBERIAN 4.0 36 -32 28 24 20 -.6 -1.2 -08 -04 -0.0 04 08 24 28 32 36 Discriminant Function Fig. 3. Distribution of the sample analysed according to the discriminating function considering five craniometric sizes (CBL, T-M3, BW, IOW, ML) plus three body sizes (HFL, HBL, TL) In this case the efficiency of the function obtained is relatively high since it provides an accurate classification percentage, 87.3% (Fig. 3). The ordination of the centroid va- lues of the groups places the Iberian population on the negative side of the function (-2.82) and the Pyrenean population on the positive side (1.25), while the Occidental 2 population is intermediate in location (-0.44). Since the selected skull lengths correlate negatively with the function and IOW and HFL have a positive correlation, this ordina- tion assigns a shorter and stronger skull and longer foot to the Pyrenean population as compared to the Iberian population. Fur coloration Based on an examination of all the pelts taken from the Quinto Real (Navarra), no dif- ferences in coloration were noted among the three relative age classes, if we compare ani- mals caught during the same season. However, differences were seen upon comparing an- imals of the same age caught during different seasons. This is due to the wear the guard hairs are subject to (in the desman there are two types of guard hairs, the most common being the ones PoDuscHkKA and RıcHARD (1985) call “Grannen”). After examining these hairs under the light microscope, the distal ends were seen to be intact and the coloration at first glance appears to be uniformly dark brown in young animals that have just grown a coat of fur or in adults that have recently moulted (animals caught between July and September). In contrast, in animals caught between January and March there are fine yel- lowish spots on the back. These spots are the distal ends of the “Grannen”. With the light Morphological and colour variation in Galemys pyrenaicus 9 JAN FEB_MAR APR _ MAY JUN JUL AUG SEP _OCT NOV DEC YOUNG ADULTS OLD ADULTS a 88 YOUNG es ale Fig. 4. Distribution of the area mottled in the dorsal fur. The darkest zones represent a greater yellow- ish spot density. The studied individuals appear distributed in function of their relative age and their month of capture microscope it can be seen that these ends are split and appear dark and soıled. This worn fur has a yellowish tint at first glance, which corresponds to the spots seen in the old, worn fur. The notion that this variation in coloration is due to the deterioration of the fur was corroborated, when the group of pelts was arranged by months of capture and age (Fig. 4). The distribution of the dorsal spots and the temporal sequence observed would suggest that the moulting of the dorsal fur follows a regular pattern and is completed in August and September. The coloration of the ventral part has a similar evolution. In June-August when the animal has new fur, the ventral part is white in colour; as the fur gets older it starts turn- ing yellow (with gold spots). No differences were seen between the sexes in terms of fur coloration. On comparing fur coloration between the established groups of animals, the colour of the ventral part did not show any substantial differences. It could be described as a shiny greyish white. The colour of the dorsal fur, however, differed among specimens of the dif- ferent groups. The animals from the Pyrenean population have a dark brown colouring, which in some cases is almost black. The animals from the Iberian population have a simi- lar dorsal colouring, although slightly lighter. The animals belonging to the Occidental 2 population have a reddish dorsal colouring, which is considerably lighter and they are the only specimens having yellow forepaws, feet, and nails. The latter fits the description that GRAELLS (1897) gave of the subspecies rufulus. The animals from the Occidental 1 popula- tion have a light coloration similar to those of Occidental 2 with the exception of one speci- men which is slightly darker. In specimens having dry fur it was not possible to detect dif- ferences in shade between the worn fur of the different groups studied. Discussion Based on the data analysed it is difficult to find structured patterns in relation to skull size. The craniometric values of the animals from the Pyrenees, described as pyrenaicus, 10 J. GONZALES-ESTEBAN, E. CASTIEN, and J. GOSÄLBEZ are intermediate between the specimens from Galicia and the Iberian range, both de- scribed as rufulus, which would discredit the opinion of MıLLEr (1912) and CABRERA (1914) who based the discrimination of the subspecies on size. The fur coloration does, however, show geographic differences, which would contra- dict, in this case, CABRERA’s (1914) opinion. The animals from the Pyrenees, Basque Country and the Iberian Range are dark brown (blackish) in colour, while the animals from the occidental area (Leön, Asturias, Galicia) belong to the light brown (reddish) type. The spots in the fur alluded to by GrAELLS (1897) are due to the deterioration of the fur. In its distribution area Galemys pyrenaicus presents a morphological variability that does not allow a clear differentiation to be made between the two subspecies described. It is only the coloration that fits a general geographic pattern, distinguishing the speci- mens from the Pyrenees and Iberian Range from the remainder, although within each of the two groups there is still a certain degree of heterogeneity. The morphometric and col- ouring observations obtained do not support the distributions of the two subspecies as de- scribed in the bibliography. Thus, the distribution of the population variations would be more complicated than the simple description of the two known subspecies. The desman is a species linked to the high regions of the rivers (CAstıEn and GosALBEZ 1992). The dispersive characteristics of this species have not been examined to date, but dispersion is certain to be basically re- lated to river courses. This would lead us to state that total or partial isolation between the populations of this species is a common occurrence. Moreover, if we add the territo- rial character of these anımals and their low density (Stone 1987), it would be possible to consider that genetically related phenomena occur. The existence of these phenomena may explain the presence of local variations throughout the distribution area of the spe- cies. An in-depth genetic study would make it possible to establish the taxonomie impor- tance of these varıations in the future. Acknowledgements The authors would like to thank PABLO AGIRRE-MENDI, GORKA BELAMENDIA (Museo de Ciencias Nat- urales de Alava), Estaciön Biolögica de Donana (Sevilla), Museo Luis Iglesias (Santiago de Compos- tela) and Dr. Francısco J. Purroy (Facultad de Biologia de Leön) for kindly providing the material studied. The authors are also grateful to I. LERAnoZz for her generous help in the collection of our own material. This study was partially funded by the Secciö de Ciencies Biologiques de l’Institut d’Es- tudis Catalans (Barcelona). Zusammenfassung Morphologische und Färbungsvariationen beim Pyrenäen-Desman Galemys pyrenaicus (Geoffroy, 1811) Die Arbeit enthält Angaben über Biometrie und Färbung des Pyrenäen-Desman Galemys pyrenaicus. Die Autoren untersuchen die geographische Variation der Größe und Färbung dieser Art innerhalb der Iberischen Halbinsel und kommentieren den derzeitigen taxonomischen Wissensstand. Das unter- suchte Material bezieht sich auf alle Regionen der Iberischen Halbinsel, in denen diese Art vor- kommt. Die Ergebnisse dieser Analyse zeigten in bezug äuf die Schädelgröße keine bedeutenden geo- graphischen Unterschiede. Sowohl die Schädelgröße als auch einige andere Körpermaße der in den Pyrenäen lebenden Unterart pyrenaicus liegen zwischen den Maßen der in Galicien und dem Ibe- rischen Gebirge lebenden Unterart rufulus. Dieses Ergebnis stellt die Heranziehung der Schädelgröße als Maßstab für die Unterscheidung von Unterarten in Zweifel. Eindeutigere Unterscheidungsfakto- ren bestehen bezüglich der Farbe, allerdings stimmen hierbei die chorologischen Modelle nicht mit Morphological and colour variation in Galemys pyrenaicus 11 denjenigen anderer Autoren überein. Die vorliegende Studie legt nahe, daß die einzelnen Populatio- nen von Galemys pyrenaicus zum Teil oder gänzlich isoliert sind, wodurch eine morphologische Un- terscheidung begünstig sein kann. Diese Tatsache hebt die Notwendigkeit der Erhaltung jedes einzel- nen dieser Siedlungskerne hervor. References CABRERA, A. (1914): Fauna Iberica. Mamiferos. Mus. Nac. Cien. Nat. Madrid. CASTIEN, E.; GOSALBEZ, J. (1992): Distribuciön geogräfica y häbitats ocupados por Galemys pyrenaicus (Geoffroy, 1811) Insectivora: Talpidae) en los Pirineos Occidentales. DoAana, Acta Vertebrata 19, 3744. GEOFFROY, E. (1811): Memoire sur les especes des genres Musaraigne et Mygale. Ann. Mus. Hist. Nat. 17, 169-194. GRAELLS, M. P. (1897): Fauna Mastodolögica Iberica. Mem. R. Acad. Esp. Cien. Exact. Fis. Nat. 23, 1- 806. GRULICH, 1. (1967): Zur Methodik der Altersbestimmung des Maulwurfs, Talpa europaea L. in der Peri- ode seiner selbständigen Lebensweise. Zool. Listy 16, 41-59. JUCKWER, E. A. (1990): Galemys pyrenaicus (Geoffroy, 1811) — Pyrenäen-Desman. In: Handbuch der Säugetiere Europas. Vol. 3/1. Ed. by J. NIETHAMMER and F. Krapp. Wiesbaden: Akademische Ver- lagsges. Pp. 79-92. MILLER, G. S. (1912): Catalogue of the mammals of western Europe (Europe exclusive of Russia). Lon- don: Brit. Mus. Nie, N. H.; Hurt, C. H.; JENKINS, J. G.; STEINBERGER, K.; BENT, D. M. (1975): Statistical package for the social sciences SPSS. New York: McGraw-Hill. NIETHAMMER, J. (1970): Beobachtungen am Pyrenäen-Desman, Galemys pyrenaica. Bonn. Zool. Beitr. 21, 157-182. PALMEIRIN, J. M.; HOFFMAN, R. S. (1983): Galemys pyrenaicus. Mammalian Species 207, 1-5. PoDUSCHKA, W.; RICHARD, B. (1985): Hair types in fur of Pyrenean Desman (Galemys pyrenaicus) Geof- froy, 1811 (Insectivora: Talpidae: Desmaninae). Sitzungsber. Akad. Wiss. Wien 194, 39-44. STONE, R.D. (1987): The social ecology of the Pyrenean desman (Galemys pyrenaicus) (Insectivora: Talpidae), as determined under natural conditions. J. Zool. (London) 213, 95-106. Authors’ addresses: JORGE GONZÄALEZ, Soc. Ciencias Aranzadi, Elkano Bidea, 31, E-20012 Donostia- San Sebastian, ENRIQUE CASTIEN, Servicio de Conservaciön de la Naturaleza, Go- bierno de Navarra, C./Alhöndiga 1, E-31002 Pamplona, and JoAQUIM GOSALBEZ, Departament de Biologia Animal, Universitat de Barcelona, Avgda. Diagonal 645, E-08028 Barcelona, Spain Z. Säugetierkunde 64 (1999) 12-18 © 1999 Urban & Fischer Verlag INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Eine spezielle Markierungsweise und ihre strukturelle Grundlage beim Krabbenwaschbär (Procyon cancrivorus) Von BirGit PoHL und W. MEYER Institut für Zoologie und Anatomisches Institut, Tierärztliche Hochschule Hannover, Hannover Eingang des Ms. 30. 03. 1998 Annahme des Ms. 23. 07. 1998 Abstract A specific type of scent marking and its structural basis in the Crab-eating raccoon (Procyon cancrivorus) The study describes for the first time rubbing of the nuchal region as a normal form of behaviour as well as a frequently applied territorial scent marking in the crab-eating raccoon (Procyon cancrivorus). Different objects marked with a fatty and odorous secretion by typical rotating or vertical movements of the head were olfactorily controlled by other members of the group studied. Concerning the fur, in contrast to the normal backwards orientated hairs at the dorsum those in the nuchal region with their tips pointed toward the head or straight upward. A histological analysis of the nuchal integument exhib- ited, as compared to the dorsal body region, a thickening of the cutis, distinctly enlarged sebaceous glands and apocrine tubular glands of central primary hair follicles producing the heavy guard hairs. These hair follicles were increased in length but not in thickness. The possible biological function of this “glandular organ” of the nuchal region in the crab-eating raccoon is discussed. Key words: Procyon cancrivorus, nuchal integument, scent marking Einleitung Innerhalb der Familie der Neuwelt-Kleinbären (Procyonidae) ist der nordamerikanische Waschbär (Procyon lotor) die wohl am häufigsten untersuchte Spezies. Weitaus geringer sind die Kenntnisse zur Biologie des engsten Verwandten, des südamerikanischen Krab- benwaschbären (Procyon cancrivorus). Nur wenige Arbeiten, wie z.B. diejenige von YA- NOSKY und MERcOoLLI (1993), die den jahreszeitlichen Aktivitätsrhythmus von Krabben- waschbären im EI Bagual Ecological Reserve von Nord-Argentinien darstellt, geben dabei Auskunft über das Verhalten dieser Tiere. Bedingt durch die dämmerungs- und nachtaktive Lebensweise der Krabbenwaschbären lassen sich jedoch differenzierte etho- logische Studien im Freiland nur in begrenztem Umfang durchführen. Aus diesem Grund wurden in der vorliegenden Arbeit die Verhaltensmuster von in Gehegehaltung lebenden Tieren beider Geschlechter näher analysiert. Dabei konnte eine für Krabben- waschbären bisher noch nicht beschriebene Verhaltensweise des Markierens gefunden werden, für deren strukturelles Korrelat im Integument ebenfalls noch keine Informatio- nen vorlagen. 0044-3468/99/64/01 —- 012 $ 12.00/0 Eine spezielle Markierungsweise beim Krabbenwaschbär 13 Material und Methode Von sechs im Gehege des Instituts für Zoologie der Tierärztlichen Hochschule Hannover gehaltenen subadulten und adulten Krabbenwaschbären (Procyon cancrivorus nigripes Mivart, 1886) (13; Alter 9 Jahre; 5 29; Alter 2, 6, 11 Jahre; Leihgaben des Zoo Säo Paulo, Brasilien) wurde das Verhalten über einen Zeitraum von 14 Monaten beobachtet, wobei die „one-zero-sampling“ Methode zur Anwendung kam. Für jedes einzelne Tier wurde dabei monatlich ein 24 h-Tag protokolliert (MARTIN und BATESON 1992): Von einem weiblichen (3 Jahre) und einem männlichen Tier (11 Jahre), die durch Krankheit be- dingt (Bißverletzung, Tumorerkrankung) euthanasiert werden mußten, war es möglich, mehrere Haut- proben aus der Nackenregion sowie dem gesamten Rückenbereich zu entnehmen. Das Material wurde in Bouinscher Lösung fixiert und nach Entwässerung über eine aufsteigende Ethanol-Reihe in den Kunststoff Technovit 7100 (Fa. Kulzer) eingebettet (GERRITS und SMID 1983). Dieses Glykolmethakry- lat-Gemisch verursacht keine Schrumpfungsartefakte (HANSTEDE und GERRITS 1983) und läßt sich da- her gut für Vermessungen am histologischen Schnitt verwenden. Nach der Herstellung von 3 um dicken Schnitten am Autocut-Mikrotom (Fa. Reichert-Jung) wurden diese mit Hämatoxylin (Hämalaun nach Delafield) - Eosin oder 1%igem Toluidinblau (nach RıcHArDson et al. 1960) gefärbt. Vermessungen verschiedener Hautanteile erfolgten an je fünf bis acht histologischen Schnitten von je sechs Proben aus dem Nacken und je sechs bis acht Proben aus der Rückenregion mit Hilfe eines stan- dardisierten Zeichengeräts (Fa. Zeiss). Die Mittelwerte der Meßergebnisse zur Hautdicke sowie von 10 bis 15 zentralen Primärhaarfollikeln und ihren Anhangsdrüsen je Hautprobe wurden auf Normalvertei- lung und mit Hilfe des t-Tests nach Student anschließend auf signifikante Unterschiede überprüft. Ergebnisse Im Rahmen ihres Territorialverhaltens markierten die Krabbenwaschbären im Gehege regelmäßig mit Hilfe des Analbeutelsekrets und durch Abgabe kleinerer Harnmengen. Auffallend war zusätzlich aber häufig ein Reiben der Nackenregion (Regio nuchalis) an Baumstämmen, Ästen, Steinen, Wassertrögen und dergleichen. Dabei senkten die Tiere den Kopf und drückten den Nackenbereich an eine markante Stelle dieser Objekte (Abb. 1) oder ergriffen, aufrecht stehend, mit einer Vorderpfote einen dünnen Zweig und Abb.1. Typische Stellung von Krabbenwaschbären (Procyon cancrivorus) beim Markieren mit der Nackenregion. 14 BirGIt PoHL und W. MEYER Tabelle 1. Vergleichende Messungen an Hautstrukturen zweier Körperregionen des Krabbenwaschbären. Hautdicke [mm] 2.o #201) 4,1(+0,4) (Epidermis, Dermis) Zentrales Primärhaar - Länge [mm] 16,8 (+ 1,8) Dicke [um] -Spitze 12,929) Mitte 101,193) Basis 68,1. EER1or2)) Zentraler Primärhaarfollikel - Länge [mm] 3,3 (201) 3.2 (202) 150 (+ 23) Einsenktiefe [mm] Dicke [um] Talgdrüsen-Komplex - Länge [um] 446 (+ 100) 129 (+ 30) 989 (+ 253) größter Durchmesser [um] 281 (+ 69) Apokriner Schlauchdrüsen- Komplex - Länge [mm] 1,2 (#06) 2,4 (30,3) Durchmesser-Endstück [um]]| 116 (+ 22) TH ATEEIS)) Epithelhöhe-Endstück [um] | 26 (+ 3,7) 34 (+ 3,8) preßten diesen an den Nacken. Durch kreisende oder vertikale Bewegungen des Kopfes wurde dann die Nackenregion am Gegenstand gerieben. Dieses Verhalten zeigten Krab- benwaschbären beider Geschlechter über den gesamten jahreszeitlichen Verlauf. Es hatte im Vergleich mit den anderen Markierungsaktivitäten einen durchschnittlichen Anteil von 17%. Gemeinsam in einem Gehege gehaltene Tiere überprüften gegenseitig die durch Nackenreiben markierten Stellen und kontrollierten neben der Analregion ab und an auch die Nackenregion des Artgenossen olfaktorisch. Die genaue Betrachtung der Nackenregion zeigte, daß die Haare in diesem Bereich, entgegen dem normalen Verlauf des Haarstrichs, vom Nacken bis zu den Ohren kopf- Eine spezielle Markierungsweise beim Krabbenwaschbär 15 wärts gerichtet waren (Abb. 2). Darüber hinaus wiesen Nackenhaut und basale Anteile der Haare meist einen fettigen, bräunlichen Belag auf, von dem ein intensiver Geruch ausging. Gegenstände oder Teile von Objekten, die von den Krabbenwaschbären mehr- mals durch Nackenreiben markiert wurden, waren gleichfalls mit diesem bräunlichen Sekret sowie dem typischen Geruch behaftet. Die histologische Analyse machte offenkundig, auf welcher strukturellen Basis sich die Produktion des Markierungssekrets im Integument gestaltete. Sehr deutlich und si- Abb. 2. Nackenregion und Hals mit kopfwärts gerichteter Stellung der Haare. Abb. 3. Übersicht einer großen Talgdrüse der Nackenhaut mit einem zusätzlichen, oberhalb des Aus- führungsganges (Pfeil) gelegenen Anteil (H.E. Färbung, Maßstab entspricht 200 um). Abb. 4. Sekretorisches Endstück einer apokrinen Schlauchdrüse der Nackenhaut, mit stark entwickel- ten Myoepithelzellen (Pfeil), neben einem Haarfollikel (HF) (H.E. Färbung, Maßstab entspricht 50 um). 16 Birsıt PoHL und W. MEYER gnifkant waren dabei speziell die Unterschiede (Tab. 1) in der Größe bzw. Dicke und Länge der beteiligten Strukturen des Haarfollikelkomplexes im Vergleich der Nackenre- gion mit der als Kontrolle verwendeten gesamten Rückenregion (hier finden sich bei Säu- getieren in der Regel die größten Hautdrüsen der allgemeinen Körperdecke, vgl. z.B. So- KOoLOV 1982; MEYER 1986, 1998; SCHWARZ und MEYER 1994). Zum ersten zeigte sich die Nackenhaut (Epidermis und Dermis) signifikant (p < 0,01) als fast doppelt so dick wie die Rückenhaut; zum zweiten offenbarten nicht alle Typen der Haarfollikel, sondern nur die zentralen Primärhaarfollikel und ihre Anhangsdrüsen signifikante Unterschiede (p < 0,01) zwischen beiden untersuchten Körperanteilen. Außerdem kamen im Nacken zwar keine dickeren, so doch kürzere Haare (Telogen- und ausgewachsene Anagenhaare) als am Rücken vor. Und dies, obwohl die dazugehörigen zentralen Primärhaarfollikel beinahe die doppelte Länge aufwiesen. Besonders auffällig waren die Meßergebnisse von den An- hangsdrüsen, wobei sich im Integument des Nackens mächtige Talgdrüsen entwickelt hat- ten (Abb. 3), die mehr als doppelt so lang und dick wie an dem entsprechenden Haarfolli- keltyp des Rückens hervortraten. Der gesamte sekretorische Anteil der jedem Primärhaarfollikel zugehörigen apokrinen Schlauchdrüse war am Nacken ebenfalls gut doppelt so lang wie am Rücken. Die spezifische histologische Struktur des sekretorischen Endstücks blieb dagegen im Durchmesser des gesamten Endstücks wie in der Epithel- höhe unverändert. Auffallend waren aber die im Nackenbereich immer sehr kräftig aus- gebildeten Myoepithelzellen der apokrinen Schlauchdrüsen (Abb. 4). Diskussion Die hier vorgelegten Befunde zum Nackenreiben und zur Struktur der Nackenhaut und ihrer Haarfollikelanhangsdrüsen weisen eindeutig auf die spezielle Funktion der Nak- kenregion des Krabbenwaschbären als Drüsenorgan hin. In diesem Zusammenhang muß keine besonders auffällige strukturelle Veränderung des Integuments vorausgesetzt wer- den. Eine erhebliche Zunahme der Größe bzw. Volumina der Anhangsdrüsen von Pri- märhaarfollikeln bei fehlender relativer Größenzunahme des Haarfollikels selbst ist z.B. sehr klar im Integument der Anogenitalregion des Pferdes erkennbar (TsukIısE und MEYER 1987, MEYER 1998) oder in der Skrotalhaut des Makaken (Macaca cyclopis) (MEYER und Tsukise 1989). Auch das Reiben der Hals- und Nackenregion an Bäumen und Gebüsch, wie zur Territorialmarkierung beim Reh (Capreolus capreolus) beobach- tet, konnte nicht mit außerordentlichen Abwandlungen der Hautstruktur dieser Körper- teile korreliert werden (SoKoLov 1982). Nackendrüsen, die fettige, duftende Flüssigkei- ten sezernieren und durch Haarbüschel gekennzeichnet sind, scheinen jedoch bei Fledermäusen (spez. Pteropodidae) vorzuliegen, obwohl eine histologische Verifizierung des Drüsenorgancharakters unseres Wissens nach noch fehlt (SCHAFFER 1940; QuAy 1970). Als Sonderfall seien vergrößerte apokrine Schlauchdrüsen in der Nackenregion junger nordamerikanischer Nerze (Mustela vison) erwähnt, die allerdings nur in der Saugperiode vorhanden sind und danach angeblich degenerieren. Eventuell helfen diese Drüsen der Mutter, ihre Nachkommen besser olfaktorisch zu differenzieren (YAGER et al. 1988). Von entscheidender Wichtigkeit ist im vorliegenden Fall eine — wie bei typischen Drü- senorganen — regional und zeitlich separat anzusteuernde Aktivität der Anhangsdrüsen der großen Primärhaarfollikel zur Produktion einer großen Menge an guthaftendem, fett- reichem Sekret, dessen bakterielle Zersetzung ein offenbar individual spezifisches Duftmuster hervorbringt. Dies gilt im besonderen unter der oben beschriebenen Beob- achtung, daß nicht nur die markierten Objekte olfaktorisch überprüft wurden, sondern gelegentlich die Nackenregion der jeweiligen Gruppenmitglieder miteinbezogen war. Das typische Verhaltensmuster des Nackenreibens an Gegenständen ist von KAMPMANN Eine spezielle Markierungsweise beim Krabbenwaschbär 197, (1972) und LÖHMER (1973) auch bei in Gehegehaltung lebenden nordamerikanischen Waschbären gefunden worden. Diesen Autoren zufolge trat es bei Procyon lotor aller- dings vornehmlich während der Fortpflanzungsperiode der Tiere auf und wurde vor allem von männlichen Waschbären ausgeführt. Beim nordamerikanischen Waschbären zeigt sich - im Unterschied zum Krabbenwaschbär - im Bereich des Nackens zudem keine dem normalen Haarstrich entgegenlaufende Ausrichtung der Haare. Welche spezielle bio- logische Bedeutung das beim Krabbenwaschbären sehr ausgeprägte Markieren mit Hilfe der Nackenregion wirklich hat, kann nur vermutet werden. Eventuell deutet es auf ein stärker ausgeprägtes Territorialverhalten als bei Procyon lotor hin. Unterstützt wird diese Annahme durch die beobachtete höhere innerartliche Aggressionsbereitschaft der Krab- benwaschbären. Desweiteren besteht die Möglichkeit, daß die mit einem individualspezi- fischen Geruch behaftete Nackenregion eine Rolle bei der Auswahl des Sexualpartners spielt. Danksagung Wir danken Frau CLAUDIA BÖDEKER und Frau INES BLUME für ihre wertvolle technische Assistenz. Zusammenfassung Die Arbeit beschreibt zum ersten Mal das Reiben der Nackenregion an Gegenständen als regelmäßige und häufig angewendete territoriale Markierungsweise bei im Gehege gehaltenen Krabbenwaschbären (Procyon cancrivorus nigripes). Der Nacken der Tiere wies eine entgegen dem normalen Haarstrich verlaufende Ausrichtung der Haare auf. Die mittels typischer Kreisender oder vertikaler Kopfbewegun- gen mit einem fettigen und duftenden Sekret markierten Objekte wurden von Artgenossen olfakto- risch kontrolliert. Eine histologische Analyse des Integuments der Nackenregion zeigte im Vergleich zum gesamten Rückenbereich eine dickere Cutis und erheblich größere Komplexe der Talgdrüsen und apokrinen Schlauchdrüsen von zentralen Primärhaarfollikeln, die in Dicke und Länge allerdings denje- nigen im Rückenbereich entsprachen. Die mögliche biologische Bedeutung dieses „Nackendrüsen- organs“ beim Krabbenwaschbär wird diskutiert. Literatur GERRITS, P. O.; SMiD, L. (1983): A new, less toxic polymerization system for the embedding of soft tis- sues in glycol methacrylate and subsequent preparing of serial sections. J. Microsc. 132, 81-85. HANSTEDE, J. G.; GERRITS, P. ©. (1983): The effect of embedding in water soluble plastics on the final di- mensions of liver sections. J. Microsc. 131, 79-86. KAMPMANnNN, H. (1972): Der Waschbär in Deutschland. Diss. Univ. Göttingen. LÖHMER, R. (1973): Vergleichende ethologische und sinnesphysiologische Untersuchungen an Wasch- bären und Krabbenwaschbären. Diss. Tech. Univ. Hannover. MEYER, W. (1986): Die Haut des Schweines. Hannover: Schlütersche Verlagsanstalt. MEYER, W. (1998): Haut und Hautorgane. In: Praxisorientierte Anatomie des Pferdes. Hrsg. von H. WıssDoRF, W. GERHARDS und B. HuUsKAMP. Hannover: Schaper Verlag. Pp. 19-48. MEYER,W.; TsSuKISE, A. (1989): Histochemistry of complex carbohydrates in the scrotal skin of the mon- key, Macaca cyclopis (Swinhoe). Z. Säugetierkunde 54, 9-21. MARTIN, P.; BATEson, P. (1992): Measuring behaviour —- An Introductory Guide. Cambridge, New York: Cambridge Univ. Press. Quay, W. B. (1970): Integument and derivatives. In: Biology of Bats. Vol. 2. Ed. by W. A. Wımsatt. New York, London: Academic Press. Pp. 1-56. RICHARDSoN, K. C.; JARETT, L.; FINKE, E. H. (1960): Embedding in epoxy resins for ultrathin sections in electron microscopy. Stain Technol. 35, 313-323. SCHAFFER, J. (1940): Die Hautdrüsenorgane der Säugetiere. Berlin, Wien: Urban und Schwarzenberg. 18 Birsıt PoHL und W. MEYER SCHWARZ, R.; MEYER, W. (1994): Haut und Hautanhangsorgane. In: Die Anatomie von Hund und Katze. Hrsg. von J. FREWEIN und B. VOLLMERHAUS. Berlin. Hamburg: P. Parey Verlag. Pp. 316-340. SOKOLoVv, V. E. (1982): Mammal Skin. Berkeley, Los Angeles, London: Univ. California Press. TSUKISE, A.; MEYER, W. (1987): Histochemical analysis of carbohydrates in the scrotal skin of the horse, with special reference to glandular appendages. Zool. Anz. 219, 129-140. YAGER, J. A.; HUNTER, D. B.; Wırson, M.R.; ALLEn, O.B. (1988): A source of cutaneous maternal semiochemicals in the mink? Experientia 44, 79-81. YANOSKY, A. A.; MERCOLLI, C. (1993): Activity pattern of Procyon cancrivorus (Carnivora: Procyoni- dae) in Argentina. Rev. Biol. Trop. 4, 157-159. Anschrift der Verf.: Dipl.-Biol. BırsIt PoHL, Institut für Zoologie, Tierärztliche Hochschule Hannover, Bünteweg 17, D-30559 Hannover; Prof. Dr. WILFRIED MEYER, Anatomisches Insti- tut, Tierärztliche Hochschule Hannover, Bischofsholer Damm 15, D-30173 Han- nover Z. Säugetierkunde 64 (1999) 19-29 = x © 1999 Urban & Fischer Verlag SÄUGETIE RKÜ NDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Daylight behaviour of Humpback dolphins Sousa chinensis in Algoa Bay, South Africa By L. KARCZMARSKI and V. G. COCKCROFT Centre for Dolphin Studies, Port Elizabeth Museum and Department of Zoology, University of Port Elizabeth, South Africa Receipt of Ms. 17. 12. 1997 Acceptance of Ms. 19. 08. 1998 Abstract Data on the daylight behaviour of humpback dolphins Sousa chinensis were collected during sea- and land-based surveys undertaken in Algoa Bay, Eastern Cape, South Africa, throughout a three year pe- riod. Dolphin activities/behaviour were categorised as: “feeding”, “travelling”, “opportunistic feeding”, “socialising and playing”, “resting” and “other”. It seems apparent that behaviour determines the spa- tial geometry of the dolphin group, but not the group size. The surfacing-breathing interval is similar for “feeding”, “opportunistic feeding”, and slow “travelling”, but differs considerably from the pattern displayed during fast “travelling”. Daylight behaviour of humpback dolphins is dominated by “feeding” and shows a regular pattern which is probably governed by the diurnal cycles of their prey. Generally, “feeding” peaks in the morning and gradually decreases through the rest of the day. As “feeding” de- creases, “travelling” and “opportunistic feeding” increase, both peaking in the afternoon. “Resting” and “socialising and playing” occur with similar frequency throughout the day. This pattern varies little between summer and winter, as does the overall proportion of daylight behaviours. The only significant seasonal difference is in the frequency of “social/sexual” behaviour which peaks in summer. Although tidal cycle influences to some extent the behaviour of humpback dolphins, in Algoa Bay their daylight activity/behaviour is predominantly governed by time of day. „6 Key words: Sousa chinensis, daylight behaviour pattern, seasonal variation. Introduction Activity rhythms of animals represent an adaptation to seasonal and diurnal variations of environmental factors and are a result of a complex compromise between optimal for- aging/feeding time, social activities, and environmental constraints (CLOUDSLEY-THOMPSON 1961; NıiELSEn 1983). Diurnal and seasonal patterns of activity and behaviour have been described in detail for several terrestrial mammals. There is, however, a disproportionate lack of similar information concerning cetaceans (for review see KLINOWSKA 1986; SHANE et al. 1986). One reason for this lies in the practical difficulty of studying the behaviour of free-ranging cetaceans. Humpback dolphins Sousa chinensis inhabit Indo-Pacific coastal waters and are known to occur along the east and south coast of South Africa (Ross et al. 1994). De- spite its inshore occurrence, there has been little study of this species and much of our knowledge is based on fragmentary information. Only recently has the natural history of humpback dolphins been investigated in detail in the Algoa Bay region on the south Eastern Cape coast of South Africa (KARCZMARSKI 1996). This long term study, although 0044-3468/99/64/01 — 019 $ 12.00/0 20 L. KARCZMARSKI and V. G. COCKCROFT not strietly behavioural in its design, included many hours of observations and provided a good opportunity to collect observational data on dolphin behaviour. It was possible to quantify the daylight activity pattern of humpback dolphins and examine the daytime, seasonal, and tidal varıations in their behaviour. Not only are these observations impor- tant in themselves, but they provide a valuable insight into the daily lives of this coastal dolphin. Furthermore, a better understanding of the ecological determinants of hump- back dolphin behaviour may help in the development of appropriate protective measures for this little known and apparently threatened (KLınowsKA 1991) species. Material and methods Algoa Bay is the easternmost and largest of several shallow (mean depth <50 m), log spiral bays found on the south-east coast of South Africa (Fig. 1). The Bay, flanked on the western side by Cape Recife (34°02’S; 25°42’E) and on the eastern side by the less prominent Cape Padrone (33°46' S; 26°28' E), is located along a generally exposed coastline and represents an open habitat with few surface geographi- cal boundaries. The behaviour of humpback dolphins was recorded during land- and sea-based surveys undertaken in the south-western part of Algoa Bay between May 1991 and May 1994. Daily land-based surveys usually started 1-2.5 hours after sunrise (weather permitting) and observations of the inshore waters, to approximately 1 km offshore, were carried out from several visually overlapping vantage points. Sea- based surveys were opportunistic, limited by both the presence of dolphins and weather conditions and were conducted using a 3.5 m inflatable boat powered by a 30 HP outboard engine. The activity/behaviour of the focal group of dolphins (sensu ALtMAnn 1974) was usually recorded at the commencement of each sighting and, thereafter, randomly for five minute intervals throughout the survey. For each of the five minute intervals, the lenght of time spent in different behaviours (see below) was estimated in the form of percentage. The raw field data were subsequently grouped into hourly intervals according to the time of day and into four tidal periods (1/2 low, low, 1/2 high, high) in which they were collected. During boat surveys, recording of dolphin behaviour began only after the animals were assumed to have become habituated to the presence of the boat - in most cases at least 30 minutes from the initial sighting and with the boat at a distance of 10-20 meters from the group. It was assumed that this gave the animals time to resume their normal activity (see also AcEvEpo 1991; BALLANcE 1992). Generally, because aggregations of humpback dolphins were small (mean = 7 dolphins, sd = 2.52; KARCZMARSKI 1996; KARCZMARSKI et al. 1998) the whole group was often the focus of observations and it was possible to categorise the behaviour of the group as a whole (the predominant activity of the ma- jority of the group members). Six broad categories of behaviour were distinguished: The first category consisted of frequent and asynchronous dives, in varying directions, in one loca- tion, with an evident lack of directional movement; surfacing and respiration displayed no obvious pat- tern. During this activity dolphins often chased fish and occasionally fish capture was seen. Conse- quently, it is likely that this complex of behaviours represents feeding and is therefore referred to as “feeding”. The second category was characterised by persistent, directional movement, with all group mem- bers diving and surfacing synchronously. Chasing of fish or even social behaviour was extremely uncom- mon during this behaviour and, consequently, it is referred to as “travelling”. The third category appeared to be a combination of the previous two. Dolphins moved slowly, but usually in a fairly consistent direction. However, surfacing and diving were apparently less synchronised than during apparent “travelling”. Furthermore, the directional movement was frequently interrupted by short bouts of localised movement (frequent changes in direction with no evident overall direction), during which some individuals performed long dives. Occasionally chasing of fish was seen. Social activ- ity during this behaviour was uncommon. It was assumed that this pattern of activity represented “opportunistic feeding” and it is referred to as such. The fourth category consisted of various vigorous activities including leaping out of the water, rid- ing waves in the surf zone, high speed movement with frequent direction changes, and prolonged body contact with other dolphins. These were frequently accompanied by prolonged bouts of almost constant physical contact between two or more dolphins, which seemed to have a sexual meaning (see also Daylight behaviour of Humpback dolphins in Algoa Bay, South Africa za rs (1.7 "k---" SOUTH AFRICA Algoa Bay a P& | SEREBE Cape Recife FE B Sardinia ; 0 10 20 and (km : oT CAPE . SLARE: ;;, PADRONE B Sundays River !\ Es REICHEN CoegaRiver ...:.,7 ET Skoenmakerskop ur 25°45'E Fig. 1. The Algoa Bay study area on the south Eastern Cape coast of South Africa. KARCZMARSKI et al. 1997). It seems most likely that these activities served a social function and are re- ferred to as “socialising and playing”. The fifth category consisted of a low level of activity, with the dolphins apparently floating station- ary and motionless at the surface, with some occasional slow forward movement. This is referred to as Sresung. Activities which could not be easily assigned to any of the above categories were termed “other”. No underwater observations were conducted and dolphin behaviour is described as it was observed from the research boat. The spatial geometry of the dolphin group (after SHANE 1990 a), its size, compo- sition, and locality were noted at the commencement of each sighting and when any variable changed. Two seasons. summer and winter, are distinguished here. “Summer” is defined as the period when the mean temperature of the inshore surface water is higher than the annual mean (18°C). The period when the surface water temperature drops below this annual mean is referred to as “winter”. In gener- al, the first days of May mark the beginning of “winter” and late October marks the beginning of “summer”. The term “group” refers to any aggregation of more than one dolphin, including all age classes, within visual range of the survey team. Typically, these animals were in apparent association and en- DD L. KARCZMARSKI and V. G. COCKCROFT gaged in similar activities for most of the observation period. Each time a group was observed, it was re- corded as a “sighting”. The term “sighting”, however, has a wider meaning and includes solitary ani- mals. Results Groups and solitary humpback dolphins were observed 104 times during more than 300 hours of observations. Dolphin behaviour was specifically recorded for a minimum of one hour during 83 sightings, for a total of 270 hours. In most instances, behaviour was re- corded for between three and four hours (mean = 3.2 h), though the longest session ex- ceeded sıx hours. Groups of humpback dolphins varied in size from three to 24 animals with a mean of seven (sd = 2.52). Solitary individuals were seen frequently and constituted 15.4 % (n = 16) of sightings (see also KARCZMARSKI et al. 1998). In most cases the size of groups remained unchanged throughout observations and was not affected by the anımals’ behaviour (Kruskal -— Wallis ANOVA, KW = 39.57, n = 361, p = 0.53). Group geometry, however, was not random but varied according to activity. During “feeding”, the dolphins were usually widely dispersed, with the distances between individuals varying constantly and ranging between approximately Im and at least 100 m. When “travelling”, hump- back dolphins formed a tight “single-file group”, or fairly compact, oval shaped aggrega- tion with the distance between individuals seldom exceeding a body-length of an adult dolphin (circa 2.5 m). The position of individuals within such groups, however, changed continually. During “opportunistic feeding” group geometry was not well defined, with dolphins moving in the same direction, but in a fairly dispersed aggregation; less dis- persed, however, than during “feeding” (the max. distance between individuals usually <50 m). Similarly, humpback dolphins did not display any consistent group geometry dur- ing “socialising and playing” or “resting”. The distances between individuals changed con- tinually, ranging between a “touching distance” when body contact was performed and circa 25 m. Fig. 2. The mean proportion (%) of daylight hours spent by humpback dolphins in each of six beha- viours (E:) - feeding, && - travelling, — opportunistic feeding, — socialising and playıng, [__)-resting and - other) in Algoa Bay between May 1991 and May 1994. Daylight behaviour of Humpback dolphins in Algoa Bay, South Africa 25 The surfacing-breathing interval did not vary significantly between “feeding” (mean 2 Sr0;secun 0 5s5d2 773), 2. opportunistiesteedineZ (mean B0Alsee., n= 27, sd=11.8) and slow “travelling” (mean = 23.3 sec., n=29, sd=7.1 ); (Kruskal — Wallis ANOVA, KW = 17.44, n=9, p=0.61). During these behaviours humpback dolphins ventilated by rolling at the surface with an overall mean interval of 26.3 sec. (sd = 12.7). This pattern changed considerably during fast “travelling” (n = 27), when several (4 to 9, mean=6, sd=1.7) rapid ventilations separated by only a few seconds (mean = 8.6 sec., sd=4.1) alternated with a long submergence of mean duration = 101.3 sec. (sd = 20.9), during which a long distance was travelled at high speed. The daylıght activity of humpback dolphins in Algoa Bay was dominated by “feed- ing” (Fig. 2). Behaviours classified as “feeding” and “opportunistic feeding”, if combined, contributed well over 50 % of all recorded activities. “Travelling” represented the second most frequently seen behaviour, while “resting” or “socialising and playing” were infre- quent and accounted for less than 10 % of the dolphins’ daylight activities. Humpback dolphins in Algoa Bay displayed little seasonal difference in the propor- tion of diurnal behaviours (Fig. 3). Although dolphins appeared to spend more time “feeding” in winter and, inversely, less time on “opportunistic feeding”, none of these dif- ferences were significant (Mann-Whitney, U = 87.50, n=83, p=0.38 and U = 95.00, n = 83, p= 0.20, respectively). The proportion of combined “feeding” and “opportunistic feeding” was similar for both summer (63.0 %) and winter (64.3 %). The only signifcant seasonal difference was for “socialising and playing” (Mann-Whitney, U = 105.00, n = 83, p = 0.05) (Fig. 3). The summer frequency of “socialising and playing” was double that of winter. The sexual component of thıs behaviour was also significantly greater in summer (47.2 %) than in winter (11.3 %) (Mann-Whitney, U = 240.00, n=65, p< 0.0001). The proportion of daylight hours spent in different behaviours was well defined and varied significantly throughout the day (Fig. 4) for both summer (Kruskal - Wallis ANO- VA, n=181, KW = 57.08, p< 0.0001 for “feeding”: KW = 56.72, p< 0.0001 for “travel- ling”; KW = 25.89, p = 0.007 for “opportunistic feeding”; KW = 35.69, p = 0.0002 for “so- 55 50 45 40 35 30 Percentage 0 EEESEN SR EISEN = 3 Feed Trav OppFeedSoc/Play Rest Othe Categories of behaviour Fig. 3. Seasonal ([__] - summer and — winter) variation in the mean proportion (%) of diurnal be- haviours displayed by humpback dolphins in Algoa Bay between May 1991 and May 1994. Sexual behaviour (ZZ) as a proportion (%) of socialising and playing is also shown. 24 L. KARCZMARSKI and V. G. COCKCROFT [A] SUMMER ZITKIHHHH Percentage of behaviour categories DDR LEERE ORKKREEREREERRERREEKEELERELEKKEHNTL Percentage of behaviour categories 6 7 8 9 10 11 12 13 14 15 16 17 Time of day Fig. 4. The mean hourly proportion (%) of the diurnal behaviours (E37) - feeding, &&% - travelling, - opportunistic feeding, - socialising and playing, [__] - resting and 8 - other) of humpback dolphins observed in Algoa Bay between May 1991 and May 1994 for both summer (A) and winter (B). cialising and playing” and KW = 31.87, p = 0.0008 for “resting”) and winter (Kruskal - Wallis ANOVA, n = 102, KW = 62.15, p< 0.0001 for “feeding”; KW = 59.20, p < 0.0001 for “travelling’; KW =27.96, p=0.003 for “opportunistic feeding”; KW=33.21, p = 0.0005 for “socialising and playing” and KW = 24.31, p = 0.01 for “resting”). Gener- ally, “feeding” peaked in the morning and gradually decreased through the rest of the day. As “feeding” decreased, “travelling” and “opportunistic feeding” increased, both peaking in the afternoon. During winter evenings the frequency of “travelling” and “op- portunistic feeding” decreased again, with a corresponding secondary increase in “feed- Daylight behaviour of Humpback dolphins in Algoa Bay, South Africa 25 100 = 8 OÖ O) - = EERIDN \ (6) ER 3 60 En >) OÖ 3 = © 20 [®) O0) & € } © 20 Q. 1/2 Low Low 1/2 High High Tide Fig. 5. Tidal influence on the diurnal behaviours (GE) - feeding, && - travelling, — opportunistic feeding, — socialising and playing, [__] - resting and - other) of humpback dolphins in Algoa Bay observed between May 1991 and May 1994. ing”. Prolonged bouts of “resting” and “socialising and playing” occurred with similar fre- quency throughout the day, but were slightly less evident in the morning and evening. There was, however, a clear summer increase in “socialising and playing” behaviour around midday. The behaviour of humpback dolphins in Algoa Bay did not appear to be significantly related to the tides (Fig. 5), except for “feeding” which increased during high tide (Krus- kal - Wallis ANOVA, KW = 27.85, n = 102, p = 0.05). Discussion The six categories of behaviour distinguished in the present study are generally consistent with the types of behaviour observed and classified in several other studies (SAAYMAn and TAYLER 1979; SHANE 1990 a, b; BALLANCE 1992; BRAGER 1993). Behaviour of humpback dol- phins in Algoa Bay appeared to determine the spatial geometry of the group in a similar way as has been observed for other cetaceans (SHANE 1990 a; 1995) which suggests a func- tional significance. The tight structure of a “travelling” group may possibly reduce the like- lihood of incidental separation of an individual from the group, increase the sensory invigi- lation of the area being travelled through and possibly (as observed by SAAYMAN and TAYLER 1979) facilitate an active, co-operative defence. Furthermore, because swimming at the surface seems to require 4.5 times more energy than swimming at a depth of about one-half of body length (Au and Weıns 1980; Hur 1987) the long distance covered under- water during fast “travelling” is likely to be energetically beneficial (Huı 1989). The close proximity of group members during “resting” could possibly increase safety of an individual due to sensory integration of a group (e.g. Norrıs and Don 1980 a). The 26 L. KARCZMARSKI and V. G. COCKCROFT dispersed geometry of groups engaged in “feeding” and “opportunistic feeding” on the other hand, appears to be related to the foraging/feeding behaviour used by the animals (Würsıs 1986). The widely dispersed “feeding” groups indicate that individual, rather than co-operative, feeding is likely to be the norm for humpback dolphins in Algoa Bay. Feeding dominated the daylıght actıvity of humpback dolphins in Algoa Bay, as has also been observed for bottlenose dolphins in the bay system of Galveston, Texas (BRA- GER 1993). However, the overall proportions of daylight behaviours recorded for hump- back dolphins during the present study vary considerably from those described by SAay- MAN and TAYLER (1979) for humpback dolphins in Plettenberg Bay. The proportions of time dolphins were seen feeding (26.5 %) and travelling (50 %) in Plettenberg Bay are al- most the reverse of those in Algoa Bay. One reason for this could be that, despite an ap- parently similar definition of behaviour categories, the actual classification of behaviour observed in the field differed considerably between the two studies. Alternatively, it is possible that there are several other factors which differ between Algoa Bay and Pletten- berg Bay. The time spent on non-feeding behaviours appears to be proportional to feeding effi- ciency (HERBERS 1981). As feeding efficiency increases, less time is spent searching for or capturing prey and more time is available for less active behaviour. Feeding efficiency is likely to increase with richness of habitat and, inversely, more time is likely to be re- quired for feeding where food is not so plentiful. Consequently, the overall proportion of diurnal behaviours is likely to be a function of the habitat and biological needs of the ani- mals and the considerably smaller proportion of daylight hours occupied by feeding in Plettenberg Bay may reflect a greater abundance of the inshore prey resources. Furthermore, as discussed in KARCZMARSKI (1996), the Plettenberg Bay region houses a multitude of shallow rocky reefs which facilitate feeding for humpback dolphins. In con- trast, only the south-westerly bight of Algoa Bay has abundant shallow reefs. Conse- quently, it is possible that humpback dolphins use the areas of Algoa Bay and Pletten- berg Bay differently, with several feeding sites in the Plettenberg Bay region; but apparently only one (limited in size) primary feeding ground in Algoa Bay, where feeding is particularly intensive. By comparison, the overall proportion of daylight behaviours ob- served for bottlenose dolphins Tursiops truncatus on their estuarine feeding grounds in the Gulf of California, Mexico (BALLANcCE 1992) is strikingly similar to that of humpback dolphins in Algoa Bay. On the other hand, the proportion of time humpback dolphins spent feeding and travelling in Plettenberg Bay is strikingly similar to that reported by BALLANCE (1992) for bottlenose dolphins when the animals were away from their estuar- ine feeding grounds. Humpback dolphins in Algoa Bay displayed little seasonal difference in the overall proportion/frequency of daylight behaviours. Only “social” and “sexual” behaviour showed a seasonal difference, increasing in summer. This corresponds with the summer peak of calving observed for humpback dolphins in Algoa Bay (KARCZMARSKI 1996) and, consequently, supports the one year gestation period suggested for this species (V. G. CockCROFT, unpubl. data). Similarly, a seasonal (spring and summer) increase in so- cial behaviour, as well as abundance of calves, was observed for bottlenose dolphins off the Texas coast (SHANE 1990 b). The general lack of seasonal variation in the overall frequency of other behaviours — particularly feeding - is surprising, considering possible changes in energy requirements of the dolphins due to declining water temperature (CockcroFT and Ross 1990; Ross and CockcRoFT 1990). There was, however, an apparent increase in feeding behaviour during winter evenings in Algoa Bay, which may possibly reflect an increased energetic demand of the dolphins or, alternatively, an increase in prey abundance on winter evenings Be- cause of early nightfall in winter, all evening observations were discontinued between 17h00 and 18h00 (compared to 19h00-20h00 in summer). In winter, however, the second- Daylight behaviour of Humpback dolphins in Algoa Bay, South Africa 27 ary feeding peak increased at this time, while no feeding was observed during evenings in summer. Consequently, it is possible that in winter feeding occupies a larger proportion of humpback dolphin activity than was apparent during the present study. The proportion of daylight which humpback dolphins used for different behaviours varied considerably throughout the day and formed a distinct diurnal pattern. This pat- tern seems to follow the solar day and is, possibly, to a large degree shaped by the diurnal cycles of the prey species. A similar phenomenon is apparent for several populations of coastal bottlenose dolphins (SaAyman et al. 1973; Würsıs and Würsıc 1979; SHANE 1990 b; BRAGER 1993; HAnson and DEFRAN 1993), as well as other cetacean species (e.g. Norris and Donr 1980b; Würsıc and Würsıc 1980; KLınowsKA 1986). Data on the diur- nal variability in abundance, density or distribution of the inshore prey resources in Al- goa Bay are, however, scarce. The only reported fluctuation in the biomass of fish in the surf zone of Algoa Bay is thought to be related to the tidal cycle and increases during low tide; while the species diversity apparently increases just after twilight (LAsıa 1982, 1984). A better understanding of the diurnal cycles of the inshore fish and squid species in Algoa Bay could contribute substantially to our understanding of humpback dolphin diurnal activity/behaviour patterns. Feeding was the only behaviour of humpback dolphins apparently affected by the tides in Algoa Bay. However, the increase in feeding during high tide in Algoa Bay was less evident than that reported by SAaAyMmAn and TAYLER (1979) in Plettenberg Bay. SAAY- MAN and TAYLER (1979) speculated that an apparent increase in shoaling behaviour of some reef associated prey species during high tide could increase their “relative accessi- bility” for dolphins and consequently shape the entire daylıght activity pattern of hump- back dolphins. This appears less so in Algoa Bay where the biomass of fish in the surf zone is reported to increase at low tide (LAsıak 1982, 1984). Several other studies conducted in a number of coastal habitats showed various de- grees of influence of the tidal cycle on dolphin movement and activity (e.g. WÜRSIG and Würsıc 1979; SHANE 1990b: Hanson and DEFRAN 1993; FELIX 1994). It seems apparent that the influence of tides, although relatively strong in enclosed bays, passes and narrow channels, generally decreases with the openness of habitat. As the Algoa Bay region is a part of an exposed coastline where wave energy is considerably greater than tidal energy, a limited tidal impact on dolphin activity/behaviour could be expected. Overall, humpback dolphin behaviour appears similar to that described for other coastal dolphin species like the bottlenose dolphin. The present study, however, was not designed to be strictly behavioural; data on dolphin behaviour were collected opportunis- tically, as part of a larger scale research project. Consequently, a clear understanding of the behaviours of humpback dolphins and the relevance of these to the fulfilment of their biological and social needs requires further investigation. Acknowledgements This work would not have been possible without the financial assistance of the Foundation for Re- search Development (FRD), WWF South Africa and the University of Port Elizabeth, which is grate- fully acknowledged. We wish to express our thanks to MEREDITH THORNTON, JIMMY HENDERSON, and SHANE NORRIS for their help during fieldwork. We thank RAnDALL Weııs (Dolphin Biology Research Institute, Mote Marine Laboratory, Sarasota, Florida, USA), BERND Würsıc (Texas A&M University at Galveston, USA), ANTON McLACHLAN and DEO WINTER (University of Port Elizabeth, South Africa) and anonymous reviewers for their comments on the earlier draft of the manuscript; and Nına Woz- NIAK (German Department, University of Port Elizabeth) for her help with translation of Zusammen- fassung. 28 L. KARCZMARSKI and V. G. COCKCROFT Zusammenfassung Tagesgang im Verhalten von Buckeldelphinen Sousa chinensis in Algoa Bay, Südafrika Daten zum Tagesverhalten von Buckeldelphinen Sousa chinensis wurden über einen Zeitraum von drei Jahren in der Algoa-Bucht, Ostkapprovinz, Südafrika, in Aufnahmen auf See und vom Land aus er- stellt. Die Aktivitäten und Verhaltensweisen der Delphine wurden eingeteilt in: „Nahrungsaufnahme“, „Fortbewegung“, „opportunistische Nahrungsaufnahme“, „gesellschaftliches Verhalten und Spiel“, „Ausruhen“ und „Anderes“. Das Verhalten der Buckeldelphine bestimmt die räumliche Geometrie der Delphingruppe, nicht aber die Gruppengröße. Die Intervalle des Einatmens an der Wasseroberfläche sind ähnlich bei „Nahrungsaufnahme“, „opportunistischer Nahrungsaufnahme“ und bei langsamer „Fortbewegung“, weichen jedoch stark von dem Verhaltensmuster bei schneller „Fortbewegung“ ab. Das Tagesverhalten des Buckeldelphins wird von der „Nahrungsaufnahme“ dominiert und weist eine starke Regelmäßigkeit auf, die wahrscheinlich von den Tageszyklen der Beutetiere gesteuert wird. All- gemein hat die „Nahrungsaufnahme“ morgens ihren Höhepunkt und nimmt im Laufe des Tages all- mählich ab. Während die „Nahrungsaufnahme“ abnimmt, nehmen „Fortbewegung“ und „opportuni- stische Nahrungsaufnahme“ zu und zeigen am Nachmittag gleichsam Höchstwerte. Die Frequenz der Aktivitäten „Ausruhen“ und „gesellschaftliches Verhalten und Spiel“ bleibt ganztägig etwa gleich, ist aber morgens und abends etwas geringer. Dieses Verhaltensmuster variiert nur geringfügig zwischen Sommer und Winter, ebenso wie das gesamte Verhältnis der Tagesverhaltensweisen. Der einzige be- merkenswerte Unterschied für die Jahreszeiten ist die Frequenz des „gesellschaftlichen/Paarungsver- haltens”, das im Sommer seinen Höhepunkt hat. Daraus scheint hervorzugehen, daß die Gesamtpro- portion der Tagesverhalten der Delphine von ihrem Habitat und ihren biologischen Bedürfnissen determiniert wird. Obwohl der Gezeitenzyklus das Verhalten der Buckeldelphine in einigem Maße beeinflußt, sind Tagesaktivitäten und -verhalten vornehmlich von der Tageszeit bestimmt. Bei zukünfti- gen Forschungsaufgaben sollten nächtliches Verhalten wie auch das Verhältnis zwischen verschiedenen Verhaltensformen und potentiellen Störfaktoren berücksichtigt werden. References ACEVEDO, A. (1991): Behaviour and movements of bottlenose dolphins, Tursiops truncatus, in the en- trance to Ensenada De La Paz, Mexico. Aquat. Mamm. 17, 137-147. ALTMANN, J. (1974): Observational study of behaviour: sampling methods. Behaviour 49, 227-266. Au, D. W.; Weiss, D. (1980): At high speeds dolphins save energy by leaping. Nature 284, 548-550. BALLANGCE, L. T. (1992): Habitat use patterns and ranges of the bottlenose dolphin in the Gulf of Cali- fornia, Mexico. Mar. Mamm. Sci. 8, 262-274. BRAGER, S. (1993): Diurnal and seasonal behavior patterns of bottlenose dolphins (Tursiops truncatus). Mar. Mamm. Sci. 9, 434-438. CLOUDSLEY-THOMPSoN, J. L. (1961): Rhythmice activity in animal physiology and behavior. Ist ed. New York: Academic Press. CockcROFT, V. G.; Ross, G. J. B. (1990): Food and feeding of the Indian Ocean bottlenose dolphin off southern Natal, South Africa. In: The Bottlenose Dolphin. Ed. by S. LEATHERWooD and R.R. Reeves. San Diego: Academic Press. Pp. 295-308. FELIX, F. (1994): Ecology of the coastal bottlenose dolphin Tursiops truncatus in the Gulf of Guayaquil, Ecuador. Invest. Cet. 25, 235-250. Hanson, M. T.; DEFRAN, R.H. (1993): The behaviour and feeding ecology of the Pacific coast bottle- nose dolphin, Tursiops truncatus. Aquat. Mamm. 19, 127-142. HERBERS, J. M. (1981): Time resources and laziness in animals. Oecologia 49, 252-262. Hu1, €. A. (1987): Power and speed of swimming dolphins. J. Mammalogy 68, 126-132. Hur, €. A. (1989): Surfacing behavior and ventilation in free-ranging dolphins. J. Mammalogy 70, 833- 835. KARCZMARSKI, L. (1996): Ecological studies of humpback dolphins Sousa chinensis in the Algoa Bay re- gion, Eastern Cape, South Africa. Ph. D. thesis, Port Elizabeth: University of Port Elizabeth. KARCZMARSKI, L.; THORNTON, M.; CocKCkOFT, V. G. (1997): Description of selected behaviours of hump- back dolphins Sousa chinensis. Aquat. Mamm. 23, 127-133. KARCZMARSKI, L.; COCKCROFT, V. G.; MCLACHLAN, A. (1998): Group size and seasonal pattern of occur- Daylight behaviour of Humpback dolphins in Algoa Bay, South Africa 29 rence of humpback dolphins Sousa chinensis in Algoa Bay, South Africa. S. Afr. J. mar. Sci. (in press) KLınowska, M. (1986): Diurnal rhythms in Cetacea - a review. Rep. Int. Whal. Commn, Special Issue 8, 75-88. KLınowskA, M. (1991): Dolphins, porpoises and whales of the world. The IUCN red data book. Gland, Cambridge: IUCN. LASIAK, T. A. (1982): Structural and functional aspects of the surf-zone fish community in the Eastern Cape. Ph. D. thesis, Port Elizabeth: University of Port Elizabeth. LASIAK, T. A. (1984): Structural aspects of the surf-zone fish assemblage at King’s Beach, Algoa Bay, South Africa: short-term fluctuations. Estuar. coast. Shelf Sci. 18, 347-360. NIELSEN, E. T. (1983): Relation of behavioral activity rhythms to the changes of day and night. A revi- sion of views. Behaviour 89, 147-173. Norris, K. S.; DoHL, T. P. (1980 a): The structure and functions of cetacean schools. In: Cetacean behav- ior: mechanisms and functions. Ed. by L. M. HErMAn. Chichester: John Wiley and Sons. Pp. 211- 261. NoRRIS, K. S.; DoHL, T. P. (1980 b): Behavior of the Hawaiian spinner dolphin, Stenella longirostris. Fish. Bull 77, 821-849. Ross, G. J.; CocKckOFT, V. G. (1990): Comments on Australian bottlenose dolphins and the taxonomic status of Tursiops truncatus (EHRENBERG, 1832). In: The Bottlenose Dolphin. Ed. by S. LEATHERWooD and R. R. REEvES. San Diego: Academic Press. Pp. 101-128. Ross, G. J. B.; HEINSOHN, G. E.; CocKCROFT, V. G. (1994): Humpback dolphins Sousa chinensis (Osbeck, 1765), Sousa plumbea (G. Cuvier, 1829) and Sousa teuszii (Kükenthal, 1892). In: Handbook of mar- ine mammals, Vol. 5: The first book of dolphins. Ed. by S. H. Rıpaway and R. HARRIson. San Die- go: Academic Press. Pp. 23-42. SAAYMAN, G. S.; TAYLER, C.K. (1979): The socioecology of humpback dolphins (Sousa sp.). In: Beha- viour of marine animals. Vol. 3: Cetaceans. Ed. by H. E. Wınn and B. L. OrLLA. New York: Plenum Press. Pp. 165-226. SAAYMAN, G. S.; TAYLER, C. K.; BoweER, D. (1973): Diurnal activity cycles in captive and free-ranging In- dian Ocean bottlenose dolphins (Tursiops aduncus Ehrenburg). Behaviour 44, 212-233. SHANE, S. H. (1990 a): Behavior and ecology of the bottlenose dolphin at Sanibel Island, Florida. In: The Bottlenose Dolphin. Ed. by S. LEATHERWooD and R.R. REEvES. San Diego: Academic Press. Pp. 245-265. SHANE, S. H. (1990 b): Comparison of bottlenose dolphin behavior in Texas and Florida, with a critique of methods for studying dolphin behavior. In: The Bottlenose Dolphin. Ed. by S. LEATHERWOoD and R.R. REEvES. San Diego: Academic Press. Pp. 541-558. SHANE, S. H. (1995): Behavior patterns of pilot whales and Risso’s dolphins off Santa Catalina Island, California. Aquat. Mamm. 21, 195-197. SHANE, S. H.; WELLS, R. S.; Würsıg, B. (1986): Ecology, behavior and social organization of the bottle- nose dolphin: a review. Mar. Mamm. Sci. 2, 34-63. Wüssıs, B. (1986): Delphinid Foraging Strategies. In: Dolphin cognition and behavior: a comparative approach. Ed. by R. J. SCHUSTERMAN, J. A. THoMAs, and F. G. Woop. London: Lawrence Erlbaum Ass. Publ. Pp. 347-359. Würsıs, B.; Würsıc, M. (1979): Behavior and ecology of the bottlenose dolphin, Tursiops truncatus, in the south Atlantic. Fish. Bull 77, 399-412. Würsıs, B.; Würsıs, M. (1980): Behavior and ecology of the dusky dolphin, Lagenorhynchus obscurus, in the south Atlantic. Fish. Bull 77, 871-890. Authors’ addresses: L. KARCZMARSKI, Oceanic Society, Midway Spinner Dolphin Research Project, Midway Island Station #2, P.O. Box 294.60, Honolulu, HI 96820 - 1860, USA and V.G. CockckoFT, Centre for Dolphin Studies, PO. Box 1856, Plettenberg Bay 6600, South Africa Z. Säugetierkunde 64 (1999) 30-35 ZEITSCHRIFT FÜ © 1999 Urban & Fischer Verlag SAUG ETl ERKU NDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Allozyme variation and taxonomic status of Calomys hummelincki (Rodentia, Sigmodontinae) By BEATRIZ A. GARCIA, ANGELA MARTINO, MARINA B. CHIAPPERO, and CRISTINA N. GARDENAL Catedra de Ouimica Biolögica, Facultad de Ciencias Medicas, Universidad Nacional de Cördoba, Cördoba, Argentina and Centro de Inverstigaciones en Ecologla y Zonas Aridas, Coro, Estado Falcon, Venezuela Receipt of Ms. 16. 05. 1994 Acceptance of Ms. 12. 10. 1998 Abstract The level of genetic polymorphism was analyzed in a population sample of Calomys hummelincki from Venezuela. Enzymes and proteins studied by means of gel electrophoresis give information on 30 loci. The proportion of polymorphic loci was 40%, and mean expected heterozygosity (H.) was 12.7 %. These values are higher than those reported for most species of rodents in the northern hemisphere, but are comparable to those observed in other Sigmodontinae species from Argentina. Nei’s genetic dis- tance (Dn) with the species C. laucha, C. venustus, and C. musculinus ranged from 0.289 to 0.494. Dx values between populations of different Sigmodontinae species are below 0.09. A distance Wagner tree based on modified Rogers’ distances shows that C. hummelincki is more closely related to C. venustus than to C. laucha. Our data support the proposal that C. hummelincki and C. laucha are fully distinct species. Key words: Calomys hummelincki, Sigmodontinae, allozymes, polymorphisms Introduction Among the Sigmodontinae rodents (family Muridae) of the tribe Phyllotini, the genus Calomys shows a wide distribution in South America. The systematics of this genus, as well as the geographic distribution and ecology of some of its species are still poorly known. The taxonomic status of one of these species, Calomys hummelincki (Venezuelan pigmy mouse) has been a matter of controversy. The species was first described by Hus- son (1960) on the basis of specimens collected on islands of the Caribbean sea and in- cluded in the genus Baiomys. Later, HERSHKOVITZ (1962) reported on this rodent also from Venezuela, assigning it to the genus Calomys as synonym of C. laucha. In 1976, HANDLEY recognized the species Calomys hummelincki as being different from C. laucha and described its geographic distribution in the Orinoco plains and the deserts around the Gulf of Venezuela. The first study on karyologic differences between these species was conducted by PEREZ ZapartA et al. (1987), who reported a karyotype of 2n = 60, FN = 64 for C. hummelincki, fairly distinct from that of C. laucha (2n = 64, FN = 68). As a contribution to the knowledge of the systematics and evolution of South Ameri- can murids of the subfamily Sigmodontinae, we present here an analysis of allozymic polymorphism in Calomys hummelincki and an estimation of its genetic distance from the species C. laucha, C. venustus, and C. musculinus. 0044-3468/99/64/01 — 030 $ 12.00/0 Allozyme variation in Calomys hummelincki 31 Material and methods Seventeen specimens of C. hummelincki collected in Planicie Costera de Adicora (11°55’N; 69°49' W) Estado Falcön, Venezuela, were studied. Preparation of tissue homo- genates, electrophoretic and staining procedures were carried out as described by GAR- DENAL et al. (1980) and GARDENAL and Branco (1985). The following enzymes were ana- lyzed: soluble esterases (ES-1 to ES-6; E.C. 3.1.1.1), aspartate aminotransferases (AAT-1 and AAT-2; E.C. 2.6.1.1), catalase (CAT, E.C. 1.11.1.6), adenylate kinase (AK; E.C. 2.7.4.3), phosphoglucomutases (PGM-1 and PGM>2; E.C. 2.7.5.1), superoxide dismutases (SOD-1 and SOD-2; E.C. 1.15.1.1), liver acid phosphatase (ACP}; E.C. 3.1.3.2), kidney acid phosphatase (ACPx; E.C. 3.1.3.2), malic enzyme (ME; E.C. 1.1.1.40), malate dehy- drogenases (MDH-1 and MDH-2, E.C. 1.1.1.37), lactate dehydrogenase (LDH-1 and LDH-2, E.C. 1.1.1.27), NADP-isocitrate dehydrogenases (IDH-1 and IDH-, E.C. 1.1.1.42), 6-phosphogluconate dehydrogenase (6-PGDH, E.C. 1.1.1.44), glucose-6-phos- phate dehydrogenase (G6PDH, E.C. 1.1.1.49), glycerophosphate dehydrogenase (GPDH, E.C. 1.1.1.8), alcohol dehydrogenase (ADH, E.C. 1.1.1.1), and NAD-linked nonspecific dehydrogenase (NDH). In addition, other proteins were studied in serum: transferrin (Tf) haptoglobin (Hpt) and albumin (Alb). Altogether, these proteins give information on ge- netic variation at 30 locıi. Statistics Average heterozygosity per individual and proportion of polymorphic loci were estimated from the 30 locı analyzed. Genetic distance (Dn) indices were calculated according to NEI (1972) and RoGErs modified by WRIGHT (1978). Calculations were based on the allele frequencies at 20loci reported previously for comparisons between the species C. musculinus, C. laucha, and C. venustus (GARDENAL et al. 1990). A distance Wagner tree (rooted at midpoit of longest path) was constructed on the basis of modified Rogers’ dis- tances (WRIGHT 1978) between the four species. All calculations were performed by using the BIOSYS-1 program (SWOFFORD and SELANDER 1981). Results and discussion Table 1 shows allele frequencies for 12 polymorphic loci in C. hummelincki. Expected average heterozygosity (H.) was 12.7 % and observed average heterozygosity (H,) was 11.8%. The proportion of polymorphic locı was 40 %. These values are clearly higher than those reported for most species of rodents in the northern hemisphere (NEvo et al. 1984). WARrD et al. (1992) reported an average H value of 6.7 % for 172 species of mam- mals. ParTon et al. (1989) found values ranging from 1.1 to 7.1% for H and between 7.7 to 21.3 % for P in different species of the tribe Akodontini (subfamily Sigmodontinae, fa- mily Muridae) from Peru. The relatively high level of polymorphism observed in C. hummelincki ıs comparable to that of other sigmodontine species of the genera Ca- lomys (GARDENAL et al. 1980; GARDENAL and BrLAnco, 1985; GARDENAL et al. 1990; Garcia et al. 1990) and Akodon (APFELBAUM and Branco 1985), and the species Gra- omys griseoflavus (THEILER and GARDENAL 1994) and Eligmodontia typus (DE Sousa et al. 1996) from Argentina (Tab. 2). BARRANTES et al. (1993) found expected H between 3.8 and 11 % in populations of eight species of Akodon from Argentina. The last value corre- sponded to one population of A. longipilis. The observed H value was much lower (3.9 %), a result not explained by the authors. Table 3 presents genetic distance values between C. hummelincki, C.laucha, C. musculinus, and C. venustus. On the basis of modified Rogers’ distances (WRIGHT 1978) 32 BEATRIZ A. GARCIA etal. Table 1. Allele frequencies in the population sample of Calomys hummelincki Allele Allele Frequency Frequency * Alleles at the loci not analyzed in other species of Calomys are designated by letters. The loci Ak, G6pdh, Pgm-1, Pgm-2, Sod-2, Hpt and Alb were monomorphic and not analyzed in other species of Calomys. Alleles at the remaining loci are designated by numbers, indicating electrophoretic mobilities of bands relative to those observed in other Calomys species by GARDENAL et al. (1990). Table 2. Proportion of polymorphic loci (P) and expected heterozygosity (H.) in different species of South American sigmodontine rodents. Species C. hummelincki C. laucha C. musculinus C. venustus Akodon dolores Akodon azarae Akodon (Peru) Akodon (Argentina) Eligmodontia typus Graomys griseoflavus Oligoryzomys flavescens 40 62.5-77.3 61.0-73.0 66.7 27.8-38.9 16.6-30.0 7.1-21.3 6.7-26.7 68.0 46.0-66.0 34.6-61.5 127 11.8-16.3 14.9-20.0 14.6 13.8-19.2 9.9-11.8 1.1-7.1 6.7-11 16.0 16.0-18.0 3.8-9.7 * Ner’s (1972) genetic distance between populations 0.002-0.010 0.005-0.015 0.075-0.093 0.0016-0.0088 Reference this study GARDENAL et al. (1990) Garcıa et al. (1990) GARDENAL et al. (1980) GARDENAL and BLANCO (1985) GARDENAL et al. (1990) GARDENAL et al. (1990) APFELBAUM and BLANCO (1985) APFELBAUM and BLANcCO (1985) PATTon et al. (1989) BARRANTES et al. (1993) DE SousA and GARDENAL (1996) THEILER and GARDENAL (1994) CHIAPPERO et al. (1997) Allozyme variation in Calomys hummelincki 33 Table 3. Values for modified Rogers’ distance (Dg; above the diagonal) and Nei’s genetic distance (Dn; below the diagonal) between species of Calomys. C. hummelincki C. laucha C. venustus C. musculinus C. hummelincki C. laucha C. venustus C. musculinus C. musculinus C. laucha C. venustus C. hummelincki 0.00 0.05 0.10 0.15 0.20 0.25 0.30 Distance from root Cophenetic correlation = 0.994 Fig. 1. Distance Wagner procedure tree based on modified Rogers’ distances between the species Ca- lomys hummelincki, C. laucha, C. venustus, and C. musculinus. the distance Wagner tree of figure 1 was constructed. C. hummelincki appears more clo- sely related to C. venustus than to C. laucha. Table 2 includes values of Dx between popu- lations of C. laucha, E. typus, and G. griseoflavus. These species belong to the same tribe (Phyllotini) as C. hummelincki. The highest Dx between populations of the same species was 0.09, while all comparisons between species gave values above 0.29. The taxonomic status of the Venezuelan pigmy mouse (C. hummelincki) has been the subject of controversy (PEREZ ZAPATA et al. 1987). HERSHKOVITZ (1962) did not recognize C. hummelincki as a separate species and considered it as C. laucha. It was known that C. laucha was distributed in a wide area of South America compris- ing southeastern Brazil, southern Bolivia, Paraguay, Uruguay, and central Argentina. This is very far from the south of the region where C. hummelincki ıs found. When HERSHKO- vırz (1962) proposed this species as a synonym of C. laucha, he assumed that its presence in Venezuela could be due to accidental transportation by man. Data presented here strongly support the proposal that C. laucha and C. hummelincki are distinct species, providing thus a more rational explanation for the geographic distribu- tion of the species. The genetic distance between C. laucha and C. hummelincki (Dx = 0.36) is within the range accepted for species which have completed their reproductive isolation (Ayara 1982). Five loci (Es-1, Es-4, Adh, Aat-1 and Gpdh) can be utilized as “diagnos- tic“, since the species do not share alleles at these loci. Genetic distances for intraspecific comparisons between populations of C. laucha, Eligmodontia typus, and Graomys griseo- flavus, species closely related to C. hummelincki, gave values below 0.09 (Tab. 2). 34 BEATRIZ A. GARcIA etal. In an analysis of karyological relationships among species of Calomys, VITULLoO et al. (1990) described different “groups“ of species on the basis of chromosomal characteris- tics (2n, fundamental number, morphology). According to these criteria, C. hummelincki (2n = 60; FN =64) and C. laucha (2n = 64; FN =68) were included in the same group (Group I), while C. venustus (2n =56; NF=66) and C. musculinus (2n =38; FN = 56) were included in different groups (II and III, respectively). The study of differentiation in structural genes presented here indicates different relationships between species of Ca- lomys than those inferred from cytogenetic analysis. Nevertheless, it has been suggested that in mammals the rate of evolution of morphology, karyotype, and structural genes can be independent, and so the relationships assigned on the basis of these criteria may dis- agree (SCHNELL and SELANDER 1981; APFELBAUM and ReıG 1989). Aknowledgements The authors thank Dr. Antonio BLAnco for advice and critical revision of the manuscript. This work was supported, in part by grants from the Concejo Nacional de Investigaciones Cientificas y T&cnicas (CONICET) of Argentina, the Consejo de Investigaciones Cientificas y Tecnolögicas de la Provincia de Cördoba (CONICOR) and the Antorchas Foundation. C.N.G. and B.A.G. are Career Investigators of CONICET and M.B.C. is a Fellow of CONICOR. Zusammenfassung Allozymvariation und taxonomische Stellung von Calomys hummelincki (Rodentia, Sigmodontidae) Die genetische Variabilität einer Population von Calomys hummelincki aus Venezuela wurde mittels Gelelektrophorese von Enzymen und Serumproteinen untersucht. Insgesamt wurden 30 Genloci erfaßt. Die Polymorphierate betrug 40 % und der durchschnittliche erwartete Heterozygotiegrad (H,) 12,7%. Diese Werte liegen höher als jene der meisten bisher untersuchten Nagetierarten der Nordhe- misphäre. Sie sind jedoch den Angaben über andere Arten der Unterfamilie Sigmodontinae aus Argen- tinien vergleichbar. Die genetischen Distanzen nach Nei (Dx) zu den Arten €. laucha, C. venustus und C. musculinus reichten von 0,289 bis 0,494, während jene zwischen Populationen der jeweiligen Arten einen Wert von 0,09 nicht überschritten. Ein auf modifizierten Rogers-Distanzen beruhender Wagner- Baum zeigt, daß C. hummelincki mit C. venustus näher verwandt ist als mit C. laucha. Unsere Daten stützen die Hypothese, daß C. hummelincki und C. laucha zwei verschiedene Arten sind. References APFELBAUM, L. I.; BLANco, A. (1985): Genetic polymorphism in populations of Akodon rodents. Canad. J. Genet. Cytol. 27, 12-17. APFELBAUM, L. I.; REIG, ©. A. (1989): Allozyme genetic distances and evolutionary relationships in spe- cies of akodontinae rodents (Cricetidae: Sigmodontinae) Biol. J. Linnean Soc. 38, 257-280. AYALA, F. J. (1982): Gradualism versus punctualism in speciation. In: Mechanisms of Speciation: Repro- ductive Isolation, Morphology, Genetics. Ed. by €. BARIıGoZZ1. New York: Alan R. Liss, pp. 51-66. BARRANTES, G. E.; ORTELLS, M. O.; REıG, O. A. (1993): New studies on allozyme genetic distance and variability in akodontine rodents (Cricetidae) and their systematic implications. Biol. J. Linnean Soc. 48, 283-298. CHIAPPERO, M. B.; CALDERÖN, G. E.; GARDENAL, C.N. (1997): Oligoryzomys flavescens (Rodentia, Muri- dae): gene flow among populations from central-eastern Argentina. Genetica 101, 105-113. DE SousA, G. B.; DE Rosa, N.; GARDENAL, C. N. (1996): Protein polymorphism in Zligmodontia typus. Genetic divergence with other phyllotine cricetids. Genetica 97, 47-53. GARCIA, B. A.; GARDENAL, C. N.; BLANco, A. (1990): Heterozygosity and gene flow in populations of Calomys laucha (Rodentia, Cricetidae). Hereditas 112, 179-186. Allozyme variation in Calomys hummelincki 35 GARDENAL, C. N.; BLANCO, A. (1985): Polimorfismo enzimätico en Calomys musculinus: nueva estima- ciön. Mendelıana 7, 3-12. | GARDENAL, C. N.; SABATTINI, M. S.; BLANco, A. (1980): Enzyme polymorphism in a population of Ca- lomys musculinus (Rodentia, Cricetidae). Biochem. Genet. 18, 563-575. GARDENAL, C. N.; GARCIA, B. A.; SABATTINI, M. S.; BLANco, A. (1990): Protein polymorphism and genet- ic distance in South American cricetid rodents of the genus Calomys. Genetica 80, 175-180. HANDLEY, C. ©. (1976): Mammals of the Smithsonian Venezuelan project. Brigham Young Univ. Science Bull. Biol. Ser. 20, 5: 1-91. HERSHKOVITZ, P. (1962): Evolution of neotropical cricetine rodents (Muridae), with special reference to the phyllotine group. Fieldiana, Zool. 46, 1-524. Husson, A.M. (1960): A new species of the rodent Baiomys from Aruba and Curacao. Stud. Fauna Curacao 10, 33-40. Neı, M. (1972): Genetic distance between populations. Am. Nat. 106, 238-292. NEVo, E.; BEILES, A.; BEN-SHLOMO, R. (1984): The evolutionary significance of diversity: ecological, de- mographic and life history correlates. In: Evolutionary Dynamics of Genetic Diversity. Ed. by G. S. Mani. Lecture Notes in Biomathematics 53, 13-213. PATTON, J. L.; Myers, P.; SMITH, M. F. (1989): Electromorphic variation in selected South American ako- dontine rodents (Muridae: Sigmodontinae), with comments on systematic implications. Z. Säuge- tierkunde 54, 347-359. PEREZ ZAPATA, A.; VITULLO, A. D.; Reıc, O. A. (1987): Karyotypic and sperm distinction of Calomys hummelincki from Calomys laucha (Rodentia, Cricetidae). Acta Cientif. Venezolana 38, 90-93. SCHNELL, G. D.; SELANDER, R. K. (1981): Environmental and morphological correlates of genetic varia- tion in mammals. In: Mammalian Population Genetics. Ed. by H. SMITH and J. JouLe. Athens: Univ. Georgia Press. p. 60-90. SWOFFORD, D. L.; SELANDER, R.K. (1981): BIOSYS-1 a FORTRAN program for the comprehensive analysis of electrophoretic data in population genetics and systematics. J. Hered. 72, 281-283. THEILER, G. R.; GARDENAL, C.N. (1994): Patterns of evolution in Graomys griseoflavus (Rodentia, Cri- cetidae). I. Protein polymorphism in populations with different chromosome number. Hereditas 120, 225-229. VITULLO, A. D.; Espinosa, M. R.; MERANI, M.S. (1990): Cytogenetics of Vesper mice, Calomys (Ro- dentia, Cricetidae): Robertsonian variation between Calomys callidus and Calomys venustus, Z. Säugetierkunde 55, 99-105. WARD, R. D.; SKIBINSKI, D. O. F.;, WoODWARD, M. (1992): Protein heterozygosity, protein structure and taxonomic differentiation. In: Evolutionary Biology. Ed. by M.K. Hecht et al. Vol. 26, New York: Plenum Press. pp. 73-158. WRIGHT, S. (1978): Evolution and the Genetics of Populations. Vol. 4: Variability within and among na- tural populations. Chicago: Univ. Chicago Press. Authors’ addresses: CRISTINA N. GARDENAL, BEATRIZ A. GARCIA, and MARINA B. CHIAPPERO, Cätedra de Quimica Biolögica, Facultad de Ciencias MEdicas, Universidad Nacional de Cördo- ba, Casilla de Correo 35 Sucursal 16, 5016 Cördoba, Argentina and ANGELA MAR- TINO, Centro de Investigaciones en Ecologia y Zonas Aridas, Apartado 7506, Coro, Estado Falcön, Venezuela. Z. Säugetierkunde 64 (1999) 36-50 ZEITSCHRIFT De rüR © 1999 Urban & Fischer Verlag SÄUG EnNl ERKÜNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY The karyotypes of Cryptomys anselli sp. nova and Cryptomys kafuensis sp. nova: new species of the common mole-rat from Zambia (Rodentia, Bathyergidae) By H. BURDA, J. ZIMA, A. SCHARFF, M. MACHOLAN, and M. KAWALIKA University of Essen, Essen, Germany, Academy of Sciences, Brno, Czech Republic; City Council of Nadola, Ndola, Zambia Receipt of Ms. 04. 05. 1998 Acceptance of Ms. 19. 10. 1998 Abstract Two new species of a “small form” (i.e. with body mass of about 90 g) of the African mole-rat Cryp- tomys (Rodentia: Bathyergidae) are described from Zambia: C. anselli sp. nova characterized by di- ploid chromosome number 2n = 68 from Lusaka Province and C. kafuensis sp. nova with 2n = 58 chro- mosomes from Itezhi-Tezhi, Kafue National Park, Southern Province. Conventionally stained, C- and G-banded karyotypes, and localisation of NORs are described for both species. Whereas classical morphological and morphometrical traits cannot be used for diagnosis of Cryptomys species, karyo- types and allozymes enable distinction of both new species from each other and from other known Cryptomys species. Nevertheless, also the thickness of the external wall of the infraorbital foramen (relative to the breadth of the opening) seems to be species-specific: it is thicker in the examined spe- cimens of C. anselli sp. nova but thinner in C. kafuensis sp. nova. Key words: Cryptomys, Bathyergidae, chromosome, subterranean mammals, taxonomy Introduction African mole-rats of the genus Cryptomys Gray, 1864 (Bathyergidae) are subterranean rodents occurring from semi-arid to mesic habitats in different soil types over a wide geo- graphic range from Ghana to the Cape Province in South Africa. Although it is not a pro- blem to recognize Cryptomys as Cryptomys, extreme variation in many morphological traits (cranial parameters, body size, pelage coloration) tradıtionally employed in alpha- taxonomy of rodents, makes taxonomic treatment of this genus very difficult. Accord- ingly, different authors recognize different numbers of species. Thus, for instance, 44 to 49 species of Cryptomys have been named by ALLEn (1939) or ELLERMANN (1940), re- spectively, whereas only three have been considered by Nowak (1991). More recently, HoNnEycUTT et al. (1991) recognized seven species: Cryptomys bocagei (De Winton, 1897); Cryptomys damarensis (Ogilby, 1838); Cryptomys foxi (Thomas, 1911); Cryptomys hotten- totus with subspecies C. h. hottentotus (Lesson, 1826), C. h. natalensis (ROBERTS, 1913), C. h. darlingi (THuomas, 1895); C. h. amatus (WROUGHTON, 1907), and C. h. whytei (THOMAS, 1897); Cryptomys mechowi (PETERS, 1881); Cryptomys ochraceocinereus (HEUGLIN, 1864); Cryptomys zechi (MATSCHIE, 1900). It has been repeatedly demonstrated (ROSEVEAR 1969; AnseLL 1978; WILLIAMS et al. 1983; NEvo et al. 1986, 1987; HoNnEycuTT et al. 1987, 1991; JANECEK et al. 1992) that classi- 0044-3468/99/64/01 — 036 $ 12.00/0 New species of Cryptomys 37 cal morphological qualitative and quantitative traits are not sufficient for the diagnosis of Cryptomys species, and additionally, cytology, serology, and molecular genetics should be taken into account. Subsequent karyological studies demonstrated that at least Crypr- omys darlingi and Cryptomy amatus should be considered distinct species (AGUILAR 1993; MACHOLAN et al. 1998). In addition, our allozyme and karyotype studies (FiLippuccı et al. 1994, 1997; MACHOLAN et al. 1993) identified two additional species of the small form of Cryptomys in Zambia characterized by karyotypes 2n=58 (the “Itezhi-Tezhi popula- tion”) and 2n = 68 (the “Lusaka population”). Particularly the Lusaka population (2n = 68) has been subjected to intensive research on various aspects of its biology: reproduction and social behaviour (BurpDA 1989, 1990, 1995; BEGALL 1997; BEGALL and BURDA 1998; WILLINGSTORFER et al. 1998), hearing, car morphology, and vocalization (MÜLLER and BurpA 1989; BURDA et al. 1992; MÜLLER et al. 1992; LINDENLAUB and BURDA 1993, 1994; LINDENLAUB et al. 1995; Kössı et al. 1996; BrÜCKMANN and BurDA 1997; CREDNER et al. 1997), magnetic compass orientation (BUR- DA et al. 1990; MARHOLD et al. 1997), aspects of neuroanatomy (OELSCHLÄGER and BURDA 1992, Misex et al. 1996), physiology of metabolism (MARHOLD and NAGer 1995). Parasites in both species (and C. mechowi) have been studied by SCHARFF et al. (1996, 1997). In the meantime, the Zambian Cryptomys has become a well established model in many further biological studies. This fact calls for an unambiguous denomination of this species. While it has been obvious to us (and we have repeatedly stated it in all our publications) that these Cryptomys represent species distinctly different from South African Cryptomys hot- tentotus (for which they had been previously taken), a formal description has not been possible until the taxonomic status of neighbouring populations of Cryptomys amatus and C. darlingi had been clarified (MACHOLAN et al. 1998; AGuıLAr 1993). While the species status and their distinction from other Cryptomys species have been proven in allozyme studies (FıLıppuccı et al. 1994, 1997), in this study we denominate both species and de- scribe their karyotypes in detail. Material and methods Altogether nine individuals of the Lusaka population (see below and Tab. 1) and three individuals (one male, two females) of the Itezhi-Tezhi population were karyotyped. Mitotic metaphases were ob- tained directly from bone marrow. Slides were differentially stained using the trypsin digestion (G- banding) technique by SEABRIGHT (1971) and the C-banding technique by SuMNER (1972). Nucleolus organizer regions (NORs) were visualized by the silver-staining method of HowELL and BLACK (1980). Results Cryptomys anselli sp. nova Holotype Adult male, whole body ethanol-preserved, in Senckenberg Museum, Frankfurt am Main, Germany, allocation number SMF 87018 (specimen’s field number L-45). Collected on 15. 07. 1996 by ANDREAS SCHARFF. Paratype Adult female, SMF 87019 (L-46); sample data as in holotype. Type locality Court of the Chainama Hills Golf Club in the north-eastern part of Lusaka, Zambia. 38 H. Burpa etal. Etymology The species name recalls late Mr. W. F. H. AnseLL and his merit in the study of taxonomy and distribution of mammals of Zambia. Measurements and diagnosis Body sıze and cranial measurements in Cryptomys have no taxonomic-diagnostic value and are not provided here (for reasoning see Discussion). The body mass of adult wild- caught individuals (which is the best parameter for comparing body size among different Cryptomys species) in C. anselli sp. nova amounts to 76 +12 g (range 65-102 g, n = 66) in females and to 96 +13 g (range 80-126 g, n= 20) in males. Pelage coloration is age- and body mass-dependent: ıt ıs dark slate grey and metallic black in sucklings, greyish brown in weaned pups, brown in juvenile and subadult animals, and eventually golden ochre in adults. There is a remarkable variatıon in the size and shape of the white head spot, never- theless, ıt ıs well developed in most individuals. The infraorbital foramen is thick-walled (1.e. the external wall is thicker than the breadth of the opening), elliptical or drop-shaped (reference is made to specimens L6, L13, L15, L25, L48, L50, L54, Kenson 4, Kenson 9, Kenson 10, Kenson x, LX-3, LX6, and LX-13 deposited at the Department of General Zoology, University of Essen). However, it should be noted that lateral asymmetry in the shape of the infraorbital foramen was found in some specimens (L19, L23). The analysıs of allozymic variation allows clear separation of Cryptomys anselli sp. nova from C. kafuensis sp. nova, C. mechowi, C. damarensis, C. h. hottentotus, and C. h. natalensis and warrants attributing of a species status (FiLıppuccı et al. 1994, 1997). The diploid chromosome number in all the individuals examined (n=9) is 2n = 68. The proportion of acrocentric and biarmed chromosomes is variable. The karyotype con- sists of mainly (56-59) acrocentric chromosomes, 2-4 large subtelocentric autosomes, 0- 3submeta- or metacentric autosomes, and 4-6 small biarmed autosomes (cf. Fig. 1, Tab. 1). The X chromosome is variable in size and centromeric position, and the two X chromosomes in female sets are often heteromorphic. The Y chromosome is dot-like, probably uniarmed. Consequently, NF is variable and ranges from 79 to 82. C-positive heterochromatic arms are observed in a varied number of large biarmed autosomes and in a small pair of subtelocentric autosomes. Centromeric dark bands are present in most chromosomes. Distinct telomeric dark C-bands are situated in one large submetacentric chromosome (presumably the X) and in six autosomal pairs. The Y chromosome stains positively in C-banded slides (Fig. 2). G-banding cannot reveal any clear homology among the large biarmed autosomes and the X chromosomes (Fig. 3). Ag-NORs are situ- ated ın the telomeric areas of one large biarmed chromosome, two small metacentric, and in about 10 acrocentric autosomes. Table 1. Composition of individual karyotypes in the examined specimens of Cryptomys anselli sp. nova. No. protocol large large small ST M/SM M/SM oO PWNH+H POANN NN NN ON M 3 E 3 F - M )2 Ei 2 F 4 E 4 M 3 IE + OO 090 I {9% DO NDHD HH O New species of Cryptomys 39 Fig. 1a. Chromosomal slides are deposited in the Institute of Vertebrate Biology, Academy of Sciences CR, Brno, Czech Republic. Distribution and habitat The animals of this species were collected in cultivated fields, gardens, golf courses, and savannah-bushland habitats in Lusaka, Zambia, and its north-eastern suburbs (within and near the University of Lusaka campus), in Ngwerere (10 km north of Lusaka), Mungule (about 30 km north-west of Lusaka) and Chinunyu (about 90 km east of Lusaka), i.e., within the degree squares 1528 Al, 1528 A4, and 1529 A1 (following the mapping of An- SELL 1978). The collecting sites are characterized by the mean annual rainfall of 822 mm (monthly precipitation amounts to 68 + 81 mm, range 0-207 mm). Cryptomys kafuensis sp. nova Holotype Adult female, whole body ethanol-preserved, in Senckenberg Museum, Frankfurt am Main, Germany, allocation number SMF 87124. Collected on 27. 05. 1991 by Jırf KoCkA. Paratype Adult females, SMF 87125 and SMF 87126; sample data as in holotype. 40 H. BurDA etal. BEIEEr »B BEN 28 Be Aa 5 | "N fo an an 00 no a0 ne ao An On ad ün ea ür an no NS 2a 02 an A an Bü un Aa sa & Fig. 1b. Fig. 1. Conventionally stained karyotypes of Cryptomys anselli sp. nova. a= individual No. 5 in table 2, b = individual No. 6. Type locality “Hot Springs” in Itezhi-Tezhi, Kafue National Park, Southern Province, Zambia, within the degree square 1526 C1 (following the mapping of AnseLL 1978). Etymology The name of the species refers to the locality, the Kafue National Park in Zambia. Measurements and diagnosis The body mass of adult wild-caught individuals in C. kafuensis sp. nova amounts to 73+9g (range 61-77, n= 10) in females and to 113 +28 g (range 84-139, n = 3) in males. Pelage coloration and its age-dependent changes correspond to the situation described above for C. anselli sp. nova. The white head spot is well developed in most individuals and tends to be more prominent than in C. anselli sp. nova; nevertheless, its size and shape are individually very variable. The infraorbital foramen is thin-walled (1. e. its exter- nal wall is thinner than the breadth of the foramen), drop-like to elliptical (reference is made to specimens K6, K8, K10, K14 deposited at the Department of General Zoology, University of Essen). The analysis of allozymic variation allows clear separation of C. kafuensis sp. nova from Cryptomys anselli sp. nova, C. mechowi, C. damarensis, C. h. hottentotus, and C. h. natalensis and warrants allocation of a species status (FıLıppuccı et al. 1994, 1997). New species of Cryptomys 41 .* as a ® * 4 .“ es: 2 . » : ‘ -® PM: ı 8 1: 2 j . = s * ® . 8 e| 7 . ° ; ’® * 8 e. ä Ss Si ae a; > n zz; _ & » , * & .. ä > Fig. 2a. The diploid chromosome number in all the individuals (n=3) examined is 2n =58, NF = 82. The karyotype consists of 11 biarmed and 17 acrocentric autosomal pairs. Four biarmed (one metacentric, two submetacentric, and one subtelocentric) autosomal pairs can be distinguished according to their larger size. The other biarmed (six meta- or sub- metacentric and one subtelocentric) pairs of autosomes are distinctly smaller. The two largest acrocentric pairs are approximately as large as the largest biarmed autosomes. One of these large acrocentric pairs possesses very short second arms. The other acro- centric autosomes are distinctly smaller and they form a continuum of decreasing sizes. The X chromosome is metacentric and its size is similar to the largest autosomes. The Y chromosome is dot-like, probably uniarmed (Fig. 4a). The C-banded karyotype reveals considerable amounts of positively stained heterochromatin. One arm and the broad peri- centromeric area of the largest metacentric autosome are completely heterochromatic. A heterochromatic small arm is visible also in the small subtelocentric pair. Centromeric dark bands are found in certain biarmed and in most’ of the acrocentric chromosomes. A telomeric C-positive band in one arm is situated in three pairs of biarmed and five pairs of acrocentric autosomes. Intercalary dark bands are situated in two acrocentric auto- somes. The X chromosome is not positively stained in C-banded preparations, whereas the Y chromosome has a prominent dark band in the pericentromeric area (Fig. 4b). The large metacentric autosome with the C-heterochromatic arm stains mainly negatively in G-banded preparations, and it possesses only one large dark band situated in the euchro- matic arm. The G-banding pattern enables identification of most of the homologous chro- 42 H. BurpA etal. “es Es & a #* rr r = a .. - s » ee; x * ö . ”-» u! 8 . 8 .® .. x . 9% r .. e* Fig. 2b. Fig. 2. C-banded karyotype of Cryptomys anselli sp. nova. a = individual No. 5, b = individual No. 3. mosomes (Fig. 4c). The Ag-NORs positive signals are observed in the telomeric areas of several (10-12) small metacentric and acrocentric autosomes. Chromosomal slides are deposited in the Institute of Vertebrate Biology, Academy of Sciences CR, Brno, Czech Republic. Distribution and habitat The animals of this species were collected in grassland habitats at the locality Hot Springs and cultivated fields of nearby villages, in Itezhi-Tezhi, Kafue National Park, Zambia, within the degree square 1526 C1 (following the mapping of AnseLr 1978). The collect- ing site is characterized by the mean annual rainfall of 787 mm (monthly precipitation amounts to 66 + 78 mm, range 0-199 mm). Discussion Morphology and morphometry As stated earlier Cryptomys mole-rats are remarkably polymorphic, so that it is not possi- ble to provide unambiguous diagnostic morphological traits or measurements. As in other rodents, Cryptomys is characterized by indeterminate growth. However, the growth is not continuous and its rate is subjected to accelerations and periods of stasis depending on di- New species of Cryptomys 43 4 = “ » 6 " Be DI 8 7) 8 = 9 35 23 s . & Sa - ar n ” . zZ» TE Eee 5 En . > . & & “ “. & Je um... ti 2 “ wi a a , - _ n: = = - .- % .- ;ı®* A = -. rn Fig. 3. G-banded karyotype of Cryptomys anselli sp. nova individual No. 5. verse factors (reproductive and social status, age, and unknown factors). Due to these facts, the generally slow growth rate, and remarkable longevity (15 years and more), the body sıze and form and consequently also cranial proportions are subject to progressive and regressive changes (i.e. they fluctuate) during individual life (cf. BEGALL and BURDA 1998). For counting the mean adult body mass we selected individuals from our sample which weighed at least 60. g in females and 80. g in males. This arbitray limit is based on our long-term observation (BurpDA 1989; 1990; BEGALL and BurpA 1998) of the lowest body mass of breeding animals in captivity. Whereas there is sıgnificant sexual dimorph- ism in body mass in both species, there is no significant difference in body mass of males or females between both species. Whereas in all the examined skulls (n = 14, juveniles and adults, females and males being represented in the sample) of C. anselli sp. nova, the external wall of the foramen infraorbitale was relatively thick, all the examined skulls (n = 4, 2 adult males, 2 adult fe- males) of C. kafuensis sp. nova were characterized by a thin-walled foramen. HoNnEYCUTT et al. (1991) considered thick-walled outer foramina to be characteristic of the C. damarensis, C. mechowi, and C. bocagei group (and west and central African species), while thin-walled foramina should characterize the C. hottentotus group. Consequently, C. anselli sp. nova should be grouped with C. damarensis and C. mechowi, whereas C. kafuensis sp. nova should be closer related with C. hottentotus. However, the results of allozymic studies (FıLippuccı et al. 1994, 1997) do not support such distinction. Moreover, it should be noted that our specimens of C. mechowi from Ndola exhibit the thin-walled condition. 4 ee RAKU AAN xx nz um xx an xx 8x HOND an oAnn nn onan Dann 0mnnANnoNnN nn nn Fig. 4a. 7. Fig. 4b. New species of Cryptomys 45 Fig. 4c. Fig. 4. Karyotypes of a male of Cryptomys kafuensis sp. nova. a = conventional staining, b = C-banding, c = G-banding. Table 2. Characteristics of known karyotypes (representing different species) of Cryptomys. M — me- tacentric, SM - submetacentric, ST — subtelocentric, A — acrocentric, NF - fundamental number of chromosome arms in a female karyotype. Species Occurrence Karyo- Autosomes Sexchro- Arms Reference type mosomes (NF) (2n) MSMST A X C. mechowi Zambia MACHOLAN etal. (Copperbelt Province) (1993) C. amatus Zambia MACHOLAN et al. (Central Province) (1998) C. h. hotten- South Africa NEvo et al. (1986) totus (Transvaal) C. h. nata- South Africa NEvo et al. (1986) lensis (Natal) C. darlingi Zimbabwe 82 _AGUILAR (1993) (Harare) C. kafuensis Zambia 82 present study (Southern Province) C. foxi Cameroon 66 (70) 130 WiLLIamsetal. (138) (1983) C. anselli Zambia 68 6-9 2-4 56-59 79-82 present study (Lusaka Province) C. dama- Botswana 78 16 60 96 Nevoetal. (1986) rensis (Kalahari) 46 H. BURDA et al. Karyotype In addition to the results of allozymic studies (FıLıppuccr et al. 1994, 1997), also distinct numbers and morphology of the chromosomes substantiate distinguishing of Cryptomys anselli sp. nova and C. kafuensis sp. nova from other species of the genus and from each other. The variation in the number of biarmed autosomes in Cryptomys anselli sp. nova is presumably due to changes in the number of heterochromatic arms. Regarding the diffi- culty in establishing homologies between the affected pairs according to the G-banding pattern, it is probable that also other unknown mechanisms were involved. Differences in the number of biarmed autosomes probably result from additions and/or deletions of the C-heterochromatic arms. The variation is interindividual and no consistent differences were found between the specimens collected in different localities. The G-banding pat- tern in the metacentric autosome with the whole-heterochromatic arm in Cryptomys an- selli sp. nova seems similar to the analogous chromosome in C. kafuensis sp. nova. The similar fundamental numbers of chromosomal arms found in Cryptomys anselli sp. nova, C. kafuensis sp. nova, C. darlingi, and C. mechowi (cf. Tab. 2) suggest Robertso- nıan rearrangements as a possible mechanısm of chromosome speciation and indicate that different chromosomal fusions might have taken place in the evolution of individual lineages. Quantitative heterochromatin changes certainly played an important role in kar- yotype differentiation in this group. This is demonstrated also by an unusual extent of in- terindividual heterochromatin variation within the Cryptomys anselli sp. nova popula- tions. A large metacentric autosome with the whole-heterochromatic arm is apparently stable in the 2n =58 karyotype; however, its presumable homologue in the 68-chromo- some karyotype is polymorphic. Unfortunately, the high chromosome number and low G-band resolution level achieved in the preparations studied do not allow direct comparison between both karyo- types, or between them and other karyotypes known to date in the genus Cryptomys. Taxonomy of Zambian Cryptomys Following species and subspecies of Cryptomys have been formally described and named by previous authors from (what is now) Zambia and across borders (cf. ALLEn 1939, and see Eie>))): 1. Cryptomys darlingi (Thomas, 1895) from surroundings of Harare (Salisbury), Zim- babwe, reference grid 1731-C. Cryptomys darlingi was considered a subspecies of C. hottentotus by HoNnEycUTT et al. (1991). Although having the same chromosome num- ber (2n = 54), composition of its karyotype is distinctly different (cf. Tab. 2) and warrants a species status (AGUILAR 1993). The karyotype is distinct from karyotypes of Zambian Cryptomys studied to date. 2. Cryptomys micklemi (Chubb, 1909) from the Kataba river region, reference grid 1523-A, was considered a subspecies of C. damarensis by all subsequent authors. The anı- mals from the type locality should be examined to check their taxonomic status. 3. Cryptomys molyneuxi (Chubb, 1908) from Luano Valley. This is a valley through which the combined Lunsemfwa and Mulungushi rivers flow after breaking through the Muchinga Escarpment (reference grids 1429-C to 1430-C, the exact type locality remains unknown - cf. AnsELL 1978). A. SCHARFF (1996 unpubl. results) has found no evidence of Cryptomys in the reference grid 1429, where Cryptomys is obviously replaced by Helio- phobius (cf. also AnseıL 1978). This would imply that the type locality has to be searched for in the eastern part of the Valley, actually nearer to the type locality of C. amatus than to C. anselli sp. nova. It should be noted that all the subsequent authors have considered C. molyneuxi a synonym of C. amatus. New species of Cryptomys 47 Fig. 5. Type localities of Cryptomys taxa described from (what is now) Zambia and extralimitally. H=C. darlingi, D=C. damarensis micklemi, K=C. kafuensis sp. nova, L=C.anselli sp. nova, ?=C. hottentotus molyneuxi=C. amatus, A=C. amatus, N=C. mechowi (2n=40, from Ndola), M = C. mechowi mellandi, W = C. whytei. 4. Cryptomys amatus (Wroughton, 1907) from the Alala Plateau (reference grid 1330- C) was considered a subspecies of C. hottentotus by subsequent authors. FAULKES et al. (1997) and BENNETT and co-authors in their studies on Cryptomys contributed to the puz- zle in calling Cryptomys from Lusaka C. h. amatus, even when citing our studies, implying that this is aname used by us to denominate Lusaka populations. This is, however, not true as we have reported these mole-rats as Cryptomys sp. (2n = 68, Lusaka population) and always mentioned the taxonomic and nomenclature problems. Recently, we have col- lected mole-rats from the type locality of C. amatus and showed that they are different from Cryptomys from Lusaka and from C. hottentotus and deserve a species status of their own (MACHOLAÄN et al. 1998). 5. Cryptomys mellandi (Thomas, 1906) from Mpika (reference grid 1131-C) was con- sıdered a subspecies or synonym of the giant mole-rat, C. mechowi. We have collected C. mechowi in Ndola (refrence grid 1328-B). It has still to be checked whether the Ndola giant mole-rats and C. (mechowi) mellandi are taxonomically identical. Giant mole-rats are not only morphologically (body size) but also karyologically (MACHOLAN et al. 1993), though less allozymatically (FırLıppuccı et al. 1997), distinct from the smaller forms of Cryptomys. 6. Cryptomys whytei (Thomas, 1897) from Karonga, Malawi (reference grid 0933-D) was considered a subspecies of C. hottentotus by subsequent authors. Although we have not examined mole-rats from the type locality, we have studied karyotypes of single indivi- duals from Kasama (ref. grid 1031-A) (BURDA and KAwWALIKA unpubl.) and from Malawian Nyika (1033-B) (BURDA and CHITAUKALI unpubl.). Animals from both localities are chro- mosomally clearly distinct from each other and from all other Cryptomys studied to date. Based on these facts we can exclude the possibility that Cryptomys anselli sp. nova from Lusaka and Cryptomys kafuensis sp. nova from Itezhi-Tezhi would represent just sy- nonyms of already described species or subspecies. 48 H. BURDA et al. Speciation “hotspot” in Zambia? The earlier studies of karyotypes in bathyergids indicated, in contrast to the situation in many other subterranean rodents (particularly spalacids and ctenomyids) remarkable chromosome stability and conservatism. Thus, only one karyotype (2n = 60, GEORGE 1979) was described in the eusocial naked mole-rat (Heterocephalus glaber), distribution of which covers 14 latitude degrees; two karyotypes (2n = 60, GEORGE 1979; 2n = 62, own unpubl. data) are known in solitary Heliophobius argenteocinereus, distributed across 18 latitude degrees, and three chromosome species of Cryptomys were defined in the South- ern African subregion, covering about 17 latitude degrees: 2n=78 (or 74) in C. damarensis and 2n = 54 in C. hottentotus (Nevo et al. 1986); and 2n = 54 in C. darlingi (AGUILAR et al. 1993). Contrary to those earlier findings on bathyergids from other regions of Africa, only in Zambia, within a relatively narrow belt covering 3 degrees of latitude, we have identified already four distinct karyotypes, representing four different species of Cryptomys: 2n=40 (MAcHOLAN et al. 1993), 2n =50 (MACHOLAN et al. 1998), 2n =58, and 2n = 68 (present study). Since only few populations were studied within the given belt and since Zambia itself extends from north to south over ten latitudes, many more karyotypes are expected to occur there (and our pilot studies confirm this prediction). Systematic faunis- tic, taxonomic, and ecological study of Cryptomys in Zambia (and neighbouring Malawi) will be of high interest for assessment of chromosomal evolution in this “hotspot” region and its historical/ecological causes, compared to relative stability in the Southern Africa subregion. Acknowledgement We thank D. Kock for his comments on the manuscript and his advice. Assistance in the field by J. KockA and late S. Krunic is highly appreciated. This study was partly supported by a travel grant from the DAAD to A. S., the grant (no. VS97102) of the Ministry of Education of the Czech Repub- lic to J. Z. and M. M., and the grant from the research pool of the University of Essen to H.B. Zusammenfassung Die Karyotypen von Cryptomys anselli sp. nova und Cryptomys kafuensis sp. nova: neue Arten des Graumull von Sambia (Rodentia, Bathyergidae) Zwei neue Arten von Graumullen, Cryptomys (Rodentia: Bathyergidae), der „Kleinform” (um ca. 90 g) werden von Sambia beschrieben: C. anselli sp. nova, charakterisiert durch die diploide Chromo- somenzahl ?2n = 68 von der Lusaka-Provinz, und C. kafuensis sp. nova mit 2n =58 Chromosomen von Itezhi-Tezhi, Kafue-Nationalpark, Süd-Provinz. Konventionell gefärbte Karyotypen, einschließlich der C- und G-Bänderungsmuster, als auch die Lokalisation der NORs werden für die beiden Arten be- schrieben. Während die klassischen morphologischen und morphometrischen Merkmale eine Artdiag- nose bei der Gattung Cryptomys nicht ermöglichen, unterscheiden die Karyotypen und Allozyme die beiden neuen Arten voneinander und von anderen bekannten Arten der Gattung Cryptomys. Die Dicke der Außenwand (verglichen mit der Breite der Öffnung) des Foramen infraorbitale scheint art- spezifisch zu sein: die Wand ist dicker bei allen untersuchten Exemplaren von C. anselli sp. nova und dünner bei C. kafuensis sp. nova. New species of Cryptomys 49 References AGUILAR, G. H. (1993): The karyotype and taxonomic status of Cryptomys hottentotus darlingi (Roden- tia: Bathyergidae) S. Afr. J. Zool. 28, 201-204. ALLEN, G.M. (1939): A checklist of African mammals. Bull. Mus. Comp. Zool. Harvard Coll. 83, 425-433. ANSELL, W. F. H. (1978): The mammals of Zambia. Chilanga, Zambia: National Parks and Wildlife Ser- vice, 1-237. BEGALL, S. (1997): The application of the Gompertz model to describe body growth. Growth, Dev. Aging 61, 61-67. BEGALL, S.; BURDA, H. (1998): Reproductive characteristics and growth rate in the eusocial Zambian common mole-rat (Cryptomys sp., Bathyergidae). Z. Säugetierkunde 63, 297-306. BRÜCKMANN, G.; BURDA, H. (1997): Hearing in blind subterranean Zambian common mole-rats (Cryp- tomys sp., Bathyergidae, Rodentia). J. Comp. Physiol. A 181, 83-88. BURDA, H. (1989): Reproductive biology (behaviour, breeding, and postnatal development) in subterra- nean mole-rats, Cryptomys hottentotus (Bathyergidae). Z. Säugetierkunde 54, 360-376. BuUrDA, H. (1990): Constraints of pregnancy and evolution of sociality in mole-rats. J. Zool. Syst. Evol. Research 28, 26-39. BURDA, H. (1995): Individual recognition and incest taboo, and not reproductive suppression by parents in eusocial common mole-rats. Experientia 51, 411-413. BURDA, H.; BRUNS, V.; HickMANn, G. C. (1992): The ear in subterranean Insectivora and Rodentia in comparison with ground-dwelling representatives. I. Sound conducting system of the middle ear. J. Morphol. 214, 49-61. BURDA, H.; MARHOLD, S.; WESTENBERGER, T.; WILTSCHKO, W.; WILTSCHKOo, R. (1990): Magnetic compass orientation in the subterranean rodent Cryptomys hottentotus (Bathyergidae, Rodentia). Experien- tia 46, 528-530. CREDNER, S.; BURDA, H.; LUDESCHER, F. (1997): Acoustic communication underground: Vocalization characteristics in subterranean social mole-rats (Cryptomys sp., Bathyergidae). J. Comp. Physiol. A 180, 245-255. ELLERMANN, J. R. (1940): The families and genera of living rodents. London: Brit. Mus. Nat. Hist. FAULKES, C. G.; BENNETT,N. C., BRUFORD, M. W.; O’BRIEnN, H.P., AGUILAR, G.H.; Jarvıs, J. U.M. (1997): Ecological constraints drive social evolution in the African mole-rats. Proc. R. Soc. Lond. B 264, 1619-1627. FıLıppuccı, M. G.; BURDA, H.; NEvo, E.; KockA, J. (1994): Allozyme divergence and systematics of com- mon mole-rats (Cryptomys, Bathyergidae, Rodentia) from Zambia. Z. Säugetierkunde 59, 42-51. FıLıppuccı, M. G.; KAWALIKA, M.; MACHOLAN, M.; SCHARFF, A.; BURDA, H. (1997): Allozyme differentia- tion and taxonomic status of Zambian giant mole-rats, Cryptomys mechowi (Bathyergidae, Roden- tia). Z. Säugetierkunde 62, 172-178. GEORGE, W. (1979): Conservatism in the karyotypes of two African mole rats (Rodentia, Bathyergidae). Z. Säugetierkunde 44, 278-285. HOoNnEYCUTT, R. L.; ALLARD, M. W.; EDWARDS, S. V.; SCHLITTER, D. A. (1991): Systematics and evolution of the family Bathyergidae. In: The Biology of the Naked Mole-Rat. Ed. by P. W. SHERMAN, J. U. M. Jarvıs, and R. D. ALEXANDER. Princeton, New Jersey: Princeton Univ. Press, Pp. 46-65. HoNEYCUTT, R. L; EDWwARDS, S. V.; NELSoN, K.; NevoEE. (1987): Mitochondrial DNA variation and the phylogeny of African mole-rats (Rodentia: Bathyergidae). Syst. Zool. 36, 280-292. Howe, W. M.; BLAcK, D. A. (1980): Controlled silver-staining of nucleolus organizer regions with a protective colloidal developer: a one-step method. Experientia 36, 1014-1015. JANECEK, L. L.; HONEYCUTT, R. L.; RAUTENBACH, 1. L.; ERASMUS, B. H.; REIG, S.; SCHLITTER, D. A. (1992): Allozyme variation and systematics of African mole-rats (Rodentia: Bathyergidae). Biochem. Syst. Ecol. 20, 401416. Kössı, M.; FRANK, G.; BURDA, H.; MÜLLER, M. (1996): Acoustic distortion products from the cochlea of the blind African mole rat, Cryptomys spec. J. Comp. Physiol. A 178, 427-434. LINDENLAUB, T.; BURDA, H. (1993): Morphometry of the vestibular organ in neonate and adult African mole-rats Cryptomys species. Anat. Embryol. 188, 159-162. LINDENLAUB, T.; BURDA, H. (1994): Functional allometry of the semieircular ducts in subterranean mole-rats Cryptomys (Bathyergidae, Rodentia). Anat. Rec. 240, 286-289. LINDENLAUB, T.; BURDA, H.; NEvo, E. (1995): Convergent evolution of the vestibular organ in the sub- terranean mole-rats, Cryptomys and Spalax, as compared with aboveground rat, Rattus. J. Morphol. 224, 303-311. 50 H. BurDpA et al. MACHOLAN, M.; BURDA, H.; ZIMA, J., MiIsEkK, I.; KAwALikA, M. (1993): Karyotype of the giant mole-rat, Cryptomys mechowi (Rodentia, Bathyergidae). Cytogenet. Cell. Genet. 64, 261-263. MACHOLAN, M.; SCHARFF, A.; BURDA, H.; ZIMA, J.; GRÜTJEN, O. (1998): The karyotype and taxonomic status of Cryptomys amatus (Wroughton, 1907) from Zambia (Rodentia, Bathyergidae). Z. Säuge- tierkunde 63, 186-190. MARHOLD, S.; NAGEL, A. (1995): The energetics of the common mole rat Cryptomys, a subterranean eu- social rodent from Zambia. J. comp. Physiol. B 164, 636-645. MARHOLD, S.; WILTSCHKO, W.; BURDA, H. (1997): A magnetic polarity compass for direction finding in a subterranean mammal. Naturwissenschaften 84, 421-423. Misek, 1.; TicHYy, F.; BURDA, H. (1996): SEM-structure of the olfactory epithelium in newborn mole-rat (Cryptomys sp., Bathyergidae, Rodentia). Acta Vet. Brno 65, 321-328. MÜLLER, M.; BURDA, H. (1989): Restricted hearing range in a subterranean rodent, Cryptomys hottento- tus (Bathyergidae). Naturwissenschaften 76, 134-135. MÜLLER, M.; LAUBE, B.; BURDA H.; Bruns, V. (1992): Structure and function of the peripheral auditory system in the African mole rat (Cryptomys hottentotus): Evidence for a low frequency acoustic fo- vea. J. Comp. Physiol. A 171, 469-476. NEVvo, E.; BEN-SHLOMO, R.; BEILES, A.; JARVIS, J. U. M.; HicKkMANn, G. C. (1987): Allozyme differentiation and systematics of the endemic subterranean mole rats of South Africa. Biochem. Syst. Ecol. 15, 489-502. NEVvo, E.; CAPANNA, E.; CoRTI, M.; JARVIS, J. U. M.; HicKMANn, G. C. (1986): Karyotype differentiation in the endemic subterranean mole-rats of Sourth Africa (Rodentia, Bathyergidae). Z. Säugetierkunde 51, 36-49. NOWAK, R. M. (1991): Walker’s Mammals of the World. 5th ed. Baltimore, London: John Hopkins Univ. Press. OELSCHLÄGER, H. A.; BURDA, H. (1992): LHRH-Immunocytochemistry in the nervus terminalis of mammals. In: Chemical Signals in Vertebrates VI. Ed. by R.L. Dorty and D. MÜLLER-SCHWARZE. New York: Plenum Press, Pp. 31-35. ROSEVEAR, D. R. (1969): The Rodents of West Africa. London: British Mus. Nat. Hist. SCHARFE, A.; BURDA, H.; TENORA, F.; KAWALIKA, M.; BARUS, V. (1997): Parasites in social subterranean Zambian mole-rats (Cryptomys spp., Bathyergidae, Rodentia). J. Zool. (London) 241, 571-577. SCHARFE, A.; TENORA, F.; KAWALIKA, M.; BARUS, V.; BURDA, H. (1996): Helminths from Zambian mole- rats (Cryptomys, Bathyergidae, Rodentia). Helminthologia 33, 105-110. SEABRIGHT, M. (1971): A rapid banding technique for human chromosomes. Lancet 2, 971-972. SUMNER, A. T. (1972): A simple technique for demonstrating centromeric heterochromatin. Exptl. Cell. Res. 75, 304-306. WILLIAMS, S. L. D.; SCHLITTER, D. A.; RoBBIns, L. W. (1983): Morphological variation in a natural popu- lation of Cryptomys (Rodentia: Bathyergidae) from Cameroon. Ann. Mus. Roy Afr. Centr. Sc. Zool. 237, 159-172. WILLINGSTORFER, W.; BURDA, H.; WINCKLER, J. (1998): Ovarian growth and folliculogenesis in breeding and non-breeding females of a social rodent, the Zambian common mole-rat, Cryptomys sp. J. Mor- phol. 237, 33—41. Authors’ addresses: HynEK BURDA and ANDREAS SCHARFE, Department of General Zoology, ; FB 9 - Biology, University of Essen, D-45117 Essen, Germany; Jan ZımA and MıLos MA- CHOLAN, Laboratory of Genetics and Embryology, Institute of Animal Physiology and Genetics, Academy of Sciences of the Czech Republic, Veveri 97, CZ-60200 Brno, Czech Republic; MATHIAS KAWALIKA, P.O. Box 73796, Ndola, Zambia. Z. Säugetierkunde 64 (1999) 51-53 = FÜ © 1999 Urban & Fischer Verlag SÄUG ET] ERKÜUN BE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY WISSENSCHAFTLICHE KURZMITTEILUNGEN Spacing among Harbour seals (Phoca vitulina vitulina) on haul-out sites in the Wadden Sea of Niedersachsen By ILonA M. TRAUT, EDITH H. Rıes, Britta DonAr, and E. VARESCHI Department of Aquatic Ecology, University of Oldenburg, Oldenburg, Germany and Institute for Forestry and Nature Research, Department of Aquatic Ecology, Den Burg, The Netherlands Receipt of Ms. 25. 05. 1998 Acceptance of Ms. 16. 10. 1998 Key words: Phoca vitulina, spacing, haul-out, Wadden Sea Harbour seals haul out on sandbanks, on rocks or on ice, where they form loosely gre- garious herds with little visible social interaction (BıGG 1969; GoDsELL 1988; DA SıLva and TERHUNE 1988). In contrast to other pinniped species, such as elephant seals or walrus, they avoid close body contact when hauled out and defend their individual space by ag- gressive behaviour (BıGG 1981). Their grouping together is believed to be an anti-preda- tor adaptation, granting more safety as well as more rest for the individual (TERHUNE 1985). In the Wadden Sea no natural predators are known and hunting was stopped in 1972. However, during the summer months most of the Wadden Sea is used intensively as a re- creational area. This leads to high antrophogenic disturbance pressure at a vulnerable time when the seals breed and moult. The objective of the present study was to document the spacing of harbour seals in order to provide basic information for future conservation measures. In 1989 and 1990 a total of 17 survey flights were performed in Niedersachsen. All seal groups on the sandbanks were photographed at altitudes of 150 m. A Canon AE1 and Nikon F801 with a zoom of 35-200 mm and 100-200 ASA films were used. The short- est distance of each seal to its nearest neighbour was determined in “seal length” (SL). The length of the 3 largest seals was averaged on each projected slide, one SL being ap- proximately 1.5 m (TrAuT 1997). Within each group, the distance to the nearest neighbour was classified as <1 SL, 1-2 SL, >2-3 SL, >3-5 SL and >5-10 SL. The midpoints of the classes were used for further calculations. All seals lying up to 10 SL apart were defined as belonging to one group. To investigate group size related variations in spacing, groups of 2-10, 11-20, 21-30, 31-50 seals (n = 20 each) and of >50 seals (n = 13) were analysed (Spearman rank correlation). Regional differences were investigated by comparing groups of 50-60 seals in the east, middle, and west (n = 4 groups for each region) of the study area. Because mother-pup pairs keep close body contact, they were regarded as one unit. Therefore, distances to the nearest neighbour were measured from the mother seal only. The orientation of seals in relation to the waterline was examined by allocating the direction of the seals body to one of four 90° segments. The segments were numbered clockwise, with segment I (315°-45°) being opposite the waterline, segment II (45°-135°) and segment IV (225°-315°) being sideways. A seal facing the water would be allocated to segment III (135°-225°). 0044-3468/99/64/01 — 051 $ 12.00/0 32 ILona M. TRAUT et al. The spacing of hauled out harbour seals resulted in an average distance of 2.7 +0.5 SL (n = 2584) in 1990 and 2.5 +#0.2 SL (n = 2094) in 1989. In general, we found that 27% of the seals kept a distance of less than 1 SL, 42% were 1-3SL, and 31% were more than 3SL apart. We found that mother-pup pairs remained within the seal groups, but they kept a distance of 3.8 +2.1 SL (n = 439) on average. In our study area the seal groups were predominantly small with an average of 11 +16 seals (range 1-138 seals). The influence of group size on spacing was investigated by comparing the average distance in groups of various sizes. In groups of more than 50 seals the individuals were significantly (rs =-0.6) closer to each other than in smaller groups (3.2SL for <10seals, 2.4-2.5 SL for 11-20, 21-30, 31-50 seals and 1.9SL for >50 seals). Regional differences were ınvestigated by comparing groups of similar size (50- 60 seals) on haul-out sites in the east, middle, and west of the study area. In the western part the individual distance was 1.1+0.2SL on average. About double the distance was found in the middle and the eastern part with 2.4+0.3SL and 2.9 #0.8 SL, respectively. The seals occupied only a fraction of the emerged sandbanks. Generally they hauled out in a single line along the water edge. Often track marks in the sand indicated that the seals had moved from higher positions towards the waterline. We found that 54.6% of the seals (n = 3364) were facing the water (section II). Seals in our study area kept an average distance of 2.5 SL (approximately 3.7 m) to their nearest neighbours. This varied in different areas, with seals being closer together in the western part of Niedersachsen. Even though quantification of disturbance pressure was not subject of thıs study, there is evidence, that especially in this area disturbance pressure is quite high during summer (TrAUT 1997). In Schleswig-Holstein, BAcH and Crauss (1989) reported a decrease in spacing with an increase of disturbance level. Female seals with pups did not separate from the group but the distance to other seals was larger. This is not surprising since it is known that females with pups are more sus- ceptible to disturbances and react more aggressively than other seals (DRESCHER 1979). We found an inverse relationship of individual spacing to group size in that seals were more densely packed in large groups. Similar results were found in a study in Schleswig- Holstein, where individual spacing decreased with increasing seal numbers (BAcH and Crauss 1989). Since we did not observe a limitation of haul-out space, the closer group- ing together could be related to higher disturbance pressure. When anımals are closer packed, alarm signals from other seals could be detected faster. In Canada, large groups always had at least one vigilant animal, whereas groups of eight or fewer seals were ob- served to have no scanning individuals on occasion (DA SıLva and TERHUNE 1988). Taking this information, we can presume that with an average of 11 seals in Niedersachsen, at least always one of the seals is scanning the area. In our study about half of the observed seals were facing the water, which is higher than one would expect if the seals were or- iented randomly. This could be another indication for disturbance pressure because it al- lows the seals a fast escape into the water. However, in order to quantify this, comparison with undisturbed areas would be needed, which, at this stage, we do not have. Disturbance is an important criteria of habitat quality. If the level of disturbance influ- ences the haul-out behaviour of seals, individual spacing as well as orientation could be used (additionally to other criteria) in judging the quality of haul-out sites. We are obliged to the Bezirksregierung Weser-Ems and Landesjägerschaft Niedersachsen for the per- mission to participate in their survey flights. We thank the people who helped photographing the seals. The project was financed by Niedersächsisches Ministerium für Wissenschaft und Kunst Kap. 0608, FT 74. Spacing among Harbour seals in the Wadden Sea of Niedersachsen 53 References BAcH, L.; CLauss, E. (1989): Studien zum Verhalten von Seehunden (Phoca vitulina L.) unter beson- derer Berücksichtigung der Auswirkung von Krankheiten und Störungen. In: Zoologische und ethologische Untersuchungen zum Robbensterben. Vorl. Endbericht, BMU, Vorh. Nr. 10805017/06. Pp. 233-271. Biss, M. A. (1969): The Harbour seal in British Columbia. Fish. Res. Board of Canada. Bull. 172, 33 pp. Biss, M. A. (1981): Harbour seal — Phoca vitulina and P. largha. In: Handbook of Marine Mammals. Seals. Vol. 2. Ed. by S. H. Rıpaway and R. J. Harrıson. London: Academic Press. Pp. 1-29. DA SILVA, J.; TERHUNE, J. M. (1988): Harbour seal grouping as an anti-predator strategy. Anim. Behav. 36, 1309-1316. DRrEsCHER, E. H. (1979): Biologie, Ökologie und Schutz der Seehunde im schleswig-holsteinischen Wat- tenmeer. Beiträge zur Wildbiologie 1. Meldorf: LJV Schleswig Holstein. GoDSELL, J. (1988): Herd formation and haul-out behaviour in harbour seals (Phoca vitulina). J. Zool. (London) 215, 83-98. TERHUNE, J. M. (1985): Scanning behavior of harbor seals on haul-out sites. J. Mammalogy 66, 392-395. TRAUT, I. M. (1997): Das aktuelle Verhalten von Seehunden (Phoca vitulina vitulina) im heutigen Wat- tenmeer. Diss. thesis, University of Oldenburg. Authors’ addresses: Dr. ILoNA M. TRAUT, BRITTA DoNAT, Prof. Dr. EKKEHARD VARESCHI, Department of Aquatic Ecology, University of Oldenburg, P.O. Box 2503, D-26111 Oldenburg, Germany; EDITH H. Rıss, Institute for Forestry and Nature Research, Department of Aquatic Ecology, P.O. Box 167, NL-1790 AD Den Burg, The Netherlands Z. Säugetierkunde 64 (1999) 54-58 ZEITSCHRIFT ”, © 1999 Urban & Fischer Verlag SÄUGETI ERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Observations on the occurrence of Irrawaddy dolphin, Orcaella brevirostris, in the Mahakam River, East Kalimantan, Indonesia By DANIELLE KREB Institute for Systematics and Population Biology, Zoological Museum, University of Amsterdam, Amsterdam, The Netherlands Receipt of Ms. 16. 06. 1998 Acceptance of Ms. 30. 10. 1998 Key words: Orcaella brevirostris, Irrawaddy dolphin, Mahakam River, encounter rates, distribution The Irrawaddy dolphin, Orcaella brevirostris (Gray, 1866), is considered a ‘facultative’ riv- er dolphin of which distinct riverine and coastal, marine populations exist. The species is mainly found in shallow coastal waters of the tropical Indo-Pacific, but also in major river systems, in particular: Irrawaddy, Mekong, Mahakam, and the estuaries of the Ganges and Brahmaputra (THomAs 1892; LLozE 1973; LEATHERWOOD et al. 1984; MARSsH et al. 1989). Relatively few published studies exist pertaining specifically to the population of Irra- waddy dolphins, in the local vernacular referred to as Pesut, in the Mahakam River, East Kalimantan, Indonesia. Studies so far have focused on the distribution and daily move- ment pattern of the species in Semayang-Melintang Lakes and connecting Pela and Me- lintang tributaries (Prıvono 1994) and on bioacoustics (KAMMINGA et al. 1983). Although no systematic surveys on their abundance have been conducted so far, the Indonesian Di- rectorate General of Forest Protection and Nature Conservation reported the existence of a population of 100-150 individuals for Semayang Lake, Pela River, and adjacent Ma- hakam River (Tas’an and LEATHERWOooD 1984) while an unpublished estimate of 68 in- dividuals in the Mahakam River was reported by Prıyono (1994). In this study, I present results of a preliminary survey, which was conducted on the Mahakam River, its tributaries, and adjacent lakes in East Kalimantan, Indonesia. Two surveys were conducted, both at medium to low waterlevels, the first from 27 February till 9 March 1997 and the second from 21 March till 6 April 1997. The river was surveyed using a small motor boat, occasionally by large public boat and by large and small motor- ized canoes, from Muara Kaman (ca. 200 km upstream) to Burit Hacau, at the rapids past Long Bagun (ca. 600 km upstream). In addition, the Semayang, Melintang, and Jempang Lakes were surveyed as well as the Pela, Melintang, and Kedang Pahu tributaries. The to- tal survey length was 1085 km. For analysis of sighting frequencies, the river was divided into a lower (from Samarinda, ca. 100 km upstream, until Muara Kaman), middle (from Muara Kaman until Long Iram, ca. 490 km upstream), and upper section (from Long Iram until the rapids after Long Bagun). Tributaries and lakes surveyed were also ana- lysed separately. Encounter rates were calculated for each section by dividing the number of observed dolphins by the number of kilometers searched. For testing whether the sighting frequen- cies are homogeneously distributed over all sections, and whether significant differences exist between different sections, G-tests of goodness of fit for single classification fre- 0044-3468/99/64/01 — 054 $ 12.00/0 Irrawaddy dolphins of the Mahakam River, East Kalimantan, Indonesia 55 quency distributions were used. To obtain a better approximation to x”, Williams’ correc- tion to G was applied (G,aj; SOKAL and RoHLr 1981). G values were compared with criti- cal values of the chi-square distribution (table C in SIEGEL and CASTELLAN 1988). Because multiple tests were performed, a corrected alpha of 0.01 was used in place of the nominal alpha of 0.05 (Rıce 1989). Dolphins were spotted by eye and by means of binoculars. Group composition, location, diving times, respiration rates, and behaviors were recorded and photos taken. Additional data on the occurrence and status of Pesut were collected by interviewing local inhabitants, mainly fishermen. During the present study, a total of 32 dolphins were observed, of which four were ju- veniles. During the first survey 29 individuals were encountered while during the second only 3 were observed, presumably because more time was spent in the upper section of the Mahakam, where no dolphins were observed. Group size varied from 3 to 7 animals with a median group size of 4 individuals. No minimum estimate of abundance could be made as only three dolphins were identifiable individually on the basis of their dorsal fin (no systematic photos of their dorsal fin were made). Also, there is the possibility that the dolphins might have been encountered more than once during each survey, in case they were heading in the same direction during the night as we were heading during the day. Irrawaddy dolphins were found to be rather inconspicuous; they do not leap high out of the water and may stay submerged for up to 12 minutes, surfacing only briefly. Except for some noises produced with their blow holes, which could be heard over 100 m dis- tance, no audible whistles or pure tones were heard. Pesuts appeared to be very social, continously staying in close contact with one another, regardless of whether they were milling (feeding), travelling, or resting. Table 1 shows the encounter rates, i.e. the number of dolphins per km of river searched, for different sections of the Mahakam River sytem. The dolphins are not homo- genously distributed over the whole length of different river sections, tributaries, and lakes (G,a; = 47.8, df=4, p< 0.01). The encounter rates of the middle river section are significantly higher than those of the upper section (Gaa; = 39.2, df = 1, p< 0.01). Signifi- cantly higher encounter rates were also found for the tributaries when compared to the combined main river sections (Gaga; = 8.3, df = 1, p< 0.01). However, all tributary observa- tions of Pesut were made in the relatively short Pela trıbutary (only 8 km search effort), a connecting tributary to Semayang Lake and the Mahakam River. No sightings were made in the longer tributary Kedang Pahu of which 65 km in total was searched. No significant differences were found between encounter rates of middle river section and tributaries. As all tributary observations were made in the Pela trıbutary connecting to the middle section of the main river, and observations in the middle section of the Mahakam were significantly higher than in the upper section (with a higher search effort), this section presumably forms the primary habitat for the dolphins, when waterlevels are medium to low. Encounter rates for the Semayang and Melintang Lakes, though lower, were not sig- nificantly so, when compared to the combined rates of the river and tributaries Table 1. Encounter rates — dolphins observed per km of river searched. Section Search effort N of individuals Encounter rate (km) N of indiv./km Lower Section Middle Section Upper Section Tributaries Lakes DANIELLE KREB 56 "EISIUOPUJ ‘ueJueunjes] Jse ‘I9ALg WEyeyeN OU} UI S14SO1A2A1q 0]]20940 Jo uonngLnsIq °T '31y uop>wony] 001 08 09 07 07 JJUAaUN))O JNOIE UONEBUHOJUT JANEZIN OIgefIEAB uoneunoJur OU JO [npqnod JUALMII0 JNOIE UONEULIOJUT JATISOA 9 F vanıuvwvs®% F > ! E AN UONEAIISIO [ENJOY ANYOI1I (ueyuewiey4) OINHOA 4 —— — — Irrawaddy dolphins of the Mahakam River, East Kalimantan, Indonesia I (Gaaj = 3.9, Go.oı = 6.6). The significant difference in encounter rates between these sec- tions is probably a result of treating dolphins sightings in the Pela tributary as tributary observations. However, the dolphins’ presence in either the Pela tributary or in Se- mayang Lake might depend on time of the day, as the dolphins are reported to migrate daily between these areas (Prıvono 1994). The absence of observations of dolphins in the lakes most certainly is due to the fact that only 50km were surveyed of the 10.300 hectares and 8.900 hectares large Semayang and Melintang Lakes, respectively. No significant differences in encounter rates were found between lower and other river sec- tions, possibly due to the low search effort in this section. The encounter rates found for Orcaella brevirostris in the Mahakam River are in the same order of magnitude as that reported for Lipotes vexilifer in the Yangtze River (0.09 dolphins/km), a population considered to have a high exctinction risk (HuA and CHEN 1992). However, the encounter rate of 0.06 dolphins/km in the mainstem Mahakam River, is considerably lower than those recorded, at similar medium-low water level con- ditions, for /nia geoffrensis and Sotalia fluviatilis in the mainstems of the Amazon- Marafon-Ucayali (0.18 and 0.27 dolphins/km, respectively; LEATHERWOoD 1996). In the present study, Pesuts were observed up till Tering, 400 km upstream (Fig. 1), but they are said to occur up till the waterfalls after Long Bagun. Although no sightings were made in any of the lakes visited, Pesut has frequently been recorded in Semayang and Melintang Lakes (TAs’an and LEATHERWOooD 1984), but the dolphins are said to be absent from Jempang Lake. Whether the Pesut occurs between Samarinda (near the mouth of the river) and the open sea, and in which of the river’s tributaries, remains un- clear. When water levels are high, dolphins are often observed by local inhabitants high up the Kedang Pahu tributary, past the village of Damai. Although the dolphins always moved away from our research vessel, they were observed twice near two villages (Muara Pahu and Tering) with high levels of boat traffic. According to local fishermen, they were said to frequent these places almost on a daily basis, presumably because of the higher availability of fish. In conclusion, the results from this preliminary survey seem to indicate that encounter rates of the Irrawaddy dolphin in the Mahakam River are relatively low and fall in the same class of those recorded for the seriously threatened Lipotes vexilifer. Furthermore, middle sections of the river seems to be the primary habitat of Pesut, at least at medium to low water levels. Given the many factors contributing to possible deterioration of dol- phin habitat (e.g. pollution from mining, forest fires, logging, and siltation), these obser- vations of low encounter rates merit further study. Acknowledgements I wish to thank the East Kalimantan nature conservation authorities, sub Balai Konservası Sumber Daya Alam, WWF-Indonesia, and WWF-Netherlands for their support and cooperation. My special thanks go to Ir. A. M. RACHMAT, D. SUPRIJONO, F. BUTCHER, and boatsman PAK SEGA, who all partici- pated in the survey. I would like to thank Dr. P. J. H. van BRrEe for all his help and support throughout the study, as well as for his comments on the manuscript. Dr. C. J. HAZEVoET, Dr. A. ©. Mooers, V. Nijman, and an anonymous reviewer are also thanked for their comments on this manuscript. I. Lysenko and the World Centre for Monitoring Cambridge are thanked for drawing of the map. References HUA, Y.; CHEN, P. (1992): Investigation for impacts of changes of the lower reach of Gezhou Dam be- tween Yichang and Chenglingji on the Baiji, Lipotes vexilifer after its key water control project founded. J. Fish. China 16, 322-329. 58 DANIELLE KREB KAMMINGA, C.; WIERSMA, H.; DUDoK VAN-HEEL, W. H. (1983): Investigations on cetean sonar VI. Sonar sounds in Orcaella brevirostris of the Mahakam River, East Kalimantan, Indonesia; First descrip- tions of acoustic behaviour. Aquatic Mammals 10, 83-94. LEATHERWOOD, J. S. (1996): Distributional ecology and conservation status of river dolphins (Inia geof- frensis and Sotalia fluviatilis) in portions of the Peruvian Amazon. Diss. thesis, Texas University, Texas. LEATHERWOOD, S.; PETERS, C. B.; SANTERRE, R.; CLARKE, J. T. (1984): Observations of cetaceans in the northern Indian Ocean Sanctuary, November 1980-May 1983. Rep. Int. Whal. Commn. 34, 509- s20: LLoZE, R. (1973): Contributions a l’Etude anatomique, histologique et biologique de /’Orcaella breviros- tris (Gray — 1866) (Cetacea-Delphinidae) du Mekong. Diss. thesis Toulouse, France. MARSH, H.; LLOZE, R.; HEINSOHN, G. E.; KAsuyA, T. (1989): Irrawaddy Dolphin Orcaella-brevirostris (Gray, 1866). In: Handbook of marine mammals. River dolphins and the larger toothed whales 4. Ed. by S. H. RıpgEwaAy and R. J. HARRISon. London: Academic Press, Pp. 101-118. PrIYono, A. (1994): A study on the habitat of Pesut (Orcaella brevirostris Gray, 1866) in Semayang-Me- lintang Lakes. Media Konservasi 4, 53-60. Rice, W. R. (1989): Analyzing tables of statistical tests. Evolution. Int. J. Org. Evol. 43, 223-223. SIEGEL, $.; CASTELLAN, N. J., Jr. (1988): Nonparametric statistics for the behavioral sciences. Sec. ed. New York: McGraw-Hill. SOKAL, R. R.; RoLr, F. J. (1981): Biometry. The principles and practice of statistics in biological re- search. Sec. ed. New York: W. H. Freeman. THomas, ©. (1892): Viaggio di L. Fea in Birmania e regioni vicine. XLI. On the Mammalia collected by Signor Leonardo Fea in Burma and Tenasserim. Annali del Museo Civico di Storia Naturale di Genova 1892, 913-949. TAs’An; LEATHERWOOD, S. (1984): Cetaceans live-captured for Jaya Ancol Oceanarium, Djakarta, 1974- 1982. Rep. Int. Whal. Commn. 34, 485-489. Author’s address: DANnIELLE KreEB, Mammal Department: (M. Sc), Institute for Systematics and Po- pulation Biology/Zoological Museum, University of Amsterdam, P.O. Box 94766, NL-1090 GT Amsterdam, The Netherlands Z. Säugetierkunde 64 (1999) 59-62 © 1999 Urban & Fischer Verlag a Ne ZEITSCHRIFT * a SÄUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY First cytogenetic analysis of the genus Bibimys (Cricetidae, Rodentia) By F. J. DYzEncHAUZ and ALıcıA I. MASSARINI Centro Nacional de Genetica Medica, Dpto. de Genetica Experimental and Grupo de Investigaciön en Biologia Evolutiva, Dpto. de Ciencias Biolögicas, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina Receipt of Ms. 07. 01. 1998 Acceptance of Ms. 20. 04. 1998 Key words: Bibimys, Scapteromyini, cytogenetics The South American Scapteromyini (Rodentia: Cricetidae), comprise three genera: Scap- teromys with two species (S. tumidus and S. aquaticus), Kunsia with two species (K. prin- cipalis and K. fronto), and Bibimys, also with two species (Bibimys torresi and Bibimys labiosus) (WALKER 1964; BRUM-ZORRILLA et al. 1986). \ ENTRE RIOS 33 : 34 2 % N ‘ N N 35 36 BUENOS AIRES 37 38 39 40 41 59 Fig. 1. Map of collection locality of Bibimys torresi. 0044-3468/99/64/01 — 059 $ 12.00/0 60 F. J. DYZENCHAUZ and ALIıcIA I. MASSARINI Fig. 2. Karyotype of Bibimys torresi. Giemsa stained (top), C-band pattern (bottom). Scapteromys and Bibimys share the same semi-aquatic niche, whereas the giant rat Kunsia inhabits a semi-subterranean niche. It is mentioned that, since the Scapteronyini share a great affinity with the Akodontini in the pattern of the molars, the adaptation to different habitats of Scapteromys, Bibimys, and Kunsia, might have developed from an akodontine ancestral stock which invaded the lowlands of the Chaco region and later on expanded towards the east (REıG 1984). Nowadays there are no records of living spe- cimens of Bibimys labiosus whereas Bibimys torresi has been collected in sites restricted to the delta of the Parana river in the Buenos Aires Province of Argentina. Neverthe- First cytogenetic analysis of the genus Bibimys 61 less, the finding of fossils of Bibimys specimens at Lagoa Santa, Brazil (Voss and MyERS 1991) and Cueva Tixi and Centinela del Mar, Argentina (PArDINAs 1995), shows that in the late Pleistocene and Holocene, the area of distribution of this genus was larger than today. The taxonomic relationships among the genera of the Scapteromyini, were based on the evaluation of morphological similarities. Until recently, cytogenetic information was available only for Scapteromys species, while the karyotypes of Kunsia and Bibimys were still unknown. Here we present the standard and C-banded karyotype of Bibimys torresi, a species described for the first time by MassoiIA (1979). Two females and one male of Bibimys were collected at Otamendi, Buenos Aires Pro- vince (Argentina), on an island of the delta of the Parana river (Fig. 1). A period of three years was necessary to collect these specimens, considering the fact that they are very rare to find. Skin and skull vouchers of the studied specimens were catalogued in the col- lection of mammals of the Mar del Plata Municipal Natural History Museum. Cytogenetic analysis was based on mitotic metaphase chromosomes from bone mar- row of animals previously injected with yeast (LEE and ELDER 1980). Standard karyotypes were stained with Giemsa and C-bands were performed according to Hsu (1974). The karyotype of Bibimys torresi shows a diploid number of 2n=70 and AN = 76. This karyotype comprises 30 pairs of acrocentric autosomes: one medium sized and the remaining small sized, and four pairs of small metacentric autosomes. The X and Y are small sized submetacentric chromosomes (LEvan et al. 1964) (Fig. 2, top). The C-band pattern performed in the females, shows a faint centromeric heterochromatic band in the X chromosome, whereas in the autosomes heterochromatin is absent except in pair 3 which presents a very faint intercallary C-positive band (Fig. 2, bottom). It is noteworthy that Bibimys torresi and Scapteromys species display a similar C- banded pattern, characterized by low amounts of heterochromatin, compared with that of other species of South American cricetid genera. This pattern was explained by a reduc- tion in the amount of satellite DNA (BRUM-ZORRILLA et al. 1986; FREITAS et al. 1984). The available cytogenetic data show that Scapteromys species karyotypes range from 2n = 24 to 2n = 36 (BRUM-ZORRILLA et al. 1986; FREITAS et al. 1984), whereas Bibimys tor- resi presents a markedly high chromosomal number. In this respect, previous reports sug- gest that in other South American cricetid-related lineages there are evidences of a direc- tional trend towards a reduction in the chromosomal number, in the course of chromosomal evolution. Therefore, the higher diploid numbers within a lineage may re- present the most primitive condition (GARDNER and PArton 1976). Hence, if the reduc- tional trend existent in other cricetid lineages counts also for the Scapteromyinı, then Bi- bimys would represent an ancestral form of this group. However, the karyotypes of Bibimys and Scapteromys differ so much in their diploid number (2n) and autosomal number (AN), that it is not possible to establish a compari- son between them. The feasibility of obtaining G-bands in karyotypes of Bibimys would help to determine homologies and rearrangements with other Scapteronyini species. Acknowledgements We dedicate this paper to the memory of Dr. O. A. REıG without whom this work would not have been feasible. We thank Dr. EsTELA BoNAVENTURA for providing the animals employed in this study, and Dr. E. MassoiA for his helpful comments. This research has been partially funded by CONICET (grant PID No 3085300), to ©. A. REıg and UBACYT (grant No EX228), to M. MuDkry. 62 F. J. DYZENCHAUZ and ALICIA I. MASSARINI References BRUM-ZORRILLA, N.; OLIVER, G.; GENTILE DE FRONZA, T.; WAIMBERG, R. (1986): Karyological studies of South American rodents (Rodentia: Cricetidae). I. Comparative chromosomic analysis in Scapter- omys taxa. Caryologia 39, 131-142. FREITAS, T. R. ©; MATTEVI, M. S.; OLIVEIRA, L. F. B. (1984): Unusual C-band patterns in three karyotypi- cally rearranged forms of Scapteromys (Rodentia, Cricetidae) from Brazil. Cytogenet. Cell Genet. 38, 3944. GARDNER, A.L.; PATToN, J. L. (1976): Karyotypic variation in Oryzomyine rodents (Cricetinae) with comments on chromosomal evolution in the neotropical cricetine complex. Occas. Pap. Mus. Zool. Lousiana St. Univ. 49, 1-48. Hsuv, T. C. (1974): Procedures for inducing C-bands and G-bands in mammalian chromosomes. Mam- malıan Chrom. Newsl. 15, 88-96. LEE, M. R.; ELDER, F. F. B. (1980): Yeast stimulation of bone marrow mitosis for cytogenetic investiga- tions. Cytogenet. Cell Genet. 26, 3640. LEvAn, A.; FREDGA, K.; SANBERG, A. A. (1964): Nomenclature for centromeric position on chromo- somes. Hereditas 32, 1-11. MassoIA, E. (1979): Description of a new genus and species: Bibimys torresi (Mammalia-Rodentia-Cri- cetidae-Sigmodontinae-Scapteromyini). Physis 38, 1-7. PARDINAS, U. F. J. (1995): Novedosos cricetidos (Mamalia: Rodentia) en el Holoceno de la Regiön Pam- peana Argentina. Ameghiniana 32, 197-203. Reıc, ©. A. (1984): Geographic distribution and evolutionary history of South American Muroids, Cri- cetidae: Sigmodontinae. Brazil. J. Genetics 7, 333-365. Voss, R.S.; Myers, P. (1991): Pseudoryzomys simplex (Rodentia: Muridae) and the significance of Lund’s Collections from the Caves of Lagoa Santa, Brazil. Bull. Amer. Mus. Nat. Hist. 206, 414- 432. WALKER, E. P. (1964): Mammals of the world. Baltimore: John Hopkins Press. Authors’ addresses: FERNANDO J. DYZENCHAUZ, Centro Nacional de Gen£tica Me&dica, Dpto. de Gene- tica Experimental, Av. Las Heras 2670, 4to. piso, 1425, Buenos Aires and ALICIA MASSARINI, Grupo de Investigaciön en Biologia Evolutiva, Dpto. de Ciencias Bio- lögicas, Universidad de Buenos Aires, Pabellön II, 4° piso, Cdad. Universitaria, Nunez, 1428 Buenos Aires, Argentina Z. Säugetierkunde 64 (1999) 63-64 ZEITSCH RIFT® ur "FÜ UR © 1999 Urban & Fischer Verlag SÄUGETI ERKU INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY MITTEILUNGEN DER GESELLSCHAFT 73. Jahrestagung der Deutschen Gesellschaft für Säugetierkunde vom 26.-30. September 1999 in Bad Dürkheim Einladung Auf Einladung des Direktors des Pfalz-Museums für Naturkunde, Herrn Dr. REINHARD FLÖSsseER, findet die 73. Jahrestagung der Deutschen Gesellschaft für Säugetierkunde von Sonntag, dem 26. September bis Donnerstag, dem 30. September 1999 im Pfalzmuseum für Naturkunde in Bad Dürkheim statt. Vorläufiges Programm Sonntag, 26. September: Anreise ab 16.00 Uhr: Vorstandssitzung ab 19.00 Uhr: Zwangloser Begrüßungsabend Montag, 27. September: 9.00 Uhr: Begrüßung 9.30 Uhr: Hauptvortrag und Kurzreferate zum Themen- schwerpunkt „Wildbiologie“ 14.00 Uhr: Posterdemonstration 15.00 Uhr: Kurzvorträge 16.00 Uhr: Mitgliederversammlung abends: Empfang durch die Stadt Bad Dürkheim Dienstag, 28. September: 9.00 Uhr: Hauptvortrag und Kurzreferate zum Themen- schwerpunkt „Sinnesphysiologie“ 14.00 Uhr: Posterdemonstration 15.00 Uhr: Kurzvorträge Geselliger Abend Mittwoch, 29. September: 9.00 Uhr: Hauptvortrag und Kurzreferate zum Themen- schwerpunkt „Ökologie und Verhalten der Rodentia“ 14.00 Uhr: Posterdemonstration 15.00 Uhr: Kurzvorträge 18.00 Uhr: Posterprämierung Donnerstag, 30. September: Exkursion Alle Interessenten, Mitglieder und Nichtmitglieder, sind zu dieser Jahrestagung 1999 der Deutschen Gesellschaft für Säugetierkunde in Bad Dürkheim herzlich eingeladen. Kongreßsprachen werden Deutsch und Englisch sein. Das Programm mit der Vortrags- folge wird den Mitgliedern — auf Anforderung auch Nichtmitgliedern — rechtzeitig vor der Tagung zugesandt. Sollten Sie eine persönliche Einladung benötigen, so wenden Sie sich bitte direkt an den 1. Vorsitzenden der Deutschen Gesellschaft für Säugetierkunde, Prof. Dr. H. ERKERT, Zoologisches Institut, Morgenstelle 283, D-72076 Tübingen (Tel. 070 71/2 972958; Fax-Nr. 0 70 71/29 46 34). 64 Mitteilungen der Gesellschaft Wir bitten um die Anmeldung von Tagungsbeiträgen. Außer Beiträgen zu den ge- nannten Themenschwerpunkten werden auch dieses Mal wieder verstärkt Kurzvorträge und Posterdemonstrationen zu anderen Fachgebieten der Säugetierkunde berücksichtigt. Bitte melden Sie Kurzvorträge (15 min + 5 min Diskussion) sowie Posterdemonstra- tionen möglichst frühzeitig, spätestens jedoch bis zum 30. April (Ausschlußfrist) beim Geschäftsführer der DGS, Prof. Dr. R. SCHRÖPFER, Fachbereich Biologie/Chemie, Barba- rastr. 11, D-49069 Osnabrück (Tel. 05 41/9 69-2848, Fax: 05 41/9 69-2862) an. Der Anmel- dung ist eine maximal einseitige informative Kurzfassung (Abstracts) (1,5-zeilig) beizufü- gen. Aus ihr sollen die Fragestellung, Methoden, Ergebnisse und die daraus gezogenen Schlußfolgerungen hervorgehen. Alle angenommenen Kurzfassungen werden wieder in einem Sonderheft der Zeitschrift für Säugetierkunde publiziert. Sie sind nach folgendem Schema abzufassen: Deutscher Titel, Leerzeile, englische Titelübersetzung (kleine An- fangsbuchstaben im Text; bitte ggf. einen „native speaker“ konsultieren!), Leerzeile, In- itialen und Familienname(n) des/der Autors(in) bzw. der Autoren(innen) in Großbuch- staben, Adresse, Leerzeile, Text. Aus arbeitsökonomischen Gründen bitten wir dringend darum, zusätzlich zum ausgedruckten Abstract möglichst noch eine Fassung auf Diskette (5.25 oder 3.5”, IBM-kompatibler DOS-PC) in Form eines Microsoft Word-, Word Per- fect- (5.1 oder 6.0) oder ASCII-Files mitzuschicken. Bitte formatieren Sie diesen File nicht! Die Maße für Poster werden im Juli-Rundschreiben der DGS bekanntgegeben. Mit Fragen zum Tagungsort und zur Organisation wenden Sie sich bitte an Herrn Dr. REINHARD FLÖSSER, Direktor des Pfalzmuseums für Naturkunde, Hermann-Schäfer- Str. 17, D-67098 Bad Dürkheim, Tel. 0 63 22/94 13-22, Fax: 0 63 22/94 13-11. III. International Symposium Semiaquatic Mammals and their Habitats May 25-27, 1999 in Osnabrück, Germany The 3rd symposium will give the opportunity to present new observations on the biology, the conservation, and the habitats of semiaquatic mammals. We also look for presenta- tions that deal especially with the problem of structure and species diversity of inland- bank-ecotones regarding the presence of semiaquatic mammals. A separate volume with abstracts will be printed, and it is planned to publish the con- tributions to the symposium. Two round-table discussions will take place. One regarding the situation of the beaver in the European agricultural landscape and the other concerning the situation of the Euro- pean mink. An excursion to lakes and water courses will be organized. For registration and further information contact: Prof. Dr. RÜDIGER SCHRÖPFER, Univer- sität Osnabrück, FB Biologie/Chemie: Ethologie, Barbarastr. 11, D-49069 Osnabrück. Tel. +49/(0)5 41/9 69-2847, Fax +49/(0)5 41/9 69-28 62, e-mail: schroepfer@biologie.uni-os- nabrueck.de Redaktionelle Hinweise Einsendung und Annahme von Manuskripten: Ms. zur Veröffentlichung sind einzusenden an den 1. Schriftleiter: Herrn Prof. Dr. Dieter Kruska, Institut für Haustierkunde, Christian-Albrechts-Universität, Olshausenstr. 40-60, D-24118 Kiel. e-mail: dkruska@ifh.uni-kiel.de Über die Annahme von Ms. zur Veröffentlichung wird gemäß Geschäftsordnung der Deutschen Gesellschaft für Säugetierkunde entschieden. Der Eingang von Ms. wird sofort bestätigt, und nach Erhalt der Gutachten wer- den die Autoren über die Entscheidung umgehend informiert. Schriftwechsel, der die technische Drucklegung betrifft, ist zu richten an: Zeitschrift für Säugetierkunde, Urban & Fischer Verlag, Villengang 2, D-07745 Jena. Äußere Form der Manuskripte: Die Ms. sind 3fach und vollständig als Original mit 2 Kopien in deutscher oder englischer Sprache einzureichen. Sie müssen mit Schreibmaschine oder als deutlich lesbare Computerausdrucke auf DIN A4-Blättern (21x 29,7 cm), mit einem linken Heftrand von 4cm und durchgehend mit einem doppelten Zeilenabstand geschrieben sein. Jedes Ms. soll auf der 1. Seite folgende Angaben enthalten: Titel der Studie; Name(n) der Autoren (männliche Vornamen nur in Initialen, weibliche ausgeschrieben); Institution(en), an der (denen) die Studie durchgeführt wurde. Ferner soll bei längeren Titeln ein kurzer Kolumnentitel (Seitenüber- schrift) von maximal 72 Anschlägen angegeben werden. Die Schrifttypen müssen einheitlich sein, ohne halbfette, gesperrte, kursive etc. Hervorhebungen. Fußnoten auf den Seiten oder Appendices am Ende sind nicht zugelassen. Autorennamen im Zusammenhang mit Literatur- zitaten sind im Text mit Bleistift zu unterstreichen. Wissenschaftliche zoologische und botanische Gattungs-, Art- und Unterartnamen sind mit einer Schlangenlinie zu unterstreichen. Der Name von Erstbeschreibern bei wis- senschaftlichen Namen wird nicht unterstrichen. Im Ms. ist am Rand mit Bleistift zu vermerken, an welcher Stelle die einzelnen Abbildungen und Tabellen ein- gefügt werden sollen. Am Ende der Studie, im Anschluß an das Literaturverzeichnis, sind die genauen postalischen Adressen des (der) Autoren anzugeben. Umfang und Aufbau der Manuskripte: Manuskripte können als Originalarbeiten oder als wissenschaftliche Kurz- mitteilungen veröffentlicht werden. a) Originalarbeiten: Originalarbeiten sollen einschließlich Abbildungen, Tabellen und Literaturverzeichnis einen Gesamtumfang von 30 Ms.-Seiten nicht überschreiten. Der Text soll eine klare Gliederung aufweisen in: Abstract (in englischer Sprache), Einleitung, Material und Methode, Ergebnisse, Diskussion (oder zusammengefaßt als Er- gebnisse und Diskussion), Danksagungen, Zusammenfassung und Literatur. Bei deutschsprachigen Studien ist über dem englischen Abstract der Titel in englischer Sprache anzugeben. Bei englischsprachigen Studien ist entsprechend über der deutschen Zusammenfassung der Titel in deutscher Sprache anzugeben. b) Wissenschaftliche Kurzmitteilungen: Wissenschaftliche Kurzmitteilungen sollen einen Gesamtumfang von 5 Ms.-Seiten nicht überschreiten. Sie enthalten kein Abstract und keine Zusammenfassung. Bei deutschsprachigen Ms. ist jedoch auf der 1. Seite zusätzlich eine englische Übersetzung des Titels anzugeben, bei englischsprachigen entsprechend eine deutsche Übersetzung. Wissenschaftliche Kurzmitteilungen sind äußerlich nicht in die Kapitel Einleitung, Material und Methode, Ergebnisse und Diskussion gegliedert, müssen aber in der textlichen Darstel- lung diese Strukturierung erkennen lassen. Key words: Bis zu fünf Key Words sollen dem Abstract folgen. Abbildungen: Anzahl und Größe der Abbildungen sind unbedingt auf das notwendige Mindestmaß zu beschrän- ken. Es können nur kontrastreiche, reproduktionsfähige Originale verwendet werden, farbige Abbildungen wer- den nicht angenommen. Beschriftungen müssen einheitlich und wegen späterer Verkleinerung groß und kräftig sein. Auf der Rückseite sind laufende Numerierung, Zeitschrift und Autorennamen anzugeben. Alle Bildunter- schriften sollen auf einem gesonderten Blatt gelistet werden. Tabellen: Tabellen müssen auf ein Mindestmaß beschränkt werden. Jede Tabelle muß auf einem gesonderten DIN A4-Blatt angelegt sowie mit Überschrift und laufender Numerierung versehen sein. Umfangreichere Tabel- len müssen auf Folgeblättern als Fortsetzung erscheinen. Literaturverzeichnis: Jede zitierte Arbeit ist in alphabetischer Reihenfolge aufzuführen. Es soll nur international erreichbare Literatur zitiert werden. Die Zeitschriften werden mit ihrem Kurztitel angegeben. Beispiele: a) Zeitschriftenbeiträge: KALkKo, E. K. V; HANDLEY, C. O. Jr. (1994): Evolution, biogeography, and description af a new species of Fruit-eat- ing bat, genus Artibeus Leach (1821), from Panamä. Z. Säugetierkunde 59, 257-273. b) Bücher: HERRE, W.; Röhrs, M. (1990): Haustiere — zoologisch gesehen. 2. Aufl. Stuttgart, New York: Gustav Fischer. c) Handbuchbeiträge: NIETHAMMER, J. (1982): Cricetus cricetus (Linnaeus, 1758) — Hamster (Feldhamster). In: Handbuch der Säugetiere Europas. Ed. by J. NIETHAMMER und F. Krapp. Wiesbaden: Akad. Verlagsges. Vol. 2/L, 7-28. Drucklegung: Der jeweilige Hauptautor erhält sowohl die Andrucke der Bilder zur Prüfung wie auch den kom- pletten Umbruch seines Artikels zusammen mit dem Ms. Es dürfen nur tatsächliche Satzfehler korrigiert wer- den. Vom Ms. abweichende Änderungen werden dem Autor in Rechnung gestellt. Es sind die üblichen Korrekturzeichen zu verwenden. Ein als druckfertig bezeichneter Abzug ist zu senden an: Herrn Prof. Dr. Peter Langer, Institut für Anatomie und Zellbiologie, Justus-Liebig-Universität, Aulweg 123, D-35385 Giessen. e-mail: peter.langer@anatomie.med.uni-giessen.de Sonderdrucke: Verfasser von Originalbeiträgen und Wissenschaftlichen Kurzmitteilungen erhalten 50 Sonder- drucke gratis. Mehrbedarf kann gegen Berechnung bei Rücksendung des Umbruchs bestellt werden. Vorbehalt aller Rechte Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, ins- besondere 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 gesetz- lichen Bestimmungen des Urheberrechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmun- gen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Ver- vielfältiger 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 in- ternal 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: Publishers Fee List” of the CCC. 2 @ . % Publisher: Urban.& Fischer Verlag GmbH & Co. KG, Branch Office Jena, P.O. Box No. 10 05 37, D-07705 Jena, Germany. Phone ++49(0)36 41/6 26-3; Fax ++49(0)36 41/62 65. 00; e-mail: journals@urbanfischer.de Type setting, printing and binding: druckhaus köthen GmbH Advertisement bookings: Enquiries should be directed to Urban & Fischer Verlag, Advertising Manager: Sabine Schröter, P.O. Box 10.05 37, D-07705 Jena, Germany. Phone ++49(0)36 41/62 6445, Fax ++49(0)36 41/62 64 21 Advertising rates: The price list from January 1, 1999 is effective at present. Distribution and subscriptidn agency: For customers ig the German speaking countries: Please turn to your bookshop or to SFG-Servicecenter Fachverlage, Zeitschriftenvertrieb: Bärbara Dressler, Villengang 2, D-07745 Jena, Germany. Phone ++49(0)36 41/62 64 44, Fax ++49(0)36 41/62 64 43; For customers in all other countries: Please contact your local bookshop or Stockton Press., c/o Debbie Floyd, The Sub- scription Dept., Houndmills, Basingstoke, Hants RG 21 6XS, UK. Phone ++44(0)12 56/32 92 42. Fax ++44/1256/812358. Subscribers in the USA can call toll free on ++1/800/ 747 3187. Terms of delivery (1999): 1 volume consisting of 6 issues. Er Subscription rate (1999): Price per volume: DM 478.00*; UK Sterling 192.00 STOCKTON Single issue price: DM 95.00*; UK Sterling 39.00 * Suggested list prices. Prices are subject to change without notice. Reduced membership rates are available on request. Please turn to our publ. house. All prices mentioned above are exclusive postage. USdollar prices are converted at the current exchange rate. We accept the following credit cards: VISA Euro-/Mastercard/American Express, (Please note card-No. and expiry date). Banking connections: For customers in the German speaking countries: Deutsche Bank AG Jena, Konto-No. 6 284 707, BLZ 820 700.00; Postbank Stuttgart, Konto-No. 92 372 704, BLZ 600 100 70. For costumers in the other countries: UK Post Office Giro Account No. 5 192 455. Full details must accompany the payment. Subsciption information: If not terminated the subscription will be effected until recalled. If no cancellation is given until October, 31st the delivery of the journal will be continued. Copyright: The articles published in this journal are protected by copyright. All rights are reserved. No part of the journal may be reproduced in any form without the written permission of the publisher. Authorization to photocopy items for internal or personal use of specific clients, is granted by Urban & Fischer, for libraries and other users registered with the Copyright Clearance Center (CCC) Transaction Reporting Service, provided that the base fee of $12.00 per copy is paid directly to CCC, 222 Rosewood Drive, Danvers, Ma 01923, USA, 0044-3468/99/064/ $ 12.00/0 This consent does not extend to copying for general distribution, for advertising or promotional purposes, for creating new collective works or for resale and other enquiries. In the cases, specific written permission must be obtained from the pub- lisher Urban & Fischer. Printed in Germany Printed on acid-free paper effective with vol. 61, no. 1, 1996. © 1999 Urban & Fischer Verlag New: For detailed journal information see our home page: http://www.urbanfischer.de/journals Abstracted/Indexed in Animal Breeding Abstracts; Current Contents Agriculture, Biology Environmental Sciences; Biological Abstracts; BIOSIS database; Current Advances in Ecological and Environmental Sciences; Dairy Science Abstracts; Elsevier BIOBASE/Cur- rent Awareness in Biological Sciences; Fisheries Review; Helminthological Abstracts; Index Veterinarius; South Pacific Periodicals Index; Veterinary Bulletin; Key Word Index to Wildlife Research; Wild Review (Fort Collins); Zoological Record 492 2484 mamn ZEITSCHRIFT FÜR SÄUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Reinhardt, Ilka; Halle, S.: Time of activity of a female free-ranging Lynx (Lynx lynx) with young kittens in Slovenia. - Aktivität eines Luchsweibchens (Lynx lynx) während der Jungenaufzucht in Slowenien..............22uosseaeeenecinene 65 Ruckstuhl, Kathreen E.; Ingold, P.: Aspects of mother-kid behaviour in Alpine chamois, Rupicapra rupicapra rupicapra. - Aspekte der Mutter-Kindbeziehung bei der Alpengemse (Rupicapra rupicapra rupicapra) «........u0000#400nne ee len 76 Hale, Molly O.; Fuller, T. K.: Estimating porcupine (Frethizon dorsatum Linnaeus, 1758) density using radiotelemetry and replicated mark-resight techniques. - Dichteschätzungen von Baumstachlern (Frethizon dorsatum Linnaeus, 1758) mittels Radiotelemetrie und Sichtungen markierter Individuen..............uu.-.ussss00ssennennennnnnnnnen anne 85 Gimenez, Mabel D.; Bidau, C.J.; Argüelles, Carina F.; Contreras, J. R.: Chromosomal characterization and relationship between two new species of Ctenomys (Rodentia, Ctenomyidae) from northern Cördoba province, Argentina. - Chro- mosomale Charakterisierung und karyologische Beziehungen zwischen zwei neuen Arten von Ctenomys (Rodentia, Ctenomyidae) aus dem Norden der Provinz Cördoba, Argentinien .........-...-..---ruosessnensennnnnnennnnnennnnannnennnn 91 _ Wissenschaftliche Kurzmitteilungen Fiedler, W.; Alder, H. U., Wohland, Pia: Zwei neue Nachweise der Weißrandfledermaus (Pipistrellus kuhli) für Deutsch- land. - Two new records of Kuhl's pipistrelle (Pipistrellus kuhli) in Germany.........-.uu00024@s00 essen nnnnnnnnnnenanen 107 Lode, Th.: Comparative measurements of terrestrial and aquatic locomotion in Mustela lutreola and M. putorius. - Ver- gleichende Messungen von terrestrischer und aquatischer Lokomotion bei Mustela lutreola und M. putonus.......... 110 Angelici, F.M.; Luiselli, L.; Politano, E.: Distribution and habitat of selected carnivores (Herpestidae, Mustelidae, Viverridae) in the rainforests of southeastern Nigeria. - Verbreitung und Habitat ausgewählter Carnivora (Herpesti- dae, Mustelidae, Viverridae) in den Regenwäldern des südöstlichen Nigeria...........--...---u---uunsuu0nssenenenuneennn 116 Anderson, R. P.; Soriano, P. J.: The occurrence and biogeographic significance of the southern Spiny pocket mouse Het- eromys australis in Venezuela. - Verbreitung und biogeographische Bedeutung der Südlichen Stacheltaschenmaus Rs ilVerlezielan. 2 en nennen na n en aa ane e 121 Colak, E.; Yigit, N.; Verimli, Reyhan: On the karyotype of the Long-clawed mole vole, Prometheomys schaposchnikovi Satunin, 1901 (Mammalia: Rodentia), in Turkey. - Über den Karyotypen der Prometheusmaus, Prometheomys scha- DOBEnKOV Satunin, 1901 (Mammalia: Rodentia), in der Türkei .......- +... .ouusaeesosseansnneessnennerenanneena nennen 126 nr EEE A a 128 ISSN 0044-3468 - Z. Säugetierkunde - 64(1999)2 - S. 65-128 - April 1999 URBAN & FISCHER 1999 EGES S Herausgeber/Editor Deutsche Gesellschaft für Säugetierkunde Schriftleitung/Editorial Office D. Kruska, Kiel - P. Langer, Giessen Wissenschaftlicher Beirat/Advisory Board W. Fiedler, Wien - S. Halle, Jena - G. B. Hartl, Kiel - R. Hutterer, Bonn - H.-G. Klös, Berlin - E. Kulzer, Tübingen - P. Lüps, Bern —- W. Maier, Tübingen - O. Anne E. Rasa, Bonn -H. Reichstein, Kiel - M. Röhrs, Hannover - H. Schliemann, Hamburg - G. Storch, Frankfurt Deutsche Gesellschaft für Säugetierkunde Altvorsitzende/Living Past Presidents D. Starck, Frankfurt (1957-1961, 1967-1971) — H. Frick, München (1972-1976) — M. Röhrs, Hannover (1977-1981) — H.-J. Kuhn, Göttingen (1982-1986) - E. Kulzer, Tübingen (1987-1991) — U. Schmidt, Bonn (1992-1996) Amntierender Vorstand/Managing Committee Vorsitzender/President: H. G. Erkert, Tübingen Mitglieder/Board Members: H. Frädrich, Berlin - R. Hutterer, Bonn -— D. Kruska, Kiel — Marialuise Kühnrich, Hamburg - R. Schröpfer, Osnabrück - Angela Wöhrmann-Repenning, Kassel Z. Säugetierkunde 64 (1999) 65-75 FÜ © 1999 Urban & Fischer Verlag SÄU GETIER KÜ NDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Time of activity of a female free-ranging Lynx (Lynx Iynx) kittens in Slovenia By ILKA REINHARDT and S. HALLE Institute of Ecology, Ludwig-Maximilians-University München, München Receipt of Ms. Acceptance of Ms. 14. Abstract We investigated the activity behaviour and time-budget of a female free-ranging Iynx (Lynx lynx) in Slovenia during the first five months of post-partum from June-October 1995. Activity was monitored by means of a radio-collar with activity sensor, employing continuous automatic recording and discontinuous time-sampling. A total of 1818 data hours was analysed with respect to the prey status, distinguishing between days with and without kill, and age of the kittens. On average the female was active for 8.5 h per 24-h day. Activity at twilight and during daytime was generally higher than during night. On days when the female hunted she was more active and more diurnal than on days when she had access to a kill. During the later post-partum mobile phase the female covered a considerably larger home range, was much more active, and showed higher twilight and daytime activity than during the earlier stationary phase. Movements to and from a kill occurred irregularly throughout the night during the stationary phase, while the whole fa- mily went to a kill around sunset and returned back around sunrise during the mobile phase. When a kill was available, the female spent on average 81 % of the 24-h day with her kittens, but only 63% on days with no kill. Times of absence from the den did not in- crease during the stationary phase as the kittens grew older. Activity timing is interpreted as a highly differential temporal adaptation to meet various contrasting challenges, 1. e., hunting, defence of kills against competitors, protection of young, and home range patrol- ling. Comparison with data from Switzerland suggest that habitat structure in addition is likely to shape Iynx activity in different areas. Key words: Eurasian Iynx, activity, radiotelemetry, time budget, breeding Introduction Recently, the Eurasian Iynx (Lynx !ynx) appaers to be regaining parts of its former areas of distribution (see BREITENMOSER-WÜRSTEN et al. 1998) due to re-introduction and legal protection. In this connection it is particularly interesting to know to which extent the species is able to adapt its behaviour to different ecological conditions. Two important as- pects in this regard are hunting activity and time budgets, since flexible responses to the environment are to be expected with these behavioural traits in particular. However, data on lynx activity behaviour are rather fragmentary. In general, the species is assumed to be primarily nocturnal and crepuscular (MATIUSCHKIN 1978, HEMMER 1993), although diurnal activity is also known to occur (HALLER and BREITENMOSER 1986). 0044-3468/99/64/02 — 065 $ 12.00/0 66 ILKA REINHARDT and S. HALLE Even less information is available on the relation between activity and breeding. The female lynx restricts its movements to a small part of its home range during the early rearing period, as also known from other Felidae (SEIDENSTICKER 1977; Sunauist 1981; SCHMIDT et al. 1997). Restricted space use is likely to be reflected in the activity pattern because the daily time schedule is closely linked to spatial behaviour. However, a feasible ad-hoc assumption about the total activity level is not possible since a breeding female has to balance the requirements of food intake with the specific challenges of rearing kit- tens. Increased hunting effort due to limited use of space during the early stationary phase may cause higher activity, while the extended use of space during the later mobile phase may also account for this as well. In general, predators are known to time their activity in accordance to the activity pattern of their prey (e.g. ABLes 1969; Sunauiıst 1981; FERGUSON et al. 1988; BELTRAN and DELIBES 1994). However, when a Iynx manages to kill a large prey like a roe deer, which is the main prey of the European Iynx (PULLIAINEN 1981; SosTak and BUNEVIC 1986; BREITENMOSER and HALLER 1987; JEDRZEJEWSKI et al. 1993; PULLIAINEN et al. 1995), it may feed on it for up to one week (REINHARDT, pers. obs.). Having a kill or not can be supposed to affect the activity behaviour directly, with the assumption that the level of ac- tivity will be lower when the animal has a kill to feed on. Furthermore, also the temporal pattern of activity over the 24-h day may vary between days with and without kill, be- cause time can be allocated to different types of activity depending on the short-term food supply. Almost nothing is known about this aspect of Iynx behaviour. In this study we present data on the activity and time budget of an intensively studied free-ranging female lynx in Slovenia during the first five post-partum months. In particu- lar, we focus on the effect of prey status on temporal behaviour during two distinctive rearing periods. Material and methods Study area The study was conducted in the Kocevska region in Southern Slovenia (45°35’N, 45°20'E). The 620 km” study area is part of the Dinara Mountain Range. Elevations range from 300 to 1300 m. The fine-scale relief is typical of high karst regions with numerous dolines, caves, boulder fields, and rock faces. Forest covers up to 90 % of the terrain with Abieti-Fagetum-dinaricum as the dominant forest community. The climate is temperate continental with annual precipitation from 1400 to 1800 mm. The average mean temperatures in January and July are -2.9 and 17.8°C, respectively. Potential larger prey species of Iynx are red deer (Cervus elaphus; annual harvest quote of 2-3 an- imals/100 ha), roe deer (Capreolus capreolus; annual harvest 0.7/100 ha), and a few chamois (Rupica- pra rupicapra). Species that may compete or otherwise interact with Iynx include brown bear (Ursus arctos), wolf (Canis lupus), red fox (Vulpes vulpes), golden jackal (Canis aureus), wild cat (Felis silves- tris), badger (Meles meles), wild boar (Sus scrofa, annual harvest 0.8/100 ha), and a variety of raptor species. Activity reading The Iynx female was caught in a box-trap in April 1994. It was tranquillised with Zoletil 100 (Virbac, France, a mixture of Tiletamin and Zolazepam), ear-tagged, and fitted with a radio-collar equipped with a tip switch as activity sensor (Wagener, Germany). The tip switch caused the pulsing rate of the transmitted signal to alter between slow and fast, depending on collar position. In addition, signal strength varied due to changes in transmitter orientation relative to the Iynx body, and relative to the receiver antenna. Since we were mainly interested in locomotor activity, lynx activity was indicated by changes in both signal strength and/or pulsing rate. Time of activity of a female Iynx with kittens 67 Activity was recorded in two different ways: Continuous recording was performed by means of an automatic recording station (B + R Ingenieurgesellschaft, Germany), which allowed determination of lynx activity for each single minute. In addition, discontinuous activity recording (time-sampling) was used when the Iynx was outrunning the receiving area of the recording station. It was then followed by car, and radio signals were monitored every 5th minute for 60 sec with a hand-held Yagi-antenna and a portable receiver. A sample was classified “active” when three or more signal changes occurred during one minute, while inactivity was indicated by a steady signal pulse. In this way we followed the “predominant activity sampling procedure” which is considered most accurate for temporal activity assessment (TYLor 1979). Compatibility of the two methods of activity reading was tested by simulta- neous automatic recording and time-sampling for 20 h, which revealed a high agreement of 96.7 %. To make both methods directly comparable in data analysis the tapes from continuous recording were time-sampled to simulate discontinuous activity recording. The lynx gave birth to two kittens on the 1 June 1995 and was then followed during the first five months of rearing, i.e., from June until October 1995. During the first weeks after birth Iynx kittens remain stationary at the den-site to which the female always returned from hunting (called “station- ary phase” hereafter). During this time the recording station was kept near to the den-site, fitted with a directional antenna oriented towards the den. This allowed to record female activity near the den as well as times and length of den attendance. When Iynx kittens have grown up to an age of seven to eight weeks, mother and young left the natal area and the kittens started accompany their mother (called “mobile phase” hereafter). Data analysis Lynx activity was analysed with respect to the overall level of activity, the activity distribution over the 24-h day, and the relative activity allocation to different light phases. Twilight was defined as Ih before and after sunrise and sunset, respectively. The basic time unit was hours of the clock. The pro- portion of samples scored active during each hour yielded the percent activity per hour, with the re- strietion that only hours with at least six activity samples were regarded in the analysis. The activity le- vel was measured as the average activity per hour, per day, or per light phase. Relative diurnality and crepuscularity indices were employed to compare the activity allocation to light phases between periods with different levels of activity and with different day length. The di- urnality index Ip (modified after HALLE 1995) reflects the proportion of diurnal as compared to noc- turnal activity and was calculated by (2 AD)/hD (2 AD)/hD + (D AN)/hN | en Ip = in which 2 AD and ZAN are the times of activity summarised for day and night, respectively, and hD and hN are day length and night length, respectively. Ip is positive if diurnal activity prevails (maxi- mum: +1 when exclusively diurnal) and negative when nocturnal activity prevails (minimum: -1 when exclusively nocturnal). The crepuscularity index Ic (modified after HALLE 1995) reflected the relative proportion of twilight activity as compared to the average activity over the 24-h day and was valued by where ZAC is the time of activity during the hours of twilight (SR+1h and SS # Ih) and ZA is the total time of activity during an entire day. Ic is positive when activity during twilight is increased com- pared to the average activity, and negative when it is decreased. Lynx activity behaviour was analysed for the total study period, the two rearing periods, and in re- lation to the prey status. For the latter all days of the field season (separated by midnight) were cate- gorised according to prey status by locating the female several times a day using the methods of trian- gulation (or “homing in”, WHITE and GARROT 1990). Locations where the Iynx remained for an evening or night were searched for carcasses of prey killed by Iynx the following day with a trained dog. Days were classified as “day with kill” if the kill was actually found, or if the Iynx returned to the same place for several nights. If there was no indication of a kill, i.e., when the Iynx changed its position considerably in successive nights, the status “day without kill” was given. When data where ambiguous, days were classified as “status unknown”. Note that this classification, however, only ap- 68 ILKA REINHARDT and S. HALLE plies to large ungulate prey and not to voles and other small prey items, which may be food resources as well (e.g. PULLIAINEN 1981; JEDRZEJEWSKI et al. 1993; PULLIAINEN et al. 1995). Differences in activity distribution over the 24-h day were tested with the Kolmogorov-Smirnov- test. When testing for differences depending on prey status, data from days with unknown status were excluded. Wilcoxon-signed-rank-tests for related samples were used for comparisons between days with and without kill, and between the two rearing periods. In the same way we tested whether the fe- male was more often absent from the den during the second half of the stationary phase than during the first. We used Kruskal-Wallis one-way analysis of variance for independent samples, adjusted by a Bonferroni procedure (c.f. TOUTENBOURG 1994), to test for differences among the average activity le- vel at daytime, night-time, and twilight. All P-values are for two-tailed test. Results General activity pattern and prey status The total base of activity data consisted of 1818 data hours of which 1275 h (70 %) were covered by automatic recording and 543 h (30%) by time-sampling. Each hour of the clock (1:00-24:00) was represented by 59 to 95 data hours with an average of 11.1 activity samples per hour. For 1490 h from 103 days the prey status was known, of which 824 h from 49 days were classified as “days with kill” and 666 h from 54 days as “days without kill”. On average, the Iynx was active for 35.5 % (N = 1818, SD = 32.18, SE = 0.75) or 8.5h of the 24-h day. Averaged over the entire study period the daily activity pattern revealed no significant deviation from an even distribution over the 24-h day (N = 1818, Kolmo- gorov-Smirnov, zZ = 12.29, P < 0.001). However, after splitting days into three distinct light phases, the overall level of activity differed significantly among them (N = 1818, Kruskal- Wallis, d{ =2, P = 0.001), being highest during twilight, intermediate during daytime, and lowest during the night. The Iynx was more active on “days without kill” (45.7%) than on „days with kill” (28.3 %; N = 24, Wilcoxon, P<0.001). Twilight and diurnal activity always prevailed but particularly so on “days without kill” (Ic = +0.12, Ip = +0.15 as compared to Ic = +0.03, Ip = +0.06 on “days with kill”). This indicates predominant hunting during broad day light. Rearing periods During the stationary phase the Iynx used only 8km” of her home range as compared to 110 km? during the later mobile phase. Correspondingly, the female was significantly more active during the mobile phase (39%) than during the stationary phase (29.7 %; N = 24, Wilcoxon, P = 0.004). This activity increase was above all caused by much higher day and twilight activity, whereas nocturnal activity decreased during the mobile phase, resulting in a substantial change in the daily activity pattern (Fig. 1). Activity distribution over the three light phases was fairly even during the stationary phase (N = 683, Kruskal- Wallis, P = 0.241), while the difference twilight > day > night was highly significant during the möbile phase (N = 1135, Kruskal-Wallis, P< 0.001; twilight vs. daytime: N = 706, P < 0.001; daytime vs. night: N = 964, P< 0.001). Accordingly, diurnality and crepuscular- ity indices were higher for the mobile phase (Ip = +0.13, Ic = +0.13) than for the station- ary phase (Ip = -0.04, Ic = -0.06). For the effect of prey status on activity behaviour (Fig. 2), only “days with kill” could be compared between the two rearing periods, because too few data from “days without kill” were available for the stationary phase. On “days with kill” the activity level was about 28.6 % during the stationary phase with an almost even distribution over all light number of events 25 20 15 10 Time of activity’of a female Iynx with kittens ----- — 160 12-1 time of day 16 18 20 22 24 69 Fig. 1. Diel distribution of occasions when the Iynx was recorded to move more than 500 m per hour. Solid line: stationary phase, broken line: mobile phase. The bars at the top indicate twilight and night- time hours. stationary phase mobile phase % act. % act. with kill without kill day twilight night light phase © >) 2 day twilight night day twilight night light phase light phase Fig. 2. Mean values (+ SE) of activity for each light phase (daytime, twilight, night-time), measured as the percentage of an hour with activity. Data are broken down to the two rearing periods and to the effect of prey status. For the stationary phase too few data were available for “days without kill” to in- clude this category in the analysis. conditions (N = 344, Kruskal-Wallis, P = 0.188; Ip = -0.09 and Ic = +0.04). The level of ac- tivity was almost the same during the mobile phase on “days with kill” (28.1 %), but now activity differed significantly among light phases (N = 480, Kruskal-Wallis, P < 0.001), de- creasing in the order twilight > day > night (daytime vs. twilight: N = 407, P< 0.001; night vs. daytime: N = 257, p = 0.003; Ip = +0.13, Ic = +0.19). Crepuscular activity was by then about 1.6 times higher than the 24-h average. 70 ILKA REINHARDT and $S. HALLE stationary phase o Ko = en} c 0:2, 4, 602% )16 185207 922024 time of day mobile phase o OD E ee} c 0....2.-. 4.6 -8...10..-12 142.16 .-.18 20: 22-024 time of day Fig. 3. Diel distribution of movements between the daytime location and the kill during the two rear- ing periods. Black bars: moving to the kili, shaded bars: returning from the kill. The bars at the top indi- cate twilight and night-time hours. The reason for this difference in behavioural parameters became obvious when look- ing at the times when the Iynx went to or returned from the kill (Fig. 3). As long as the kittens were stationary the female went to the kill and returned to the den at variable times. Later, when the kittens followed their mother, the pattern became much more reg- ular, i.e., the family went to the kill in the evening and early night, and returned back to their resting place in the late night and early morning. During the mobile phase the level of female activity almost doubled on “days without kill” (51.5 %) as compared to “days with kill” (28.1 %). When she had no kill and was hunting, significantly more activity was performed during daytime and twilight than during the night (daytime vs. night: N = 383, P < 0.001; twilight vs. night: N = 265, P < 0.001; Fig. 4, c.f. Fig. 2). To determine whether the female had preferred times for sleeping we computed the averages of the longest inactive period per hour of the day. The resultant 24-h pattern Time of activity of a female Iynx with kittens 71 mobile phase 100 % act. 0727743776 38710 .,12 14 i6 18 20.22.24 time of day Fig. 4. Pattern of female activity at “days with kill” (black) and “days without kill” (shaded) during the mobile phase. The bars at the top indicate twilight and night-time hours. was compared with the total distribution of inactivity (60 min. - X min. active/h) for the two rearing periods as well as for days with and without kill. The distributions always cor- responded well with each other (N = 24, Spearman rank correlation, all P< 0.001, station- ary phase: rs = 0.72; mobile phase: rs = 0.92; “days with kill”: rs = 0.81; “days without kill”: rs = 0.86), indicating that sleep was an integrated part of the main rhythm of activ- ity and inactivity, respectively. Den attendance From 1 June the animal returned constantly to the same small area where she had her breeding den the previous year (HUBER, unpubl. data). After a disturbance on 23 June the female moved to a second den, approximately 400 m up the slope, where two kittens aged about 4 weeks were found on 2 July. Also later during the stationary phase the Iynx family repeatedly changed location a few hundred metres, so obviously the female had ac- cess to several auxilliary dens that where used alternatively. The female always returned to the place where she started from, so she probably was not accompanied by her kittens during the excursions. On 19 July the Iynxes moved 3.5 km (straight line) to a place well outside the area where the female has been hunting during the previous seven weeks. During the stationary phase the female spent on average 17.9h of the day (74.5 %) near her kittens at the den. The time of presence, however, differed significantly between “days with kill” (80.6 %) and “days without kill“ (62.6 %), being on average 4.5 h longer on “days with kill” (N = 24, Wilcoxon, P < 0.001). Relatively, presence at the den-site oc- curred more often during daytime on “days with kill” than on “days without kill” (N = 400, Mann-Whitney-U, P< 0.001, Fig. 5). However, differences in den attendance during night and twilight hours between “days with kill” and “days without kill” were not statistically significant (Mann-Whitney-U, night: N= 120, P=0.44; twilight: N = 180, P = 0.27). On “days with kill” the times of absence from the den corresponded generally well with the activity records (N = 24, Spearman rank correlation, r, = 0.71, P< 0.001). How- ever, activity during daytime was not closely related to the pattern of absence from the den, indicating high activity in the close vicinity of the den, probably performed as play 72 ILKA REINHARDT and $S. HALLE den attendance 100 presence [7] © % 0.2.2 6.8 ,.10,.12,.14 16,18 207227524 time of day Fig. 5. Diel distribution of den attendance by the female at “days with kill” (black) and “days without kill” (shaded) during the stationary phase. The bars at the top indicate twilight and night-time hours. and comfort behaviour. The time of absence from the den did not increase as the kittens grew older, instead it decreased slightly from 28 % to 24% of the day (N = 24, Wilcoxon, Pr 02P): Discussion We are well aware that the data presented here are restricted to only one individual, hence the important aspect of behavioural variation among individuals had to be ignored. However, since there is only scarce information about the activity and early maternal phase of free-ranging Eurasian Iynx, we feel that our approach to focus on one breeding female and to follow her closely through a whole summer may be justified. According to the lumped data from the entire study period, activity appeared initially to be acyclic. However, a more detailed scrutiny of the data revealed a highly differen- tiated activity behaviour, in which prey status and rearing period had substantial effects on the pattern. In fact, our analysis verified that the flexible pattern of the females activ- ity allowed for short-term adaptive responses to her and her kittens requirements. The highest activity of the female was recorded during twilight and daytime, which is in accordance with a short survey of re-introduced lynxes in Austria (FEstETics 1981). In contrast, MATJUSCHRIN (1978) observed the Iynx in Russia to be mostly or predominantly nocturnal with only little diurnal activity. The general activity level of about 36 % in our study was considerably lower than the 58% Aupama et al. (1991) reported for the Iber- ian lynx (ZL. pardinus). In particular, the Iynx was less active on “days with kill” than on “days without kill”, so obviously a Iynx that already has a kill can lower its hunting effort and, as a rule, does not move over greater distances. Behaviour during the stationary phase The level of activity was almost the same on “days with kill” during the two rearing peri- ods, portraying the basic activity level of a female Iynx with cubs. During the stationary phase the Iynx fed mostly at night, and also movements between den-site and kill oc- curred predominantly at night. This resulted in higher nocturnal activity on “days with kill” as compared to the mobile phase. Time of activity of a female Iynx with kittens 73 Time of absence from the den-site per day did not gradually increase with age of the kittens. This contradicts the assumption that the restricted hunting area during the sta- tionary phase will demand increasing hunting efforts as kittens grow older and need more food, which in turn would lead to shorter den attendance and higher activity. So probably, the area close to the den-sites was a hunting ground sufficient to cover food demands during the entire stationary phase. This assumption is reasonable, since the large fields of dolines offer rich grazing for roe deer, while a stalking predator benefits from good cover. This estimate of habitat quality, however, raises the question why behaviour at all changes during the transition to the mobile phase, if not for limited food supply. A suit- able explanation may be a need for home range patrolling. Home range occupation was reported by Sunauist (1981) for a tigress while the activity of the former resident female was restricted to a small area during the early rearing period. The same resulted for Iynx in Switzerland BREITENMOSER et al., 1993). Hence, staying with the kittens too long may result in a loss of the home range, while starting to move with the kittens too early may endanger the kittens life. An observation made one week before the female moved with her kittens a longer distance for the first time may point to this difficult trade-off. The lynx made a remarkably extensive excursion over 15 h to the northern end of her range, and almost the same way was then taken accompanied by the kittens when they left the natal area. Thus, the mother most probably pre-explored the travelling route before. Another reason for giving up the stationary phase may be due to energy constraints. The energy costs of lactation are high and exceed all other reproductive costs in euthe- rıan mammals (Loupon 1985). Deag et al. (1987) proved that nursing cats (Felis domesti- ca) lost weight at an increasing rate over the first eight weeks after parturition. Increasing costs of lactation may, therefore, demand to provide the kittens with solid food. To our knowledge there is no report of Iynx carrying large prey or parts of it back to the den, and felids are not able to regurgitate food to their offspring as known from canids. In- stead, they lead their young to the kill, resulting in the vagabond behaviour of the mobile phase. Behaviour during the mobile phase Activity behaviour during the mobile phase was bimodal (AscHorr 1957), and daytime was the preferred activity period irrespective of prey status. When the female had a kill, she went there together with her kittens in the evening, spent the whole night close to the kill, and left in the morning. Accordingly, the highest locomotor activity was recorded during twilight in this situation. This pattern may reflect a temporal adaptation to protect the kill against potential scavengers. JEDRZEWSKI et al. (1993) found eight species of sca- vengers feeding on lynx kills in Poland, with wild boar having the greatest impact. In our study area the brown bear is probably an important competitor, which is mainly noctur- nal (KACZENSKY pers. comm.). It may be advantageous to stay near the kill at night to se- cure that the pay-back from time and energy investment is not taken away by another species. Against visually orientated diurnal scavengers like raptors blinding with leaves and branches was an efficient technique. Even on “days with kill” activity was higher during daytime than during the night, suggesting that play and comfort behaviour mainly occurred during daytime. The activity rhythm of the kittens may thus partly explain diurnal activity of the female. Many mam- malian species exhibit diurnal behaviour as juveniles and shift to nocturnality as adults, which is, for instance, known for Iberian Iynx (BELTRAN and DELIBEs 1994), wild boar (BRIEDERMAN 1971), and badger (EıßL-EiIBELSFELDT 1954, cited in AscHorF 1957). If kit- ten activity would be a decisive factor, increased diurnality should exclusively be shown by females with young offspring. 74 ILKA REINHARDT and S. HALLE After having consumed a kill completely, the female normally moved with her kittens during broad daylight to another area before hunting. Hunting times of the Iynx were very variable during the mobile phase, but as a general rule tended to be higher during twilight and daytime. In the Swiss Jura Mountains, however, nocturnal and twilight activ- ity prevailed distinctively (BERNHART pers. comm.). The different temporal hunting pat- terns in Slovenia and Switzerland may reflect adaptations to hunting on roe deer, the main prey of lynx in both areas (BREITENMOSER et al. 1993; HUBER unpubl. data) in differ- ently structured habitats. In Slovenia, forest covers more than 90 % of the study area, so that deer can stay under cover for long periods of time. In this situation temporally flex- ible roaming may give the best hunting success. In the Swiss study area, however, forest cover is only about 39 % (BERNHART pers. comm.) and roe deer emerge predominantly in the evening to browse on pastures. Accordingly, Iynx hunt especially along the forest edges during the evening and night (BREITENMOSER and HALLER 1987) with frequent indi- cations of ambushing (KACZENSKY pers. comm.). Since founders of both populations stem from the same population of origin in Slovakia (HALLER and BREITENMOSER 1986; Cop and Frkovıc 1998), behavioural differences between Swiss and Slovenian Ilynx are likely to reflect responses to the two areas rather than individual or genetic variance. This indi- cates that behavioural flexibility of Eurasian Iynx allows to adapt foraging to different ecological situations. Acknowledgements We gratefully thank P. KAczEnsky and T. HUBER for support and critical discussion, and J. DIEBERGER from the Institut für Wildbiologie und Jagdwirtschaft at Vienna as well as J. STANISA for technical sup- port. We also want to thank R. MÜLLER and D. Wirz for assistance with data analysis and K. HINDEN- LANG and two anonymous referees for their constructive comments. Funding was provided by the Austrian Federal Ministry of Science and Research (G.Z. 30.435/-23/92). Zusammenfassung Aktivität eines Luchsweibchens (Lynx Iynx) während der Jungenaufzucht in Slowenien Aktivitätsverhalten und Zeitbudget eines Luchsweibchens (Lynx !ynx L.) in Slowenien wurden wäh- rend der ersten fünf Monate der Jungenaufzucht von Juni-Oktober 1995 mittels Radiotelemetrie un- tersucht. Die Aktivität wurde teils kontinuierlich mit einer automatischen Registrierstation, teils dis- kontinuierlich durch “time-sampling” aufgenommen. Insgesamt wurden 1818 Datenstunden hinsicht- lich des Alters der Jungen analysiert, wobei zwischen Tagen mit und ohne Riß unterschieden wurde. Im Durchschnitt war die Luchsin pro 24 Std.-Tag für 8,5 Std. aktiv, wobei die meiste Aktivität in der Dämmerung und tagsüber erfolgte. An Tagen ohne Riß war sie stärker diurnal und insgesamt deutlich aktiver als an Tagen mit Riß. Verglichen zur frühen, stationären Phase der Jungenaufzucht, in der die Luchsin die Jagdausflüge auf einen kleinen Teil ihres Streifgebietes beschränkte, war sowohl die Ge- samtaktivität als auch die Dämmerungs- und Tagaktivität in der späteren, mobilen Phase höher. Orts- wechsel zwischen Riß und Tagesstandort erfolgten in der stationären Phase unregelmäßig über die Nacht verteilt, während Ortswechsel in der mobilen Phase sehr regelmäßig in der Morgen- bzw. Abenddämmerung stattfanden. An Tagen mit Riß verbrachte die Luchsin durchschnittlich 81 % des Tages bei ihren Jungen, jedoch nur 63 % an Tagen ohne Riß. Die Zeit der Abwesenheit von der Wurf- höhle nahm während der stationären Phase nicht mit dem Alter der Jungen zu. Die Verhaltensände- rungen während der Jungenaufzucht können vor dem Hintergrund gegensätzlicher Erfordernisse (Jagd, Rißverteidigung gegen Konkurrenten, Schutz der Jungen, Streifgebietskontrolle) interpretiert werden. Ein Vergleich der Ergebnisse mit Studien aus Gebieten unterschiedlicher Habitat- struktur läßt darauf schließen, daß das flexible Jagd- und Aktivitätsverhalten des europäischen Luchses eine Anpassung an verschiedenartige ökologische Bedingungen erlaubt. Time of activity of a female Iynx with kittens 73 References ABLES, E. D. (1969): Activity studies of red foxes in Southern Wisconsin. J. Wildl. Manage. 33, 145-153. ALDAMA, ]. J.; BELTRAN, J. F.; DELIBES, M. (1991): Energy expenditure and prey requirements of free- ranging Iberian lynx in southwestern spain. J. Wildl. Manage. 55, 635-641. ASCHOFF, J. (1957): Aktivitätsmuster der Tagesperiodik. Naturwiss. 13, 361-367. BELTRAN, J. F.; DELIBEs, M. (1994): Environmental determinants of circadian activity of free-ranging Iberian lynxes. J. Mammalogy 75, 382-393. BREITENMOSER, U.; HALLER, H. (1987): Zur Nahrungsökologie des Luchses Lynx Iynx in den schweizer Nordalpen. Z. Säugetierkunde 32, 168-191. BREITENMOSER, U.; KACZENSKY, P.; DÖTTERER, M.; BREITENMOSER-WÜRSTEN, C.; CAPT, S.; BERNHART, F.; LIBEREK, M. (1993): Spatial organization and recruitment of lynx (Lynx !ynx) in a re-introduced po- pulation in the Swiss Jura Mountains. J. Zool. (London) 231, 449-464. BREITENMOSER-WÜRSTEN, C.; ROHNER, C.; BREITENMOSER, U. (Eds., 1998): The re-introduction of the lynx into the Alps. Proc. 1° SCALP Conf., Engelberg, Switzerland, 1995. Environmental encoun- ters, No. 38, Council of Europe Publ. (in press). BRIEDERMANN, E. (1971): Ermittlungen zur Tagesperiodik des Mitteleuropäischen Wildschweines (Sus s. scrofa L.). Zool. Garten 40, 302-327. Cop, J.; FRKoVIc, A. (1998): The re-introduction of the Iynx in Slovenia and Croatia. Proc. I SEAPBP Conf., Engelberg, Switzerland, 1995. Environmental encounters, No. 38, Council of Europe Publ. (in press). DEAG, J. M.; CANDACE, E. L.; MAnning, A. (1987): The consequences of differences in litter size for nur- sing cat and her kittens. J. Zool. (London) 213, 153-179. FERGUSON, J. W. H.; GALPIN, J. S.; DE WET, M. J. (1988): Factors affecting the activity pattern of black- backed jackals Canis mesomelas. J. Zool. (London) 214, 55-69. FESTETICS, A. (1981): Die Wiederansiedlung des Luchses am Beispiel der Ostalpen. Natur und Land- schaft 56, 120-122. HALLE, S. (1995): Diel pattern of locomotor activity in populations of root voles, Microtus oeconomus. ]. Biol. Rhythms 10, 211-224. HALLER, H.; BREITENMOSER, U. (1986): Zur Raumorganisation der in den Schweizer Alpen wiederange- siedelten Population des Luchses (Zynx Iynx). Z. Säugetierkunde 51, 289-311. HEMMER, H. (1993): Felis (Lynx) Iynx Linnaeus 1758 — Luchs, Nordluchs. In: Handbuch der Säugetiere Europas. Bd. 5 (2). Ed. by M. STuBBE and F. Krapp. Wiesbaden: AULA-Verlag, Pp. 1119-1167. JEDRZEJEWSKI, W.; SCHMIDT, K.; MILKOWSKI, L.; JEDRZEJEWSKA, B.; OKARMA, H. (1993): Foraging by lynx and its role in ungulate mortality: the local (Bialowieza Forest) and the Palaearctic viewpoints. Acta Theriol. 38, 385-403. Loupon, A. S.1I. (1985): Lactation and neonatal survival of mammals. Symp. Zool. Soc. Lond. 54, 183- 207. MATJUSCHKIN, E.N. (1978): Der Luchs. Neue Brehm-Bücherei. Bd. 517. Wittenberg Lutherstadt: A. Ziemsen Verlag. PULLIAINEN, E. (1981): Winter diet of Felis lynx L. in SE Finland as compared with the nutrition of other northern Iynx. Z. Säugetierkunde 46, 249-259. PULLIAINEN, E.; LINDGREN, E.; TUNKKARI, P.S. (1995): Influence of food availability and reproductive status on the diet and body condition of the European Iynx in Finland. Acta Theriol. 40, 181-196. SCHMIDT, K.; JEDRZEJEWSKI, W.; OKARMA, H. (1997): Spatial organization and social relations in the Eu- rasian Iynx population in Bialowieza Primeval Forest, Poland. Acta Theriol. 42, 289-312. SEIDENSTICKER, J. (1977): Notes on the early maternal behavior of the leopard. Mammalia 41, 111-113. SOSTAR, S. V.; BunEvic, A.N. (1986): The impact of Iynx on red deer population. IV S’ezd Vsesojuzno- go Teriologiceskogo Obscestva, Moskwa, Vol. 3, 76-77. Sunauist, M.E. (1981): The social organisation of tigers (Panthera tigris) in Royal Chitawan National Park, Nepal. Smithsonian Contr. Zool. 336, 1-98. TOUTENBOURG, H. (1994): Versuchsplanung und Modellwahl. Heidelberg: Physia Verlag. TYLor, S. (1979): Time-sampling: A matter of convention. Anim. Behav. 27, 801-810. WHITE, G. C.; GARROTT, R. A. (1990): Analysis of wildlife radio-tracking data. San Diego: Academic Press. Authors’ address: ILKA REINHARDT and Prof. Dr. STEFAN HALLE, Institute of Ecology, Friedrich-Schil- ler-University Jena, Dornburger Str. 159, D-07743 Jena, Germany Z. Säugetierkunde 64 (1999) 76-84 © 1999 Urban & Fischer Verlag SÄUG EIl au, „Aell http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Aspects of mother-kid behavior in Alpine chamois, Rupicapra rupicapra rupicapra By KATHREEN E. RUCKSTUHL and P. INGOLD Department of Biology, University of Sherbrooke, Sherbrooke, Quebec, Canada and Department of Zoology, University of Bern, Bern, Switzerland Receipt of Ms. 02. 10. 1998 Acceptance of Ms. 14. 01. 1999 Abstract We studied mother-kid associations for 9 mother-kid pairs of alpine chamois (Rupicapra rupicapra ru- picapra) from May to October 1991. Mother-kid distance was studied from birth to weaning of the young. We further investigated the spatial relationship between the kid and the closest alien chamois within a group throughout the 6 months. The synchronization of activities between the mother and her kid was also analyzed. When mother and kid were in the same group, they were next neighbors in 90 % of all observations. Mother and kid were closest to each other when Iying, while they were fur- thest apart when mothers were grazing and kids lying. Mothers and kids spent most of their daytime in the same group. The synchronization of activities between the mother and her kid increased with increasing age of the young. Mother and kid maintained close contact throughout weaning. The close association of mother and kid throughout the first 6 months of life of the young likely evolved as an anti-predator behavior and is first maintained through suckling and later through synchronization of activities between mother and kid. Key words: Rupicapra, mother-kid bond, ontogeny, synchronization Introduction Suckling behavior and particularly mother-young interactions such as synchronization of activities and maintenance of proximity are central for the development and survival of young ungulates, especially in the first few week after birth (EpsmArK 1971; Geıst 1971; SHACKLETON and HAywooDp 1985). Although lactation is the most important care mothers give to their offspring in the first few months of their life, guidance, transmission of knowledge, and the learning of social behavior can further benefit the young (LEnt 1974; RICHARD-HANSEN and CAMPAN 1992; RICHARD-HANSEN 1993). Especially during eagle (Aquila chrysaetos) attacks proximity to the mother or to a defending female can be cru- cial for chamois kids (KrÄmER 1969; LocArı 1990). The time the young spend in the vicin- ity of their mother also seems to depend on the frequency with which they suckle (SHACKLETON et al. 1984; SHACKLETON and Haywoop 1985). As suckling frequency de- clines, the mother-kid bond may loosen as well. In Rocky Mountain bighorn sheep, Ovis canadensis, lambs spend less time with their mother and more time in the company of other lambs when they are weaned (BERGER 1979), although post-weaning mother-daugh- ter associations occasionally occur at high population densities (L’HEUREUX et al. 1995). Chamois kids belong to the ‘follower’ type (LEnT 1974) and kids follow their mothers within a few hours of birth (CouTURIER 1938). Despite its evolutionary importance and ef- 0044-3468/99/64/02 - 076 $ 12.00/0 Aspects of mother-kid behavior in Alpine chamois 7 fects on social ontogeny and group structure, the development of mother-offspring bonds in ungulates has received little attention, apart from studies on suckling behavior (Ro- BINSON 1980; CLUTTON-Brock et al. 1982, 1989; OFTEDAL 1985; FESTA-BIANCHET 1988; WHITE et al. 1989; RucksSTUHL and InGoLD 1994; Hass 1995). Although mother-kid bonds seem to be the strongest associations in chamois, descriptions on the ontogeny and strength of these associations are rare (KRÄMER 1969; RıcHARD-HANSEN and CAMPAN 1992; RiCHARD-HAnsEN 1993). We do not know how proximity between mother and kid is maintained through ontogeny and how close mother-kid associations are compared to associations between the kid and an alien chamois. The aim of this study was to obtain quantitative information on the ontogeny of spa- tial relationships and the synchronization of activities in mother-kid pairs. We further in- vestigated the spatial relationship between kids and their closest neighbor, to evaluate the strength of the mother-kid bond in comparison to non-mother-kid associations. Material and methods Study area and animals Nine individually marked female chamois and their kids were observed between May and ÖOcto- ber 1991, on the Augstmatthorn, Switzerland (see detailed description of the study site in KRÄMER 1969). The study area is in a game sanctuary where hunting is prohibited. Focal females were all tagged with yellow-colored plastic stripes glued around the horns. Kids were unmarked, but individu- ally recognizable through pelage characteristics, scars, and size differences. Mothers were identified during suckles, as chamois only suckle their own offspring (KRÄMER 1969). Female home ranges were between 1400 and 2137 m above sealevel (INGoOLD and MARBACHER 1991). Date of birth was estimated to be mid-way between the last observation of a female without a kid and the first observation with a kid. One kid was born between June 3 and 12. The 8 other kids were born between May, 8 and 22. All observations were made from a point where most of the slope used by the focal females was visible. Observations were made with binoculars (10x40) and spotting scopes (30x 60). Ages of females were estimated through counting of horn annuli at capture. The fe- males ranged in age between 4 and 13 years (see Tab. 1). Data collection Each mother-kid pair was observed for 2 to 9 days, depending on presence and visibility. A total of 52 days (346 hours) focal scan sampling observations (ALtMAnN 1974) were carried out. If several marked females were visible at the beginning of observation sessions, the female with the least obser- vation hours or days was chosen as focal anımal. Observations lasted between 2 and 14 hours (Tab. 1). Long observation hours (>8 hours) on the same mother-kid pair lasted usually from dawn to dusk, shorter observation periods (<8 hours) were distributed evenly throughout the day. The activity (walking, standing, lying, or grazing) of the kid, the mother, and the kid’s closest neighbor, as well as the distances between mother and kid and the kid and its closest neighbor were written down every 15 minutes. From these scan samples we calculated the percentage of time mother and kid were next neighbors, and the percentage of time an alien chamois and the kid were next neighbors. Focal observations shorter than 5 hours were discarded from the analysis, because short ob- servation periods may not reflect average activity budgets or inter-individual distances of females and their kid (especially because sample sizes per female are relatively small). Distances between individuals were estimated in animal lengths (referred to as chamois lengths). The next neighbor is defined as the adult chamois, which was closest to the kid at the moment of sam- pling. Animals were determined to be in the same group if they were closer than 50 chamois lengths from each other. Distances or proximity between mother and kid were estimated, when both were Iy- ing or grazing or, when the kid was lying and the mother was grazing. Distances between the kid and next neighbor were only estimated when both were grazing. Most chamois kids in our study area were weaned by November (RuckstuHL and InGoLD 1994). We therefore describe the ontogeny of the mother-kid relationship from birth throughout weaning. 78 KATHREEN E. RUCKSTUHL and P. INGOLD Statistical analyses Mean distances between mothers and kıds were calculated separately for each mother-kid pair for the first 6 months of life of the kid. Differences in proximity between mother-kid and kid-next neighbor were tested with ANOVA (SokAL and RoHLr 1995). Differences in proximity were then tested each month using Mann-Whitney U-tests (SIEGEL and CASTELLAN 1988) and Bonferroni adjusted signifi- cance levels. The effect of kid age (in months) and individual differences on the distances between mother-kid pairs were calculated using 2-way-ANOVA. Some mother-kid pairs were sampled more than once per month. We therefore calculated the mean percent time mothers and kids had the same activity in a given month, to reduce pseudo-repli- cation (MAcHuis et al. 1985; LEGER and DipricHsons 1994). We calculated the percent time females and their kids had the same activity. Percentages were arcsine square-root transformed (ZArR 1984). Medians are given with interquartile ranges, means are given with standard deviations. Results If mother and kid were in the same group, the kid was always closer to its mother than to any other adult throughout the first 6 months of life, except in September (F = 41.33, df=1, p<0.001; Fig. 1). Mother and kid were neighbors in 90% of all observations throughout the summer (Fig. 2). When grazing, the kid always followed its mother. The distances between mother and kid were therefore small (Fig. 3a). The apparent increase in distance between mother and kid from the 4" to the 6°" month was due to individual differences (F=3.06, df=8, p<0.001) in mother-kid pairs and was not age-related (F = 0.63, df=1, p = 0.44). 50 U] mother UZ2 alien 40 30 20 10 distance in chamois lenghts n=99 n=136 n=50 n=25 (9) (6) (4) (2) | | | | | | May June July August Sept. Oct. Fig. 1. Distances in chamois lengths between mother and kid, and between the kid and the next alien female from May to October 1991, Augstmatthorn, Switzerland. Number in brackets = number of mother-kid pairs, n = number of estimated distances. *=p<0.05, **=p<0.001, n.s.= no significant difference (Mann-Whitney U-test, two-tailed). Box plots represent maxima, medians and minima with interquartile ranges. Aspects of mother-kid behavior in Alpine chamois 70) U] mother 100 | alien 80 : | [(o) Re} zZ @2 60 © [= P) x = o 40 E ee) Rn 20 June July August Sept. Oct. Fig. 2. Percent of observations when the next neighbor of a chamois kid was either its own mother or an alien female. n = observation days. Box plots represent maxima, medians and minima with interquar- tile ranges. During the first 6 months of life of the kid mother and young most often lay in body contact with each other. The median distance therefore was often Ochamois lengths (Fig. 3b). In general it was the kid who actively searched body contact with its mother. Except in two observations the mother was the first to bed down and the kid then lay down beside her. Adult females grazed for longer and more frequently than their kids. After a longer lying phase, but also when the mother was grazing, the kid often remained bedded. Therefore the mother automatically increased the distance between her and her young while grazing (Fig. 3c). Hence greater distances were observed when the kid was Iying and the mother grazing than when both had the same activity. As distances between mother and kid showed, they usually stayed in close proximity to each other when they were in the same group. Mother and kid did not only spend most of their time in the same groups, but they also were closest to each other (Figs. 1 and 4). The percent time mother and kid spent in the same group during a day varied between and within mother-kid pairs (Tab. 1). While chamois number 4 was always seen in the same group as her kid during all observations (6 days), chamois number 6 was with her kid in less than 50 % of the observation period in 2 of 3 days. Longer periods of separation between mother and kid were rare but hap- pened sometimes when mothers went to natural salt licks a few hundred meters from their preferred grazing grounds. Kids then stayed in the company of another mother and her kid. Separations between mother and kid could last up to 6 hours (2 observations). After a separation the mother always returned to the place where she had left her kid. If it was no longer there she started walking around in search of her kid often calling, and looking around. After the mother and kid reunited the kid attempted to suckle immedi- ately. 80 KATHREEN E. RUCKSTUHL and P. INGOLD a) Ben n=10 n=55 n=81 n=124 n=51 n=11 ee n=16 n=108 n=40 n=122 n=12 n=14 b) Distance in chamois lengths Distance in chamois lengths „Ust: n=5 n=14 n=20 n=24 n=9 Distance in chamois lengths May June July August Sept. Oct. ı nn 9 6 (4 (2) Fig. 3. Distances in lengths between chamois mother and kid when a) both were grazing or b) lying, and c) when the mother was grazing and the kid Iying. n = number of estimated distances. Number in brackets = number of mother-kid pairs. Box plots represent maxima, medians and minima with inter- quartile ranges. A kid’s age affected the percentage of time mother and kid had the same activity (F4>3 = 5.09, p < 0.005). Mothers and kids were least synchronized in their activities in June (61 +16 % of observation time) and July (69+8%), more synchronized in August (78+5%) and September (78+9%) and most in October (86 + 6%; significant differ- ence between June, July, and October: Scheff& post-hoc, p < 0.05). There were no indivi- dual differences in percent time spent in the same activity (Fs 2 = 0.85, p = 0.53). Aspects of mother-kid behavior in Alpine chamois 81 100 80 60 40 % time in same group 20 n=1 n=10 n=17 n=17 n=6 n=3 (1) (6) (7) (7) (3) (3) May June July August Sept. Oct. Fig. 4. Percent time chamois mothers and kids spent together in the same group during observations from May to October 1991, Augstmatthorn, Switzerland. n = observation days. Number in brackets = number of mother-kid pairs. Box plots represent maxima, medians and minima with interquartile ranges. Discussion Mother and kid maintained a very close proximity to each other in the first 6 months of the kid’s life, as described for isard, Rupicapra pyrenaica pyrenaica, (RICHARD-HANSEN and CAamPpan 1992). They spent most of their time in the same group and were mostly next neighbors and therefore were closely associated. With increasing age the kid moved more freely within a group, probably in accordance with the increased time it spent graz- ing. The kid therefore sometimes was in the vicinity of an alien female, although it stayed most often close to its mother. The chamois is a follower type and the mother defends her kid against predators instead of relying on concealment as in the hiding types (LENT 1974). It is therefore important for the mother and her kid to stay in close proximity to each other (KRÄMER 1969). Suckling further increases the benefit to the young and main- tains the strong bond between mother and offspring. Chamois often live in rugged terrain and the kid probably also depends on guidance through difficult terrain (KRÄMER 1969; Geist 1971). Kids did not graze often in their first month of life and usually fed close to their mothers, decreasing the average mother-offspring distance. Interestingly, the distance be- tween mother and kid on average was always shorter than the distance between the kid and an alien chamois, as suggested by PACHLATKO and NIEVERGELT (1987). This is contrary to what we would expect if the process of weaning weakened the mother-offspring bond (GEIST 1971). When mother and kid were separated the kid was never alone, but was with other kids in a ‘kindergarten’ or in the company of a ‘baby-sitter’ (RUCKSTUHL and IncoLp 1998). This might explain why kids on average are less than 10 animal lengths 82 KATHREEN E. RUCKSTUHL and P. INGOLD Table 1. Percent time mother-kid pairs spent in the same group during observation hours, at the Augstmatthorn, Switzerland, 1991. Age = estimated age of the mother. % in same group = percent of the total observation time/day, when mother and kid were together in the same group. Mother # Observation day Observation hours % in same group July 7 August 4 August 3 September 19 October 3 June 26 August 12 August 20 September 16 July 1 July 2 July 28 July 30 August 15 August 22 July 11 July 20 August 7 August 13 August 27 September 20 August 14 August 21 October 10 June 11 July 20 August 5 August 22 July 30 August 16 October 4 June 22 July 23 July 30 August 13 August 16 October 11 1 1 2 2 22 3 3 3 3 4 4 4 4 4 4 3 5 5 5 5 5 6 6 6 7 7 7 U 8 8 8 9 9 9 9 9 9 SINN 26090 00 oo un un N away from the next alien chamois within their group. As suggested by RiCHARD-HANSEN (1993) kids seem to be attracted by peers and often also approach adult females. Adult females often respond to such an approach by threatening with their horns or attacking (Locarı 1990), leading to the greater observed distances between the kid and alien fe- males compared to the kid with its own mother. Synchronization of activity is crucial for group cohesion (JARMANn 1974; BENHAM 1982; Rook and PEnnInG 1991; AGESUMA 1995). Mothers and kids were least synchronized in their activities (grazing/lying) in June and July and most synchronized in October. In the first few weeks kids do not spend much time grazing and mainly depend on their mothers’ milk. As the kids grow older they increase the time spent grazing and therefore likely become more synchronized with their mothers. On the other hand kids suckle less often and for shorter periods in September and October (RucksTUHL and InGoLD 1994), Aspects of mother-kid behavior in Alpine chamois 83 and the mother-kid bond could consequently decrease. Nevertheless, we observed a strong mother-kid bond in our study area throughout the season; which may have two dif- ferent reasons: 1) in the first few months of life, the kid depends on the mother’s milk and it therefore should have a strong incentive to stay close to its mother, 2) as the mother-young bond loosens up because of weaning, proximity may be maintained through increased synchronization of activities. Acknowledgement This study was financed through the Federal Game Department and the Game Department of the Canton of Berne. We are thankful to M. FEstA-BıAancHET and P. NEUHAUS and an anonymous referee for comments on an earlier draft of this manuscript. Zusammenfassung Aspekte der Mutter-Kindbeziehung bei der Alpengemse (Rupicapra rupicapra rupicapra) Am Augstmatthorn im Berner Oberland, Schweiz, wurde vom Mai bis Oktober 1991 die Mutter-Kind Beziehung von Gemsen (Rupicapra rupicapra rupicapra) untersucht. Wenn Mutter und Kind in der- selben Gruppe waren, so waren sie meistens nächste Nachbarn (außer im September) und verbrach- ten die meiste Zeit nahe beisammen. Die Aufrechterhaltung der Nähe schien von dem Kitz aus zu kommen. Die größten Distanzen zwischen Geiß und Kitz wurden gemessen, wenn die Geiß äste und das Kitz lag. Obwohl die Kitze mit zunehmendem Alter unabhängiger wurden, blieb die enge räum- liche Beziehung bestehen. Mit zunehmendem Alter der Kitze waren Mutter und Kind in ihren Aktivi- täten stärker synchron. References AGESUMA, N. (1995): Foraging synchrony in a group of Yakushima Macaques (Macaca fuscata yakui). Folia Primatol. 64, 167-179. ALTMANN, J. (1974): Observational study of behavior: sampling methods. Behaviour 49, 227-267. Benham, P. F. J. (1982): Synchronization of behaviour in grazing cattle. Appl. Anim. Ethol. 8, 403-404. BERGER, J. (1979): Weaning conflict in desert and mountain bighorn sheep (Ovis canadensis): an ecologi- cal interpretation. Z. Tierpsych. 50, 188-200. CLUTTON-BROocK, T. H.; ALBON, S. D.; GUINNESS, F. E. (1989): Fitness costs of gestation and lactation in wild mammals. Nature 337, 260-262. CLUTTON-BRrock, T. H.; IAson, G. R.; ALBON, S. D.; GUINEss, F. E. (1982): Effects of lactation on feeding behaviour and habitat use in wild Red deer hinds. J. Zool. (London) 198, 227-236. EPSMARK, Y. (1971): Mother-young relationship and ontogeny of behavior in reindeer (Rangifer taran- dus L.). Z. Tierpsych. 29, 42-81. FESTA-BIANCHET, M. (1988): Nursing behaviour of bighorn sheep: correlates of ewe age, parasitism, lamb age, birthdate, and sex. Anim. Behav. 36, 1445-1454. GEIST, V. (1971): Mountain Sheep: A Study in Behaviour and Evolution. Chicago: Univ. Press. Hass, C. C. (1995): Gestation periods and birth weights of desert bighorn sheep in relation to other ca- prinae. Southw. Naturalist 40, 139-147. INGOLD, P.; MARBACHER, H. (1991): Dominance relationships and competition for resources among cha- mois Rupicapra rupicapra in female social groups. Z. Säugetierkunde 56, 88-93. JARMAN, P. J. (1974): The social organisation of antelope in relation to their ecology. Behaviour 48, 215- 267. KRÄMER, A. (1969): Soziale Organisation und Sozialverhalten einer Gemspopulation (Rupicapra rupi- capra) der Alpen. Z. Tierpsych. 26, 889-964. LEGER, D. W.; DiDRICHsons, I. A. (1994): An assessment of data pooling and some alternatives. Anim. Behav. 48, 823-832. 84 KATHREEN E. RUCKSTUHL and P. INGOLD Lent, P. C. (1974): Mother-infant relationships in ungulates. In: Behaviour of Ungulates and its Relation to Management. Morges: IUCN 24, 14-55. U’HEUREUX, N.; LUCHERINI, M.; FESTA-BIANCHET, M.; JORGENSON, J. T. (1995): Density-dependent mother-yearling association in bighorn sheep. Anım. Behav. 49, 901-910. Locartı, M. (1990): Female chamois defends kids from eagle attacks. Mammalia 54, 155-156. MACHLIS, L.; DoDD, P. W. D.; FENTRESS, R.; FENTRESS, J. C. (1985): The pooling fallacy: problems arising when individuals contribute more than one observation to the data set. Z. Tierpsych. 68, 201-214. OFTEDAL, O. T. (1985): Pregnancy and lactation. In: Bioenergetics of Wild Herbivores. Ed. by R. J. Hupson and R. G. WHITE. Boca Raton: CRC Press, Inc. Pp. 215-238. PACHLATKO, T.; NIEVERGELT, B. (1987): Time budgeting, range use pattern, and relationships within groups of individually marked chamois. In: The Biology and Management of Capricornis and Re- lated Mountain Antelopes. Ed. by H. SomA. London: Croom Helm. Pp. 93-101. RICHARD-HANSEN, C. (1993): Social interactions in Isard kids, Rupicapra pyrenaica pyrenaica (Artio- dactyla: Bovidae): possible influence on development of social bonding. Ethology 94, 291-305. RICHARD-HANSEN, C.; CAMPAN, R. (1992): Social environment of Isard kids, Rupicapra pyrenaica p., during their ontogeny. Z. Säugetierkunde 57, 351-363. ROBINson, J. J. (1980): Energy requirements of ewes during late pregnancy and early lactation. Vet. Re- cord 29, 282-284. Rook, A. J.; PEnNING, P. D. (1991): Synchronisation of eating, ruminating, and idling activity by grazing sheep. Appl. Anım. Behav. Science 32, 157-166. RUCKSTUHL, K.; INGOLD, P. (1994): On the suckling behaviour of Alpine chamois Rupicapra rupicapra rupicapra. Z. Säugetierkunde 59, 230-235. RUCKSTUHL, K. E.; InGoLD, P. (1998): Baby-sitting in chamois: a form of cooperation in females? Mam- malıa 62, 125-128. SHACKLETON, D. M.; HAywoop, J. (1985): Early mother-young interactions in California bighorn sheep, Ovis canadensis californiana. Can. J. Zool. 63, 868-875. SHACKLETON, D. M.; PETERSON, R. G.; HAYwoopD, J.; BOTTRELL, A. (1984): Gestation period in Ovis cana- densis. J. Mammalogy 65, 337-338. Siegel, S.; CASTELLAN, N. J. (1988): Nonparametric statistics for the behavioral sciences. 2. ed. New York: McGraw-Hill. Sokal, R. R.; RoHLF, F. J. (1995): Biometry. 3. ed. New York: W. H. Freeman and Company. White, R. G.; HOLLEMAN, D. F.; TiPLADy, B. A. (1989): Seasonal body weight, body condition, and lacta- tıonal trends in muskoxen. Can. J. Zool. 67, 1125-1133. ZAR, J. H. (1984): Biostatistical Analysis. 2. ed. Englewood Cliffs: Prentice-Hall. Authors’ addresses: Dr. KATHREEN E. RUCKSTUHL, Dept. of Zoology (LARG), University of Cam- bridge, Downing Street, Cambridge CB2 3EJ United Kingdom and Prof. Dr. PAUL InGoLp, Ethology and Nature Conservation, Department of Zoology, University of Berne, Länggassstrasse 27, CH-3012 Berne, Switzerland. Z. Säugetierkunde 64 (1999) 85-90 ZEITSCHRIFT FÜR © 1999 Urban & Fischer Verlag SAUG ETIERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Estimating porcupine (Erethizon dorsatum Linnaeus, 1758) density using radiotelemetry and replicated mark-resight techniques By Morıy O. Hare and T. K. FULLER Department of Forestry and Wildlife Management, University of Massachusetts, Amherst, USA Receipt of Ms. 31. 07. 1998 Acceptance of Ms. 15. 10. 1998 Abstract Quantitative estimates of the density of North American porcupines (Erethizon dorsatum Linnaeus, 1758) were obtained in two adjacent study areas in central Massachusetts (northeastern U.S.). Using mark-resight (with radio-marked porcupines) estimators with data collected on one single and four re- plicated surveys, none of the estimates (range = 10-42 porcupines/km’) provided the relative precision needed to detect area-specific differences in density. This was because of the small samples of marked individuals (range= 5-12/survey; 4-6/km’), low observability of porcupines during surveys (x = 15% of marked animals seen; range = 0-40%), and low numbers of surveys. Porcupines are more reclusive than we previously thought, and intensive survey efforts are needed to obtain reasonably precise den- sity estimates in forested habitats. Key words: Erethizon dorsatum, porcupine, mark-recapture, density estimation Introduction Despite the apparent abundance and wide geographic distribution of the porcupine in North America, there exist few quantitative estimates of population density for this spe- cies. In addition to helping make sense of demographic data, such estimates are useful for interpreting the role of porcupines in a forest ecosystem (e.g. KREFTING et al. 1962; KEITH and Cary 1991) or gauging the effectiveness of an eradication program (e.g., Dopce 1959; BRANDER and Books 1973). We wanted estimates of porcupine population density for use in demographic (HALE and FuULLER 1996) and habitat (GRIESEMER et al. 1995, 1996, 1998) studies we were con- ducting in central Massachusetts. Censuses using tracks in snow have been used in a num- ber of studies (e.g., Currıs 1944; BRANDER 1973; PowELL and BRANDER 1977; SmıTH 1977; Roze 1984), but when we followed tracks in snow we often encountered networks of in- tersecting and overlapping porcupine trails in concentrated denning areas where porcu- pines shared feeding trees and/or dens (GRIESEMER et al. 1996). This precluded our suc- cessful use of this technique. Other methods besides censuses have been employed to estimate porcupine numbers, but most reports seemed unsatisfactory or incomplete. Several researchers (TAyLor 1935; REEKS 1942; GoLLEY 1957; KrEFTING et al. 1962; DopDGeE 1967) based population estimates on the number of porcupines seen or shot over a time period. These counts may have provided minimum numbers, but did not account for missed animals or immigrants. SHA- PIRO (1949) used line strip techniques but did not detail his methods, did not account for 0044-3468/99/64/02 - 085 $ 12.00/0 86 Morıy O. HALe and T. K. FULLER potential bias due to low sighting probability, and did not provide an estimate of standard error. BRANDER (1973) used mark-recapture methods to estimate porcupine numbers and to generate an estimate of standard error. His study, however, lacked evidence that as- sumptions of mark-recapture methods (e.g., a closed population, equal catchability, and no loss of marks) were not violated. The mark-resight method (Orıs et al. 1978) described in this study was used in asso- ciation with radio-marked animals (cf. MıLLEr et al. 1997) in an attempt to assess its rela- tive precision and feasibility as applied to the study of porcupines. We present our results, discuss the shortcomings of our efforts, and suggest means by which porcupines might be more rigorously counted. Study area We sampled 2 survey areas on the Prescott Peninsula at the Quabbin Reservoir in central Massachu- setts, an area covered by Transition Hardwoods-White Pine-Hemlock forest (WESTVELD et al. 1956). Elevation on the study areas ranged from 162 to 351 m and the terrain was hilly, including some areas with steep rocky slopes. The East and Central survey areas covered 2.2-2.6 km? and 2.2-3.1 km? respectively, and were lo- cated about 1 km apart. The East area included a very steep rocky ridge where numerous porcupine dens were concentrated, and had significantly fewer white pines (Pinus strobus) than the Central area (GRIESEMER et al. 1996, 1998). The Central area was only moderately hilly and lacked rocky slopes. There were fewer den sites and usually they were hollow trees or logs. We established these survey areas to encompass porcupine winter and summer home ranges as determined from radio-tracking 50 different individuals from July 1991 to September 1993 (HALE and FULLER 1996); this allowed us to maximize the number of radio-marked porcupines in the areas at the time of the surveys. Material and methods Porcupines were captured by hand or in live traps by focusing efforts in two portions of the Prescott Peninsula that eventually became our study sites. Captured individuals were marked with 40-g radio- collars and small (3 mmx10 mm) yellow plastic eartags (HALE et al. 1994; HALE and FULLER 1996). Some individuals also had one to three 1.5x1.5 cm pieces of colored vinyl taped affıxed to the end of their 20-cm radio antennas, but we assumed that none of the marking devices were so conspicuous as to cause the initial sightability of marked animals to differ from that of unmarked ones (in fact, at least 4 marked porcupines were initially classified as unmarked by survey observers and only sub- sequently verified as marked by an independent observer with a telemetry receiver). Data for the mark-resight estimates were collected concurrently during nine surveys carried out during spring and autumn 1992, and spring 1993. To maximize the sightability of porcupines, each sur- vey was conducted during these parts of the year when most porcupines were out of dens and the for- est canopy was relatively open (GRIESEMER et al. 1998). Of the nine surveys, two surveys were made in the East area during each of the three seasons. The remaining surveys were conducted in the Cen- tral area, once in fall 1992 and twice in spring 1993. In each study area, we mapped twelve 1.5- to 3-km long parallel transects 100-m apart; the first transect line was begun at a random point no nearer than 50 m, and no farther than 100 m, from the survey area boundary. Each transect was walked in approximately 2 hours by solitary persons ex- perienced in looking for porcupines or by pairs of less experienced persons (e.g., at a rate of about 1 km/h). Transects walked in a single day comprised one survey. When a porcupine was sighted it was classified as marked or unmarked, based on observation with binoculars: The location of each sighted porcupine was marked with flagging. Immediately after the transects were completed, all radio-marked porcupines in or near the survey area were tracked and visually located to verify those reported as being seen from the transect lines, and to identify those in the survey area but not seen. Mark-resight density estimates were calculated using NOREMARK (WHıTE 1996) which calcu- Porcupine density estimations 87 lated simple Lincoln-Petersen estimates for single surveys, and joint hypergeometric maximum-likeli- hood estimates (JHE) and 90% confidence intervals (CI) for user-specified alphas for replicated sur- veys (e.g., MILLER et al. 1997). To investigate the actual effort needed to identify differences between mark-resight estimates of porcupines, we identified several combinations of observability, marked ani- mals available, and number of repeated surveys for a hypothetical population, and then estimated the population density using NOREMARK. Because our porcupines were marked with radio-collars, we could monitor deaths and emigration and determine how many marked individuals were present on the survey area on each survey day; thus, we assumed our population was closed. Because these marked animals were previously captured opportunistically within the study area and monitored for up to 2 years, we recognize that our marked sample may not have been representative of the population either at the time of capture or at the time of the survey. However, because there is no feasible way to make such an evaluation, we as- sumed the sample was representative for the purposes of these analyses. We also recognize that a commonly violated assumption of mark-reobservation studies is that of “equal catchability”, resulting in a negative bias in the estimate. In this study porcupines sighted in trees and on the ground might have unequal sighting probabilities. Population densities for each such sub-population, as well as for different ages and sexes, should ideally have been estimated separately, but the small sample size in this study ruled out this possibility. However, the different methods used for initial captures and later resights should cause any bias produced by unequal catchability to be relatively small. We found that the sighting location (ground or tree) was, in fact, independent of the marked/unmarked status of por- cupines (X? = 0.03, d.f.= 1, P = 0.86). Results During each of the nine surveys, 2-10 porcupines (x = 6.4) were seen by observers (Tab. 1); this comprised 4-32% (x = 16%) of the estimated population. Although 5-12 marked porcupines were available to be seen during each of the surveys (x = 7.2, this comprised 5-27%, x = 18% of the estimated population), only 0-2 of these were seen (x = 1.0; x proportion seen = 0.15, range = 0.00-0.40). During the five area-specific survey periods, the minimum number of porcupines known to be alive (MNA) varied from 9-16. Point estimates of density determined by mark-resight methods varied 4-fold (range = 10-42/km?) in the East area and by 1.5-fold (range = 11-17/km?) in the Central area, but significant differences among seasons or areas could not be detected (Tab. 1). Our simulations, given hypothetical population of 100, indicated that increasing the proportion of the population seen from 0.10 to 0.30 resulted in a 57% reduction in confi- dence interval length (Tab. 2). Similarly, increasing the proportion of the population marked from 0.20 to 0.40, or increasing sighting occasions from 2 to 4, resulted in 40 and 38% reductions, respectively. Improving all three parameters simultaneously resulted in an 83% reduction. Discussion Even with the major effort we expended in trying to enumerate porcupines, our quantita- tive estimates using the mark-resight method were unsatisfactory. Though our mark-re- sight density estimates for both sites and all surveys (10-42/porcupines/km”) are compar- able to other relatively recent estimates (5-18/km°; PowELL and BRANDER 1977; RozE 1984) from mixed forests, unrealistic variation among survey estimates make us reluctant to say much about differences in porcupine density. At the time of our surveys, we had no good feel for porcupine abundance, much less the proportion of animals marked or the likelihood of seeing animals while slowly walking through the woods. This, combined with the unavailability of good computer models to augment our efforts, resulted in im- precise, though probably not inaccurate, population estimates. (19-6) LI (78-5) II (Zs-L) rl (rSE-EI) Zr Moriıy ©. HALE and T. K. FULLER (ST-9) OT (ID %06) „us/’oN oyewmso Ayısuaq 0,0) 00 -IOJeuINs9 PooyıjoyI-wnwmxewm 9LN9Wo9gLSdAy-Jurol Zursn uonepndod p9sop e 107 ayewnso uonepndod ysIssı-NIeW 5 („uy) e9Ie A9A -INS JO 9ZIS uonpndoq oyewmsd ‘OU WNWIUN "Syeumsa UISI9IOAT-UJOIUTT . 97T [eJOL £6 AeW 8 £6 keWI 6 [EJOL 76 PO r7 [eu9) rI [eJOL £6 ke Ss £6 Adv 87 97 [eJOL 76 PO TE PO LI IT [eIoL 76 AeW EI 76 ıdV 67 u99s ge sourdnsıod uoassaurd -[reae sourd p>yıewu -naıod pay -na1od p9Y jo uonıodoıg -ıewjJo ‘ON -IewJo ON A9Aıms Jo SJeq u99s sourd -n910d poy.ıeuu -un JO 'ON (VNN) 9AITe UMOUY BIIV ‘(9661 Am :AMVWIMON we.ıdord 191ndwo9 pue) sIoyewmss Jygısd1-yızuı Jursn Aq pourut -1979p se sYOSnyVesseW [enus9 ur ejnsurusg N09s91d >y) uo sea1e Apnys om} ur sojeumss Ayısusp uonerndod Funpnssı ay) pue eyep AaAıns aurdn91og "T 3IQEL Porcupine density estimations 89 Table 2. Variation in precision of mark-resight porcupine population estimates given a) different sightability, b) proportion of animals marked, c) number of repeated surveys, or a combination of all three. All estimated are based on a hypothetical population of 100 porcupines, and use joint hyper- geometric maximum likelihood estimates (JHE) as calculated by the computer program NORE- MARK (WhrTE 1996). Sight- Proportion Number of porcupines seen No. of Total (90%CI) abılyaR 201 DOLCU, er Tr TEE ET EEE eDeated pines Total Marked Unmarked surveys marked 99 (57-235) 99 (74-150) 99 (57-235) 99 (70-176) 99 (57-235) 99 (65-176) 100 (87-117) Our simulation modeling that incorporated realistic data supported the notion that the precision of mark-resight estimates can be improved by increasing the number of surveys, the number of marked porcupines, and/or the number of porcupines seen per survey (e.g., Rogson and REGIErR 1964; Rıce and HARDER 1977; PoLLock et al. 1990). From a practical point of view, increasing the number of sighting occasions and increasing the number of marked porcupines might be the best way to increase precision of estimates. Increased search effort might help, but we have no data to test this potential improvement. Acknowledgements We are grateful to S. GRIESEMER, D. BROWNLEE, E. YORK, S. Kot, M. BRADY, G. GARNETT, T. LEONARDO, E. HETRICK, S. HOKENSON, S. SCHALLER, L. SALAS, T. OLIVER, and numerous undergraduate students for field assistance. We received funding and support from the U.S. Department of Agriculture (USDA) Forest Service Northeastern Forest Experiment Station; a UUSDA MclIntire-Stennis grant; the College of Food and Natural Resources and Department of Forestry and Wildlife Management at the Univer- sity of Massachusetts, Amherst; the Metropolitan District Commission (especially P.Lyons); and the Massachusetts Cooperative Fish and Wildlife Research Unit, U.S. Fish and Wildlife Service (now Bio- logical Resources Division of the U.S. Geological Survey). L. MACDonALD, G. SMITH, and 2 anony- mous reviewers provided many helpful suggestions on an earlier draft of this study. Zusammenfassung Dichteschätzungen von Baumstachlern (Erethizon dorsatum Linnaeus, 1758) mittels Radiotelemetrie und Sichtungen markierter Individuen Quantitative Schätzungen der Dichte von nordamerikanischen Baumstachlern (Zrethizon dorsatum Linnaeus, 1758) wurden in zwei benachbarten Untersuchungsgebieten von Zentral-Massachusetts (nordöstl. USA) durchgeführt. Die Daten wurden hauptsächlich an markierten Tieren über Teleme- trie und Sichtungen während einmaliger und wiederholter Kontrollgänge erhoben. Keine der Ab- schätzungen (10-42 Ind./km?) lieferte eine hinreichende Genauigkeit, um Regionen-spezifische Un- terschiede in der Dichte zu ermitteln. Gründe dafür könnten in der geringen Anzahl markierter Individuen liegen (5-12 Ind./Kontr.; 4-6 Ind./km?), in der geringen Sichtung von Baumstachlern wäh- rend der Kontrollgänge (x = 15% der markierten Individuen; Schwankung = 0-40%) und in der gerin- gen Anzahl von Kontrollgängen. Baumstachler leben offenbar stärker zurückgezogen als angenom- men. Intensivere Kontrollen sind notwendig, um präzise Daten über Dichteschätzungen in Wald- habitaten zu erhalten. 90 Mortıy ©. HaALe£ and T. K. FULLER References BRANDER, R. B. (1973): Life history notes on the porcupine in hardwood-hemlock forest in upper Michi- gan. Mich. Acad. 5, 425-433. BRANDER, R. B; Boos, D. J. (1973): The return of the fisher. Nat. Hist. 82, 52-57. Curtis, J. D. (1944): Appraisal of porcupine damage. J. Wildl. Manage. 8, 88-91. Done, W. E. (1959): An effective poison and repellent for porcupine control. J. Forestry 57, 350-352. Done, W. E. (1967): The biology and life history of the porcupine (Erethizon dorsatum) in western Massachusetts. Ph.D. dissertation, Univ. Massachusetts, Amherst. GOLLEY, F. B. (1957): Distribution of the porcupine in upper Michigan. J. Mammalogy 38, 526-527. GRIESEMER, S. J; FULLER, T. K; DE GRAAE,R.M. (1995): Effects of excluding porcupines from estab- lished winter feeding trees ın central Massachusetts. Northeast Wildl. 51, 29-33. GRIESEMER, S. J; FULLER, T. K; DEGRAAF, R. M. (1996): Denning patterns of porcupines, Erethizon dor- satum. Can. Field-Nat. 110, 634-637. GRIESEMER, S. J.; FULLER, T. K.; DEGRAAF, R. M. (1998): Porcupine habitat use in western Massachu- setts. Am. Midl. Nat. (in press). HALE, M. B.; GRIESEMER, S. J.; FULLER, T. K. (1994): Immobilization of porcupines with tiletamine hy- drochloride and zolazepam hydrochloride. J. Wildl. Dis. 30, 429-431. HALE, M. B.; FULLER, T. K. (1996): Population demography of porcupines in western Massachusetts. Can. J. Zool. 74, 480-484. KEITH, L. B.; CAry, J. R. (1991): Mustelid, squirrel and porcupine population trends during a snowshoe hare cycle. J. Mammalogy 72, 373-378. KREFTING, L. W.; STOEKLER, J. H.; BRADLE, B. J.; FITZWATER, W. D. (1962): Porcupine-timber relation- ships in the lake states. J. Forestry 60, 325-330. MILLER, S. D.; WHITE, G. C.; SELLEERS, R. A.; REYNOLDS, H. V.; SCHOEN, J. W.; Titus, K.; BARNES, V. G. Jr.; SMITH, R. B.; NELSON, R. R.; BALLARD, W. B.; SCHWARTZ, C. C. (1997): Brown and black bear density estimation in Alaska using radiotelemetry and replicated mark-resight techniques. Widl. Monogr. 133, 1-55. Orıs, D. L.; BURNHAM, K. P.; WHITE, G. C.; ANDERSON, D. R. (1978): Statistical inference from capture data on closed animals populations. Wildl. Monogr. 62, 1-135. PoLLock, K. H.; NicHots, J. D.; BROWNIE, C.; HıneEs, J. E. (1990): Statistical inference for capture-recap- ture experiments. Wildl. Monogr. 107, 1-97. PowELL, R. A.; BRANDER, R. B. (1977): Adaptations of fishers and porcupines to their predator prey system. In: Proceedings of the 1975 predator symposium. Ed. by R.L. PhırLıps and C. JONKEL. Montana Forest and Conservation Experiment Station, Univ. Montana, Missoula, Pp. 45-53. REEKS, W. F. (1942): Notes on the Canada porcupine in the Maritime Provinces. Forest Chronicles 18, 182-187. RıcE,W. R., HARDER, J. D. (1977). Application of multiple aerial sampling to a mark-recapture census of white-tailed deer. J. Wildl. Manage. 41, 197-206. Rosson, D. S.; REGIER, H. A. (1964): Sample size in Petersen mark-recapture experiments. Trans. Am. Fish. Soc. 93, 215-226. RoZze, U. (1984): Winter foraging by individual porcupines. Can. J. Zool. 62, 2425-2428. SHAPIRO, J. (1949): Ecological and life history notes on the porcupine in the Adirondacks. J. Mammal- ogy 30, 247-257. SMITH, G. W. (1977): Population characteristics of the porcupine in north-eastern Oregon. J. Mammal- ogy 58, 674-676. TAYLoR, W. P. (1935): Ecology and life history of the porcupine (Erethizon epixanthum) as related to the forests of Arizona and the southwestern United States. Univ. Arizona Bull. 6, 1-177. WESTVELD, M.; AsHMAN,R.1I.; BALDWwIN,H. I.; HOLDSWORTH, R.P; JOHNSON, R.S.; LAMBERT, J. H.; Lutz, H. J.; Swaın, L.; STANDISH, M. (1956): Natural forest vegetation zones of New England. J. Forestry 54, 332-338. WHITE, G. C. (1996): NOREMARK: population estimation from mark-resighting surveys. Wildl. Soc. Bull. 24, 50-52. Authors’ addresses: MoLıy ©. HALE, Department of Forestry and Wildlife Management, University of Massachusetts, Amherst, MA 01003-4210 USA, and Topp K. FULLER, Department of Forestry and Wildlife Management and Graduate Program in Organismic and Evolutionary Biology, University of Massachusetts, Amherst, MA 01003-4210 USA Z. Säugetierkunde 64 (1999) 91-106 ZEITSCHRIFT * = vo © 1999 Urban & Fischer Verlag SÄUGETIE RKÜN DE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Chromosomal characterization and relationship between two new species of Ctenomys (Rodentia, Ctenomyidae) from northern Cordoba province, Argentina By MABEL D. GIMENEZ, C. J. BIDAU, CARINA F. ARGÜELLES, and J. R. CONTRERAS Laboratorio de Genetica Evolutiva, Facultad de Ciencias Exactas, Ouimicas y Naturales, Universidad Nacional de Misiones, Posadas, Misiones, Argentina Receipt of Ms. 06. 03. 1998 Acceptance of Ms. 14. 10. 1998 Abstract Karyotypes of two recently described species of Ctenomys from northern Cördoba province (Argenti- na) were studied. C. osvaldoreigi is only known from the type locality in the high valleys of the Sierras Grandes at more than 2000 m above sea level. The karyotype consists of 2n =52 chromosomes with FN = 56 and includes 22 pairs of telocentric autosomes that decrease gradually in size, a pair of subtelo- centric autosomes (n °8), two pairs of small metacentrics and a pair of sex chromosomes. Three popula- tions from the northeastern plains of Cördoba province (including one from the type locality) of C. rosendopascuali were analyzed. All individuals were 2n = 52 but FNs of the three populations were different. Individuals from Los Mistoles showed FN = 62 and the karyotype consists of a large subtelo- centric autosomal pair, a medium-sized subtelocentric (n 8), twenty telocentric and three small meta- centric pairs plus a pair of sex chromosomes. Candelaria specimens had FN = 64; the karyotype in- cludes a second large subtelocentric pair which replaces a large telocentric, the remainder of the complement being similar to Los Mistoles. A further large subtelocentric occurs in the Mar Chiquita population, thus FN = 66; the remainder of the karyotype does not differ from the two other popula- tions. In order to compare the new species to a known species of the same general geographical area, four populations of C. bergi from northwestern Cördoba were karyotyped. All specimens had 2n = 48, FN = 90. The three karyotypes found in C. rosendopascuali are remarkably similar and obviously re- lated to that of C. osvaldoreigi through relatively simple chromosomal rearrangements, which confirms their morphological and molecular proximity. Key words: Ctenomys rosendopascuali, C. osvaldoreigi, C. bergi, northern Cördoba, karyo- type Introduction The South American Octodontoidea are a remarkable group of mammals with respect to their extraordinary karyotypic diversity. Diploid chromosome numbers range from 2n =10 in the Bolivian species Ctenomys steinbachi (Ctenomyidae) (ANDERSoN et al. 1987; RuEDAS et al. 1993) to 2n = 102 in Tympanoctomys barrerae (Octodontidae) (Con- TRERAS et al. 1990). Fundamental numbers (FN) also vary enormously (16-202). Most of this chromosomal diversity is due to karyotypic variation within a single genus: Cienomys (BıpAu et al. 1996; CoNTRERAS et al. 1990; GIMENEZ et al. 1997; ORTELLS 1995; ORTELLS et al. 1990; Reıc et al. 1990, 1992). Ctenomys, with more than 60 extant species, is one of the best examples of “explo- sive” speciation accompanied by extensive karyotype repatterning (Bıpau et al. 1996; 0044-3468/99/64/02 -091 $ 12.00/0 92 MABEL D. GIMENEZ etal. Reıc 1984, 1989; Reıc et al. 1990). According to fossil data, the Ctenomys radiation is thought to have occurred 1.5 MY ago (ORTELLS 1990; Reıc et al. 1990). These evidences strongly suggest that the main mode of speciation has been (and is) chromosomal. The subterranean mode of life plus the populational characteristics of most of the species (small deme size, low vagility) support the chromosomal speciation hypothesis (BIDAU et al. 1996; Kıng 1993). In this study we investigate Karyotypes of Ctenomys from northern Cördoba province (Argentina). The analyzed populations belong to two new biological species, C. rosendo- pascuali and C. osvaldoreigi (CONTRERAS 1995 a, b). Our results are compared to previous ones and discussed within the frame of a model for the evolution of the genus which in- corporates molecular, morphological, and paleobiogeographical data. Material and methods This work is based on the individuals of Ctenomys indicated in table 1 and figure 1. All specimens were deposited in the collection of the PROBBAS (CONICET, Corrientes, Argentina), with the following catalogue numbers (sex in parentheses): C. rosendopascuali. Mar Chiquita: C-03363 (F), C-03364 (M). Candelaria: C-03464 (M), C-03465 (F). Los Mistoles: C-03509 (M), C-03510 (F). C. osvaldoreigi. Estan- cia San Luis (Sierras Grandes): C-03462 (F), C-03463 (F), C-03977 (F), C-03978 (F), C-03979 (F), C-03980 (F). C. bergi. Cruz del Eje: C-03460 (M), C-03461 (M). Las Toscas: C-03506 (M). Salinas Grandes: C-03507 (F). Guanaco Muerto: C-03508 (F). Mitotic metaphases were obtained following two protocols: direct bone-marrow preparations ac- cording to a modified version of FORD and HAMERTOoN’s (1956) technique, and short-term bone-marrow in vitro culture (GIMENEZ and BıpAu 1994). In the first case, bone marrow was incubated in 0.1 ml 0.05 % colchicine plus 9.9 ml 0.075 MKCl for 55 min at 37°C and subsequently fixed in 3:1 metha- nol:glacial acetic acid. For short-term culture, the tissue was incubated in RPMI 1640 medium supple- mented with 15 % foetal calf serum for 20.5 h at 37 °C. A drop of 0.005 % colchicine was then added to the culture, and 15 min later the cells were hypotonized in 0.075 M KCl and fixed in 3:1. Nondifferen- tial chromosome staining was performed in phosphate buffered Giemsa (pH = 6.8). G- and C-banding followed the protocols of SEABRIGHT (1971) and SumneERr (1972), respectively. NORs were stained ac- cording to HowELL and BLAck (1980). Meiotic preparations for the observation of sperm morphology were made by the technique of Evans et al. (1964). Table 1. Localities, number and sex of the Ctenomys rosendopascuali, C. osvaldoreigi and C. bergi indi- viduals studied. Locality N’ of specimens Male Female C. rosendopascuali Contreras, 1995 Mar Chiquita! (30°55’ S-62°41’W) Candelaria (29°49' S-63°21’ W) Los Mistoles (30°38° S-63°54' W) C. osvaldoreigi Contreras, 1995 Estancia San Luis! (31°24’° S-64°48' W) (Sierras Grandes) C. bergi Thomas, 1902 Cruz del Eje! (30°44’ S-64°48’ W) Guanaco Muerto (30°27 S-65°01’ W) Salinas Grandes (30°03’ S-65°05’ W) Las Toscas (30°08 S-64°53’ W) ! Type localities Karyotypes of new species of Ctenomys Cördoba 66° 69% 64° 63° 62° . Catamarca - & Santiago del Estero . N - province az nl ® „‚province | A . 1 am DEP Sl . 1 -omrsapan. } e } ® \ SALINAS GRANDES | ,* ei CANDELARIA \ : 2 . Ber \ P Be 204 A\ 2N=52, FN=64 Sn 30° La Rioja ; LAS TOSCAS \ sin province/ en anta Fe # 2n=48, |GUANACO MUERTO 2n=52, FN=62 « FN=90 A 2 ) g CRUZ DEL EJE [ Or [0] I Est. SANLU MAR CHIQUITA ; 31 ı 2n=52, FN= 2n=52, FN=66 + 8 © r ı Ei ’ ' U 8 1} A knn=-" San Euis province | 8 8 U 8 8 8 8 U a [ [) “ [) ” [) . » [) 2 „’ r 3 jr 8 i Cördoba|province % 33 £ | I 2 .. BEN. : | er | 2 „34° 1 i [I U [1 [ ereuausan Buenos Aires province Fig. 1. Geographic distribution of the populations of Ctenomys rosendopascuali, C. osvaldoreigi, and C. bergi analyzed. 94 MABEL D. GIMENEZ etal. 1} 117 LITTETEIELEIETET 2 3 4 5 6 7 8 10 r nr os 0“. an 00 His .. na 12 MR ss Ib 7 i8..00 90 .. as mn ih as hs gi 1 2 ma 2.2 5 XY xX a saas ah Pa .s re am IWIEWIEIWIE IE IE, ar ı 2 13 4 5 6 ı 83 SE DR an 20 on "En >: 21 22 23 24 25 XY xXX en er Ka] c Fig. 2. Karyotype of C. rosendopascuali from Mar Chiquita; a. Giemsa stained, b. G-banding, c. C- banding. Bar = 10 um. Karyotypes of new species of Cienomys 95 KA NARn 00 anno an oa nn an 1a 2 3 4 3 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Se aan HR ne. Ku it 21 22 23 24 25 XY XX ’st 6 7 8 9 10 G Bo un an nu De m on an ie 2 an >» m © 6 i2 08 19.20 ” u { en BB anna. 28% An DI 23 au > XY XX b L a de RW N Wa mn a N N RB Nee ee Kr Ihe Irre Ihe lie hear Ihr Ihe Ihe Ike lhr (Hr ww DH HD +2000 In MW Rn ron PO OO [Io WM DD WM WM Denon ın ir ko ae ip Ihe Ike Iren Far Ike (Ar lie D D N DHNDOoOoWHr RUND WM WW W DD Ar Ike Ar m (aelkır nn Kinematic analysis of treadmill locomotion of Tupaia glis 149 Discussion On a treadmill, animals perform only a portion of their locomotion repertoire. Treadmill locomotion or restrained locomotion can be different from that on a normal ground (un- restrained locomotion), as described, e.g., in horses (BARREY et al. 1993) and humans (ELLior and BLAnksBY 1976). Our present observations of unrestrained locomotion dur- ing force plate recordings on Tupaia glis at gallop show that the limbs are more extended than on the treadmill, the animal jumps higher and gains longer distances during the swing phase (0.3 m-0.4 m). However, only locomotion on the treadmill allows to analyse kinematics properties using cineradiography, and especially to record series of steps. JENKINS (1974 a) primarily investigated movements of the distal elements in his study on tree-shrew locomotion. His results were based on a small number of observations. He reported only one value for each of the analysed angles during exploratory activity of the anımal. During bounding runs, he analysed only the hindlimbs; no data were given for the forelimbs. Only a detailed analysis of many steps, however, allows to recognise and assess the degree of variability or stereotype of parameters in locomotion. A major defi- ciency in JENKINS’ work (1971, 1974 a) is his neglect of the scapula, which contributes to body propulsion of up to more than 40 % in T. glis and more than 60 % in other mam- mals. According to JENKINS (1974 a), rhythmic flexions and extensions restricted to the in- tervertebral articulations between Thll and L1 occur during exploratory activity and bounding run: “... the lumbar series remains rigid and [does] not contribute to even the most extreme flexion observed”. His observations are in sharp contrast to ours on Tupaia but also on other small mammals. We found movements in the caudal thoracic spine, but the highest intervertebral amplitudes occur in the lumbar region. The sagittal lumbar spine movements in Procavia capensis (FISCHER 1994) and Ochotona rufescens (FISCHER and LEHMANN 1998) contribute extensively to body propulsion during in-phase gaits. ‘Pelvic movement’ during symmetrical gaits in 7. glis is low compared to other small mammals (P. capensis < 20°, FiıscHEr 1994; Monodelphis domestica 9°, unpubl. obs.). Most other studies consider only angles of larger segments of the vertebral column to the hori- zontal line (e.g., JEnKıns 1974 a; HECKNER 1982; Hurov 1987). FiscHEr (1994) calculated intervertebral joint angles between reconstructed foot down and lift off positions on freshly dead or anaesthetised animals. The present study is the first cineradiographic ana- lysis that measured sagittal spinal movements in intervertebral joints in animals. Tupaia is comparable to other small mammals in its limb geometry (FIscHEr and LEH- MANN 1998). Especially at gallop we found almost right angles in shoulder, elbow, hip, and knee joint. Limb segments that are in horizontal orientation at foot down or lift off may contribute with their whole length to step length. Such a positioning is found in: Di- delphis virginia (JENKINS 1971), Rattus norvegicus (JENKINS 1974 b), Procavia capensis (FiscHER 1998), Ochotona rufescens (FISCHER and LEHMANN 1998), Eulemur fulvus (SCHMIDT and FiscHEr 1999) for humerus and tibia at foot up and for femur at foot down. In T. glis, the ulna is also nearly parallel to the ground at foot down during symmetrical gaits. In Tupaia and those other animals so far analysed with the ‘overlay method’ (O. rufes- cens, FISCHER and LEHMANN 1998; Eulemur fulvus, SCHMIDT and FiscHER 1999) scapular movements contribute substantially to body propulsion (T. glis: 42 %-43 %, O. rufescens: 67 %, E. fulvus: 63 %). Contribution of the humerus ante- and retroversion is comparable between these animals (O. rufescens: 21%, E. fulvus: 31 %). In T. glis, the effects of hu- merus movements are different during symmetrical gaits (17 %) from gallop (45 %). The ulna contributes 32 % to body propulsion because of the large extension in the elbow joint at the end of stance during symmetrical gaits, but only 3% at gallop. The contribu- tion of the hand is similar during all gaits (9%) and larger in Tupaia than in other ani- 150 NADJA SCHILLING and M. S. FISCHER mals (E. fulvus: 1%, O. rufescens: 3%). In the hindlimb, propulsive effects of the seg- ments change with gaits, caused by the additive sagittal spine movement at gallop. Spinal contribution is low during symmetrical gaits (2%) and high (42%) at gallop. The femur contributes 49% and the foot 21% to propulsion at gallop. The animal looses some stance length through the movements of the tibia. In O. rufescens, sagittal spine move- ments contribute the major component to propulsion (55 %-65 %) during in-phase gaits, followed by the tibia (19 %-35 % ), femur (8 %-10 %) and foot (4 %-7 %). HEGLUunD and TAyLor (1988) postulated size-related modes of acceleration. Small mammals should increase step frequency and larger animals step length, but step fre- quency should remain nearly constant even with increasing speed during in-phase gaits. Tupaia glis belongs to intermediate forms as well as Procavia capensis (FISCHER 1998) and Eulemur fulvus (SCHMIDT and FiscHEr 1999), which increase both parameters. An in- crease in step duration and length is reported for Rattus norvegicus (CoHEN and GAnNS 1975). Increase of step frequency is attained by abbreviation of the stance phase (hind- limbs of T: glis; E. fulvus, SCHMIDT and FIscHeEr 1999; Loris tardigradus, Nycticebus cou- cang, DEMES et. al. 1990; Macaca mulatta, Felis catus, humans, VILENSKY and GEHLSEN 1984), of the swing duration (Dasyuroides byrnei at in-phase gaits, KÜHNAPFEL 1996) or both parameters (forelimbs of T: glis; R. norvegicus, CoHEN and Gans 1975; P. capensis, FiscHER 1998). JENKINS (1974 a) described a shortening of the stance duration without dif- ferences in the footfall pattern during symmetrical gaits (v = 1.5m/s-1.75 m/s). Swing duration seems to be increased (JENKINS 1974 a), but no details were given. In the pre- sent study, a decrease of the stance duration up to 75% was found during symmetrical gaits; at gallop step frequency is almost constant. All mammals analysed so far have speed independent swing durations (CoHEN and Gans 1975; ELLIOT and BLANKSBY 1976; VILENSKY and GEHLSEN 1984; Hoy and ZERNICKE 1985; DEMES et al. 1990; van WEEREN et al. 1993; FiscCHER 1994; KÜHNAPFEL 1996; SCHMIDT and FISCHER 1999). The measured values for Tupaia are in the range of animals of com- parable size (70 ms-140 ms). A remarkable result is the almost constant horizontal distance between the scapular fulcrum and the finger tips at foot down (symmetrical gaits: 68+4mm, gallop: 62+7 mm). In contrast to this, JENKINnS (1974 a) figured foot down beneath the shoulder joint. Such a positioning of the foot seems to be a speciality of exploratory activity, but is never observed at faster gaits. Also in Monodelphis domestica and Dasyuroides byrnei, the distance between the fulcrum of the scapula and finger tips scatters only slightly (M. domestica: 35 +3 mm, N = 34; D. byrnei: 32 +5, N = 19; own observ.). Foot down is below the eye point in T. glis, M. domestica, D. byrnei as well as in Galea musteloides, Rattus norvegicus (unpubl. observ.) and Ochotona rufescens (FISCHER and LEHMANN 1998). In contrast, the point of foot down lies in front of the eyes in Eulemur fulvus (SCHMIDT and FiscHER 1999) which has elongated limbs. The onsets of the various flexion and extension movements are not synchronous with foot down and lift off. The movements of nearly all joints on the fore- and hindlimb start before foot down or lift off in Tupaia glis. A beginning of scapula retroversion well be- fore foot down was also described for Procavia capensis (FISCHER 1994), Felis catus (Boc- ZEK-FUNcKE 1996), Ochotona rufescens (FiscHER and LEHMANN 1998), Eulemur fulvus (Scumipr and FiscHEr 1999), Dasyuroides byrnei, Monodelphis domestica (unpubl. ob- serv.). In contrast, anteversion of the scapula begins at different times e.g., in the last quarter of stance (T. glis, E. fulvus, F catus, M. domestica) or at lift off (P. capensis, Cer- copithecus aethiops (WHITEHEAD and LArson 1994). Retroversion before foot down could reduce deceleration forces (FiscHER 1994). Measurements of ground reaction forces prove that deceleration forces in T. glis are relatively small (unpubl. observ.). For the first time, a gait-dependent kinematic behaviour of a limb joint is observed. During symmetrical gaits, the shoulder joint movements in Tupaia comprise two flexions Kinematic analysis of treadmill locomotion of Tupaia glis 151 and two extensions per step. Only one flexion and one extension occur at gallop. With change of gaits, shoulder joint movements are reduced from a biphasic to a monophasic pattern. The works of WHITEHEAD and Larson (1990) and SCHMIDT and FiIscHEr (1999) are the only cineradiographic studies on primate locomotion (Cercopithecus aethiops and Eule- mur fulvus, respectively) available at present. Shoulder joint amplitudes are significantly larger in the terrestrial C. aethiops than in small non-primates. A remarkable scapular movement was described for E. fulvus, including a distinct mediad rotation. Both species as well as other primates have elongated limbs. Tupaia glis does not share these features with primates. The locomotion of T. glis is similar to that of other small mammals in its kinematic and metric parameters. It appears that these parameters are common in mam- mals of a small to medium size class, independent of their taxonomic group. In all analysed small to medium-sized mammals body propulsion is mainly achieved by actions of the proximal limb segments. The flexed limb posture enables the animal to react to obstacles and reduce vertical displacement of the centre of gravity. The horizon- tal orientation of the upper arm and lower leg at lift off, and the upper leg at foot down seem to be standard parameters of the locomotion of small mammals. Additive sagittal spine movements contribute substantially to body propulsion during in-phase gaits. All these features probably occurred in the most recent common ancestor of therian mam- mals. Acknowledgements We are indebted to Prof. Dr. H.-R. Duncker (Gießen) and Prof. Dr. H.-J. Kunn (Göttingen) for kindly providing us with the animals. We thank Dr. D. HAARHAUS and his staff (IWF Göttingen) for their patience and competence in cineradiography. We also thank all members of our working group in Jena for many discussions, and especially Dr. A. Haas (Jena) and PD Dipl. Ing. Dr. Med. H. WiTTE (Jena) for thoroughly revising the manuscript. M. Roser (Jena) skilfully helped with the illustrations. This research was supported by the Deutsche Forschungsgemeinschaft (DFG) (Fi 410/ 1-3; Innova- tionskolleg “Bewegungssysteme”, Teilprojekt B 1). Zusammenfassung Kinematische Analyse der Fortbewegung von Tupaia glis (Scandentia: Tupaiidae) auf dem Laufband Mit Hilfe der Röntgenkinematographie wurde die Fortbewegung von Tupaia glis auf dem Laufband in verschiedenen Gangarten (Schritt, Trab und Galopp) untersucht. 7. glis erhöht seine Geschwindig- keit in den symmetrischen Gangarten (Schritt, Trab) durch eine Steigerung der Schrittfrequenz, im Galopp wird durch eine Flugphase Schrittlänge gewonnen. Die Vorderextremität fußt in allen Gang- arten unter dem Auge auf, der Abfußpunkt liegt in den symmetrischen Gangarten hinter und im Ga- lopp meist vor dem Lot des Scapula-Drehpunktes. Humerus und Tibia werden beim Abfußen in allen Gangarten horizontal positioniert. Beim Auffußen wird das Femur parallel zum Untergrund gestellt. Die proximalen Extremitätenabschnitte sind maßgeblich am Rumpfvortrieb beteiligt (Scapula in allen Gangarten: 42-43 %, Femur in den symmetrischen Gangarten: 82 % und im Galopp: 49 %). Im Ga- lopp trägt die additive Sagittalbewegung 42 % zum Vortrieb des Körpers bei. Ellbogen- und Kniege- lenk werden in den symmetrischen Gangarten am Ende der Stemmphase deutlich weiter geöffnet (30-40°) als im Galopp. Erstmals konnte ein von der Gangart abhängiger Ablauf der Gelenkbewe- gungen beobachtet werden. Der biphasische Bewegungsablauf des Schultergelenkes mit zwei Beugun- gen und Streckungen pro Schrittzyklus in den symmetrischen Gangarten wird auf einen monopha- sischen Ablauf im Galopp reduziert. Die Dorsal- und Ventralflexionen der Wirbelsäule wurden zum ersten Mal auf der Basis der Röntgenkinematographie untersucht und dabei nachgewiesen, daß es sich im Unterschied zu der bei anderen Säugetieren beschriebenen additiven Lumbalbewegung bei 152 NADJA ScHILLING and M. S. FISCHER T. glis um eine additive Thorako-Lumbalbewegung handelt. Die untersuchten kinematischen und metrischen Parameter von T. glis stimmen in wesentlichen Punkten mit denen anderer kleiner und mittelgroßer Säugetiere überein. Die Kinematik ist abhängig von der Körpergröße und unabhängig von der systematischen Stellung der Tiere. Literature BARREY, E.; GALLOUX, P.; VALETTE, J. P.; AUVINET, B.; WOLTER, R. (1993): Stride characteristics of over- ground versus treadmill locomotion. Acta Anat. 146, 90-94. BOCZEK-FUNCKE, A.; KUHTZ-BUSCHBECK, J. P.; ILLERT, M. (1996): Kinematic analysis of the cat shoulder girdle during treadmill locomotion: an X-ray study. Eur. J. Neurosci. 8, 261-272. COHEN, A.H.; Gans, C. (1975): Muscle activity in rat locomotion: movement analysis and electromyo- graphy of the flexors and extensors of the elbow. J. Morph. 146, 177-196. DEMES, B.; JUNGERS, W. L.; NIESCHALK, U. (1990): Size- and speed-related aspects of quadrupedal walk- ing in slender and slow lorises. In: Gravity, posture, and locomotion in primates. Ed. by F.K. Jour- FROY, M. H. STAcK and D. NIEMITZ. Firenze: Sedicesimo. Pp. 175-197. ELLIOTT, B. C.; BLANKSBY, B. A. (1976): A cinematographic analysis of overground and treadmill run- ning by males and females. Med. Sci. Sports. 8, 84-87. ENGLISH, A. W. (1978): Functional analysis of the shoulder girdle of cats during locomotion. J. Morph. 156, 279-292. FISCHER, M. S. (1998): Die Lokomotion von Procavia capensis (Mammalia: Hyracoidea). Ein Beitrag zur Evolution des Bewegungssystems der Säugetiere. Verhdl. naturwiss. Verein Hamburg (NF) 32, 1-207. FISCHER, M. S. (1994): Crouched posture and high fulcrum, a principle in the locomotion of small mam- mals: The example of the rock hyrax (Procavia capensis) (Mammalia: Hyracoidea). J. Hum. Evol. 26, 501-524. FISCHER, M. S.; LEHMANN, R. (1998): Application of cineradiography for the metric and kinematic study of in-phase gaits during locomotion of the pika (Ochotona rufescens, Mammalia: Lagomorpha). Zo- ology 101, 148-173. HECKNER, U. (1982): Vergleichende Untersuchungen des Bewegungsverhalten einiger ursprünglicher Ruminantia. Zool. Anz. 209, 283-293. HEGLUND, N. C.; TAYLoR, C. R. (1988): Speed, stride frequency and energy cost per stride: how do they change with body size and gait? J. exp. Biol. 138, 301-318. HILDEBRAND, M. (1976): Analysis of tetrapod gaits: general consideration and symmetrical gaits. In: Neural control of locomotion (Advances in behavioural biology). Ed. by R. M. HERMAN, S., GRILL- NER, P. S. G. STEIN, and D. G. STUART. New York, London: Plenum Press. 18, 191-203. HILDEBRAND, M. (1977): Analysis of asymmetrical gaits. J. Mammalogy 58, 131-156. Hoy, M. G.; ZERNICKE, R. F. (1985): Modulation of limb dynamics in the swing phase of locomotion. ]. Biomechanics 18, 49-60. Hurov, J. (1987): Terrestrial locomotion and back anatomy in vervets (Cercopithecus aethiops) and pa- tas monkeys (Erythrocebus patas). Am. J. Primatol. 13, 297-311. JENKINS, F. A. Jr. (1971): Limb posture and locomotion in the Virginia opossum (Didelphis marsupialis and other non-cursorial mammals. J. Zool. (London) 165, 303-315. JENKINS, F. A. Jr. (1974 a): Tree shrew locomotion and the origins of primate arborealism. In: Primate Locombotion. Ed. by F. A. Jr. JEnKINS. New York: Academic Press. Pp. 85-115. JENKINS, F. A. Jr. (1974 b): The movement of the shoulder in claviculate and aclaviculate mammals. J. Morph. 144, 71-83. JENKINS, F. A. Jr.; CAMAZINE, S. M. (1977): Hip structure and locomotion in ambulatory and cursorial carnivores. J. Zool. (London) 181, 351-370. KÜHnAPFEL, M. (1996): Röntgenkinematographische Untersuchungen der Lokomotion von Dasyur- oides byrnei und Monodelphis domestica (Metatheria). Diss. thesis Univ. Tübingen. MILLER, $.; MECHE, F. G. A. VAN DER (1975): Movements of the forelimbs of the cat during stepping on a treadmill. Brain Res. 91, 255-270. NovAcek, M. J. (1992): Mammalian phylogeny: shaking the tree. Nature, 356, 121-125. SCHMIDT, M.; FISCHER, M. S. (1999): Cineradiographic study of forelimb movements during quadrupedal walking in the Brown Lemur (Eulemur fulvus, Primates: Lemuridae). Amer. J. Phys. Anthropol. (in press.). Kinematic analysis of treadmill locomotion of Tupaia glis 153 VAN WEEREN, P. R.; VAN DEN BOGERT, A. ]J.; BAcK, W.; BRUIN, G.; BARNEVELD, A. (1993): Kinematics of standardbred trotter measured at 6, 7, 8, and 9 m/s on an treadmill, before and after 5 months of preface training. Acta Anat. 146, 154-161. VILENSKY, J. A.; GEHLSEN, G. (1984): Temporal gait parameters in humans and quadrupeds: How do they change with speed? J. Hum. Mov. Stud. 10, 175-188. WHITEHEAD, P. F; LARson, S. G. (1994): Shoulder motion during quadrupedal walking in Cercopithecus aethiops: Integration of cineradiographic and electromyographic data. J. Hum. Evol. 26, 525-544. Authors’ address: Dipl. biol. NADJA ScHILLIinG and Prof. Dr. MARTIN S. FISCHER, Institut für Spezielle Zoologie und Evolutionsbiologie mit Phyletischem Museum, Friedrich-Schiller- Universität Jena, Erbertstr. 1, D-07743 Jena, Germany. Z. Säugetierkunde 64 (1999) 154-168 RT, © 1999 Urban & Fischer Verlag SÄUGETI Senne http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Nahrungspräferenzen der Feldmaus Microtus arvalis in der Asgrarlandschaft unter Berücksichtigung der Pflanzeninhaltsstoffe Von LORENZA BALMELLI, W. NENTWIG und J.-P. AIROLDI Zoologisches Institut, Universität Bern, Bern, Schweiz Eingang des Ms. 13. 10. 1998 Annahme des Ms. 25. 01. 1999 Abstract Food preferences of the common vole Microtus arvalis in the agricultural landscape with regard to nutritional components of plants At high population densities the common vole Microtus arvalis may cause severe damage to agricul- tural crops. Knowledge of its food preferences could be used to promote preferred plants in agricul- tural compensation areas such as fallow fields or weedy borders to crop fields. Thus migration into ad- jacent fields may be prevented. Since such migrations are more likely to occur in winter, laboratory feeding choice tests were carried out during this season in order to investigate the food choices of the common vole. Feeding signs in the field served for qualitative comparisons. Plant nutritional compo- nents (nitrogen, sugar, starch, and phenolics) as well as energy and water content were analyzed and related to plant preferences. Clear preferences emerged from the choice trials. The most preferred plants were Hordeum vulgare (leaves), Brassica napus (leaves) and Beta vulgaris altissima (roots) among cultivated plants, and Achillea millefolium (leaves) and Trifolium pratense (leaves) among weed strip plants. There was no relationship between preferences and the analyzed nutritional para- meters. From the 5 most attractive plants in laboratory tests, feeding signs in the field were observed at high frequency only for T. pratense and B. napus. T. pratense may therefore be suitable for prevent- ing migrations of M. arvalis from bordering areas into rape fields. Key words: Microtus arvalis, food preferences, plant nutritional components Einleitung Die Feldmaus (Microtus arvalis Pallas) besiedelt verschiedenartigstes Kulturland sowie offenes, nicht zu feuchtes Grasland bis oberhalb von 2000 m ü.M. Im größten Teil ihres Verbreitungsgebietes, das sich von Nordspanien bis nach Ostasien erstreckt, spielt sie eine bedeutende Rolle als Ackerschädling (STEIN 1958). Die Intensivierung der Landwirtschaft hat zu einer drastischen biologischen Verar- mung der Agrarlandschaft geführt. Ackerkrautstreifen, die als Buntbrache mit einhei- mischen Wildkräutern seit ca. zehn Jahren auch in der Schweiz in oder am Rande von Feldern angesät werden, erhöhen die Biodiversität der Agrarlandschaft. Als positive Aus- wirkung wurde nicht nur eine erhöhte Artendiversität verschiedener Arthropodengrup- pen in den Streifen festgestellt (Lys und NentwiG 1994; FRANK und NENTWIG 1995 a), son- dern auch eine höhere Dichte von Nützlingen in angrenzenden als in weiter entfernten Feldbereichen, was zu niedrigeren Dichten schädlicher Insekten in Streifennähe führt (Lys und NEnTwIG 1992; FRANK und NENTwIG 1995b; HAUSAMMANN 1996). Außerdem 0044-3468/99/64/03 — 154 $ 12.00/0 Nahrungspräferenzen der Feldmaus Microtus arvalis 155 werden die Ackerkrautstreifen von Vögeln zum Nahrungserwerb genutzt (LiLLE 1996). Feldmäuse kommen dort mit jahreszeitlichen Schwankungen vor (BAUMANN 1996): Von Mai bis September herrscht eine relativ hohe Populationsdichte, die ab Oktober ab- nimmt, gegen März ein Minimum erreicht und danach wieder zunimmt. Im Sommer stellte BAUMANN (1996) unmittelbar nach der Getreide- und Hanfernte der angrenzenden Felder eine vorübergehende starke Zunahme der Population von M. arvalis in den Strei- fen fest, was er auf Wanderungen aus den Feldern zurückführte. Dieses wirft die Frage auf, ob sich die Feldmäuse aus den Streifen auf die Suche nach Nahrung in die Felder be- geben und insbesondere, ob sie sich hauptsächlich von Kultur- oder Buntbrachepflanzen ernähren. Im Winter sind Wanderungen aus den Streifen in die angrenzenden Felder am wahrscheinlichsten, weil die Streifen in dieser Jahreszeit verminderte Ernährungsmöglich- keiten bieten. Wintergetreide- und Rapsfelder dürften hingegen eine anziehende Wir- kung ausüben. Die Nahrungsaufnahme von M. arvalis in Agrarökosystemen unter Berücksichtigung der Wildkräuter ist bisher wenig erforscht. TRUSZKOwskI (1982) stellte in Getreide- und Rapsfeldern eine Vorliebe für Unkräuter fest; von den 29 vorhandenen Unkrautarten wurden fast alle gefressen, und zwar ohne ausgeprägte Präferenzen. Auf Dauergrünland geht die Feldmaus bei der Nahrungsaufnahme selektiv vor und zeigt deutliche Präferen- zen für bestimmte Pflanzenarten (Yu et al. 1980; LEUTERT 1983; RınkE 1990). In keiner der oben angeführten Studien zur Nahrungsökologie von M. arvalis wurden die Ursachen der dargelegten Präferenzen untersucht. Da die Nahrungspräferenzen der Wühlmäuse von einer Kombination „positiver“ (Nährstoffe, Energie und Wasser) und „negativer“ (Sekundärmetabolite und Fasern) Nahrungseigenschaften abhängig sind (BArzıı 1985), müssten bei solchen Untersuchungen Parameter beider Eigenschaften analysiert werden. Ziel der vorliegenden Arbeit war die Untersuchung der Nahrungswahl von M. arvalis in der Agrarlandschaft im Zeitraum Spätherbst-Winter mittels Futterwahlexperimenten im Labor. Als qualitativer Vergleich diente die Untersuchung der Fraßspuren im Feld. Zur Ursachenuntersuchung eventueller Nahrungspräferenzen wurden einige positive und ne- gative Parameter der getesteten Pflanzen analysiert. Die Fragestellungen lauteten wie folgt: (1) Geht die Feldmaus bei der Nahrungsauf- nahme selektiv vor? (2) Welche Buntbrache- bzw. Kulturpflanzen werden bevorzugt? (3) Werden bestimmte Pflanzenorgane vorgezogen? (4) Besteht ein Zusammenhang zwischen den Präferenzen und den analysierten Parametern? Material und Methode Mäusefang und Haltungsbedingungen Die Feldmäuse wurden von Oktober 1997 bis Januar 1998 in zwei Ackerkrautstreifen auf Ackerland in Belp (bei Bern) mit Longworth-Lebendfallen gefangen. Die Haltung erfolgte bei einer Photope- riode von 12h Licht/12h Dunkel unter folgenden Temperatur- bzw. Feuchtigkeitsbedingungen: 17- 20°C bzw. 45-60% bis zum 15. 12. 1997 und 10-16°C bzw. 48-63% danach. Die Feldmäuse wurden einzeln in Makrolonkäfigen (40x25x15 cm) mit Späne und einem Blumentopf als Unterschlupf ge- halten. Mäusezuchtfutter und Wasser standen ad libitum zur Verfügung und dreimal pro Woche wur- den zusätzlich Möhren verfüttert. Futterwahlexperimente Versuchsanordnung und -ablauf Es wurden Futterwahlexperimente als sogenannte „Cafeteria-Iests“ durchgeführt. Dabei wird den Tieren die Wahl zwischen verschiedenen Futtertypen gegeben. Durch die konsumierte Menge jedes Futtertyps können Rückschlüsse auf die Nahrungspräferenzen gemacht werden. 156 LORENZA BALMELLIU.a. Vor Versuchsanfang wurde den Tieren eine Laborangewöhnungszeit von mindestens 20 Tagen ge- lassen. Es wurden 2 Serien aus je 6 „Cafeteria-Iests“ durchgeführt. Die zwei Serien fanden vom 18. 11. bis zum 12. 12. 1997 bzw. vom 24. 1. bis zum 25.2. 1998 statt. Die einzelnen Tests dauerten um die 40h und wurden mit einem Abstand von zwei bis zehn Tagen durchgeführt. Serie 1 und 2 un- terschieden sich nur in der Auswahl der Pflanzen und nicht in der Versuchsanordnung. Es wurden ei- nige Kulturpflanzen und im Winter häufige Buntbrachepflanzen ausgewählt. In jedem Test wurden den Feldmäusen 2 verschiedene Blatt- und 2 verschiedene Wurzeltypen zur Wahl vorgelegt. In beiden Serien wurden insgesamt je 8 Pflanzenarten bzw. -organe (im Folgenden als 8 Pflanzen bezeichnet) getestet (Tab. 1a, 1b). Trifolium pratense wird in der vorliegenden Arbeit ausschließlich als Buntbra- chepflanze betrachtet, obwohl sie auch kultiviert wird. Für die Tests wurde frisches Pflanzenmaterial verwendet, das am Tag des Tests aus dem Feld geholt wurde. Einzig die Zuckerrüben wurden bereits im Oktober gesammelt und bis zur Verwendung kühl gelagert. Mit Ausnahme von Test le, für den nur 5 Tiere zur Verfügung standen, wurden alle Tests mit 6 Einzeltieren durchgeführt. Für jeden Test wurden 6 (für 1e nur 5) Makrolonkäfige (55x 35x20 cm) neu vorbereitet: Der Boden wurde mit Haushaltpapier bedeckt, vier 14x8x4 cm große Plastikfutter- näpfe (je einer pro Futtersorte) wurden nebeneinander gestellt und dazwischen Metalltrennwände (15x 11cm) befestigt. Das gesammelte Pflanzenmaterial wurde gewaschen. Vier fast gleiche Portionen (zwischen 20 und 40 g je nach Test) wurden abgewogen (Frischgewicht Anfang, FGa) und in die vier Futternäpfe jedes Käfigs gelegt. Die Verteilung der vier Pflanzen auf die vier Futternäpfe jedes Käfigs wurde so ge- wählt, daß am Ende jeder Serie jede Pflanze möglichst gleich oft in jeder der 4 Positionen angeboten worden war. Von jeder der vier Pflanzen wurde zusätzlich eine Probe bei 50°C getrocknet und an- Tabelle 1a. Versuchsanordnung der „Cafeteria-Serie“ 1. Für jeden der 6 „Cafeteria-Tests“ (1a-1f) sind die 4 getesteten Pflanzen angegeben (X); b): Versuchsanordnung der „Cafeteria-Serie“ 2. Für je- den der 6 „Cafeteria-Tests“ (2a-2f) sind die 4 getesteten Pflanzen angegeben (X). a No. Pflanzenarten und -organe Cafeteria TEE Test Oenothera Oenothera Silene Silene Trifolium Beta Brassica Dipsacus biennis biennis alba alba pratense vulgaris napus fullonum Rosettenbl. Wurzel Rosettenbl. Wurzel Blätter altissima Blätter Wurzel Rübe Pflanzenarten und -organe Cafeteria Test Brassica Brassica Hordeum Verbascum Leucan- Pastinaca Achillea Echium napus napus vulgare densiflorum theum sativa millefolium vulgare Blätter Wurzel Blätter Wurzel vulgare Wurzel Blätter Wurzel Blätter Nahrungspräferenzen der Feldmaus Microtus arvalis 157 schließend gewogen, um mit dem Wassergehalt das angebotene Trockengewicht (Trockengewicht An- fang, TGa) zu berechnen. Das nicht verwertete Material jeder Pflanze wurde für die chemischen Ana- lysen getrocknet. Die Feldmäuse wurden gewogen (Gewicht Anfang, Ga) und in die Versuchskäfige gesetzt. Nach Ablauf der Versuchsdauer wurden die Tiere aus den Versuchskäfigen genommen und gewogen (Gewicht Ende, Ge). Die Reste der Pflanzen wurden sortiert, gewogen (,„Frischgewicht“ Ende, FGe: nur in Serie 2), getrocknet und erneut gewogen (Trockengewicht Ende, TGe). Tests, in de- nen keine Reste einer oder mehrerer Pflanzen übrig geblieben waren, wurden mit einem größeren FGa wiederholt. Wegen der beschränkten Haltungsmöglichkeiten wurden 25 der insgesamt 46 ver- wendeten Feldmäuse zweimal eingesetzt. In solchen Fällen wurde darauf geachtet, daß mindestens 14 Tage Abstand zwischen den beiden Tests eingeschaltet waren, und daß die zwei Tests keine ge- meinsame Pflanze hatten (z.B. 1a und Idin Tab. 1a). Präferenzmaß Als Maß für die Präferenz wurde der „Consumption Index“ (CI) nach WALDBAUER (1968) folgender- x 1000 K ßen berechnet: CI = maßen berechne GxD wobei K = Trocken- bzw. Frischgewichtskonsum (g), G = (Ga + Ge)/2 = Durchschnittliches Tierge- wicht während des „Cafeteria-Tests“ (g) und D = Dauer des Tests (h). Der CI wurde sowohl nach dem Trockengewichtskonsum (CI TG) als auch nach dem Frischgewichtskonsum (CI FG) wie folgt berechnet: Ga — TGe TGa-TGe FGa AR Da in Serie 2 die Pflanzenreste auch vor dem Trocknen gewogen wurden (FGe), ist deren Wasserge- halt bekannt. Um der Tatsache Rechnung zu tragen, daß die Pflanzenteile (v.a. Blätter) während des Versuchs Wasser verlieren, wurde für Serie 2 der CI zusätzlich nach dem durchschnittlichen Frischge- wicht (CI FGd) folgendermaßen errechnet: TGa — TGe 1 z GxD & durchschnittlicher Anteil Trockenmaße u Durchschnittlicher Anteil Trockenmaße = Mittelwert aus iber, und ut FGa FGe Als Stichprobengrößen ergeben sich 18 CI TG-, 18 CI FG- und (nur in Serie 2) 18 CI FGd-Werte pro getestete Pflanze; lediglich vier Pflanzen aus Serie 1 weisen 17 und nicht 18 CI TG- und CI FG-Werte auf. Fraßspuren im Feld Von Oktober 1997 bis Januar 1998 wurde regelmäßig nach Fraßspuren von M. arvalis in einem Aus- fallraps-, Raps-, Zuckerrüben- bzw. Wintergerstenfeld und in vier Ackerkrautstreifen in Belp gesucht. Zuerst wurde nach Aktivitätszeichen (Laufwege, Kot, Löcher usw.) und dann nach Fraßspuren ge- sucht. Als Fraßzeichen galten: direkt an der Pflanze an- oder abgefressene Teile, ganze Pflanzen oder Teile davon in Baueingängen, Nahrungshäufchen auf Laufwegen. Mitte Juni wurden das Raps- und Gerstenfeld erneut nach Aktivitäts- und Fraßzeichen abgesucht. Hierbei wurde darauf geachtet, ob die im Winter festgestellten Vegetationsschäden im Rapsfeld noch erkennbar waren. Chemische Analysen der Pflanzen Die für die Analysen getrockneten Pflanzenproben wurden mit einem Mixer zu feinem Pulver gemah- len, in luftdichte Gläschen abgefüllt und danach im Dunkeln bei Raumtemperatur aufbewahrt. Die Analysen erfolgten folgendermaßen: Energiegehalt: Die Bestimmung des Energiegehalts erfolgte mit einem Calorimeter mit einer Einwaage von ca. 1 g und Benzoesäure als Kontrollstandard. Gesamt- stickstoff: Der Stickstoffgehalt wurde mit einem Stickstoffanalysator bestimmt. Die Einwaage betrug 158 LORENZA BALMELLIU.a. ca. 1g und als Kontrollstandard diente L-Asparaginsäure. Stärke und lösliche Zucker: Die freien Zucker wurden aus 100 mg Pulver mit 0,2M HCl extrahiert. Für die Stärkebestimmung wurde der zentrifugierte Rückstand zweimal mit Wasser gewaschen und mit 1,5M Perchlorsäure hydrolysiert. Als Nachweismethode diente der Anthrontest. Die optische Dichte wurde bei 623 nm gemessen und mit Fructose- bzw. Glucosestandards verglichen (MÜLLER-FERCH und Moucı 1995). Lösliche Phenole: Die Phenole wurden aus 1 g Pulver unter Rückflußkochen mit Methanol/H,O 2:1 (v/v) extrahiert. Das Folin-Ciocalteus Phenolreagenz diente als Nachweisreagenz. Die optische Dichte wurde bei 675 nm gemessen und mit Gallussäurestandards verglichen (ScEHovIc 1990). Von jeder der 16 getesteten Pflanzen standen drei Proben für die Analysen zur Verfügung (je eine pro „Cafeteria-Test“); für jede Pflanze wurden somit 3 unabhängige Werte pro Parameter be- stımmt. Datenauswertung Der Durbin-Rang-Test diente zum Nachweis eines globalen signifikanten Unterschieds zwischen den „Consumption Indices“ der acht Pflanzen jeder „Cafeteria-Serie“. Es wurde dieser Test verwendet, weil a) die Versuchsanordnung einem „incomplete block design“ entspricht, b) keine Normalvertei- lungen als Voraussetzung für parametrische Tests vorlagen. Bei einem „incomplete block design“ wird nicht jeder Block (hier 6 Tiere) allen Behandlungen (hier 8 Pflanzen) sondern nur einem Teil davon (hier 4 Pflanzen) unterzogen. Der Tukey-Test diente zur Bestimmung der signifikanten Unterschiede (MARASCUILO und MCSWEENEY 1997). Für jede Pflanze wurde aus den entsprechenden drei Tests der durchschnittliche Rang berechnet. Zur Überprüfung auf Korrelationen zwischen den durchschnittli- chen Rängen, die auf dem CI TG, CI FG und CI FGd basieren, wurde der Spearman-Rang-Korrela- tionskoeffizient berechnet. Der Wilcoxon-Paardifferenzen-Test diente zum Vergleich der „Consump- tion Indices“ der Blätter mit jenen der Wurzeln. Unterschiede bezüglich der Nahrungsbestandteile zwischen Blättern und Wurzeln bzw. Kultur- und Buntbrachepflanzen wurden mit dem Mann-Whit- ney-U-Test auf Signifikanz geprüft. Ergebnisse Futterwahlexperimente: Präferenzen In Serie 1 zeigten die Feldmäuse sowohl auf der Grundlage vom CI TG als auch vom CI FG signifikant unterschiedliche Präferenzen (Durbin-Test: X’ =56,16 für den CI TG, Y = 66,80 für den CI FG; df = 7, p< 0,001 für beide Tests; Abb. 1a). Die durchschnittli- chen Ränge der Pflanzen, die auf dem CI TG basieren, waren stark mit jenen korreliert, die auf dem CI FG basieren (Spearman-Rang-Korrelation, r, = 0,98, N=8, p< 0,001). Brassica napus, Trifolium pratense und Beta vulgaris waren sehr beliebt. Die restlichen fünf Pflanzen wurden hingegen wenig bis kaum gefressen. Als besonders unbeliebt zeig- ten sich die Rosettenblätter von Oenothera biennis. In Serie 2 stellten sich sowohl hinsichtlich des CI TG als auch des CI FG (Abb. Ib) und CI FGd ebenfalls signifikant unterschiedliche Präferenzen heraus (Durbin-Test: X =83,36 für den CI TG, y = 83,94 für den CI FG, x? = 82,27 für den CI FGd; df = 17, p < 0,001 für alle drei Tests). Auf eine Darstellung des CI FGd wurde in Abb. 1b verzich- tet, weil die Signifikanzen (Tukey-Test) wie jene der ersten Darstellung (CI TG) waren. Es ergab sich eine sehr starke Korrelation zwischen den durchschnittlichen Rängen der Pflanzen, die auf dem CI TG bzw. CI FG bzw. CI FGd beruhen (Spearman-Rang-Korre- lation, r, = 0,99, N= 8, p< 0,001 zwischen dem CI TG und dem CI FG bzw. CI TG und CI FGd bzw. CI FG und CI FGd). Hordeum vulgare, B. napus (Blätter) und Achillea mil- lefolium waren sehr beliebt. Die übrigen fünf Pflanzen wurden hingegen wenig bis kaum gefressen. Als besonders unbeliebt zeigte sich Verbascum densiflorum. In beiden Serien waren zwei der drei sehr beliebten Pflanzen Kulturpflanzen. Die ein- zige den zwei Serien gemeinsame Pflanze (Brassica napus Blätter) rangierte in beiden Serien hoch oben in der Präferenzrangfolge. Unter den Buntbrachepflanzen erwies sich Nahrungspräferenzen der Feldmaus Microtus arvalis 159 EEG Br.na.: Brassica napus (Blätter) Tr.pr.: Trifolium pratense (Blätter) Be.vu.: Beta vulgaris (Rübe) Si.al.. Silene alba (Rosettenblätter) Oe.bi.: Oenothera biennis (Wurzel) 4 be Di.fu.: Dipsacus fullonum (Wurzel) 2 Sial* Silene alba (Wurzel) O Oe.bi.*:Oenothera biennis (Rosettenblätter) m Kulturpflanzen Buntbrachepflanzen Br.na. Tr.pr. Be.vu. Stal. Oe.bi. Di.fu. Si.al.* Oe.bi.* a CIFG Br.na.: Brassica napus (Blätter) Tr.pr.: Trifolium pratense (Blätter) Be.vu.. Beta vulgaris (Rübe) Si.al.: Silene alba (Rosettenblätter) Oe.bi.: Oenothera biennis (Wurzel) Di.fu.: Dipsacus fullonum (Wurzel) Si.al.*: Silene alba (Wurzel) Oe.bi.*:Oenothera biennis (Rosettenblätter) Kulturpflanzen Buntbrachepflanzen Br.na. Tr.pr. Bevu Si.al. Oe.bi. Di.fu. Si.al.* Oe.bi.* Abb. 1a: „Consumption Indices“ auf der Grundlage vom Trockengewicht (CI TG) bzw. Frischgewicht (CI FG) der acht in Serie 1 getesteten Pflanzen. Die Box-Plots zeigen die 10%-, 25%-, 50%-(Median); 75%- und 90%-Quantile (N = 17 bzw. N = 18). Stichproben von Box-Plots mit gemeinsamen Buchsta- ben unterscheiden sich nicht signifikant voneinander (Tukey-Test, p > 0,05). nur T. pratense bzw. A. millefolium als besonders attraktiv, während sich alle anderen ge- gen Ende der Präferenzreihenfolge befanden. Auch unterschiedlichen Pflanzenorganen gegenüber verhielt sich M. arvalis selektiv und fraß signifikant mehr an Blättern als an Wurzeln (Wilcoxon-Paardifferenzen-Test, p < 0,001 für beide Serien). 160 LORENZA BALMELLIU.a. Ho.vu. Br.na. Ac.mi. Pa.sa. Levu. Br.na.* Ec.vu. Ve.de. b CIFG Ho.vu. Br.na. Ac.mi. Pa.sa. Le.vu. Br.na.* Ec.vu. Ve.de. Ho.vu.: Hordeum vulgare (Blätter) Br.na.: Brassica napus (Blätter) Ac.mi.: Achillea millefolium (Blätter) Pa.sa.: Pastinaca sativa (Wurzel) Le.vu.: Leucanthemum vulgare (Blätter) Br.na.*: Brassica napus (Wurzel) Ec.vu.: Echium vulgare (Wurzel) Ve.de.: Verbascum densiflorum (Wurzel) Kulturpflanzen Buntbrachepflanzen Ho.vu.: Hordeum vulgare (Blätter) Br.na.: Brassica napus (Blätter) Ac.mi.: Achillea millefolium (Blätter) Pa.sa.: Pastinaca sativa (Wurzel) Le.vu.: Leucanthemum vulgare (Blätter) Br.na.*: Brassica napus (Wurzel) Ee.vu.: Echium vulgare (Wurzel) Ve.de.: Verbascum densiflorum (Wurzel) Kulturpflanzen Buntbrachepflanzen Abb. 1b: „Consumption Indices“ auf der Grundlage vom Trockengewicht (CI TG) bzw. Frischgewicht (CI FG) der acht in Serie 2 getesteten Pflanzen. Die Box-Plots zeigen die 10%-, 25%-, 50-(Median), 75%- und 90%-Quantile (N = 18). Stichproben von Box-Plots mit gemeinsamen Buchstaben un- terscheiden sich nicht signifikant voneinander (Tukey-Test, p > 0,05). Nahrungspräferenzen der Feldmaus Microtus arvalis 161 Fraßspuren im Feld Im Feld wurden Fraßspuren an insgesamt 24 Pflanzenarten beobachtet. Die verschiede- nen Pflanzenarten bzw. -organe ließen sich je nach Fraßhäufigkeit in vier Kategorien ein- teilen (Tab. 2). Unter den am häufigsten (Kategorien 3 und 2) angefressenen Pflanzen be- fanden sich folgende Arten und Organe: Silene alba (alle vorhandenen Organe mit Ausnahme der Wurzeln), Brassica napus (Blätter), Verbascum densiflorum (Wurzel), Me- lilotus officinalis (Wurzel), Agrostemma githago (Samen), Trifolium pratense (Blätter), Dipsacus fullonum (Wurzel). Die Fraßspuren an B. napus im Winter waren besonders auffällig. Im Rapsfeld hatte M. arvalis auf zahlreichen größeren Flächen an den meisten Pflanzen die Blätter am Stiel abgefressen. Die Beweise, daß es sich dabei um Feldmäuse handelte, waren Laufwege mit Kot sowie Rapsblätter in Baueingängen rundherum. Bei der späteren Kontrolle im Juni wurden im Rapsfeld weder Aktivitäts- noch Fraßzeichen entdeckt und die im Winter festgestellten Vegetationsschäden waren nicht mehr erkenn- bar. Viele Pflanzenarten bzw. -teile wurden nur selten angefressen (Kategorie 1). An Beta vulgaris, Hordeum vulgare und Echium vulgare konnten im Feld keine Fraßzeichen beobachtet werden. Das Gerstenfeld war auch im Juni unbeschädigt. In fol- genden Fällen stimmen die Nahrungspräferenzen, dıe sich aus den Tests ergaben, nicht mit jenen überein, die aufgrund der Fraßhäufigkeiten im Feld resultierten: S. alba (Ro- settenblätter), V. densiflorum und D. fullonum schienen im Feld beliebter zu sein als in den Laborversuchen. Umgekehrt erwiesen sich FH. vulgare, B. vulgaris und Achillea mille- folium in den Wahlversuchen viel attraktiver als im Feld. Im Falle von B. vulgaris dürfte es sich jedoch um eine Fehleinschätzung handeln, da beim Ernten des Feldes viele Feld- mäuse zum Vorschein kamen (GyGui, pers. Mitt.). Die Tiere hatten dort wahrscheinlich Fraßschäden angerichtet, die aber nicht entdeckt wurden. Beim Vergleich von Blättern mit Wurzeln kam im Feld die im Labor festgestellte Vor- liebe für Blätter nicht zum Vorschein. Nahrungskomponenten Die Werte der untersuchten Komponenten sind in Tabelle 3 zusammengestellt. Werden die zwei Serien zusammen betrachtet, ergibt sich in der Variabilität der einzelnen Para- meter folgendes Bild: Phenole, lösliche Zucker und Stärke waren die am stärksten vari- ierenden Parameter. Die größte Variation ergab sich im Phenolgehalt: Die Rosettenblät- ter von Oenothera biennis hatten einen ca. 20fach größeren Wert als die Wurzel von Beta vulgaris, Silene alba, Pastinaca sativa und Brassica napus. S. alba (Rosettenblätter) hatte mit 9,7% den niedrigsten, B. vulgaris mit 67,4% den höchsten Zuckergehalt. P. sativa war mit 7,7% besonders reich und Trifolium pratense mit 2,0% besonders arm an Stärke. Ab- gesehen von B. vulgaris bestanden zwischen den Pflanzen nur mäßige Unterschiede im Stickstoffgehalt; im Durchschnitt lagen die Werte um 3,5%. Die Werte des Energie- und Wassergehalt, zeigten die kleinste Variation und lagen um 17,5 kJ bzw. 80%. Blätter waren signifikant reicher an Energie und Stickstoff (Mann-Whitney-U-Test, p< 0,01) sowie an Wasser und Phenolen (p < 0,05) als Wurzeln. Wurzeln enthielten hinge- gen signifikant mehr Zucker (p< 0,001) und tendenziell auch mehr Stärke (p = 0,059). Zwischen Kultur- und Buntbrachepflanzen bestand hingegen bezüglich keiner der Kom- ponenten ein signifikanter Unterschied (p > 0,05). Für einen Vergleich der „Consumption Indices“ mit den untersuchten Parametern müssen Serie 1 und 2 getrennt betrachtet werden. Der Vergleich der Werte der Nahrungs- bestandteile der drei Pflanzen mit dem höchsten CI mit jenen der restlichen fünf läßt in keiner Serie eine Tendenz erkennen, Pflanzen mit einem höheren Gehalt an positiven Komponenten (Wasser, Energie, Stickstoff, Zucker, Stärke) bzw. mit einem niedrigeren Phenolgehalt zu bevorzugen. LORENZA BALMELLIU.a. 162 „(uswes pun uposdeyusues) 08pynS punuajsoady „(uposdeyuswes) pgjp auopıs „(uIopeIg run [98UusIS) pgjp auoms „(SsIseq[o3uslS pun [pZzıny) sıypumı[fo snopJoW 9°E (PZIn A) wnaojfısuap wnaspgqaaA zur] ([91S u 1opelg]) Sndou vaısspag TI/ID Tz (ONSSOJeIg) pgpp auaps 19 e (pzıny pun usdsouyusInig) pıp[ın[ odpssn] „(ujosdeyuswes) undofmuos [ 'n svooya aoandn „(SweH) Sıypraıı Dog „(sIseq[odusJs pun [9Zzın \) 24pdjna wınjn21uao] „(oNDSOAMEeIg) dolpıu odpıun]d +(USIUV up Sne I9UION]) wnAansop wundy1A] LE (pzıny) wnuoym/ snovsdiq ID y] (uionelg yru [93u9JS) osuamwad wnmofurL ID T (199eI Sdıpue)spunıg) Doonl poAnpjua) (uıopelgl nu [9USIS) pryofnaıa sıyadagouo (SN9SOAYeI]) PIDJOOEDJUD] Odpyunyg „(u1opelgl yıuu 98U9IS) DLIODUN snuoyruy „(sIsegJogus]s) panns 0o8pM1pa „(OAyy) wnioyfupnu un]07 „(uıonelg yru [93U9JS) sıypumı/fo smomj>W 67 (ONE) 2andjna wnweyyunandT XD 67 (PZIınM) panps D9Vunsdd DD 27 (operg) unmoforpmua voymypy D Le (pzınqy) sndou vaissvag DD se (ONPSOAJEeI) Suuaıg vaayyuaO ID 9° (PZnM) Pgjp auons TI 97 (pZıny) stuuaig vaaylouaO ID SE (PZınM) 94Ddna una DD TI (oneIg) o4pdjna wunap4oH DD 07 .(qna) Ssıumajna Drag ID 0 "USP-I9IM 19JU1993 19pO USALOISIE IOPIMJU9 Isgioyyeds un SfrsjuszuejJd "MZg uazuejJg 9SOIp ep “uspınm Jypewog »porodsgunyansisjun A9p Suejuy we nu Jıp “usgunyydegosgl !x -u9gaFagur (Jy9IMaFYISLıJ WOP ypeu „xopuj uondumsuoy“) HI ID WOA 9Fejpunın) I9p Jne duey Syaıpyruydsysınp I9P „OZULJJI-ELISISTEY“ Opal any Ist y9IOJSA19A SIV PPU9MIOA 7 'MZQ | „OLIOS-ELIIISFTEI“ U !7I "ID YNJ9STNE U9ZULJJA SPPUIMIOA USYINSI9AJUEM UI INU PUIS 0. 2IuN '(%00T-89 10 Ay9s = € ‘(%LY-FE) 170 = 7 (u9uonyodsumpfag I9P YEE-T) UPS = [ 'DTu = () :uospınm JoYy9egoag usandsgeig 19p rw ‘NOYSINEH I9p ne Jynıoq usyfeds 1914 Hp ur Junjisyurg SIG “uapınm JoJydegoag usındsgjerg IJ0 Iy9s SIQ U9JJ9S MZG Julsy pad WI usuap ur ‘S]Io)- pun USIEU9ZULJJA *T AIPAEL 163 Nahrungspräferenzen der Feldmaus Microtus arvalis 6TFIE sSorrl SIFLI VEFIY TOFITZ coFr] 00FYO DOFYIE 00FrO LOFLE 00F61 VOFSI rOF60I OFFO TOFL's TOFLT goFrsSt 00FJI0 SOoFTY TOFET (%) 9JOoUusyd UIFO0r SOrTeE sIırrr LOFOE LIFEV 0IFS9 LOF6E LOFLT SIFLL EOFTLE LoOFL7 60FTr LOFEE SOFSE LoFre TOFTE TOFYE 00OFST TOF07 TOFLT (%) YABIS LIT FO'SE gLIFTLE SHFTor SIFITZ goFSEer 6TFS'SS SIFOLE SITFrTIe TOFrES STFSST TcHsSsT STFTLST LIFIeE STFIEH SorFsTr IOF6Hrr TIFL6 ToFrL9 E0F09 OL FSOE (%) I9YNZ LOFTE SIFSE 60797 Ee0F0r LoFrEe TOFLT LOF6E ToFOr TOFLT LoF/LE LoOFrr EOFEr TOFSE LOFLI7Z L0F67 LOFT7 LOFEer 00F60 roFsie zor0r (%) YOISNOHS SOFELI SOFOSI SOFSI IOFTSI LOFZLl E00 FIT roOFgL] r00FL8I EOOFOLI LOF8_I TOFISI ToOFLS] TOFTZ/] roF9y9 LOF6SI rOF69I L0OF64/1 rOFs'II ToF6S8I ‘ ‘ I0O+1381 (sans ‘310 3/4) SIS19ug LSFLT8 rEeF618 sr ry8L YTF0Ss8 IOFSSL SOFErHL 90 F 908 I0F098 60 FS6L LO FESS 607678 SOFTHS LO F 598 SOFSOL E0 F 998 EOFLIS TIF068 SOFTLL roOFLI8 Lo +8r8 (%) J9sseM Eu) uaIpoN uszuegdaydeigyung uszueggdanyıny up>zunAaA 1opeld (Dzıny) wnAaoyfisusp wnaspqaaA (pzıny\) 9403jna wunıyaF7 (Ppzıny) Sndou vaısspag (1O91eJg]) 91pdjna wnwoyyunandaT] (PZıny) panps D9DunsDg (OneIg) wnmofomu voy>W (109eIg) sndou va1sspvag (10YeIg) 91D8jna wınap4of z 291195 (19YEJqUSNISoNy) Suuaıg DpA2yJouU9aO (pzıny) pgjp auons (PZın My) suuoig vAayoU2O (Zın y) wnuom/ snavsdıq (19YEIqusyasoy) pgjp auos (sqny) sıupajna miag (191eIg) asuowad wnmofia] (10Yeg) sndou va1sspag KIN JUBF1O- pun US] IBUSZUL]JA -usgagadue AffogeL I9p apug we yaıjzyes -nz puis uszuejpdoypeigyung pun -ınyynyy ınJ 9ImMos ujoziny pun IoYejg In] USLIOS uSpIog IOP I9M[YIN ssewusy>9oAL >Ip Jne 'syeyasıasseq SOp Juyeu -Sny JIuu ‘yaıs uSy9IZOgq pun usgagadue JunyammgepIepuels F SLIOM[ONIMN Se puIs Sfrsjpuejsagsduniygen 9Iq "Iynjosjne HA ID ueipam wapuswysuge y9eu uszue]Jd DIp puIs 9LIOS »pal ang "uaZuejJd U9J9I89%93 19p usyuouodwoy] Iıyansıoyun pun (YA ID) Iy9IMmaSyasıLı] wop yoeu „sasıpuj uondumsuoy“ € AdqeL 164 LORENZA BALMELLIU.a. Diskussion Futterwahlexperimente: Präferenzen Die Ergebnisse der Wahlversuche bestätigen die Erkenntnisse anderer Autoren in der ausgeprägten Selektivität von M. arvalis bei der Nahrungswahl (Yu et al. 1980; LEUTERT 1983; RınkE 1990, 1991). Der Vergleich der mittels „Cafeteria-Tests‘“ festgestellten Präfe- renzen mit jenen aus anderen Untersuchungen ist jedoch durch folgende Faktoren er- schwert: Zur Untersuchung der Nahrungsökologie der Feldmaus wurden bisher verschie- dene Methoden angewendet. Die meisten davon weisen bestimmte Schwachpunkte auf. Mageninhalts- bzw. Kotuntersuchungen zeigen nur die Zusammensetzung der letzten Nahrungsaufnahme bzw. führen zur Unterschätzung von leicht verdaulichen Pflanzentei- len wie z.B. fleischigen Wurzeln (PHiLLıpson et al. 1983). Bei der Bestimmung der Arten- zusammensetzung von Nahrungsresten werden Pflanzen, die in situ gefressen wurden, nicht erfaßt. Futterwahlexperimente berücksichtigen nicht die relative Ressourcenverfüg- barkeit, die eine wichtige Komponente der Nahrungswahl in der Natur ist (GODFREY 1953; ZIMMERMAN 1965; RıEwE 1973). Die untersuchten Feldmauspopulationen stammen aus verschiedenen Habitaten (Agrarökosysteme, Fett- bzw. Trockenwiesen usw.), was 2. T. zu völlig unterschiedlichen erfaßten Pflanzenarten führt. Zwischen Populationen aus dem- selben Habitattyp können beträchtliche Unterschiede bezüglich der Nahrungspräferenzen bestehen. Die verschiedenen Studien wurden z.T. zu unterschiedlichen Jahreszeiten durchgeführt. Daraus wird ersichtlich, warum sowohl Übereinstimmungen als auch Dis- krepanzen zu Literaturangaben vorliegen können. In der Folge werden die Ergebnisse der Futterwahlversuche mit den Befunden an- derer Autoren, die das Nahrungswahlverhalten von M. arvalis quantitativ untersucht ha- ben, verglichen. Es wird nur auf die wenigen gemeinsamen Pflanzenarten eingegangen. In Einklang mit dem sehr niedrigen Median „Consumption Index“ von Leucanthemum vulgare in dieser Studie konnte RınkE (1990) diese Art nur selten und mit einem sehr geringen Volumenanteil im Mageninhalt nachweisen. Im Gegensatz zu der hier registrier- ten hohen Attraktivität von Trifolium pratense ordnete Rınke (1990) diese Art der Kate- gorie derjenigen mit minderer Präferenz zu. Die hier ermittelte Attraktivität von Achillea millefolium stimmt weder mit den Befunden von LEUTERT (1983) noch mit jenen von RınkE (1990) überein. Der erstere stellte gegenüber dieser Art Gleichgültigkeit fest und der letztere teilte sie in die Kategorie der Arten mit minderer Präferenz ein. So wie hier signifikant mehr an Blättern gefressen wurde als an Wurzeln, bestand die Diät der von TRUSZKowsKI (1982) untersuchten Population oft zu fast 100% aus grünen Pflanzenteilen, während Wurzeln mit einem Anteil von nur 0,1 bis 7,4% in der Diät vorhanden waren. Abgesehen von wenigen Ausnahmen schien M. arvalis bei der Wahl zwischen Kultur- und Buntbrachepflanzen erstere vorzuziehen. Im Gegensatz dazu stellte TRUSZKowskI (1982) in Getreide- und Rapsfeldern eine Präferenz für Unkräuter fest. Man muß aber beden- ken, daß in der vorliegenden Arbeit nur eine kleine Auswahl von Buntbrachepflanzen berücksichtigt wurde. Fraßspuren im Feld Ein sehr artenreiches Nahrungsspektrum der Feldmaus ergab sich auch aus anderen Stu- dien (Yu et al. 1980; LEUTERT 1983; RınkE 1990; BAUMANN 1996). In Einklang mit den hier erhobenen Feldbefunden stellte StTEin (1958) besonders im Winter starken Fraß an Trifolium pratense fest. Die großen Schäden an Brassica napus müssen nicht verwundern, da nach STEın (1958) M. arvalis in Winterrapsfeldern optimale Lebensbedingungen fin- det. Raps ist nur im frühen Stadium gefährdet (Stein 1958). Dies erklärt, warum bei der späteren Kontrolle im Juni keine Fraßzeichen entdeckt wurden. Die Tatsache, daß die im Nahrungspräferenzen der Feldmaus Microtus arvalis 165 Winter festgestellten Vegetationsschäden im Juni nicht mehr erkennbar waren, läßt sich auf die hohe Regenerationsfähigkeit von Raps zurückführen. Unerwartet war das Fehlen von Fraßzeichen im Gerstenfeld im Juni, da unter den Kulturpflanzen reifendes Getreide am stärksten von diesem Ackerschädling betroffen ist (Stein 1958). Mitte Juni waren die Gerstenkörner noch nicht ganz reif und vielleicht deshalb für die Feldmäuse nicht attrak- tiv. Nach Stein (1958) ist Beta vulgaris wenig begehrt, während hier weder das Vorhan- densein noch das Fehlen von Fraßspuren mit Sicherheit behauptet werden kann. Die Tatsache, daß einige der in den Wahlversuchen verwendeten Pflanzen im Feld be- liebter bzw. weniger attraktiv waren als im Labor, läßt sich mit großer Wahrscheinlichkeit auf Faktoren wie z.B. Feinde und Deckung zurückführen, die in den Laborexperimenten unberücksichtigt blieben. So konnten im Feld trotz des sehr hohen „Consumption Index“ keine Fraßzeichen an Hordeum vulgare beobachtet werden, da sich die Feldmäuse im Winter vermutlich wegen unzureichender Deckung nicht ins Gerstenfeld wagten. M. arvalis frißt nur bei guter Deckung (STEIN 1958). Nahrungskomponenten Die Erwartung bei der Bestimmung einiger Komponenten war, daß nur solche Para- meter, die unter den untersuchten Pflanzen eine deutliche Variabilität aufwiesen, eine Rolle bei der Nahrungswahl spielen. Aufgrund dieser Erwartung erwiesen sich nur die Phenole, der Zucker und die Stärke als potentiell wichtig. Angesichts seiner mäßigen Variation zeigte sich der Stickstoffgehalt hingegen als potentiell weniger wichtig. Daß der Wasser- und Energiegehalt nur eine kleine Variation aufwiesen, war vorauszusehen: Ei- nerseits haben Blätter und dicke, fleischige Wurzeln in der Regel eine nicht stark von- einander abweichende prozentuale Trockenmasse und andererseits wurden keine fettspei- chernden und somit energiereicheren Organe untersucht. Die Tatsache, daß in den „Cafeteria-Tests‘“ Blätter einen signifikant höheren „Con- sumption Index“ hatten als Wurzeln, steht mit großer Wahrscheinlichkeit nicht im Zu- sammenhang mit dem festgestellten signifikanten Unterschied im Gehalt an den unter- suchten Komponenten. Vielmehr könnte die in den Wahlversuchen fehlende Deckung die Ursache davon sein: Das Fressen aus den niedrigen und offenen Futternäpfen bzw. auf dem kahlen Käfigboden war den Tieren sicher unangenehm. Deshalb haben sie versucht, das Futter zum unmittelbaren Verzehr möglichst in den Blumentopf zu transportieren. Der kleine Durchmesser des Eingangslochs hat ihnen das Eintragen und somit den Ver- zehr von Wurzeln stark erschwert. Diese Interpretation würde somit auch erklären, wes- halb aufgrund der Fraßspuren im Feld keine Vorliebe für Blätter zum Vorschein kam. Der Vergleich der „Consumption Indices“ mit den Werten der analysierten Parameter zeigte, daß zwischen den Präferenzen und diesen Parametern keine Beziehung besteht. Auch in der Mehrheit der bisher durchgeführten Studien zur Nahrungsökologie von Kleinnagern besteht kein Zusammenhang zwischen Präferenzen und dem Gehalt an Koh- lenhydraten, Phenolen und Stickstoff. Dies gilt für die Arbeiten von Kopp (1993) und Bozınoviıc et al. (1997) bezüglich des Kohlenhydratgehalts, für jene von LiNDROTH und BAtzuı (1984), Marouis und BArtzuı (1989), Kopp (1993) sowie HJÄLTEN et al. (1996) be- züglich des Phenolgehalts und für jene von HJÄLTEN et al. (1996) und MURRAY und Dick- MAN (1997) bezüglich des Stickstoffgehalts. Im Gegensatz dazu fanden BERGERON und Jo- DOoIN (1987) sowie HARJU und HAKKARAINEN (1997) eine starke negative Korrelation zwischen Präferenzen und Phenolgehalt, während Marauıs und Barzıı (1989) und Kopp (1993) eine positive Korrelation zwischen Präferenzen und Stickstoffgehalt feststellten. Das Fehlen einer Beziehung zwischen den hier festgestellten Präferenzen und den analy- sierten Komponenten heißt aber nicht, daß sich Feldmäuse prinzipiell ohne Rücksicht auf das Vorhandensein bestimmter Nahrungsbestandteile ernähren. Folgendes muß berück- sichtigt werden: Phenole sind eine komplexe Gruppe von Sekundärmetaboliten, zu der 166 LORENZA BALMELLIU.a. mehrere Verbindungen mit vielfältigen Wirkungen auf Herbivore gehören (LINDROTH 1989): a) Einfache Phenole und Flavonoide können verschiedene toxische Effekte haben wie z.B. Hemmung der Zellatmung, der Enzymfunktion und des Membrantransports. b) Cumarine können Organläsionen verursachen sowie hämorrhagische und koagulations- hemmende Effekte zeigen. c) Tannine wirken infolge oberflächlicher Eiweißfällung auf Haut- und Schleimhautzellen adstringierend. Weiter hemmen sie Verdauungsenzyme, bil- den mit Nahrungsproteinen unverdauliche Komplexe und reduzieren die Aktivität der Mikrobenflora des Darmes. d) Lignine hemmen die Verdauung. Daraus geht hervor, daß die Bestimmung der Gesamtphenole ohne jegliche Differenzierung vermutlich nicht aus- reicht, um den Effekt dieser vielseitigen Stoffgruppe auf die Schmackhaftigkeit festzustel- len. Der Gesamtstickstoffgehalt gibt meistens keine korrekte Schätzung des Proteinan- teils an. Stickstoff ist nicht nur in proteinogenen Aminosäuren enthalten, sondern z.B. auch in mehreren Verbindungen unter den Sekundärmetaboliten wie Alkaloiden, Ami- nen, cyanogenen Glykosiden, Glucosinolaten und nicht proteinogenen Aminosäuren (LinprotTH 1989). Es wurden nur einige wenige positive und negative Parameter der Pflanzen bestimmt. Nach Barzıı (1985) richten sich Wühlmäuse bei der Nahrungswahl auch nach dem Gehalt an Kalzium, Phosphor, Natrium, Faser, Alkaloiden und Saponi- nen. Die Verdaulichkeit des Futters für Herbivore ist negativ mit dem Fasergehalt (Hemicellulose, Cellulose, Lignin) korreliert und zu viel Faser kann die Nahrungsauf- nahme von Wühlmäusen hemmen (Barzıı 1985). Alkaloide und Saponine wirken toxisch (LinDroTH 1989). Der Versuch, die Präferenzen der Feldmaus mit wenigen, einzeln be- trachteten Nahrungsbestandteilen zu erklären, ist auch deshalb schwierig, weil die Präfe- renzen vermutlich von einem Komplex interagierender Komponenten bestimmt werden. So können z.B. Giftwirkungen einiger Stoffe durch andere Bestandteile der Nahrung ge- mildert werden. Die gleichzeitige Aufnahme von Tanninen und Saponinen (im richtigen Verhältnis) mit der Nahrung kann die Absorption der Gifte im Darmtrakt verhindern (FREELAND et al. 1985). Natrium ist wahrscheinlich ein limitierendes Element für die Ent- giftung von Abwehrstoffen der Futterpflanzen (HaAnsson 1990). Sollte dies der Fall sein, könnte durch die Bestimmung des Natriumgehalts die Beziehung zwischen Präferenzen und Sekundärmetaboliten aufgeklärt werden. Um die Aufnahme chemischer Abwehr- stoffe zu vermeiden, mag die Feldmaus in einigen Fällen Pflanzen mit einem niedrigen Nährwert bevorzugen. In anderen Fällen wird sie Pflanzen mit einem hohen Gehalt an Sekundärmetaboliten vorziehen, weil die Kosten dieser Wahl bei weitem von hohen Nährstoffwerten kompensiert sind. Die einzige Pflanze, die sich sowohl in den Wahlversuchen (Serie 1 und 2) als auch im Feld als äußerst beliebt erwies, ist Brassica napus (Blätter). Außer den in den Blättern enthaltenen Proteinen und Carotinoiden mit Vitamin-A-Wirkung (Hänser et al. 1971- 1979, 1992-1994) sind keine Inhaltsstoffe der Blätter dieser Pflanze bekannt, die für diese hohe Attraktivität verantwortlich sein könnten. Das Fehlen von Glucosinolat und Eruca- säure im O0-Raps ist mit großer Wahrscheinlichkeit nicht ausschlaggebend, da bereits STEIN (1958) über die Beliebtheit von Winterraps berichtete, als O0-Raps noch nicht ange- baut wurde. Folgende Angaben über Inhaltsstoffe einiger der getesteten Pflanzen könn- ten eventuell deren Unbeliebtheit erklären: Die Blätter von Oenothera biennis, die in Serie 1 kaum angerührt wurden, enthalten Quercetin (ein Flavonoid) sowie 11% Gerbstoffe (HÄnseEL et al. 1992-1994). Am bekann- testen unter den Gerbstoffen ist Tannin, das bei einer Konzentration über 2% Säugetiere vom Fressen abhält (Swaın 1979). Die Wurzel von Pastinaca sativa enthält Pastinacin (ein Alkaloid) und Furanocumarine (HÄnseL et al. 1971-1979, 1992-1994). Leucanthemum vulgare ist schwach cyanogen und enthält Polyacetylene und Flavonoide (HEGNAUER 1962-1986; HÄnsEL et al. 1992-1994). In Verbascum densiflorum sind Flavonoide vorhan- den (KrnETA-JorDI 1998). Nahrungspräferenzen der Feldmaus Microtus arvalis 167 Danksagungen D. PaveELıc und E. VoGeEL danken wir für die Hilfsbereitschaft bei der Bestimmung der Kohlenhydrate bzw. Phenole. H. J. BACHMANnN sind wir für die Durchführung der Stickstoffanalyse dankbar. M. GyGLi stellte seine Felder als Untersuchungsgebiet und J.-P. CHARLES die Makrolonkäfige zur Verfügung. E. HASHoRVA und D. DiETRIcH danken wir für die statistische Beratung und T. FRANK sowie F. BAL- MELLI für die kritische Revision des Manuskripts. Zusammenfassung Die Nahrungswahl der Feldmaus Microtus arvalis bezüglich Pflanzen der Agrarlandschaft wurde mit- tels Futterwahlexperimenten im Labor untersucht. Als qualitativer Vergleich diente die Untersuchung der Fraßspuren im Feld. Mit der Bestimmung einiger Pflanzeninhaltsstoffe (Stickstoff, Zucker, Stärke und Phenole) sowie des Energie- und Wassergehalts wurde nach einer Ursache der Präferenzen ge- sucht. Aus den Wahlversuchen stellte sich eine stark ausgeprägte Selektivität bei der Nahrungswahl heraus. Sehr beliebte Pflanzen waren unter den Kulturpflanzen Hordeum vulgare (Blätter), Brassica napus (Blätter) und Beta vulgaris altissima (Rübe) bzw. unter den Buntbrachepflanzen Achillea mille- folium (Blätter) und Trifolium pratense (Blätter). Zwischen den Präferenzen und den analysierten Nahrungskomponenten bestand keine Beziehung. Von den fünf im Labor sehr attraktiven Pflanzen wurden im Feld nur an T. pratense und B. napus oft bzw. sehr oft Fraßzeichen beobachtet. Literatur BATZL1, G. ©. (1985): Nutrition. In: The Genus Microtus in North America. Ed. by R. TAMARIN. Am. Soc. Mammal. Spec. Publ. 6, 779-811. BAUMANN, L. (1996): The influence of field margins on populations of small mammals. A study of the population ecology of the common vole Microtus arvalis in sown weed strips. Field margins News- letter 6, 7, 26-34. BERGERON, J. M.; Jopoin, L. (1987): Defining „high quality“ food resources of herbivores: the case for meadow voles (Microtus pennsylvanicus). Oecologia 71, 510-517. Bozınovic, F.; NovoA, F. F.; NEspoLo, R. F. (1997): Effect of dietary composition on food selection and assimilation in the leaf-eared mouse (Phyllotis darwini) inhabiting central Chile. Rev. Chil. Hist. Nat. 70, 289-295. FRANK, T.; NENTWIG, W. (1995 a): Artenvielfalt von Laufkäfern (Carabidae), Schwebfliegen (Syrphidae) und Tagfaltern (Rhopalocera) in Ackerkrautstreifen und angrenzenden Feldern. Mitt. dtsch. Ges. allg. angew. Ent. 9, 685-691. FRANK, T.; NENTWIG, W. (1995 b): Ground dwelling spiders (Araneae) in sown weed strips and adjacent fields. Acta Oecol. 16, 179-193. FREELAND, W. J.; CALCOTT, P. H.; ANDERSON, L. R. (1985): Tannins and saponin: Interaction in herbivore diets. Biochem. Syst. Ecol. 13, 189-193. GODFREY, G. K. (1953): The food of Microtus agrestis hirtus (Bellamy, 1839) in Wytham, Berkshire. Säu- getierkdl. Mitt. 1, 148-151. HÄNSEL, R.; KELLER, K.; RIMPLER, H.; SCHNEIDER, G. (1971-1979): Hagers Handbuch der pharmazeu- tischen Praxis. Bde. 1-6. Berlin: Springer. HÄNSEL, R.; KELLER, K.; RIMPLER, H.; SCHNEIDER, G. (1992-1994): Hagers Handbuch der pharmazeu- tischen Praxis. Bde. 1-6. Berlin: Springer. Hansson, L. (1990): Mineral selection in microtine populations. Oikos 59, 213-224. HARJU, A.; HAKKARAINEN, O. (1997): Effect of protein and birch-bark powder on selection of food by root voles (Microtus oeconomus). J. Mammalogy 78, 563-568. HAUSAMMANN, A. (1996): The effects of weed strip-management on pests and beneficial arthropods in winter wheat fields. Z. PflKrankh. PflSchutz 103, 70-81. HEGNAUER, R. (1962-1986): Chemotaxonomie der Pflanzen. Bde. 1-7. Basel, Stuttgart: Birkhäuser. HJÄLTEN, J.; DANELL, K.; Erıcson, L. (1996): Food selection by two vole species in relation to plant growth strategies and plant chemistry. Oikos 76, 181-190. 168 LORENZA BALMELLIU.a. Kopp, R. (1993): Etude de l’impact de la forme fouisseuse du campagnol terrestre, Arvicola terrestris scherman (Shaw), sur la vege&tation d’une prairie. These de doctorat, Univ. Lausanne. KRrNETA-JorDI (1998): Der Heilpflanzengarten im Botanischen Garten der Universität Bern. Bern, Hannover: Agrarökologie. LEUTERT, A. (1983): Einfluß der Feldmaus, Microtus arvalis (PALL.), auf die floristische Zusammenset- zung von Wiesen-Ökosystemen. Veröff. Geobot. Inst. ETH, Stiftung Rübel (Zürich) 79, 1-127. LiLLE, R. (1996): Zur Bedeutung von Bracheflächen für die Avifauna der Agrarlandschaft: Eine nah- rungsökologische Studie an der Goldammer Emberiza citrinella. Agrarökologie 21, 1-150. LINDROTH, R. L. (1989): Mammalian herbivore — plant interactions. In: Plant-Animal Interactions. Ed. by W. G. ABRAHAMSonN. New York, St. Louis, San Francisco, Auckland: McGraw-Hill. Pp. 163-206. LINDROTH, R. L.; BATZLı, G. ©. (1984): Food habits of the meadow vole (Microtus pennsylvanicus) in bluegrass and prairie habitats. J. Mammalogy 65, 600-606. Lys, J.-A.; NEnTwIcG, W. (1992): Augmentation of beneficial arthropods by strip-management. Oecolo- gia 92, 373-382. Lys, J.-A.; NENTwIıG, W. (1994): Improvement of the overwintering sites for Carabidae, Staphylinidae and Araneae by strip-management in a cereal field. Pedobiologia 38, 238-242. MARASCUILO, L. A.; MCSWEENEY, M. (1997): Nonparametric and Distribution-Free Methods for the So- cial Sciences. Belmont, California: Wadsworth. MARauIıSs, R. J.; BATZLI, G. ©. (1989): Influence of chemical factors on palatability of forage to voles. ]J. Mammalogy 70, 503-511. MÜLLER-FERCH, G.; Moucı, M. (1995): Einfluß der Mahd auf die Reservestoffe und den Insektenbe- stand von Ackerkräutern. Agrarökologie 14, 1-103. MURRAY, B. R.; DIcKMANn, C. R. (1997): Factors affecting selection of native seeds in two species of Aus- tralian desert rodents. J. Arıd Environ. 35, 517-525. PHILLIPSON, J.; SARRAZIN-COMANS, M.; STOMATOPOULOS, C. (1983): Food consumption by Microtus agres- tis and the unsuitability of faecal analysis for the determination of food preference. Acta Theriol. 28, 397-416. RIEWE, R.R. (1973): Food habits of insular meadow voles, Microtus pennsylvanicus terraenovae, (Ro- dentia: Cricetidae) in Notre Dame Bay, Newfoundland. Can. Field Nat. 87, 5-13. Rınke, T. (1990): Zur Nahrungsökologie von Microtus arvalis (Pallas, 1779) auf Dauergrünland. I. All- gemeine Nahrungspräferenzen. Z. Säugetierkunde 55, 106-114. Rınke, T. (1991): Percentage of volume versus number of species: Availability and intake of grasses and forbs in Microtus arvalis. Folia Zool. 40, 143-151. SCEHOVIc, J. (1990): Tanins et autres polymeres phenoliques dans les plantes de prairie: determination de leur teneur et de leur activite biologique. Revue suisse Agric. 22, 179-184. STEIN, G. H. W. (1958): Die Feldmaus. Wittenberg Lutherstadt: Ziemsen. SwAIn, T. (1979): Tannins and lignins. In: Herbivores: Their Interaction with Secondary Plant Metabo- lites. Ed. by G. A. ROSENTHAL and D. H. JANZEn. New York: Academic Press. Pp. 657-682. TRUSZKOWSKI, J. (1982): The impact of the common vole on the vegetation of agroecosystems. Acta Theriol. 27, 305-345. WALDBAUER, G. P. (1968): The consumption and utilisation of food by insects. Adv. Insect Physiol. 5, 229-289. Yu, O.; VERGNE, Y.; GOUNOT, M. (1980): Modele d’interaction entre campagnols Microtus arvalıs et prairie permanente. Rev. Ecol. (Terre Vie) 34, 373-426. ZIMMERMAN, E. G. (1965): A comparison of habitat and food of two species of Microtus. J. Mammalogy 46, 605-612. Anschrift der Verf.: LorenzA BALMELLI und Prof. Dr. WoLrGAanG NENTWIG, Zoologisches Institut, Universität Bern, Baltzerstr. 3, CH-3012 Bern und Dr. JEAN-PIERRE AIROLDI, BES/Biologie, Universität Bern, Gertrud Wokerstr. 5, CH-3012 Bern, Schweiz EN 5 Z. Säugetierkunde 64 (1999) 169-175 ZEITSCHRIFT FÜR © 1999 Urban & Fischer Verlag SÄUGETI ERKU NDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Parental care and time sharing in the Mongolian gerbil By R. WEINAnDY and R. GATTERMANN Institute of Zoology, Martin Luther-University Halle Wittenberg, Halle Receipt of Ms. 02. 10. 1998 Acceptance of Ms. 14. 01. 1999 Abstract The biparental care behaviour of the social Mongolian gerbil (Meriones unguiculatus) was quantified from birth to weaning of the young under laboratory conditions. Nestbuilding, nest-residence, and re- trieving of the offspring were measured. The behaviour of the parents was registered per video-obser- vation on days 2, 5, 8, 13, and 20 after the birth of the young, each for 24h. To obtain control data, we additionally observed all pairs for 24 h without progeny. The objective of our study was to evalu- ate the paternal and maternal efforts in rearing the young and to focus on parental time sharing in the nest. The female made the greatest contribution to care since there was no paternal support in building of the litter-nest and retrieving of the young. However, both adults cooperated via their synchronized presence with the young (temporal coordination or time sharing in the nest). We suggest that via this mutual behavioural synchronization the physiological strains of the female caused, e.g., by lactational hyperthermia are reduced. Key words: Meriones unguiculatus, parental care, time sharing Introduction Although in mammals the females mainly care for the young, paternal investment can also increase their chances of survival via direct support, such as warming, and more in- direct assistance like nest-building or defending the young. Investigations on cooperative breeding and paternal care mainly refers to primates and carnivores (for a review see: So- LOMON and FRENCH 1997; for rodents: OSTERMEYER and ErwooD 1984; SOLOMoN and GETZ 1997; GERLACH 1998). Based on short-term observations during the light period (90 minutes per family), EL- wooD (1979) showed that certain parental activities of the social Mongolian gerbils, e.g., nest-building, are influenced by the other parent. The aim of our long-term study was to quantify in greater detail paternal and maternal efforts from birth to weaning of the altri- cial young and to focus on parental time sharing in the nest. Material and methods Animals and housing We selected six adult males and six females from a laboratory colony of Mongolian gerbils (Meriones unguiculatus). The animals came from different litters and were caged in pairs after weaning at 6-8 weeks. They were kept in climatised rooms with a photoperiod of 12:12 h light (200-300 Ix per 0044-3468/99/64/03 — 169 $ 12.00/0 170 R. WEINANDY and R. GATTERMANN cage): dark (5-10 Ix per cage) (light period: 0700-1900 h Central European Time). The room tempera- ture was 24+2°C and the relative humidity varied from 65-70 %. The cages (size: 55x 33x20 cm) were plastic with a wire mesh top and included a circular treadmill (30 cm in diameter and a running wheel area width of 10 cm). Water and food pellets (Altromin® 7024, Altromin GmbH, Lage) were provided ad libitum. The anımal bedding was provided from Altromin GmbH, Lage. To facilitate nest-building, the anımals were also provided with cellulose. We confirm that the experiments have been performed in accordance with local animal welfare legislation and the legal requirements of Germany. Data analysis and statistics The parental care behaviour of five pairs towards their first litter and of one pair towards their third litter was observed over 1600 h. The mean litter size was 5.1 pups (3-7). The behaviour of the parents was registered on days 2, 5, 8, 13, and 20 after the birth of the young (day of birth = day 0), each for 24 h. To obtain control data, we observed all pairs for 24 h without offspring, i.e., 2-3 weeks before birth or after weaning. We used the time-lapse videotechnic (Panasonic WV-CL352E u. AG-7350) and chose the 12 h mode. The analysis was performed using the software The Observer V 3.0 (Noldus, NL). For both pair partners we collected the following behaviours: nest-building (duration): time spent with the carrying-in and arranging of nesting material; nest-residence (duration): time spent in the nest; retrieving of the young (frequency): carrying the pups back to the nest. The parameters are given as mean values, the statistical measure of variance is the standard error. The Friedman analysis of variance and subsequently the two-tailed Wilcoxon test were used to assess the differences of the means. Differences were significant at p< 0.05 (* in the graphs). The computer package used for the statistical analyses was Winstat (V 3.1). Results Nest-building The time the gerbils spent on nest-building depended on whether it was used as a nest for resting (rest-nest) or as a nest for the approaching litter (litter-nest), i.e., the nest had two functions. When the adults lived without young under laboratory conditions, both males and females built a rest-nest as a depression in the animal bedding which was located in a corner of the cage (Fig. 1, males vs. females: Wilcoxon test, z=-0.94, N = 6, p > 0.05). It was only slightly enlarged during the last activity period before and completed just after birth of the pups. This litter-nest was more compact and was also built by both sexes. Nevertheless, regarding the whole observation period, the respective effort of male and female was different (Fig. 1). In the mean a male invested 15.9+7.4 and a female 27.3+4.7 minutes per day (males vs. females: Wilcoxon test, z=-1.99, N=6, p<(0.05). Whereas males did not intensify the nest-building behaviour (day 2-day 20 vs. control: Friedman ANOVA, chi-square approximation, Y =83,N =6,df=5,p > 0.05), it was signi- ficantly elevated in the females until day 13 (day 2-day 20 vs. control: Friedman ANOVA, chi-square approximation, x” = 15.64, N =6, df=5, p<0.05; Wilcoxon test, day 2 vs. con- trol: z=-1.99, N=6, p<0.05; day5 vs. control: z=-1.99, N=6, p<0.05; day 8 vs. con- trol: z=-2.2, N=6, p<(0.05; day 13 vs. control: z=-1.99, N = 6, p<0.05; day 20 vs. con- trol: z= 1.36, N 6,pr 0:09): | Nest-residence Both males and females stayed for approximately the same time alone in the commonly es- tablished nest (Fig. 1). There were no intersexual differences in that respect either in the rest-nest (Wilcoxon test, z=-1.36, N = 6, p > 0.05) or in the litter-nest (Wilcoxon test, z = -0.52, N=6, p>0.05). In the presence of the young this separate nest-residence of the Parental care and time sharing in the Mongolian gerbil Al Nest-building 60 - - Nest-residence h/ day 2 5% 8. 13% 20. control Retrieving of young * * numbers / day 2. 5! 8. 13. 20. Fig. 1. Biparental care in Mongolian gerbils. Data of six pairs. Day 2 to day 20: wıth young (day 0 = day of birth); control: without young; scattered columns = males; dotted columns = females; blank col- umns = both parents together; bars = standard error of mean. * p< 0.05; The Friedman analysis of var- iance and subsequently the two-tailed Wilcoxon test were used to assess the differences of the means (for details, see text). adults increased, i.e., the two parents showed time sharing in the litter-nest. This increase of the sole care for the progeny was always significant for the males (day 2-day 20 vs. con- trol: Friedman ANOVA, chi-square approximation, x’ = 13.38, N=6, df=5,p<0.05; Wil- coxon test, day 2 vs. control: z=-1.99, N =6, p< 0.05; day5 vs. control: z=-1.99, N=6, p<0.05; day8 vs. control: z=-22, N=6, p<(0.05; day 13 vs. control: z=—22, N=6, p< 0.05; day 20 vs. control: z=-1.99, N = 6, p< 0.05). The same applies to the females ex- 172 R. WEINANnDY and R. GATTERMANN cept for day 13 (day 2-day 20 vs. control: Friedman ANOVA, chi-square approximation, x = 17.4, N=6, df=5, p<0.05; Wilcoxon test, day 2 vs. control: z=—2.2,N=6, p < 0.05; day 5 vs. control: z=—2.2, N = 6, p<0.05; day 8 vs. control: z=-2.2, N=6, p< 0.05; day 13 vs. control: z=-1.4, N = 6, p > 0.05; day 20 vs. control: z=-2.2,N =6,p< 0.05). The common nest-residence, 1. e., the time the adults spent together in the nest as a pair, was about four times longer than the separate stay (Fig. 1). However, in contrast to this, it was not affected by the offspring (day 2-day 20 vs. control: Friedman ANOVA, chi-square approximation, Y =11.5, N=6, df=5, p<0.05; Wilcoxon test, day 2 vs. con- trol: z=-0.94, N=6, p>0.05; day5 vs. control: z= -1.57, N=6, p>0.05; day 8 vs. con- trol: z= -1.36, N = 6, p > 0.05; day 13 vs. control: z=-1.57, N=6, p > 0.05; day 20 vs. con- trol: z= -1.78, N = 6, p> 0.05). Summing up the data of the separate and the paired nest-residence results in the total time the adult gerbils spent in the nest. It ranged from 14.4 #+1.2 h per day (control) and 15.4+1.8h per day (mean of day 2-day 20). As for the common nest-residence, the statis- tical analysis showed that there was no difference between these two periods (day 2- day 20 vs. control: Friedman ANOVA chi-square approximation, x = 11.98, N=6, df=5, p < 0.05; Wilcoxon test, day 2 vs. control: z=-1.21, N=6, p> 0.05; day5 vs. control: z = -0.52, N=6, p> 0.05; day8 vs. control: z= 0.37, N=6, p > 0.05; day 3, yszeontrola72 -1.36, N = 6, p > 0.05; day 20 vs. control: z=-1.36, N = 6, p > 0.05). Retrieving of young Until day 5, the progeny were passively dragged out of the nest while attached to the mothers nipples. Nevertheless, to an even greater extent they were thrown out of the nest by digging movements of the adults. With advancing age and increased locomotor activity the pups actively left the nest and were retrieved essentially by females (Fig. 1, males vs. females: Wilcoxon test, day2: z= 22, N =6,p=<0.05; days2z - 222765220105; day&: z=—2.2, N=6, p<0.05; day 13; z=_—2.2,N=6, p<0.05; day 2022- D482N 6, p> 0.05). The mothers grabbed the young in the neck or other parts of the body with their teeth. At day 20 the retrieving behaviour of females ended. The males also tried to retrieve their offspring but they pushed the young back to the nest with their snout. However, they failed in all observed cases. Discussion In nature Mongolian gerbils live under territorial conditions in groups which are estab- lished by a founder pair (BAnnIıKov 1954; AGREN 1984; HENDRIE and STARKEY 1998). In addition to the female, the male and the other family members also participate in pro- moting the development of the offspring (ELwooD 1975; OSTERMEYER and ErwooD 1984). During their first days they have an incompleted ability to thermoregulate and are warmed and sheltered in a nest by the parents in order to stay alive. Paternal behaviour is described also in other rodent species. WoLFF and CIcIRELLO (1991) showed that Pero- myscus maniculatus males retrieved pups and nested with females and newborn pups. In the laboratory adult gerbils built plain nests for common resting (rest-nest). For rearing the altricial young, the nests were enlarged and constructed more compactly mainly by the females (litter-nest). As shown in the golden hamster, the size of the nest and the amount of the female nest-building activity do not depend on the sexual cycle or the state of gravidity but on the environmental temperature (RıcHArDS 1966; BHATIA et al. 1995). Since the room temperature in our experiments was high (23-25 °C), the nest- building activity was relatively low. The increase in the female nest-building behaviour immediately after the birth of the pups is a response to parturition and the presence of Parental care and time sharing in the Mongolian gerbil 173 the young. Besides the temperature, the nest-building behaviour in house mice and other myomorph rodents is intensified according to olfactorical and acoustical (ultrasonic-) sti- muli of the young (No1rort 1972, 1974; SALES and SMITH 1978). Two weeks after birth the coat of the pups is well developed and the reduced relative surface of the body di- minishes the loss of body heat. This shift of the young gerbils from being “heat sinks” to “heat sources” is reflected in a reduction of the nest-building effort following day 13. In contrast to the female-biased nest-building both males and females cared for the progeny in the nest. While one animal stayed in it and warmed the young, the other left it. This ensues from the increased separate nest-residence of the father and the mother after the litter, i.e., there was an intersexual time sharing in the nest. Even in the prairie vole, Microtus ochrogaster, the female does leave the nest more often when the male takes part in the care of the litter (Wang and Novak 1992). Nevertheless, in this case the data were not calculated for timed synchrony. WynnE-EDwarDs (1995) observed the care behaviour of Phodopus campbelli for 30 minutes per day during the activity phase and during rest, respectively. She was also able to prove a temporal synchronisation in the care behaviour between the parents and additionally between mother and sister, 1. e., the aunt of the offspring. The cooling down of the pups is prevented due to this temporal co- ordination of parental behaviour. Furthermore, the mother is able to satisfy her increased need for nutrients following the litter and during lactation (GALEF 1983) and to reduce the physical strain caused by the lactational hyperthermia. This phenomenon of an in- creased core body temperature while in physical contact with the young is described in various small mammals (Apeıs and LEon 1986; SCRIBNER and WYNNE-EDWARDS 1994a, 1994 b). In gerbils the daily mean values of core body temperature during the whole peri- od of lactation are elevated by 0.6 °C (WEINANDY and GATTERMANN 1995). Although the common nest-residence of the pair partners was longer than the sepa- rate ones, it was not influenced by the pups. In accordance to their nocturnal activity pat- tern (WEINANDY and GATTERMANN 1996/97) and their social behaviour gerbils rest to- gether during most of the light phase, irrespective of the presence of young. They were left alone in the nest for about 8.6 h per day, i.e., the pups were not constantly warmed by the parents. Our assumption is that this is a consequence of the relatively high tem- perature conditions in the laboratory. Furthermore, the mutual warming of the young also reduced the loss of heat. Retrieving the offspring is another direct nursing effort, which was in our study only successfully carried out by the female. The observed increase of retrieval behaviour was most likely triggered by ultrasonic vocalisation and the growing locomotor activity of the young, which left the nest more frequently. Furthermore, they were thrown out of the nest due to the species-specific stereotypic digging behaviour of both adults (WIEDEN- MAYER 1997). Nevertheless during the rearing period, the females tended to dig more of- ten (67 to 88 minutes per day; unpubl. obs.). Similar results for this species were obtained by Kapran and HyranD (1972) and they considered this phenomenon an indication of fe- male hyperactivity connected with litter and lactation. In conclusion, in gerbils there is no paternal support in the building of the litter-nest and the retrieving of the young. Both adults cooperate via their synchronised presence with the young (temporal coordination or time sharing in the nest). We suggest that via this mutual behavioural synchronisation the physiological strains of the female caused by lactational hyperthermia are reduced. Acknowledgements We thank AnJA SONNTAG and STEFFEN WACHS for technical assistance and KATE WILLIAMS for correcting the English. The work was supported by the DFG (Ga 434/1-2). 1074 R. WEINAnDY and R. GATTERMANN Zusammenfassung Elterliche Jungenpflege und zeitliche Kooperation bei der Mongolischen Wüstenrennmaus Das elterliche Pflegeverhalten der sozial lebenden Mongolischen Wüstenrennmaus (Meriones unguiculatus) wurde von der Geburt bis zur Entwöhnung der Jungen unter Laborbedingungen quantitativ erfaßt. Nestbau, Nestaufenthalt und das Eintragen der Jungtiere wurden analysiert. Die Registrierung dieser Verhaltensweisen der Elterntiere erfolgte per Videobeobachtung an den Tagen 2,5, 8, 13 und 20 nach der Geburt der Jun- gen (Tag der Geburt = Tag) für jeweils 24 Stunden. Als Kontrolle wurden alle Paare darüber hinaus einmalig für 24 Stunden ohne Nachwuchs beobachtet. Neugeborene Mongolische Wüstenrennmäuse sind typische Nesthocker. Ziel dieser Studie war es, den väterlichen und den mütterlichen Aufwand bei der Jungenaufzucht zu ermitteln. Das Weibchen leistete insgesamt den größten Anteil an der Jungenpflege, da es keine män- nliche Unterstützung beim Bau des Wurfnestes oder beim Eintragen der Jungtiere gab. Dagegen kooperierten beide Elterntiere aufgrund ihrer synchronisierten Anwesenheit bei den Nestlingen miteinander (temporale Koordination). Wir vermuten, daß durch diese wechselseitige Verhaltenssynchronisation die physiologischen Belastungen des Mut- tertieres, beispielsweise verursacht durch die Laktationshyperthermie, reduziert werden können. References ADELS, L. E.; LEon, M. (1986): Thermal control of mother-young contact in Norway rats: Factors me- diating the chronic elevation of maternal temperature. Physiol. Behav. 36, 183-196. ÄGREN, G. (1984): Pair formation in the Mongolian gerbil. Anim. Behav. 32, 528-535. BANNIKOV, A.G. (1954): The places inhabited and natural history of Meriones unguiculatus. In: Mam- mals of the Mongolian People’s Republic. Moscow: USSR Acad. Sciences. Pp. 410-415. BHATIA, A. J.; SCHNEIDER, J. E.; WADE, G. N. (1995): Thermoregulatory and maternal nestbuilding in Syr- ian hamsters. Interaction of ovarian steroids and energy demand. Physiol. Behav. 58, 141-146. Erwoop, R. W. (1975): Paternal and maternal behaviour in the Mongolian gerbil. Anim. Behav. 23, 766-772. ELwooD, R. W. (1979): Maternal and paternal behaviour of the Mongolian gerbil: a correlation study. Behav. Neural Biol. 25, 555-562. GALEF, B. G. jr. (1983): Costs and benefits of mammalian reproduction. In: Symbiosis in Parent-Off- spring Interactions. Ed. by L. A. ROsENBLUM and H. Morrtz. New York: Plenum Press. Pp. 249-278. GERLACH, G. (1998): Impact of social ties on dispersal, reproduction and dominance in feral house mice (Mus musculus domesticus). Ethology 104, 487-499. HENDRIE, C. A.; STARKEY, N. J. (1998): Pair-bond disruption in Mongolian gerbils: effects on subsequent social behaviour. Physiol. Behav. 63, 895-901. KarLan, H.; HYLAND, S. ©. (1972): Behavioural development in the Mongolian gerbil Meriones ungui- culatus. Anim. Behav. 20, 147-154. NoIROT, E. (1972): Ultrasounds and maternal behaviour in small rodents. Dev. Psychobiol. 5, 371-387. NoIRoT, E. (1974): Nest building by the virgin female mouse exposed to ultrasound from inaccessible pups. Anım. Behav. 17, 340-349. ÖSTERMEYER, M. C.; ErwooD, R. W. (1984): Helpers (?) at the nest in the Mongolian gerbil, Meriones unguiculatus. Behaviour 91, 61-77. | RICHARDS, M. P.M. (1966): Activity measured by running wheels and observation during the oestrous cycle, pregnancy and pseudopregnancy in the golden hamster. Anim. Behav. 14, 450-458. SALES, G. D.; SMITH, J. C. (1978): Comparative studies of the ultrasonic calls of infant murid rodents. Dev. Psychobiol. 11, 595-619. SCRIBNER, ST. J.; WYNNE-EDWARDS, K.E. (1994a): Disruption of body temperature and behaviour rhythms during reproduction in dwarf hamsters (Phodopus). Physiol. Behav. 55, 361-369. SCRIBNER, ST. J.; WYNNE-EDWARDS, K.E. (1994 b): Thermal constrains on maternal behaviour during re- production in dwarf hamsters (Phodopus). Physiol. Behav. 55, 897-903. Parental care and time sharing in the Mongolian gerbil 178 SOLOMoON, N. G.; FRENCH, J. A. (1997): Cooperative Breeding in Mammals. Cambridge: Univ. Press. SoLoMoN, N. G.; GETZ, L. L. (1997): Examination of alternative hypothesis for cooperative breeding in rodents. In: Cooperative Breeding in Mammals. Ed. by N. G. SoLomon and J. A. FRENCH. Cam- bridge: Univ. Press. Pp. 199-230. Wang, Z. X.; Novak, M. A. (1992): Influence of social environment on parental behaviour and pup de- velopment of meadow voles Microtus pennsylvanicus and prairie voles M. ochrogaster. J. Com. Psy- chol. 106, 163-171. WEINANDY, R.; GATTERMANN, R. (1995): Measurement of physiological parameters and activity in a Mongolian gerbil during gravidity and lactation with an implanted transmitter. Physiol. Behav. 58, 811-814. WEINANDY, R.; GATTERMANN, R. (1996/97): Time of day and stress response to different stressors in ex- perimental animals. Part II: Mongolian gerbil (Meriones unguiculatus Milne Edwards, 1867). J. Exp. Anim. Sci. 38, 109-122. : WIEDENMAYER, C. (1997): Causation of the ontogenetic development of stereotypic digging in gerbils. Anim. Behav. 53, 461-470. WOLFF, J. O.; CICIRELLO, D. M. (1991): Comparative paternal and infanticidal behavior of sympatric white-footed mice (Peromyscus leucopus noveboracensis) and deermice (Peromyscus maniculatus nubiterrae). Behav. Ecol. 2, 38-45. WYnNE-EDWARDS, K.E. (1995): Biparental care in Djungarian but not Siberian dwarf hamsters (Pho- dopus). Anim. Behav. 50, 1571-1585. Authors’ address: Dr. REnE WeEIınAanDy and Prof. Dr. ROLF GATTERMANN, Institute of Zoology, Martin-Luther-University Halle-Wittenberg, Domplatz 4, D-06108 Halle (Saale), Germany Z. Säugetierkunde 64 (1999) 176-186 = ER © 1999 Urban & Fischer Verlag SÄUGETI ERKÜUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Effects of sex and breeding status on habitat selection by feral House mice (Mus musculus) on a small Mediterranean island By I. ToOrRE and M. BoscH Museu de Granollers-Ciencies Naturals, Granollers and Departament d’Ecologia, Universitat de Barcelona, Barcelona, Spain Receipt of Ms. 03. 04. 1998 Acceptance of Ms. 16. 02. 1999 Abstract Patterns of habitat use of the house mouse (Mus musculus) in relation to sex and breeding status were studied in April and May (the early breeding season) in two structurally different habitats on a small Mediterranean island in NE Spain. Overall mice abundance increased from bare and rocky areas to areas with a dense cover of shrubs and herbaceous plants. Females were associated to shrub areas in April, shifting towards more herbaceous areas in May. Males were less selective, being only slightly related to shrub height in April, and they were not associated with structural habitat features in May. Overlap in habitat use by sexes existed in both months, being more reduced in April than in May. Female densities were significantly higher in the habitat with dense vegetation cover (suitable habitat) than in the habitat with scarce vegetation in both months, while densities of males were not. Male density decreased from April to May in the suitable habitat, and intersexual competition was ex- clusively detected in the period of higher male density. In this habitat, intraspecific competition ex- plained the spatial distribution of sexes regardless of habitat structure characteristics. In absence of competition the spatial distribution of sexes was mainly related to habitat structure. Females and males started sexual activity early in the season in the suitable habitat. Weight of fe- males was higher in suitable habitat in both months, also showing a positive association with the her- baceous cover, and the average weights of males and females at trapping stations were positively asso- ciated. Our results are finally discussed in relation to the social organisation models proposed for house mouse populations. Key words: Mus musculus, island, habitat selection, sex, breeding status Introduction The house mouse is a widespread species, living in mainland Europe as well as on Medi- terranean islands (Orsını et al. 1983; Amorı et al. 1984). Northern populations are mainly comensal to human settlements but live outdoors for a great part of the year (CARLSEN 1993). In Southern Europe, however, feral populations are found throughout the year, e.g. in Mediterranean habitats (CAassaınGg and CroseEr 1985; CacnIn et al. 1996). Also Iberian populations are mainly linked to human dwellings (SAns-ComA et al. 1987; GosSALBEZ 1987) but some local and well-established feral populations are present in moist habitats (GosALBEZ 1987). Some differences in habitat selection are related to in- terspecific competition, of which this species seems especially sensitive (BoITAnNI et al. 0044-3468/99/64/03 — 176 $ 12.00/0 Habitat selection by feral Mus musculus 177 1985; FAIRLEY and SMAL 1987). Additionally, its distribution is conditioned in insular habi- tats by competition with other rodent species rather than by habitat structure (DUESER and PoRTER 1986). | Mammal communities on islands differ in some ecological aspects from those on the mainland (BLonperL 1986). These differences are mainly related to the degree of isola- tion, which is a problem to the colonisation of non-volant mammals from the mainland. They are also related to the surface and the size of the island, where small areas are not to be colonized by medium- or large-sized mammals, i. e., carnivores (BLoNDEL 1986). As a consequence, smaller islands normally are poor in species, and small mammal commu- nities living on islands have special features of habitat use because of reduced predation pressure and interspecific competition (CROWELL 1983). In this study we investigated the physical cues that may influence habitat use of a fer- al house mouse population on a small archipelago uninhabited both by humans and also by other rodents. Our objectives are to analyse spatial distribution of the species at a cer- tain time in relation to habitat structure, sex, and breeding status, and to provide some in- formation on the social organisation of house mice in insular habitats. Material and methods The study was performed on the Medes Islands (42°0'N, 3°13’E, NE Spain) during spring 1996. These islands are a small calcareous archipelago only 0.9 km off the coast. The vegetation of the archipelago is dominated by nitrophylous communities linked to the presence of one of the largest breeding colo- nies of yellow-legged gulls (Larus cachinnans) in the Mediterranean (Bosch et al. 1994). Three main habitats differing in vegetation features are distinguished within the islands: (1) shrubby habitat, dominated by Artriplex halimus, a dense shrub which reaches 70-100 cm height; (2) grassy habitat, dominated by grassy, ruderal plants, such as Hordeum murinum; and (3) bare habitat, with very scarce vegetation, bare ground and dispersed rocks (see BoscH and Sor 1998). The archipelago was transito- rıly occupied by humans until 1923, being deserted for the last 70 years. The small mammal commu- nity of these islands is composed of house mice (Mus musculus) and white-toothed shrews (Crocidura russula) (GOSALBEZ et al. 1984). Two plots of 49 and 25 Sherman live traps (i.e. 7 rowsx 7 columns of traps, and 5 rowsx 5 col- umns of traps, respectively, equidistance between traps 16.6 m) were set during three consecutive days from 31 March to 2 April (first session), and from 26 to 28 May (second session) on the largest island of the archipelago (Meda Gran, 18.2 ha). The study was conducted during the early breeding season of the house mouse which is described to last from spring to late summer on the Medes islands (Go- SALBEZ et al. 1984). Traps were baited with a mixture of tuna fish in olive oil and flour to allow in- creasing trappability, since low trappability might explain low recapture rates (Kreps et al. 1994). Trapping effort for each trapping session was 222 trapnights/session. The large plot was 1 ha in area and included the shrubby and grassy habitats, while the small one was 0.5 ha and only included the bare habitat. The trapping plots were examined early in the morning and the animals found were identified, weighed, sexed, examined for reproductive condition and marked by toe-clipping (GURNELL and FLOWERDEW 1990), and released at the trap station. To allow comparisons between plots and months, population densities were estimated as the average number of individuals caught per trapping station during the three consecutive days. The habitat structure was characterized at each trap station at the same time when trapping was conducted, by means of estimating values of height and cover on a 5m radius circular plot centred around the Sherman trap (ALCANTARA and TELLERIA 1991). Two factorial analyses (BHATTACHARYYA 1981) were performed (one per month) with the habitat structure variables of all traps to obtain independent multivariate factors considered as gradients to which the frequencies of occurrence of the small mammals refer. To ascertain preferences of the house mouse spatial distribution, the frequency of captures at each trap station was considered as a relative measure of density in the surrounding habitat (DUESER and HALLETT 1980), and then was related to the habitat structure variables by means of non-para- 178 I. TORRE and M. BoscH metric Spearman correlation analysıs. To test for intersexual competition and its influence on habitat use, we used the method described by HALLETT and Pımm (1979). The unweighed average situations of the sexes on the factorial space were obtained by averaging the values of the factor scores of the trap- ping stations with captures on the factors extracted. To ascertain the habitat variables that best ex- plained the abundance of the house mice, stepwise multiple regression analysis was performed, with the frequencies of occurrence as dependent variables and the habitat variables as the independent ones (YAHNER 1982). To avoid autocorrelation in habitat variables, multiple regression analyses were also performed with factors as the independent variables, and the Bonferroni correction was applied when necessary to maintain &@< 0.05 (Rıce 1989). Microhabitat characterization of the house mouse feral population was estimated as the average values of the habitat variables at the trap stations where the species or sexes were trapped (selected areas). These values were compared with the average values of the habitat variables at the trap sta- tions where the species or sexes were not trapped (non selected areas), using the Kruskal-Wallis ANOVA. When possible, these tests were also used to verify sexual and temporal habitat preferences. Mann-Whitney U-tests were performed to ascertain differences in habitat preferences between sexes, and for the same sex between different trapping sessions. To simplify the statistical analysis, overlap in habitat use by sexes was estimated as the z-value obtained when testing for differences between aver- age values of both sexes on each of the multivariate factors extracted (using Mann Whitney U-test). The greater the z-value the smaller is the overlap. Possible differences in the relative abundances of house mice between plots, months or sexes were tested by Chi-square analysis (with the Yates’ correction for continuity) on standardized trapping areas. The differences between the two sampling periods in habitat structure as well as in house mouse variables at the same trapping stations were tested with the Wilcoxon signed-rank test for matched pairs. Before parametric statistical treatment, variables were log (x + 1) and arcsine transformed (ZAR 1996). Results Habitat structure The two habitats sampled were structurally different in both periods (Tab. 1). Plot 1 was characterized by higher values of vegetation cover, and plot 2 by higher values for the slope and rock cover. Monthly variation of habitat structure was only observed in the her- baceous cover and height. The factorial analysis performed with the structural variables yielded similar results in both periods. In April, two eigenvectors were extracted, explaining altogether 75.5% of the structural habitat variance (Tab. 2). The first factor was positively correlated to the rock and dead vegetation cover and to the slope. It was negatively related to shrub and herbaceous cover, and height of shrubs and herbaceous plants. This factor was inter- preted as the negative effects of the increasing slope on the establishment of vegetation strata. The second factor was positively related to shrub cover, shrub height, and slope, but negatively related to herbaceous height. Since shrub cover and height had positive loadings, and herbaceous cover had a negative loading, this factor was interpreted as a negative effect of the shrub plants on the development of the herbaceous plants. The fac- torial analysis conducted in May yielded similar results (Tab. 2). House mouse abundance in relation to sex and breeding status 61 individuals of Mus musculus (37 males, 24 females) were trapped in April, and 46 indi- viduals (22 males, 24 females) were trapped in May (Tab. 3). The relative abundance of M. musculus was greater in plot 1 than in plot 2 for both periods (April: x” = 7.88, p < 0.01, d.f.=1; May: x” = 7.54, p< 0.01, d.f.=1). Male density decreased from April to May (x° = 3.90, p < 0.05, d.f.= 1), while female density remained 179 Habitat selection by feral Mus musculus cgy0 020 191 070 80'0 el’ xt e xrxrx66 € EV sL’I 8sL’I c6 0 Kew-1udvy xrel ce 180 x» TEC Lv IE T 080 KkkkSV 7 rk 00 9 x E sco 680 0 0 dpuez I 3014 SYUIOA *L6 | 010 *06 1 Ss E00 x VE C xl € Krk 9 xrel V er )0 P sr O P x 317€ 9 rk [PS dpuez 159-N 60°T + 0091 rL'O + 6891 eier Se 610 + 950 rl0+0r0 600 + 910 LING rS’e+6Sr7 II’E FOILE l6°E +09'S ITO+9T'O ser oc 96T +CSsı (ST=u) ZIOld AeW 69°0 # 6881 r80+ OLlI SIT +S0'0T rlO+Ss80 LO'O + CEO OLOFESO tr I+rLTo 66°C + LEE 0S’T + 65°65 c6L+09LL rL'E + TON gET+SHL 960 + Sch or I SE) 3 ST 61°C L6°0 vLI srl S er CI rel 9 xrrr|/ TV xrrr8L TV rk 8 G xrrrV70 7 dpuez 189I-N) 18°0 + 00'SI SsH0+Er'SsI O0 rl0+7Ss'0 1I’0O +0r'0 9O0FTIO LEE FSOTE IST + 06°€ EL E- Vi SCH +SEs ST0+1IrO 977 +80 ch OLE ES (ST =u) ZIOId udvy SEO F+FSEI WOFSLSI LSO+09L1 9TO+FrII cLOFLIO 0LOFcrO GE H@6I6 661 F Sr Le ce + c6 99 SESHSSLL OS ar GE gET+SOL IH0FSTH [e101 sopew sopewsfF (3) 1ysıam Apog [eo] sopeuu sopewpf Aysuaq (%) uoneJ93>4 peaq (u) Jysıay sSno99egIaH (%) 9409 SNODEqIIH (u) sqnıys Jo JysıaH (%) 19409 qnIyS (%) I9A09 PoN (%) 2dojs gene‘ °((T000°0 > A) xxx (T00'0 > I) xxx “(TOO > I) xx ‘(S0'0 > I) „ :a9ueaLpusıs Jo [9AO]) SIS9J UOXOIJLM YIIM sporsad ZJurdures u99MJOgq SS9UFIIJFJIP pue ‘sIs2I-N AvuyyM uueM YNM P9]S9} sIOoJd usO9MJogq SS9UDISFJIA 'SIYSIOM pue sonIsusp 9snouL I9snoy IyJ IOJ SInJeA '9'S pue sdeısae pue ‘(Aem pue [ıdWY) sporıod yyoq ur 0oMJ pue Juo sjoJd ur suoners Jurdden je psınseau (10119 pIepues F uB9UL) sOfgelieA JeınJonnS *T afgeL 180 I. TORRE and M. BoscH Table 2. Factorial analysis performed with the habitat structure variables in both months, and level of significance of the correlations between variables and factors (see Tab. 1). Variable April Factor 1 Factor 2 Factor 1 Factor 2 Slope 068 USE 030 Rock cover OS 089.2 0.07 Shrub cover 0.03 **** —0.02**** 0.753233 Shrub height NR —0.08**** 0.032 Herbaceous cover 0.83 + *** -0.56**** 0.70**** Herbaceous height 0.88**** 0.88**** 0.14 Dead vegetation cover OO —0.49#*** 0.19 Eigenvalue 4.00 3.42 1.62 % Variance SU 48.9 | Ac. % Variance SD. 48.9 TAN Table 3. Number of house mouse individuals trapped in relation to sex (n), frequency of recapture (FR) within the same sampling period and breeding status (BS: frequencies of active males with scro- tal testes and pregnant females) in the two plots (Pl and P2) and sampling periods (April and May). males 16 2 2 66.6 83.3 females 0 0 20 30 80.9 0 2 Table 4. Spearman non-parametric correlation matrix between house mouse relative abundances and habitat structural variables in both months (significance levels as in Tab. 1). F1 = Factor 1; F2 = Factor 2; Rc= Rock cover; Sc=Shrub cover; Sh = Shrub height; Hh = Herbaceous height; only significant correlations (p< 0.007, Bonferroni correction) are shown. Slope, dead vegetation, and herbaceous cover with no significant correlations. Levels of significance as in Tab. 1. females April -0.30** 080, -0.31** 0.44#** 032 May 0.29** males April 0.05* May total April 0.377 FE 7 EOAIEEEE 0.,46%°+2, 041 0,3622 May total April+ May -0.35** 039252 0.48**** VS 0362= the same (£&° = 0.01, p > 0.05, d.f.= 1). Considering the captures on both plots, the sex-ra- tio was biased towards males in April (males:females, 1.7:1; &= = 4.22, p< 0.09, dt DB) but not in May (males:females 1.1:1; &°=2.7, p > 0.05, d.f.=1). Both males and females attained sexual activity earlier in plot 1 (Tab. 3). Later the number of active males decreased at the same time when females become pregnant. In plot 2, females could be considered transient since no recaptures were obtained. Sexually active males increased later in plot 2, and the high recapture rate obtained might be con- sidered as a degree of site-attachment. 181 Habitat selection by feral Mus musculus General patterns of abundance in relation to habitat structure M. musculus showed clear patterns of relative abundance in relation to habitat structure in April, but a more vague pattern in May (Tabs. 4 and 5). The habitat structure variables that explained most of the presence of the species were the height of shrubs in April and the herbaceous height in May. Pooling capture data from both months, M. musculus showed significant preferences for traps with higher values for height and cover of shrubs @= 10.9052n47, 0.88, respectively, n’=7, p< 0:05; Eig:l). Table 5. Results of the stepwise multiple regression analysis performed with the habitat structure variables or factors as independent variables, and house mouse relative abundance as dependent vari- ables, showing the first variables selected and the percentage of variance explained by the models. Levels of significance as in Tab. 1. April May April + May Variable selected Dependent variable males - Shrub height Factor 1 Shrub cover females Factor 2 Shrub height total R’andp 0 0.05* 0.17°%** 0.18+*** 0.23**** 0.22**** Variable selected Factor 1 Herbaceous height Herbaceous R2 andp 0 0 0.06** 0207, 0 0.06** Variable selected Factor 2 Shrub cover Factor 1 Shrub cover Factor 1 Shrub height R’andp 0.06** 0.05* 0.23 + 0.217*** 0.23% *** height 120 100 80 60 SHRUB HEIGHT (cm) 40 20 0 1 2 3 a 5 6 7. NUMBER OF MUS MUSCULUS AT TRAPPING STATIONS Fig. 1. Frequencies of capture of Mus musculus at trapping stations in relation to average (+ standard error) of shrub height. Numbers are sample sizes for each category. 182 I. TORRE and M. BoscH Microhabitat preferences in relation to sex and month A) April: Females showed marked microhabitat preferences, while males did not. Fe- males showed significant correlations with both factors; their abundance increased along factor 2 and decreased along factor 1 (Tab. 4), meaning that areas with higher vegetation cover and height were selected and rock covered areas were avoided. The shrub cover ex- plained the greater amount of variance (19%) in female correlations with the structural varlables, followed by height of shrubs (10%), rock cover (9%) and herbaceous height (8%). Males did not show any correlation with both factors and only one correlation with the structural varıables (Tabs. 4 and 5). Habitat used by both sexes overlapped in both factors (values for overlap on Fl and F2 were 0.91 and 1.18, respectively), but females were more selective. Furthermore, the centroids of males and females on both factors dif- fered significantly from the centroids of the non-capture sites (Tab. 6), suggesting the avoidance of a part of the habitat available. The distance of the factorial space defined by factors 1 and 2 from male to female centroids was 0.46. Table 6. Situation (average + standard error) of Mus musculus (males, females and both sexes pooled) on the structural factors extracted by the factorial analyses performed in both months, and average situation of non-capture sites. Asteriscs show significant differences between mean values of the species or sexes and the non-capture values on both factors (differences tested with U-Mann- Whitney tests, and level of significance as in Tab. 1). Variables Factor 1 Factor 2 Factor 1 Factor 2 males 0.16 + 0.18* VEN 0.10 # 0.20 0.34 +#0.18 (m 53) m=51) (n=22) (M=22) females 0.49 #. 0.17** 050-202, 0.42 + 0.16** 0.12 + 0.24 (n=18) (mM=18) (m 22) (n=22) total -0.25 + 0.15** DIS 0.21 # 0.14* 0.06 + 0.17 (n = 40) (n = 40) (m—32) (m 59) Non-capture sites 0.34 # 0.16 -0.41 #0.15 025.007 -0.19 # 0.15 (m=32) (n = 32) (n = 38) (n = 38) The average weight of females was higher in plot 1 (Tab. 1), and was positively corre- lated to the herbaceous cover (r, = 0.52, n=18, p<0.05). Thus, heavier females were found on areas with higher herbaceous cover. The average weight of males did not show any relation with the habitat structure variables. Otherwise, a significant and positive cor- relation was found between the average weight of both sexes at trapping stations (7.1069, 7108 =20:05)): B) May: The variance explained by the stepwise regression models performed in May with the house mouse relative abundances and the habitat structure variables or factors was derived from the presence of females alone (Tab. 5). On the other hand, the number and significance of correlations between relative abundance and structural variables de- creased (Tab. 4). The herbaceous height explained the greater amount of variance in fe- male abundance (12%), followed by rock (7%) and shrub cover (6%). Females corre- lated negatively with factor 1, but no significant correlation was found with factor 2. Males did not show any correlation with factors or structural variables (Tab. 4). Habitat used by both sexes overlapped in both factors (1.13 on Fl and 0.10 on F2), but the dis- tance on the factorial space defined by factors 1 and 2 from male to female centroids (0.37) was nearly the same as the distance observed in April (0.46). From April to May, habitat used by females shifted along factor two (0.44 units), and slightly along factor 1 Habitat selection by feral Mus musculus 183 (0.07 units). Since factor two is negatively correlated to the herbaceous cover, the in- crease in female’s mean values for this factor can be interpreted as a displacement to- wards more herbaceous areas. The shift of males was moderate (only 0.13 units for the factorial space). The average weight of both sexes did not show any relation with the habitat structure variables, but the average weight of females in plot 1 was higher than in plot 2, as oc- curred in April. Intersexual competition and habitat use Negative male-female interactions were detected in April on plot 1, and symmetrical in- tersexual competition explained the spatial distribution of sexes regardless of habitat structure. In April on plot 2 no male-female interactions were detected, and the spatial distribution of sexes was related mainly to habitat structure characteristics (Tab. 7). The same occurred in May on plot 1, and on plot 2: neither intersexual competition nor habi- tat structure influenced the spatial distribution of sexes. Table 7. Multiple regression analysis performed in April and May with the frequencies of occurence of one sex as the dependent variable and both factors extracted and the frequencies of occurence of the other sex as independent variables. The partial regression coefficients, t-values and levels of signif- icance are shown (see Tab. 1). Plot Dependent variable Variables selected Coefficient Female Male Male Female Female factor 1 Male factor 1 factor 2 Female factor 2 Male factor 1 Female - Male - Discussion On the Medes Islands the house mouse showed an increasing pattern of its relative abun- dance from bare and rocky areas to areas covered by shrubby and grassy vegetation. These results are in agreement with the pattern observed in another insular population of Mus musculus (DUEsER and PoRTER 1986). M. musculus is sensitive to interspecific com- petition (FAIRLEY and SMAL 1987), since its mainland distribution seems to be restricted by the presence of some rodent species which in sympatry exclude M. musculus from na- tural environments (Boıtant et al. 1985; AurFFRAY et al. 1990). Insular populations of M. musculus are also sensitive to interspecific competition (DUESER and PORTER 1986), and the absence of competitors from the Medes Islands could allow M. musculus to in- habit natural xeric environments, as reported for feral mainland populations in absence of M. spretus (Orsını et al. 1982; AUFFRAY et al. 1990; CAcnin et al. 1996). The pattern of house mice abundance varied when considering the sampling month, the habitat sampled, and the sex of the individuals trapped. Females density was higher in the plot with higher vegetation cover (suitable habitat). Since females start sexual ac- tivity in early spring (GoOSALBEZ et al. 1984), the higher male density in April might be a 184 I. ToORRE and M. BoscH consequence of competition for mating with sexually active females, and the lower den- sity in May as a result of the decreasing number of potential mating partners, with most of the females being pregnant or lactating. Female recapture rates increased as the breed- ing season progressed on a suitable plot (suggesting a degree of female site-attachment inherent to pregnancy or lactation, KrEBs et al. 1994), while male recapture rates de- creased at the same time, suggesting a contrary pattern with a greater mobility. The in- creasing number of active males on the non-suitable plot late in the season could be inter- preted as the displacement of active males from suitable to non-suitable habitats in search of sexually active females. Our results agree with the general pattern of habitat use found in other small mam- mals, with females selecting microhabitats that provide greater protective cover (SEAGLE 1985). They tend to shift towards more herbaceous-covered areas as the season pro- gresses (BELK et al. 1988). Males were competing for breeding females (KreBs et al. 1995), and they showed a more reduced habitat selectivity (BELK et al. 1988). This may result from a direct consequence of their greater mobility or from an indirect conse- quence of their association with females. Different habitat utilisation by sexes seems likely to exist with the consequence to decrease intraspecific competitive pressure on re- productive females (SEAGLE 1985). BowErs and SMITH (1979) documented a case for Pe- romyscus maniculatus in which such a segregation was a result of female dominance over males due to larger body size being a way to maximize reproductive effort. In spite of a general absence of sexual dimorphism concerning body size (GOSALBEZ et al. 1984), fe- male house mice were heavier (probably caused by pregnancy) than males throughout the study period on the more suitable plot, and females trapped on this plot were heavier than females trapped on the other plot, regardless time of sampling. The positive relation- ship between average weight of males and females at trapping stations suggested a hier- archical displacement of subordinates to unfavourable microhabitats by dominant indi- viduals, as has also been reported in laboratory studies (REIMER and PETRASs 1967). Finally, the characteristics of the house mouse population studied seems to be in agreement with the social organisation model proposed by NEwsoME (1969) and sup- ported by Kress et al. (1995), with feral house mouse populations not being territorial but showing social dominance through body size. Dominant females may aggregate in 'high resource quality areas, as has been reported for wood mice (MONTGOMERY et al. 1991), and energetic advantages for these females could arise as a result of habitat selec- tion, reducing predation risk (Price and Brown 1983), increasing foraging efficiency (Thompson 1982), or may be living under more favourable microclimatic conditions (WALSBERG 1985). Acknowledgements We thank A. ARRIZABALAGA for his technical support and advice throughout the study. J. L. TELLA, G. BLanco, J. L. TELLERIA and two anonymous referees kindly revised the manuscript; A. VILLARROYA helped with statistics, and M. ZABALA, J. M. SERRANO, Port Autönom de Barcelona and the Servei de Vigilancia de les Illes Medes for their help in the field work. Zusammenfassung Auswirkungen von Geschlecht und Fortpflanzungsstatus auf Habitatwahl bei freilebenden Hausmäusen (Mus musculus) auf einer kleinen Mittelmeerinsel Das Muster der Lebensraumnutzung von freilebenden Hausmäusen (Mus musculus) in bezug auf Geschlecht und Fortpflanzungsstatus wurde während des Frühjahrs (April und Mai, d.h. zu Beginn Habitat selection by feral Mus musculus 185 der Fortpflanzungsperiode) in zwei strukturell verschiedenen Lebensräumen auf einer kleinen Mittel- meerinsel im NE Spaniens untersucht. Im Allgemeinen zeigte die Zunahme in der Abundanz der Hausmaus von felsigen und pflanzenlosen Gebieten zu Bereichen mit einer dichten Deckung durch Sträucher und krautige Pflanzen eine deutliche Bevorzugung bestimmter Habitatstrukturen. Betrach- tet man die Geschlechter getrennt, so ergab sich während beider Untersuchungsperioden, daß die Weibchen deutlich wählerischer waren. Eine schrittweise multiple Regressionsanalyse zeigte, daß die Weibchen im April buschartige Gebiete bevorzugten und im Mai zu krautigen Gebieten wechselten. Die Verbreitung der Männchen zeigte im April eine schwache Beziehung zur Höhe der Pflanzen, im Mai war das Vorkommen dagegen weitgehend unabhängig von den Eigenschaften des Lebensraumes. Während beider Monate war in beiden Geschlechtern eine Überlappung in bezug auf den Lebens- raum zu beobachten, die im April geringer ausgeprägt war als im Mai. In beiden Monaten war die Dichte der Weibchen in pflanzenreichen Gebieten deutlich höher als in Gebieten mit spärlichem Pflanzenwuchs. Die Siedlungsdichte der Männchen war dagegen in beiden Gebieten gleich, und ihre Dichte nahm von April bis Mai ab. Konkurrenz innerhalb der Art (Männ- chen-Weibchen-Wechselwirkung) wurde auschließlich in dem Monat mit höherer männlicher Dichte beobachtet. In diesem Lebensraum war die Konkurrenz durch die räumliche Verteilung der Ge- schlechter bestimmt und weitgehend unabhängig von den strukturellen Eigenschaften der Gebiete. Ohne diese Konkurrenz war die räumliche Verteilung der Geschlechter hauptsächlich auf die Lebens- raumstruktur bezogen. Im günstigen Lebensraum begannen Weibchen und Männchen ihre sexuelle Aktivität zu Anfang der Fortpflanzungsperiode. Während beider Monate lag das Gewicht der Weibchen in den günstigen Lebensräumen höher als in den ungünstigen und zeigte darüber hinaus einen positiven Zusammen- hang mit der pflanzlichen Bedeckung des Gebietes. An den Fangstellen war das durchschnittliche Ge- wicht der Männchen positiv mit dem der Weibchen korreliert. References ALCANTARA, M.; TELLERIA, J. L. (1991): Habitat selection of the Wood mouse (Apodemus sylvaticus) in cereal steppes of Central Spain. Z. Säugetierkunde 56, 347-351. AMORI, G.; CRISTALDI, M.; ConToLı, L. (1984): Sui Roditori (Gliridae, Arvicolidae, Muridae) dell’Italia peninsulare ed insulare in rapporto all’ambiente bioclimatico mediterraneo. Anımalia 11, 217-269. ÄUFFRAY, J. C.; TCHERNOV, E.; BONHOMME, F.; HEITH, G.; SIMSON, S.; NEvo, E. (1990): Presence and ecolo- gical distribution of Mus “spretoides” and Mus musculus domesticus in Israel. Circum-Mediterra- nean vicariance in the genus Mus. Z. Säugetierkunde 55, 1-10. BHATTACHARYYA, H. (1981). Theory and methods of factor analysis and principal components. In: The use of multivariate statistics in studies of wildlife habitat. Ed. by D. E. CApEn Vermont: USDA For- est Service. Pp. 72-79. BELK, M. C.; DUANE SMITH, H.; LAwson, J. (1988): Use and partitioning of montane habitat by small mammals. J. Mammalogy 69, 688-695. BLONDEL, J. (1986): Biog&ographie Evolutive. Masson: Paris. BoOITANI, L.; Loy, A.; MOLINARI, P. (1985): Temporal and spatial displacement of two sympatric rodents (Apodemus sylvaticus and Mus musculus) in a Mediterranean coastal habitat. Oikos 45, 246-252. BoscH, M.; PEDROCCHI, V.; GONZALEZ-SOLIS, J.; JOVER, L. (1994): Densidad y distribuciön de los nidos de la gaviota patiamarilla Larus cachinnans en las Islas Medes. Efectos asociados al häbitat y al des- caste. Donana, Acta Vert. 21, 39-51. Bosch, M.; Sor, D. (1998). Habitat selection and breeding success in yellow-legged gulls Larus cachin- nans. Ibis 140, 415421. Bowers, M. A.; DUANE SMITH, H. (1979): Differential habitat utilization by sexes of the deermouse, Peromyscus maniculatus. Ecology 60, 869-875. CAGNIN, M; ALOISE, G.; GAROFALO, G.; MILAZZO, C.; CRISTALDI, M. (1996): Les communautes de petits mammiferes terrestres de trois “fiumare” de la Calabre (Italie du sud). Vie Milieu 46, 319-326. CARLSEN, M. (1993): Migration of Mus musculus in Danish farmland. Z. Säugetierkunde 58, 172-180. CASSAING, J.; CROSET, H. (1985): Organisation spatiale, competition et dynamique des populations sau- vages de Souris (Mus spretus Lataste et Mus musculus domesticus Rutty) du Midi de la France. Z. Säugetierkunde 50, 271-284. 186 I. TORRE and M. BoscH CROWELL, K.L. (1983): Islands - Insight or artifact?: Population dynamics and habitat utilization in in- sular rodents. O1Kos 41, 442-454. Dueser, R. D.; HALLETT, J. G. (1980): Competition and habitat selection in a forest-floor small mammal fauna. Oikos 35, 293-297. DUESER, R. D.; PORTER, J. H. (1986): Habitat use by insular small mammals: relative effects of competi- tion and habitat structure. Ecology 67, 195-201. FAIRLEY, J. S.; SMAL, C. M. (1987): Feral house mice in Ireland. Ir. Nat. J. 22, 284-290. GOSÄLBEZ, J. (1987): Insectivors i rosegadors de Catalunya. Ketres: Barcelona. GOSÄLBEZ, J.; SANS-CoMA, V.; LÖPEZ-FUSTER, M. J. (1984): Els micromamifers de les Illes Medes. In: Els sistemes naturals de les illes Medes. Ed. by J. Ros, I. OLIVELLA, and J. M. GiLı. Barcelona: Institut d’Estudis Catalans. Pp. 307-319. GURNELL, J.; FLOWERDEW, J. R. (1990): Live Trapping Small Mammals. A Practical Guide. Pub. Mam- mal. Soc. 3, 1-39. HALLETT, J. G.; Pımm, S. L. (1979): Direct estimation of competition. Am. Nat. 113, 593-600. KREBS, C. J.; SINGLETON, G. R.; KENNEY, A. J. (1994): Six reasons why feral house mouse populations might have low recapture rates. Wildl. Res. 21, 559-567. KREBS, C. J.; KENNEY, A. J.; SINGLETON, G. R. (1995): Movements of feral house mice in agricultural landscapes. Aust. J. Zool. 43, 293-302. MONTGOMERY, W. I.; WiıLson, W. L.; HAMILTON, R.; MCCARTNEY, P. (1991): Dispersion in the wood mouse, Apodemus sylvaticus: variable resources in time and space. J. Anim. Ecol. 60, 179-192. NEWSOME, A.E. (1969): A population study of house-mice permanently inhabiting a reed-bed in South Australia. J. Anim. Ecol. 38, 361-377. ORsinı, P.; BONHOMME, F.; BRITTON DAVIDIAN, J.; CROSET, H.; GERASIMOY, S.; THALER, L. (1983): Le com- plexe d’especes du genre Mus en Europe Central et Orientale. Z. Säugetierkunde 48, 86-95. ORSINI, P.; CASSAING, J.; DUPLANTIER, J. M.; CrosET, H. (1982): Premieres donne&es sur l’ecologie des po- pulations naturelles de Souris, Mus spretus Lataste et Mus musculus domesticus Rutty, dans le Midi de la France. Rev. Ecol. (Terre Vie) 36, 322-336. PRICE, M. V.; BRown, J. H. (1983): Patterns of morphology and resource use in North American desert rodents communities. Great Basıin Nat. Mem. 7, 117-134. REIMER, J. D.; PETRAS, M.L. (1967): Breeding structure of the House mouse, Mus musculus, in a popula- tion cage. J. Mammalogy 48, 85-99. Rice, W. R. (1989): Analyzing tables of statistical tests. Evolution 43, 223-225. SANS-CoMA, V.; VARGAS, J. M.; AnTÜNEZ, A. (1987): Micromamiferos (Insectivoros y Roedores) del S- SE de la Peninsula ib£rica: una revision faunistica y taxonömico-sistemätica. In: Mamiferos y Hel- mintos. Ed. by V. SANS-CoMA, S. MAs-CoMA, and J. GosSÄLBEZ. Barcelona: Ketres. Pp. 11-32. SEAGLE, S. W. (1985): Patterns of small mammal microhabitat utilization in cedar glade and deciduous forest habitats. J. Mammalogy 66, 22-35. THoMPpson, S. D. (1982): Structure and species composition of desert heteromyid rodent species assem- blages: effects of a simple habitat manipulation. Ecology 63, 1313-1321. WALSBERG, G. E. (1985): Physiological consequences of microhabitat selection. In: Habitat Selection in Birds. ed. by M. L. Copy. New York: Academic Press. Pp. 389410. YAHNER, R.H. (1982): Microhabitat use by small mammals in farmstead shelterbelts. J. Mammalogy 63, 440-445. ZAR, J. H. (1996): Biostatistical analysis. New Jersey: Prentice Hall, Englewood Cliffs. Authors’ addresses: Isnacıo TORRE, Museu de Granollers-Ciencies Naturals, Francesc Maciä 51, E- 08400 Granollers, Barcelona, Marc Bosch, Departament d’Ecologia, Universitat de Barcelona, Avda. Diagonal 645, E-08028 Barcelona, Spain. e Z. Säugetierkunde 64 (1999) 187-191 ZEITSCHRIFT FÜR © 1999 Urban & Fischer Verlag SÄUGETI ERKUN DE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY WISSENSCHAFTLICHE KURZMITTEILUNG Foraging behavior of the Indian short-nosed fruit bat Cynopterus sphinx By N. GOPUKUMAR NAIR, V. ELANGOVAN, K. SRIPATHI, G. MARIMUTHU, and R. SUBBARAJ Department of Animal Behaviour and Physiology, School of Biological Sciences, Madurai Kamaraj Uni- versity, Madurai, India Receipt of Ms. 18. 06. 1998 Acceptance of Ms. 14. 12. 1998 Key words: Cynopterus sphinx, frugivory, foraging, telemetry The Indian short-nosed fruit bat Cynopterus sphinx, one among the old world bats (Pter- opodidae), inhabiting the tropics, roosts solitarily or in small groups in the foliage (BHArT and Kunz 1995). C. sphinx exhibits elaborate tent-roosting behaviour (BALASINGH et al. 1995). The data collected to date on the foraging behaviour of C. sphinx are more biased to male individuals since these tent-roosting harem males exhibit several behavioural re- pertoires (BALASINGH et al. 1993; MARIMUTHU et al. 1998). We carried out radio-telemetry studies on the foraging behaviour of male and female C. sphinx from the same habitat. The study area, Madurai Kamaraj University campus and adjoining areas in Madurai, South India (lat 9°58’ N; long 78°10’ E) is surrounded by tall trees including Polyalthia longifolia, P. pendula, Azadirachta indica, Ficus bengalensis, F. religiosa, F. benjamina, Bassia latifolia, Cocos nucifera, Caryota urens, Borassus flabellifer, and Mimusops elengi. Five males and five females were radio-tagged. The foraging studies were carried out for 640 night hours (for 81 nights). Male C. sphinx (47 + 3g) chew and clip the twigs of the interior of the foliage of trees such as P. /ongifolia, B. flabellifer, and C. nucifera and thus make tent-roosts. Bats were captured from their tent roosts using mist nets (Avinet-Dry- den, NY. 13053-1103, U.S.A.). Their body mass was recorded to the nearest 1.0 mg with a spring balance (Avinet-Dryden, NY. 13053-1103, U.S.A.) Length of forearm was mea- sured to the nearest 0.1 mm with the help of vernier calipers. The radiotracking studies were conducted between May 1997 and March 1998. Each bat was fitted with a radio- transmitter (2.6 g) covering a range of 400-500 m, mounted over an aluminium collar cov- ered by a light reflective tape. The transmitter along with the collar weighed only 5.5 % of the average body weight of C. sphinx. We used two sets of receivers and collapsible 5- element Yagi antennae (Customs Electronics, Urbana, Illinois, U.S.A.). Radiolocations were triangulated from three tracking units. Bearings were taken in as short a time inter- val as possible and locations with a minimum angle of interaction <30° were discarded. The time duration between the first and last bearing used to estimate a bearing was usual- ly <9 min. Where ten or more locations of a bat could be triangulated, the size of the foraging area of the individual was calculated by the ‘minimum range method’ (MoHR 1947). Theoretical centres of activity within the foraging areas were estimated (HAYNE 1949), and the distance between these centres of activity and the day roost was calcu- lated. The study area map (Fig. 1) was divided into 20 grids of 1 km” area each. Horizon- tally it is marked ‘a’ to ‘e’ and vertically it is marked ‘1’ to ‘4’. This would facilitate nam- 0044-3468/99/64/03 — 187 $ 12.00/0 188 N. GOPUKUMAR NAIR et al. P==- 2.22 2 22 2212 2 2 2 2m > == =- J) -- - >-- >= --> > > >=. . oo 2 2. >. a Denn IT IOAOTADODN CS) o Fig. 1. Typical representative triangulated foraging areas of a male, M4 (©) and a female, Fl (DI) C. sphinx, in a grid map of the study area in Madurai, Southern India, Each grid covers 1 km“. Table 1. Distance between centres of foraging area and day roost for five males and five females of C. sphinx. FA - Foraging Area, FA1 - First foraging area, FA2 - Second foraging area, FA3 — Third foraging area. Distance between the day roost to FAs (km) FAl Foraging behaviour of Cynopterus sphinx 189 ing specific foraging areas and day roost of each bat. The activity of the bats was moni- tored with the night vision sniperscope (Litton Precision Product, Germany, M-972) and their activity budget was calculated by monitoring the fluctuations of beep pulses from the receiver. The constant beep signals were considered as ‘rest’. All the five radio-tagged males (MI to M5) had their respective day roost and foraging area in grid 2b, whereas, the fifth male (M5) had an additional foraging area in 3c (Fig. 1). Throughout our study only one male (M4) was found to have night roost fidelity and Guer- terda speciosa was used as night roost constantly. But the other tagged bats of both sexes used more than one night roost. Each individual had used more than one tree species such as Areca catechu, P. longifolia, A. indica, C nucifera, Achras sapota, etc. The mean travel- ling distance for males was 0.22 +# 0.19 km and the mean size of the foraging area was 0.75 + 0.27 km’. Allthe females were found to be utilising more than one foraging area. Among these, one area lies far from the day roost and the other area (s) lies nearer to the day roost. For instance, F2 and F3 foraged in grids 2b, 2c, and 3b; and 2a, 2b, and 2c respec- tively, while F1 foraged in grid 2c and two sites in 4. d. During the observation, Fl, F2, and F3 regularly utilised the three different foraging areas every night. But the sequence of vis- its to different areas changed every night. F4 foraged only at two different foraging areas (4b and 2b). Interestingly, F5 roosting in grid 2b spent < 20 % of night time (n = 68 hours observation) foraging in 2b. The other foraging area of this bat could not be located as the foraging area was beyond 4 kms. The mean travelling distance for females was 2.1 +1.0 km and the mean size of the foraging area was 0.83 + 0.12 km’. The males and females exhibit a high level of activity during the early hours of night soon after emergence and another ac- tivity peak during pre-dawn hours (Fig. 2). Our study shows that the foraging pattern of €. sphinx ıs similar to that of Artibeus ja- maicensis, Phyllostomus hastatus, and Carollia perspicillata (August 1981; FLEMING 1988; MCCRrAcKEN and BRADBURY 1981; Morrıson 1978). Bats leave the day roost shortly after sunset and fly to foraging areas while they begin to search for ripe fruits. The harvested fruits were transported to the “night roosts” for consumption. These “night roosts” might promote digestion and energy conservation, offer retreat from predators, serve as centres for information transfer about the location of fruit patches and facilitate social interac- tion (Kunz 1982). A regular travel path exhibited by M 4 between day roost and foraging area may be attributed to the constancy of resource availability. Such a “trap-lining” be- haviour minimizes search distances and energy cost (Kunz 1982). It seems clear that the male C. sphinx restricts ıts foraging areas closer to the day roost. Since the males involve in tent construction, harem formation and defence, a short distance foraging area would promote harem defence near the day roost (FLEMING 1988; MARIMUTHU et. al., 1998). This observation of short distance foraging flights of males is consistent with observations of the harem males of A. jamaicensis, P. hastatus and C. perspicillata where feeding predominantly occurs ın the vicinity of their day roost (FLEMING 1988; HARDLEY and MorrISsoNn 1991; MCCRACKEN and BRADBURY 1981; MORRI- son 1979; MoRRISON and MorrISsoN 1981). The foraging areas of males are overlaping because the day roost of most of the males lies within a rich food patch. The reasons for the commutation to longer distances, spending more time and utilization of several foraging areas of female bats are not clearly known. One of the reasons for long dis- tance commutation by females might be search for potential male tent roost and to as- sess the harem male’s parental ability. Furthermore, they change their primary foraging area in an unpredictable fashion as observed in C. perspicillata (Kunz 1982). Since not every foraging area contains the same potential food source, one reason for such unpre- dietable “visits” might be to increase dietary diversity. In the usual bimodal pattern of activity, maximum foraging bouts occurred in the early hours of the night and lesser ac- tivity during the pre-dawn hours (FLEminG 1982). C. sphinx also shows a similar pattern of activity. 190 N. GOPUKUMAR NAIR etaal. S oO | | 1900 2100 2300 oO1O00 oOo30o OS5O00 60 50 40 30 20 10 0 ! 1900 2100 2300 0100 0300 0500 TIME (h) Fig. 2. a) Duration of activity and b) duration of perching for radio-tagged C. sphinx (1 - Male and EM - Female) DURATION OF ACTIVITY (min) DURATION OF PERCHING (min) Acknowledgements We are grateful for the financial assistance from DOEn to RS (14/11/94-MAB/RE), CSIR to GM (37(861)/94/EMR-II) and SRF (CSIR) to VE. Foraging behaviour of Cynopterus sphinx 191 References AUGusrt, P. V. (1981): Fig fruit consumption and seed dispersal by Artibeus jamaicensis in the Ilanos of Venezuela. Reprod. Bot. Suppl. Biotropica 13, 70-76. BALASINGH, J.; ISAAC, S. S.; SUBBARAJ, R. (1993): Tent-roosting by the frugivorous bat Cynopterus sphinx (Vahl, 1797) in Southern India. Curr. Sci. 65, 418. BALASINGH, J.; JOHN KOILRAJ, A.; Kunz, T. H. (1995): Tent construction by the short-nosed fruit bat, Cy- nopterus sphinx (Chiroptera: Pteropodidae) in Southern India. Ethology 100, 210-229. BHAT, H.R.; Kunz, T.H. (1995): Altered flower/fruit clusters of the kitul palm used as roosts by the short-nosed fruit bat, Cynopterus sphinx (Chiroptera: Pteropodidae). J. Zool. (London) 235, 597- 604. FLEMING, T. H. (1982): Foraging strategies of plant visiting bats. In: Ecology of Bats. Ed. by T. H. Kunz. New York: Plenum Publ. Corp. Pp. 287-323. FLEMING, T. H. (1988): The short-tailed fruit bat. Chicago, London: Univ. Chicago Press. HARDLEY, JR. C. O.; MoRRISON, D. W. (1991): Foraging behaviour. In: Demography and Natural History of the Common Fruit Bat, Artibeus jamaicensis, on Barro Colorado Island, Panama. Ed. by C.0. HANDLEY; D. E. Wırson, and A. L. GARDNER. Washington, D. C.: Smithsonian Institution Press. Pp. 137-140. HAYNE, D. W. (1949): Calculation of size of home range. J. Mammal. 30, 1-18. Kunz, T.H. (1982): Roosting ecology of bats. In: Ecology of Bats. Ed. by T. H. Kunz. New York: Ple- num Press. Pp. 1-55. MARIMUTHU, G.; RAJAN, K. E.; JoHN KOILRAJ, A.; Isaac, S. S.; BALASINGH, J. (1998): Observation on the foraging behaviour of a tent roosting megachiropteran bat C'ynopterus sphinx. Biotropica. 30, 321- 324. MCCRACcKEN, G. F.; BRADBURY, J. W. (1981): Social organization and kinship in the polygynous bat Phyl- lostomas hastatus. Behav. Ecol. Sociobiol. 8, 11-34. MOoHR, C. ©. (1947): Table of equivalent populations of North American small mammals. Am. Midl. Nat. 25, 196-233. MORRISoN, D. W. (1978): Lunar phobia in a Neotropical fruit bat, Artibeus jamaicensis (Chiroptera: Phyllostomidae). Anim. Behav. 26, 852-855. MORRISoN, D. W. (1979): Apparent male defence of tree hollows in the fruit bat, Artibeus jamaicensis. ). Mammal. 60, 20-29. MORRISON, D. W.; MoRRISON, S. H. (1981): Economics of harem maintenance by a neotropical bat. Ecol- ogy 62, 864-866. Authors’ address: N. GOPUKUMAR NAIR, V. ELANGOVAN, K. SRIPATHI, G. MARIMUTHU, and R. SUBBARAJ, Department of Animal Behaviour and Physiology, Madurai Kamaraj University, Madurai 625 021, India. Z. Säugetierkunde 64 (1999) 192 ZEITSCH RIFT® Be FÜR © 1999 Urban & Fischer Verlag SÄUG Bill ERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Buchbesprechung ANnonYyMUSs: Die Nashörner. Begegnung mit urzeitlichen Kolossen. Fürth: Filander Verlag 1997. Zahl- reiche Abb. und Farbtafeln, geheftet, 258 pp. DM 68,-. ISBN 3-930831-06-6. Schon auf der Titelseite gerät der Leser in Verwunderung: Ein Herausgeber für diesen Band wird nicht genannt! Ferner werden die Nashörner im Buchtitel als „urzeitliche Kolosse‘“ bezeichnet, ob- wohl in dem vorliegenden Werk nur in einem Beitrag von C. P. Groovzs fossile Rhinocerotoidea be- handelt werden; alle anderen Kapitel beschäftigen sich mit rezenten Arten. Der Band enthält 20 Kapitel, welche von verschiedenen Autoren verfaßt wurden. In den vier ein- leitenden Abschnitten werden die Beziehungen zwischen Mensch und Nashörnern dargestellt, dann folgt der schon erwähnte Abschnitt über die Stammesgeschichte und Verwandtschaft und jeweils ei- ner über die Körperbeschaffenheit, sowie über das Verhalten der Arten. Zehn Kapitel bilden den Hauptteil des Bandes: Angaben zum Erscheinungsbild der Arten, durch Farbabbildungen illustriert, sowie Bemerkungen zur Biologie, zum Verhalten und zur Verbreitung (durch Karten gut illustriert) werden geboten. Der Leser findet in diesen Kapiteln gute Darstellungen der fünf rezenten Arten der Familie Rhinocerotidae. Die drei asiatischen Arten werden in jeweils eigenen Kapiteln behandelt: Das Sumatra-Nashorn (Dicerorhinus sumatrensis) wird vorgestellt, das Java-Nashorn (Rhinoceros sondaicus) beschrieben, und ein Artikel über das Indische Panzernashorn (Rhinoceros unicornis) folgt. Den beiden afrikanischen Nashornarten sind jeweils mehrere Kapitel ge- widmet. In drei Darstellungen wird ein umfassendes Bild des Spitzmaul-Nashorns (Diceros bicornis) geboten, wobei nicht nur die Biologie der Art, sondern auch der Gesamtbestand und sein Schutz be- handelt werden. Dem Breitmaul-Nashorn (Ceratotherium simum) sind vier Kapitel verschiedener Autoren gewidmet. Den letzten Teil des Buches füllen Darstellung des Handels mit Nashornprodukten, sowie Über- sichtsdarstellungen des Nashorn-Schutzes und der Erhaltungsmaßnahmen, ferner werden Bemerkun- gen zu „Nashörnern im Zoo heute“ gemacht. Ein entbehrliches Kapitel schildert die Tätigkeit eines kenyanischen Aktivisten, der sich dem Schutz von Nashörnern verschrieben hat. Eine kurze tabella- rische Zusammenstellung von vergleichenden Daten zu allen fünf Arten, sowie eine unnötig langat- mige Vorstellung der am vorliegenden Band beteiligten Autoren und ein nützlicher vierseitiger Index schließen den Band ab. Die Nutzbarkeit der in diesem Band gebotenen reichen Informationen wird leider dadurch be- trächtlich eingeschränkt, daß die Einzelbeiträge nicht mit Literaturverzeichnissen versehen sind und die Zusammenstellung auf Seite 245 nur acht Publikationen, von denen drei reine Bibliographien sind, umfaßt! Weiterhin stört eine große Zahl von Druckfehlern. Es handelt sich dabei häufig um „Leichen“, welche in der Textverarbeitung durch Veränderungen ohne anschließende vollständige Löschung des alten Textes entstanden sind. P. LANGER, Gießen 0044-3468/99/64/03 — 192 $ 12.00/0 Redaktionelle Hinweise Einsendung und Annahme von Manuskripten: Ms. zur Veröffentlichung sind einzusenden an den 1. Schriftleiter: Herrn Prof. Dr. Dieter Kruska, Institut für Haustierkunde, Christian-Albrechts-Universität, Olshausenstr. 40-60, D-24118 Kiel. e-mail: dkruska@ifh.uni-kiel.de Über die Annahme von Ms. zur Veröffentlichung wird gemäß Geschäftsordnung der Deutschen Gesellschaft für Säugetierkunde entschieden. Der Eingang von Ms. wird sofort bestätigt, und nach Erhalt der Gutachten wer- den die Autoren über die Entscheidung umgehend informiert. Schriftwechsel, der die technische Drucklegung betrifft, ist zu richten an: Zeitschrift für Säugetierkunde, Urban & Fischer Verlag, Villengang 2, D-07745 Jena. Äußere Form der Manuskripte: Die Ms. sind 3fach und vollständig als Original mit 2 Kopien in deutscher oder englischer Sprache einzureichen. Sie müssen mit Schreibmaschine oder als deutlich lesbare Computerausdrucke auf DIN A4-Blättern (21 x 29,7 cm), mit einem linken Heftrand von 4 cm und durchgehend mit einem doppelten Zeilenabstand geschrieben sein. Jedes Ms. soll auf der 1. Seite folgende Angaben enthalten: Titel der Studie; Name(n) der Autoren (männliche Vornamen nur in Initialen, weibliche ausgeschrieben); Institution(en), an der (denen) die Studie durchgeführt wurde. Ferner soll bei längeren Titeln ein kurzer Kolumnentitel (Seitenüber- schrift) von maximal 72 Anschlägen angegeben werden. Die Schrifttypen müssen einheitlich sein, ohne halbfette, gesperrte, kursive etc. Hervorhebungen. Fußnoten auf den Seiten oder Appendices am Ende sind nicht zugelassen. Autorennamen im Zusammenhang mit Literatur- zitaten sind im Text mit Bleistift zu unterstreichen. Wissenschaftliche zoologische und botanische Gattungs-, Art- und Unterartnamen sind mit einer Schlangenlinie zu unterstreichen. Der Name von Erstbeschreibern bei wis- senschaftlichen Namen wird nicht unterstrichen. Im Ms. ist am Rand mit Bleistift zu vermerken, an welcher Stelle die einzelnen Abbildungen und Tabellen ein- gefügt werden sollen. Am Ende der Studie, im Anschluß an das Literaturverzeichnis, sind die genauen postalischen Adressen des (der) Autoren anzugeben. Umfang und Aufbau der Manuskripte: Manuskripte können als Originalarbeiten oder als wissenschaftliche Kurz- mitteilungen veröffentlicht werden. a) Originalarbeiten: Originalarbeiten sollen einschließlich Abbildungen, Tabellen und Literaturverzeichnis einen Gesamtumfang von 30 Ms.-Seiten nicht überschreiten. Der Text soll eine klare Gliederung aufweisen in: Abstract (in englischer Sprache), Einleitung, Material und Methode, Ergebnisse, Diskussion (oder zusammengefaßt als Er- gebnisse und Diskussion), Danksagungen, Zusammenfassung und Literatur. Bei deutschsprachigen Studien ist über dem englischen Abstract der Titel in englischer Sprache anzugeben. Bei englischsprachigen Studien ist entsprechend über der deutschen Zusammenfassung der Titel in deutscher Sprache anzugeben. b) Wissenschaftliche Kurzmitteilungen: Wissenschaftliche Kurzmitteilungen sollen einen Gesamtumfang von 5 Ms.-Seiten nicht überschreiten. Sie enthalten kein Abstract und keine Zusammenfassung. Bei deutschsprachigen Ms. ist jedoch auf der 1. Seite zusätzlich eine englische Übersetzung des Titels anzugeben, bei englischsprachigen entsprechend eine deutsche Übersetzung. Wissenschaftliche Kurzmitteilungen sind äußerlich nicht in die Kapitel Einleitung, Material und Methode, Ergebnisse und Diskussion gegliedert, müssen aber in der textlichen Darstel- lung diese Strukturierung erkennen lassen. Key words: Bis zu fünf Key Words sollen dem Abstract folgen. Abbildungen: Anzahl und Größe der Abbildungen sind unbedingt auf das notwendige Mindestmaß zu beschrän- ken. Es können nur kontrastreiche, reproduktionsfähige Originale verwendet werden, farbige Abbildungen wer- den nicht angenommen. Beschriftungen müssen einheitlich und wegen späterer Verkleinerung groß und kräftig sein. Auf der Rückseite sind laufende Numerierung, Zeitschrift und Autorennamen anzugeben. Alle Bildunter- schriften sollen auf einem gesonderten Blatt gelistet werden. Tabellen: Tabellen müssen auf ein Mindestmaß beschränkt werden. Jede Tabelle muß auf einem gesonderten DIN A4-Blatt angelegt sowie mit Überschrift und laufender Numerierung versehen sein. Umfangreichere Tabel- len müssen auf Folgeblättern als Fortsetzung erscheinen. Literaturverzeichnis: Jede zitierte Arbeit ist in alphabetischer Reihenfolge aufzuführen. Es soll nur international erreichbare Literatur zitiert werden. Die Zeitschriften werden mit ihrem Kurztitel angegeben. Beispiele: a) Zeitschriftenbeiträge: KALko, E. K. V; HANDLEY, C. O. Jr. (1994): Evolution, biogeography, and description af a new species of Fruit-eat- ing bat, genus Artibeus Leach (1821), from Panama. Z. Säugetierkunde 59, 257-273. b) Bücher: HERRE, W.; RöhHrs, M. (1990): Haustiere - zoologisch gesehen. 2. Aufl. Stuttgart, New York: Gustav Fischer. c) Handbuchbeiträge: NIETHAMMER, J. (1982): Cricetus cricetus (Linnaeus, 1758) — Hamster (Feldhamster). In: Handbuch der Säugetiere Europas. Ed. by J. NIETHAMMER und F. Krapp. Wiesbaden: Akad. Verlagsges. Vol. 2/1, 7-28. Drucklegung: Der jeweilige Hauptautor erhält sowohl die Andrucke der Bilder zur Prüfung wie auch den kom- pletten Umbruch seines Artikels zusammen mit dem Ms. Es dürfen nur tatsächliche Satzfehler korrigiert wer- den. Vom Ms. abweichende Änderungen werden dem Autor in Rechnung gestellt. Es sind die üblichen Korrekturzeichen zu verwenden. Ein als druckfertig bezeichneter Abzug ist zu senden an: Herrn Prof. Dr. Peter Langer, Institut für Anatomie und Zellbiologie, Justus-Liebig-Universität, Aulweg 123, D-35385 Giessen. e-mail: peter.langer@anatomie.med.uni-giessen.de Sonderdrucke: Verfasser von Originalbeiträgen und Wissenschaftlichen Kurzmitteilungen erhalten 50 Sonder- drucke gratis. Mehrbedarf kann gegen Berechnung bei Rücksendung des Umbruchs bestellt werden. Vorbehalt aller Rechte Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, ins- besondere 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 gesetz- lichen Bestimmungen des Urheberrechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmun- gen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Ver- vielfältiger 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 in- ternal 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 CCC. Publisher: Urban & Fischer Verlag GmbH & Co. KG, Branch Office Jena, P. 0. Box No. 100537, D-07705 Jena, Germany. Phone ++49(0)36 41/6 26-3; Fax ++49(0)36 41/62 65 00; e-mail: journals@urbanfischer.de Type setting, printing and binding: druckhaus köthen GmbH Advertisement bookings: Enquiries should be directed to Urban & Fischer Verlag, Advertising Manager: Sabine Schröter, P.O. Box 10 05 37, D-07705 Jena, Germany. Phone ++49(0)36.41/62 6445, 9 Fax ++49(0)36 41/62 64 21 Advertising rates: The price list from January 1, 1999 is effective at present. Distribution and subscription agency: For customers in the German speaking countries: Please turn to your bookshop or to SFG-Servicecenter Fachverlage, Zeitschriftenvertrieb: Barbara Dressler, Villengang.2, D-07745 Jena, Germany. Phone ++49(0)36 41/62 64 44, Fax ++49(0)36 41/62 64.43; For customers in all other countries: Please contact your local bookshop or Stockton Press., c/o Debbie Floyd, The Sub- scription Dept., Houndmills, Basingstoke, Hants RG 21 6XS, UK. Phone ++44(0)12 56/32 92 42. Fax ++44/1256/812358. Subscribers in the USA can call toll free on ++1/800/ 747 3187. Terms of delivery (1999): 1 volume consisting of 6 issues. Subscription rate (1999): Price per volume: DM 478.00*; UK Sterling 192. 00 Single issue price: DM 95.00*; UK Sterling 39.00 * Suggested list prices. Prices are subject to change without notice. Reduced membership rates are available on request. Please turn to our publ. house. All: prices mentioned above are exclusive postage. -WSdollar prices are converted at the current exchange rate. We accept the following credit cards: VISA Euro-/Mastercard/American Express, (Please note card-No. and expiry date). Banking connections: For customers in the German speaking countries: Deutsche Bank AG Jena, Konto-No. 6 284 707, BLZ 820 700 00; Postbank Stuttgart, Konto-No. 92 372 704, BLZ 600 100 70. For costumers in the other countries: UK Post Office Giro Account No. 5 192 455. Full details must accompany the payment. Subsciption information: If not terminated the subscription will be effected until recalled. If no cancellation is given until October, 31st the delivery of the journal will be continued. Copyright: The articles published in this journal are protected by copyright. All rights are reserved. No part of the journal may be reproduced in any form without the written permission of the publisher. Authorization to photocopy items for internal or personal use of specific clients, is granted by Urban & Fischer, for libraries and other users registered with the Copyright Clearance Center (CCC) Transaction Reporting Service, provided that the base fee of $12.00 per copy is paid directly to CCC, 222 Rosewood Drive, Danvers, Ma 01923, USA, 0044-3468/99/064/ $ 12.00/0 This consent does not extend to copying for general distribution, for advertising or promotional purposes, for creating new collective works or for resale and other enquiries. In the cases, specific written permission must be obtained from the pub- lisher Urban & Fischer. Printed in Germany FR STOCKTON Printed on acid-free paper effective with vol. 61, no. 1, 1996. © 1999 Urban & Fischer Verlag New: For detailed journal information see our home page: http://www.urbanfischer.de/journals Abstracted/Indexed in Animal Breeding Abstracts; Current Contents Agriculture, Biology Environmental Sciences; Biological Abstracts; BIOSIS "database; Current Advances in Ecological and Environmental Sciences; Dairy Science Abstracts; Elsevier BIOBASE/Cur- rent Awareness in Biological Sciences; Fisheries Review; Helminthological Abstracts; Index Veterinarius; South Pacific Periodicals Index; Veterinary Bulletin; Key Word Index to Wildlife Research; Wild Review (Fort Collins); Zoological Record Sr FÜR SÄUGETIERKUNDE N INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Elmeros, M.; Madsen, A. B.: On the reproduction biology of otters (Lutra lutra) from Denmark. - Über die Fortpflan- Bea von kschorierm Luffe lütra). an Danemark.n......-...... nennen ee cnnenensnnaassnnensnsnde nennen 193 Kloskowski, J.: Otter Lutra lutra predation in cyprinid-dominated habitats. - Beuteerwerb des Fischotters Lutra lutra in von Weißfischen dominierten Habitaten Kasnusnssunsansnnnnnssnsansnnsnnsnnssnssnsnunsnssnnnnnsnusnnsnnsnnsnssnssnsensenenn 201 Sözen, M.; Colak, E.; Yigit, N.; Özkurt, S.; Verimli, R.: Contributions to the karyology and taxonomy of the genus Spalax Güldenstaedt, 1770 (Mammalia: Rodentia) in Turkey. - Zur Karyologie und Taxonomie der Gattung Spalax Gülden- Ba nımalıdz Rodentia) in der Türkei ......0 nes ecnesennansannasaecsnesaunennasnnnanen nn enesnen nennen 210 Kunze, Bärbel; Dieterlen, F.; Traut, W.; Winking, H.: Karyotype relationship among four species of Spiny mice (Acomys, Rodentia). - Karyotypische Verwandtschaftsbeziehungen zwischen vier Arten der Gattung Acomys (Rodentia)....... 220 Camin, S.: Mating behaviour of Ctenomys mendocinus (Rodentia, Ctenomyidae). - Paarungsverhalten von Ctenomys ER LERHENHNGDGEN) Se name ankenensmen ernten near nes sun eh ennnen sense nee 230 Wissenschaftliche Kurzmitteilungen Tschapka, M.; Wilkinson, G. S.: Free-ranging Vampire bats (Desmodus rotundus, Phyllostomidae) survive 15 years in the wild. - Freilebende Vampirfledermäuse (Desmodus rotundus, Phyllostomidae) überleben 15 Jahre in der Natur....... 239 Siemers, B. M.; Kaipf, Ingrid; Schnitzler, H.-U.: The use of day roosts and foraging grounds by Natterer’s bats (Myotis nattereri Kuhl, 1818) from a colony in southern Germany. - Nutzung von Tagesquartieren und Jagdgebieten durch Fransenfledermäuse (Myotis nattereri Kuhl, 1818) in einer Kolonie in Süddeutschland................uu-2ues#0s een 241 Szapkievich, Valeria B.; Cappozzo, H.L.; Crespo, E. A.; Bernabeu, R. O.; Comas, Cecilia; Mudry, Marta D.: Genetic relat- edness in two Southern sea lion (Otaria flavescens) rookeries in the southwestern Atlantic. - Genetische Ver- wandtschaft zwischen zwei Populationen von Seelöwen (Otaria flavescens) im südwestlichen Atlantik................ 246 Siciliano, S.; Lailson Brito Jr., J.; de F. Azevedo, A.: Seasonal occurrence of killer whales (Orcinus orca) in waters of Rio de Janeiro, Brazil. - Saisonales Vorkommen von Schwertwalen (Orcinus orca) in den Gewässern vor Rio de Janeiro, en a aa rasen Ina nn nah namen benennen nen 251 ISSN 0044-3468 - Z. Saugetierkunde - 64(1999)4 - S. 193-256 - August 1999 URBAN & FISCHER 1999 GESEL, 8°: Xi, [L >: ZEITSCHRIFT FÜR Se: SÄUGETIERKUNDE N ee ee I INTERNATIONAL JOURNAL & ® OF MAMMALIAN BIOLOGY (ee) Rn TERKUND“ Herausgeber/Editor Deutsche Gesellschaft für Säugetierkunde Schriftleitung/Editorial Office D. Kruska, Kiel - P. Langer, Giessen Wissenschaftlicher Beirat/Advisory Board W. Fiedler, Wien - S. Halle, Jena - G. B. Hartl, Kiel - R. Hutterer, Bonn -H.-G. Klös, Berlin - E. Kulzer, Tübingen - P. Lüps, Bern - W. Maier, Tübingen - O. Anne E. Rasa, Bonn -H. Reichstein, Kiel - M. Röhrs, Hannover - H. Schliemann, Hamburg - G. Storch, Frankfurt Deutsche Gesellschaft für Säugetierkunde Altvorsitzende/Living Past Presidents D. Starck, Frankfurt (1957-1961, 1967-1971) - H. Frick, München (1972-1976) — M. Röhrs, Hannover (1977-1981) - H.-J. Kuhn, Göttingen (1982-1986) - E. Kulzer, Tübingen (1987-1991) - U. Schmidt, Bonn (1992-1996) Amntierender Vorstand/Managing Committee Vorsitzender/President: H. G. Erkert, Tübingen Mitglieder/Board Members: H. Frädrich, Berlin - R. Hutterer, Bonn - D. Kruska, Kiel — Marialuise Kühnrich, Hamburg - R. Schröpfer, Osnabrück - Angela Wöhrmann-Repenning, Kassel Z. Säugetierkunde 64 (1999) 193-200 ZEITSCHRIFT * ur h © 1999 Urban & Fischer Vers — _ _ _ ___ ____SÄUGETIERKÜNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY On the reproduction biology of otters (Lutra lutra) from Denmark By M. Ermeros and A. B. MADSEN National Environmental Research Institute, Department of Landscape Ecology, Ronde, Denmark Receipt of Ms. 20. 07. 1998 Acceptance of Ms. 29. 03. 1999 Abstract The reproduction in Danish otters was inferred from examination of reproductive organs from 242 carcasses collected between 1982 and 1996. Estimated months of birth of collected cubs and evi- dence of breeding, determined in female reproductive organs, showed distinct seasonal patterns. 82 % of the cubs were born during summer and autumn months from June to November, although litters were born throughout the year. Mean litter size at birth was 1.7 +0.9 cubs per litter. Adult male ot- ters showed continuous mating preparedness. No seasonal variation in paired testes weight for adult males was determined and males with high density of spermatozoa in testes smears occurred through- out the year. Adult males with spermatozoa present had a significantly higher body condition index compared to males without spermatozoa. As the imminent factor determining the breeding chronol- ogy, fish densities peaked in autumn, coinciding with maximum energetic demands on reproductive active females during the lactation period. Key words: Lutra lutra, breeding pattern, reproduction, Denmark Introduction Otters (Lutra lutra) living in different areas and habitats show large variation in the breeding pattern varying from an even distribution of births throughout the year (STE- PHENS 1957; SıiDoRoVIcH and TuMmAnov 1994) to a strictly seasonal occurrence (ERLINGE 1967; Kruuk et al. 1987). In several populations, births have been reported throughout the year with a seasonal peak (STUBBE 1969; WIINGAARDEN and PEPPEL 1970). Temporal and spatial fluctuations in the availability of food resources determine birth patterns, tim- ing the period of highest energetic requirements of reproductive females at peak lacta- tion with maximum fish densities (Kruuk et al. 1987; OFTEDAL and GITTLEMAN 1989; HEGGBERGET and CHRISTENSEN 1994). Most studies focus on reproductive activities and status of females and cub recruit- ment (e.g. SIDOROVICH 1991; BEJA 1996; ANSORGE et al. 1997). Only short notes on male mating preparedness have been published (HEGGEBERGET and CHRISTENSEN 1994; SIDORO- vıcH and TuMAnov 1994). Further knowledge, on male reproductive capacity and mating preparedness in seasonally and non-seasonally breeding populations, is important for con- servation and management of otters. In Denmark, otters are known to breed throughout the year (JENSEN 1964), however, no detailed information on seasonal distribution of births and reproductive phases has been presented. The aim of this study is to investigate the breeding pattern in Danish ot- ters. 0044-3468/99/64/04-193 $ 12.00/0 194 M. ELmeERros and A. B. MADSEN Material and methods Otter carcasses were collected between 1982 and 1996 in northern Jutland representing the main dis- tribution area of the Danish otter population. Specimens originated from inland freshwater habitats and marine habitats with brackish waters. The majority of otters were killed in traffic. Mortality rates and the probability of finding otter carcasses were assumed to be the same throughout the year. Col- lection of adult otters was randomly distributed through seasons (n = 113, X = 5.7, n.s.). Juveniles and subadults were collected primarily during autumn (n= 129, 7 =44.3, P< 0.01). Carcasses were sub- jected to a detailed necropsy and examination of health condition (MADSEn 1996). Specimens were aged as juveniles (younger than 5 months), subadults and adults (older than 18 months) on skeletal criteria (MAson and MADpsen 1993). Body condition index (CI) for the otters was calculated as the re- lation between body weight and total length according to KrUUk et al. (1987). Age of juveniles was assessed from body weight (STEPHENS 1957), after calculating body weight from total length at normal body condition index (Kruuk et al. 1987). To establish female reproduc- tive status length and diameter of uteri were measured and ovaries and uteri examined macroscopi- cally of presence of corpora lutea, embryos, and placental scars (HEGGBERGET 1988). Uterus horns were flushed with 2 ml water to collect blastocysts. The birth time of litter born by females was as- sessed from the colour of placental scars and appearance of the uterine tissue. Initially after birth im- plantation sites have dark pigmentation, while in the final stages implantation sites gradually become orange and white (HEGGBERGET and CHRISTENSEN 1994). Based on examination of a female killed with a cub at about 3 months of age, dark placental scars are estimated to persist for at least 3 months. Male reproductive capacity was assessed by paired testes weight including epididymes, and micro- scopic examination for presence of spermatozoa in testes smear (MADSEN and RAsMmussEn 1985). Re- lative occurrence of spermatozoa in smear was valued as: none, low or high density. Determination of all parameters in all specimens was not possible. Seasonal densities of fish were determined by electrofishing at 5 freshwater localities (TAASTROM and JACoBSEN 1999). Results Juveniles Seasonal distribution of estimated time of birth for the collected cubs is shown in fig- ure 1. The seasonal distribution of births was significantly different from an even distribu- tion (n=22, X =48.8, P< 0.001); 82 % of the cubs were born in summer and autumn 60 50 40 Percentage of cubs Winter Spring Summer Autumn Fig. 1. Seasonal distribution of births for received Danish otter cubs (n = 22). Ages were assessed from body weight (STEPHENS 1957; KrUUX et al. 1987). Winter: December-February; Spring: March-May; Summer: June-August; Autumn: September-November. Reproduction in Danish otters 195 Fish density (#/m?) Percentage adult females Winter Spring Summer Autumn Mreproductive Einonreproductive females —®- fish density Fig. 2. Seasonal percentage of adult femalkes with dark placental scars and fish densities in Danish freshwater habitats (TAASTROM and JAcoBsen 1999). Percentages of reproductive females were calcu- lated from numbers of all adult females collected each season (Winter, n = 10; Spring, n = 9; Summer, n = 11; Autumn, n = 14). months from June to November. Highest frequency of births was seen in the three-month period of July, August, and September: 59 % in total. Siblings were counted as one. Females Distribution of reproductive females with dark placental scars varied seasonally (n= 13, X =733, P< 0.001) peaking in autumn and winter (Fig. 2). In all females with placental scars, uterus had regressed to normal size (HEGGBERGET 1988). Assuming parturition some two months earlier, these litters were born during the summer and autumn months corresponding with the main birth months of the cubs. Seasonal occurrence of reproduc- tive females exhibited a good correlation with seasonal fish densities in different fresh- water habitats in Denmark (Fig. 2) (TAASTROM and JACoBSEN 1999). Additional indications of reproductive activities determined from adult females corre- spond with estimated births in summer and autumn; a female killed in March had neck wounds and spermatozoa in her reproductive tract, and one pregnant female with small embryos was collected in April, and a female killed in January had old faded placental scars. Three females had placental scars of different colours possibly indicating abortion or resorption of embryos. Overall, indications of breeding were found in 34% of all adults. Based on numbers of embryos and newest placental scars, the average litter size was 1.7#0.9, range 1-4 (n=15). Including all placental scars the estimated litter size was 2.2+1.2, range 1-4 (n= 15). Counts of corpora lutea resulted in litter sizes of 2.7 #1.7, range 1-5 (n = 7). No blastocysts were recovered. Seasonal variation of the body condi- tion index of all adult females was not significant (n= 45, F= 1.01, P= 0.40) and no dif- ferences in body condition indices between reproductively active and nonreproductive fe- males were established (n = 45, t = 0.405, P = 0.69). All females classified as adults had mature reproductive organs. Alterations of repro- ductive organs were observed in four adult females. One female had severely convoluted uterine horns. One female had fibrous but normal sized uterus with two cysts on the uterus and an occlusion in the uterine body. Two females had small uterine cysts. No plac- cutal scars, embryos or corpora lutea were recorded in these four females. Immature fe- males had thin and translucent uterine tissue. Some females aged as subadults showed 196 M. ELmERoS and A. B. MADSEN maturing uteri and had large follicles in ovaries. They were collected at all seasons, but most frequently in spring (n = 16, x’ = 41.4, P< 0.01 ). Males Paired testes weights of adult, subadult, and juvenile males differed significantly (Tab. 1). Within age classes specimens with different densities of spermatozoa showed different paired testes weights. Adult males with spermatozoa in low and high densities were found throughout the year. Seasonal variation in paired testes weight for all adults was not found (n=56, F = 0.31, P= 0.81) (Fig. 3), nor was any trend apparent for males with spermatozoa pre- sent (n = 48, F = 0.09, P= 0.97). No seasonal variation in body condition index occurred (n =55, F = 0.51, P= 0.68). Paired testes weight and body condition index of adult males Table 1. Paired testes weights for different age classes of male otters (*P< 0.001 ). Paired testes weights for testes with different densities of spermatozoa within an age class(’P < 0.001 and °P < 0.01). Age class Mean +SD (g) Range (g) Adults 10.8 + 3.4° high ID D85 low 9.3 +2.2P non 7.1+2.9° Subadults ee high 9.8+1.2® low 6.1+2.6°° non ERS Juveniles 0.9 +0.5° 20%: © 8 eo = ® © | 8 © : = [) 5b $ ® U) % [ 'd eo ® ® 2 ® © 8 = 0008 a ® ö ® ) Fi o.% 17 [ x # oO “* % 2 ® 0 x () - —— Ss > % = x j %* | ja * * © KR x x 6) KUN * [n Aue * 0 o 8 0 OD ——— 4 4 991 F M A M J J A S O N D Months o juveniles %# subadults @ adults Fig. 3. Monthly variations of paired testes weights among Danish male otters (n = 115). Reproduction in Danish otters 1977 20 er ® ® © ’ oe ° 0. eh . I . ® %“ %* 0 > 2° Be) . © 10: L x 0 48 8 ® & 0% PL SEE 2. = (6) %* x 0 = (6) & Sl % A [0] Ö 0° | | I IF 1 0.6 0.8 1.0 12 1.4 1.6 Body condition index O none * low @ high density Fig. 4. Paired testes weight and body condition indices for adult Danish male otters (n = 48). Densities of spermatozoa are indicated as none, low, and high. Body condition index is in accordance with KRUUK et al. (1987) were significantly correlated (n = 48, t= 3.38, P<0.001) (Fig. 4). Adult males with sper- matozoa present showed higher body condition indices than males without spermatozoa (n = 48, F = 18.8, P< 0.001). For subadults with immature testes and no spermatozoa present, no differences be- tween seasonal testes weights were established (n = 22, F = 0.22, P = 0.88). Paired testes weight for subadults with spermatozoa present varied throughout the seasons, peaking during winter (n = 23, F=5.68, P<0.01). Subadult males with high density of spermato- zoa in the testes (n = 8) were found only among specimens collected in autumn and win- ter. The findings of subadults with high densities of spermatozoa indicate that maturation may occur at an age of 18 months. No seasonal variation in body condition index (n =53, F = 0.77, P= 0.68) was estab- lished among subadults. Despite different stages of sexual maturity in this age group, paired testes weight and body condition were significantly correlated (n = 42, t = 2.84, P<0.01) and subadults with spermatozoa present had significantly higher body condition indices (n = 42, F = 7.24, P<0.05). Discussion Reproductive phases of females and cub recruitment have been examined in other popu- lations. Litter sizes in Germany based on counts of corpora lutea reached 2.8. Based on embryos and placental scars they were 2.7 and 2.1 based on observed cubs per litter (An- SORGE et al. 1997). In a Belarussian population, litter size was 2.7 at corpora lutea phase, 2.6 cubs less than one month of age and 2.4 cubs following females (SıporovicH 1991). Si- milar numbers of corpora lutea were counted in Danish otters, but the implantation fre- quency was lower and loss of embryos during gestation higher (20 %). An equivalent fre- quency of resorption or abortion of implanted embryos has been reported in Norwegian otters (HEGGBERGET and CHRISTENSEN 1994). Litter sizes estimated from numbers of em- bryos and placental scars in females represent maximum litter size at birth (STRAND et al. 198 M. ELmERros and A. B. MADSEN 1995). Postnatal mortality rates between 12 % and 24 % further reduces litter sizes (SI- DOROVICH 1991; HEGGBERGET and CHRISTENSEN 1994; ANSORGE et al. 1997). Danish otters produce smaller litters compared to litter size estimated from observations of family groups in marine habitats (Kruuk et al. 1987; HEGGBERGET and CHRISTENSEN 1994; BEJA 1996), and noticeable smaller litters than generally reported from freshwater habitats (WIINGAARDEN and PEPPEL 1970; Mason and MAcDoNALD 1986; BEJA 1996; ANSORGE et al. 1997). Relatively low recruitment due to small litters in the Danish otter population is compensated by a higher proportion of reproductive active females (34 %), compared to 23 % of adult females in the stable high density freshwater population in the eastern part of Germany (AnsoRGe et al. 1997). Otters living in adjacent freshwater and marine habitats may have different breeding chronology and litter sizes (BEsA 1996). Within an individual otter’s home range, how- ever, utilisation of freshwater and marine centres vary depending on food availability (SIÖAsEn 1997), and a separation of Danish populations in strictly marine or freshwater living otters would be questionable. The continuous mating preparedness in adult males is consistent with unseasonal oes- trus and ovulation bouts in females (HEGGBERGET and CHRISTENSEN 1994). Adult males with poor body condition indices and low paired testes weights all suffered from various infectious diseases (MADSEN 1996) or were collected during a severe winter. The latter had probably lost conditions rapidly and still had spermatozoa present in testes. As in this study, a considerable range in paired testes weights with spermatozoa was found in a small number of adult males from Russia and Belarus (SiDorovicH and TUMA- nov 1994). Maturation at 18 months in Danish otters was equivalent to age of maturation observed in Germany (STUBBE 1969). In Russia and Belarus SıporovicH and TUMANOV (1994) found immature testes in all 1-2 year old males, and in Norway females matured between 2 and 3 years of age (HEGGBERGET 1988). Paired testes weight appears to be a sufficient measurement of testicular activities for large samples, but with apparent differences in testes weights between populations and a wide weight range within populations, conclusions on mating preparedness in male otters from testes weights data alone must be interpreted cautiously. Continuous oestrus cycle and non-seasonal breeding potential in otters may have evolved as a reproductive risk-reducing adaptation to an unpredictable, although season- ally changing environment (HEGGBERGET and CHRISTENSEN 1994). Additional adaptation to annual fluctuations of the environment include flexible population dynamics with so- cial regulation of recruitment and population density-dependent fecundity of females (Kruuk et al. 1991; SıiporovicH 1991). Seasonal variation in body condition of females correlated with food availability and breeding success on the Norwegian coast (HEGGBER- GET and CHRISTENSEN 1994). Assessed by the invariability of body condition index for Danish otters, food re- sources appear relatively stable. However, a seasonal birth peak has evolved correlating maximum energetic strain on reproductive females with peaking fish densities in Danish freshwaters. Acknowledgements We gratefully acknowledge H. H. DIETZ at the Danish Veterinary Laboratory and the taxidermists at the Natural History Museum, Ärhus and the Zoological Museum, Copenhagen, for assistance with the autopsy and preservation of specimens. We also acknowledge B. GAARDMAND and M. HAMMERSHY@J for technical assistance and K. Zaruskı for linguistic help. H. WEBER wrote the German summary. T. M. HEGGBERGET and the anonymous reviewers are acknowledged for valuable comments on the manuscript. Reproduction in Danish otters 199 Zusammenfassung Über die Fortpflanzungsbiologie von Fischottern (Lutra lutra) in Dänemark Die Geschlechtsorgane von 242 Totfunden dänischer Otter aus den Jahren 1982 bis 1996 wurden im Hinblick auf den Reproduktionszustand des jeweiligen Einzeltieres untersucht. Diese Ergebnisse ließen Rückschlüsse auf die Fortpflanzungsbiologie der dänischen Otterpopulation im Vergleich zu den aus der Literatur bekannten Mustern von Otterpopulationen anderer Länder zu. Der Nachweis erfolgter Geburten anhand von makroskopischen Untersuchungen weiblicher Ge- schlechtsorgane wies deutlich saisonale Unterschiede auf. Schätzungen des Geburtsmonats tot aufge- fundener Jungtiere konnten diese Saisonalität noch untermauern: 82 % der Jungtiere wurden in den Sommer- und Herbstmonaten zwischen Juni und November geboren, wobei grundsätzlich Geburten zu allen Jahreszeiten nachgewiesen werden konnten. Die durchschnittliche Jungtieranzahl betrug 1,7+0,9 zum Zeitpunkt der Geburt. Adulte männliche Otter scheinen im Allgemeinen das ganze Jahr über paarungsbereit zu sein. Zwar wurden Schwankungen der Spermiendichte im Hodensekret adulter Männchen festgestellt, diese konnten jedoch ebensowenig wie die ermittelten Hodengewichte saisonalen Mustern zugeord- net werden. Dagegen bestand eine signifikante Korrelation zwischen dem Hodengewicht und dem Konditionsindex, wobei adulte Männchen mit im Hodensekret enthaltenen Spermien höhere Kon- ditionsindizes aufwiesen. Saisonale Variationen der Konditionsindizes aller Otter für verschiedene Altersgruppen, Männchen oder Weibchen, konnten nicht beobachtet werden. Beziehungen zwischen der Hauptphase der Jungtieraufzucht und der Fischdichte der dänischen Gewässer waren auffällig. Letztere erreicht ihren Höhepunkt im Herbst, die Zeit des maximalen ener- getischen Bedarfs laktierender Otterweibchen. Somit ist anzunehmen, daß die Fischdichte einen be- deutenden Faktor für die Saisonalität der Reproduktion dänischer Otter darstellt. References ÄNSORGE, H.; SCHIPKE, R.; ZINKE, O. (1997): Population structure of otters, Lutra lutra. Parameters and model for a Central European region. Z. Säugetierkunde 62, 143-151. BEJA, P.R. (1996): Seasonal breeding and food resources of otters, Lutra lutra (Carnivora, Mustelidae), in south-west Portugal: acomparison between coastal and inland habitats. Mammalia 60, 27-34. ERLINGE, S. (1967): Home range of the otter Lufra lutra L. in southern Sweden. Oikos 18, 186-209. HEGGBERGET, T. M. (1988): Reproduction in the female Eurasian otter in central and northern Norway. J. Mammalogy 69, 164-167. HEGGBERGET, T. M.; CHRISTENSEN, H. (1994): Reproductive timing in Eurasian otters on the coast of Norway. Ecography 17, 339-348. JENSEN, A. (1964): Odderen i Danmark. Danske Vildtundersggelser 11, 1-64. KRUUK, H.; ConRoy, J. W. H.; MOORHOUSE, A. (1987): Seasonal reproduction, mortality and food of ot- ters (Lutra lutra L.) in Shetland. Symp. Zool. Soc. Lond. 58, 263-278. KRUUK, H.; ConRoy, J. W. H.; MOORHOUSE, A. (1991): Recruitment to a population of otters (Lutra lu- fra) in Shetland, in relation to fish abundance. J. Appl. Ecol. 28, 95-101. MADSEN, A.B. (1996): Odderens ZLutra lutra @kologi og forvaltning i Danmark Ph.D. thesis. National Environmental Research Institute. MADSEN, A. B.; RASMUSSEN, A.M. (1985): Reproduction in the Stone Marten Martes foina ın Den- mark. Natura Jutlandica 21, 145-148. Mason, C. F.; MACDONALD, S.M. (1986): Otters, conservation and ecology. Cambridge: Cambridge Univ. Press. Mason, C. F.; MADSEN, A.B. (1993): Organochlorine pesticide residues and PCBs in Danish otters (Lutra lutra). Sci. Total Environ. 133, 73-81. OFTEDAL, O. T.; GITTLEMAN, J. L. (1989): Patterns of energy output during reproduction in carnıvore. In: Carnivore Behaviour, Ecology and Evolution. Ed. by J. L. GITTLEMAn. London: Chapman and Hall. Pp. 355-378. SIDOROVICH, V. E. (1991): Structure, reproductive status and dynamics of the otter population in Byelo- russia. Acta Theriol. 36, 153-161. 200 M. ELMERoSs and A. B. MADSEN SIDOROVICH, V. E.; TUMANOv, I. L. (1994): Reproduction in otters in Belaruse and north-western Russia. Acta Theriol. 39, 59-66. SIÖÄsEn, T. (1997): Movements and establishment of reintroduced European otters Lutra lutra. J. Appl. Ecology 34, 1070-1080. STEPHENS, M.N. (1957): The otter report. London: Universities Federation for Animal Welfare. STRAND, O.; SKOGLAND, T.; Kvam, T. (1995): Placental scars and estimation of litter size: an experimental test in the arctic fox. J. Mammalogy 74, 1220-1225. STUBBE, M. (1969): Zur Biologie und zum Schutz des Fischotter Lutra lutra (L.). Arch. Naturschutz Landschaftsforsch. 9, 315-324. TAASTROM, H.-M., JACOBSEN, L. (1999): The diet of otters (Lutra lutra L.) in Danish freshwater habitats, compared to prey fish populations. J. Zool. (London) (in press). WIINGAARDEN, A. VAN; PEPPEL, M. VAN DEN (1970): De otter, Lutra lutra (L.) in Nederland. Lutra 12, 3-70. Author’s address MORTEN ELMEROS and AKSEL Bo MADSEn, National Environmental Research In- stitute, Department of Landscape Ecology, Grenävej 14, Kalß, DK-8410 Rande, Denmark Z. Säugetierkunde 64 (1999) 201-209 ZEITSCHRIFT ® Door © 1999 Urban & Fischer Verlag SÄUGETI ERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Otter Lutra lutra predation in cyprinid-dominated habitats By J. KLosKowskI Institute of Biology, Department of Nature Conservation, M. Curie-Sktodowska University, Lublin, Poland Receipt of Ms. 18. 08. 1998 Acceptance of Ms. 10. 03. 1999 Abstract Diet composition of otter Lutra lutra was quantified by analysis of 2048 spraints collected at half- month intervals concurrently with a one year culture cycle of carp Cyprinus carpio in a fish pond area (S.E. Poland) and compared with data on farm stocks. Presence of cultured fish affected the diet com- position, carp being the single most important food. The estimated biomass proportion of carp in spraints varied monthly between 15.7 % in June and 78.0 % in March. Carp abundance changes could not explain the seasonal variation in predation pressure. The factors responsible for the seasonal shifts in the use of resources are discussed. Key-words: Lutra lutra, predation, cyprinids, thiaminase, fish-ponds Introduction Freshwater populations of the Eurasian otter Lutra lutra (Linne, 1758) have suffered a substantial decline in many European countries, this being related to some aspects of hu- man management of aquatic ecosystems (Mason 1989). Although most of the man-made changes are regarded as detrimental to otters, some developments in freshwater habitats dominated by culture-based fisheries may contribute to an increase of food resources. Lo- cal strongholds for otter populations have been reported in association with aquaculture sites in central Europe (MAcpDonALD 1995), but the scientific data on the interactions be- tween these populations and farm stocks are lacking. There is a clear need for research on dietary patterns of otters in such man-modified habitats (MAson 1989), even more so because it is suggested that otter numbers are limited by fish populations (KrUUk et al. 1993). In central and eastern Europe aquaculture is dominated by cyprinids, stocked in large earthen ponds, which are very difficult to protect against predators. In Poland only the common carp C'yprinus carpio contributes to more than 95 % of the overall aquacul- ture production and there is growing concern over possible depredation of farm stocks by otters (DOBROWOLSKI et al. 1995). Cyprinid aquaculture sites create a specific habitat dominated by one prey category for otters. Abundance of food supply is governed by fish culture regime. Moreover, cypri- nids exhibit high thiaminase activity, an enzyme that catalyzes the cleavage of thiamine — vitamin B, (Harrıs 1951). The aim of this study was to quantify seasonal variation in ot- ter diet composition at cyprinid fisheries and to gather information on factors influencing the exploitation of farm stocks. The research was conducted in a carp-pond area in south- east Poland. 0044-3468/99/64/04-201 $ 12.00/0 202 J. KLOsKOwsKI Material and methods Study area The study areas were the Tysmienica and Prokop (hereafter Tysmienica) fish-ponds and adjoining bodies of water situated in the upper and middle regions of the Tysmienica River 22°50'-52’E, 51°32°- 33'N, Lublin Province, S.E. Poland, an area of about 2400 ha. Tysmienica valley is covered by grass- lands and bounded from the east by coniferous woodland. Its climate is influenced by continental fac- tors with marked seasonal temperature and percipitation changes. Semi-ıntensively cultivated ponds, up to 1km wide along the Tysmienica, are managed by two fish-farms, but spatially they constitute one complex. Farming comprised 19 ponds (1 to 18ha area, 1.0-1.5 m deep) with a total water surface of 127.5 ha, seven of them stocked with 1+ (fish in their sec- ond year of life) carp and the others with 0+ (fry) carp. The associated waterflows are up to 5m wide and up to 1.5m deep. Because of high bacteriological and organic contamination of the Tysmienica River the number of fish species has decreased dramatically during the last twenty years (RADWAN and GIRSZTOWTT 1994). Despite the obvious degradation of the Tysmienica valley, otters are widely reported by local fish- ermen to have occurred there for many years. During the present research no otter deterrence was undertaken by the fish-farms. American mink Mustela vison have not been recorded in the study area. Farm stocks As a warm-water species, carp plays a role in the riverine habitats of S.E. Poland only in artificially stocked sites. As otters normally spraint close to their feeding sites (KrRuuk 1995), it is most likely that spraints collected in the study area represented prey items consumed at Tysmienica ponds. How- ever, their ranges may have incorporated other carp farms located up to a few tens of km along the waterways away from Tysmienica. Research on otter predation covers a one-year seasonal cycle of carp (0+ and 1+) culture between April 1994 and March 1995, therefore data are presented in this order. The local fisheries’ managers provided information on fish supply for each pond. Assessment of carp abundance was based on the data from stocking and restocking operations, and routine net sampling twice a month between June and October. No fishing was undertaken between November and February as the ponds were tem- porarily frozen. The missing figures were estimated with the help of the fish-farm ichthyologists, as- suming a constant mortality rate and weight decrease of young carp during winter. The term ‘farm fish’ refers only to carp. The ‘unplanned breeding’ and the accompanying species in some of the fry ponds: wels Silurus glanis and pike Esox lucius fry (together 0.7 % of the total fish biomass at harvesting) were not taken into account in the statistical analysis. Access to ponds was assessed by measuring the ice cover and relating it to the whole pond water surface twice a month between November 1994 and March 1995. Spraint sampling and analysis Otter diet was assessed by spraint analysis. Spraints were collected at ponds and along water courses adjoining the ponds. Standard collection routes, totaling about 14 km, were visited twice each month between April 1994 and March 1995. Faecal material was wrapped in individual bags, dried, and washed through a 0.5 mm sieve. Prey remains were identified using a reference collection and existing keys. ‘Key bones’, the species-specific hard parts that occurred with the greatest frequency in spraints, were used to assess the size of prey items represented in the spraints. They were measured to the nearest 0.5 mm using an ocular micrometer. Key bones comprised pharyngeal teeth structures, maxil- lae, and dentaries (cyprinids, cobitids); preoperculae, operculae, and jaws (percids); operculae, pre- maxillae, and dentaries (gasterosteids) and dentaries, articularies, and operculae (other taxa). Mea- surements from individuals collected in the study area were regressed against original total fish lengths and these against fish fresh masses to produce conversion equations for carp (separately for both age classes) and species dominating in the otter diet. The relationships for less common fishes were taken from the literature, mainly Ligoıs et al. (1987). Amphibian key bones were ilea, frontoparietale, and jaws while body masses were estimated from Otter predation on cyprinids 203 a reference collection by adopting three weight classes for each species. Body lengths of crayfısh Asta- cus astacus individuals were estimated from chela widths and lengths of the anterior part of cepha- lothorax with the regressions of Popsıapro and OLEcH (1994) and length-weight relationship with equations from STYPINSKA (1978). Since it was often impossible to assess crayfish body length from broken pieces retrieved from spraints, data based on remains of predated crayfishes found on river banks were extrapolated to the entire half-month sample. Bird and mammal remains were assigned to the family level by examining hair and feather characteristics (Day 1966; Brom 1986). Mean weights of the most common species of these families observed at Tysmienica ponds, taken from CRrAMP and Sımmons (1977) and MyrcHA (1969) were applied. Prey proportions estimation Prey proportions in the diet were assessed by the method of relative weight percentages with total weight of individuals of a prey category, expressed as a percentage of the total weight of all prey indi- viduals (BEKKER and NoLer 1990). The number of individuals of a species represented in a spraint was scored as the highest total of any of the key bones present. Crayfish and bird numbers were defined as the highest totals of any of the identifiable parts present in the whole half-month sample. Insects smaller than 1 cm were considered an effect of secondary ingestion and omitted. Carp fraction in sam- ples was also expressed in terms of frequency of occurrence - the percentage of spraints where a prey category occurred (BEKKER and NoLET 1990). Monthly changes in calorific values of ‘averaged’ otter food per weight unit were estimated. Rela- tive calorific values of prey categories (Tab. 1) were used combined with data on digestibility (as the percentage of the total ingested food available to the predator as energy). Data on digestive effi- ciency (50 % for invertebrates, 70 % for fish and 80 % for other vertebrates) were assumed following BEJA (1996). Furthermore, prey species were divided into thiaminase-free and thiaminase-active. Data on thiaminase presence were obtained from published reviews (Tab. 1). Non-parametric tests were used throughout as most of the analyzed variables did not meet the as- sumptions of normality or homogeneity of variances. The trophic niche breadth was estimated using Levins’ measure B = 1/Zp;, where p; is the percent occurrence of a general prey taxa (Levins 1968). Table 1. Main food categories for otters in the Tysmienica valley: energy content and thiaminase pre- sence. Mean energy content of crayfish and fish prey after SCHERZ and SENSER (1989), except cobitids (KLEJMENoV 1962), ictalurids (FEUNTEUN and Marıon 1994) and gasterosteids (Massıas and BECKER 1990); values of other non-fish prey from GRIFFITHS (1977), except mammals (MyrcHA 1969). Some data have been recalculated to kJg' wet weight from the original sources. Data on thiaminase activ- ity after Harrıs (1951), TATARSKAYA et al. (1954) and ENDER (1966). If no relevant data for a given species were available from the literature, they were taken from the closest related taxa. Food items Energy content Antithiamine activity (kJg' wet weight) Insects Crayfish Pike Carp Other cyprinids Cobitids Ictalurids Gasterosteids Percids Amphibians Waterbirds Mammals (soricids) 204 J. KLOsSKOWSKI Results 8437 prey individuals were represented in 2048 spraints. Otters fed primarily on fish, composing more than 90 % of food in all months except in June-September (44.3 %- 73.6 %) and January (82.6 %). On an annual basis carp formed 40.8 % of the estimated consumed biomass being the major component in the otter diet during all months (Fig. 1) except June. However, its contribution to otter diet varied from the relatively lowest va- lues in summer (15.7 % by weight and 18.0 % by frequency in June) and in the coldest months, December-January (21.1 %-23.4% by weight and 17.2 %-19.7 % by frequency) to a peak in March when carp made up 78.0 % of the estimated consumed biomass and was found in 68.6 % of spraints. Monthly estimates of carp use in terms of biomass and frequency of occurrence were similar (r = 0.92, n = 12, p< 0.001). Percids (ruffe Gymnocephalus cernua and perch Perca fluviatilis), making up an an- nual mean of 13.7 % of biomass and cyprinds (mainly gudgeon Gobio gobio) with 12.3 % formed the largest proportion of wild fish recorded in the otter diet. Estimates of wild fish proportion ranged from 19.6 % in March to 72.3 %-59.2 % in December-January. Wild fish species represented in spraints were smaller than carp throughout the year. Even in June, when the difference was the smallest, carp was significantly larger (Mann- Whitney U-test; u = 323.0, p< 0.001). The estimated mean length of wild fish prey was 7.2+2.5 (SD) cm with the median length of 7.0 cm (n=5545). The mean length of carp was 11.4+2.6cm with the median of 10.9 cm (n = 1249). In summer, crayfish and waterbirds (Podicipedidae, Rallidae) constituted a substantial part of otter food. In June, crayfish was the principal prey species with 22.3%. Amphi- bians (frogs Rana ‘esculenta’, R. temporaria, R. arvalis) were taken by otters during all seasons, but made a significant contribution only in January reaching 16.5 %. Mammals and insects were of negligible importance (0.01 % on an annual basis). Carp abundance in terms of biomass varied up to 6.6-fold in the course of the year but these changes were not significantly correlated with estimates of the contribution to the diet of otters in terms of either weight (r=0.24, n=12, p= 0.44) or occurrence (r= 0.16, n= 12, p> 0.71; Fig. 2). There was an inverse correlation between proportion of carp in spraints and the extent of ice cover at ponds (r = -0.79, n=7, p<0.05 for both biomass and occurrence; Fig. 3), while positive and non-significant relationships were 100 # 80 BE Crayfish Waterbirds ii Amphibians um Wild fish AMJJASONDJF MM Cap Months Fig. 1. Monthly variation in the estimated proportion of major prey categories in otter diet in Tysmienica ponds area, April 1994-March 1995, quantified in terms of weight. Percentage of biomass Otter predation on cyprinids 205 80 O0 Percentage of m occurrence 30 BEE Farm stock biomass Percentage of carp in spraints 388588 Carp stocks (T) Near) »A, Sue N DE M Months Fig. 2. Frequency of occurrence (with 95 % confidence intervals) of carp in otter spraints and changes in the total abundance of farm stocks in the Tysmienica fish-ponds, April 1994-March 1995. — 100 100 = © ‚© SS Ö I > 6 co 8 = 8 Ss 40 | 107 5 ;= ) = 9 S 2 90 90 © I] Percentage by S © biomass e) 5 S ZA Frequency of Q A A A = Decl Janl Febl March| u DCZUreneS O DecIli Janli Febll —— Ice cover Spraint sampling period Fig. 3. Mean percentage of ice cover on the Tysmienica fish-ponds to the total water surface of the ponds in November 1994-March 1995 and variation in the proportion of carp in terms of biomass and frequency of occurrence in otter diet (half-month samples). Numbers over columns are sample sizes. found for wild fish (r = 0.57, p> 0.18 and r = 0.75, p > 0.05, respectively). After exclusion of the data from December-January, the period after establishment of persistent ice cov- er, the correlations between farm fish abundance and its proportion in otter diet were stronger, but still not significant (r = 0.61, n = 10, p > 0.06 for both methods of diet quanti- fication). Monthly changes in the calorific value of otter food per weight unit were inversely correlated with fluctuations in the proportion of thiaminase-free prey (r=-0.85, n=12, p < 0.001; Fig. 4). June-August was the period of the lowest calorific content of the otters’ food (the calorific equivalent of an ‘average’ digested 1 g dropped below 3.0 kJ), while the proportion of thiaminase-free prey was highest then, ranging from 41.4 % to 55.7 % of the consumed biomass. The energetic intake per ‘digested’ prey weight unit was great- est in February-March and November, when it reached 3.4kJg'. In these months the proportion of thiaminase-free prey was the lowest, ranging from 2.4 % to 4.4 %. 206 J. KLOSKOWSKI Ar DMDPOTZOR A ES IR SEN DE RE EREN Months Fig. 4. Monthly varıation in the digestible calorific value of otter ‘average’ food per weight unit and the proportions of thiaminase-free prey between April 1994 and March 1995. = 100 2 DS > ® = n == Thiaminase-free prey 3 o 80 ke) = *7 Energy content Paz 7 ° Q 60 5 = Oo = gs 4 5 5 o U © 29 © = Siruh, 2 5 08 7 = ® = Be) L O The trophic niche breadth varied throughout the year, otters specializing most in sum- mer (Levins’ B falling from 5.0 in June to 3.6 in August) with another low in March (B=4.9) while a wider range of items was eaten in December (B=8.8) and January (B = 9.2). The total biomass of farm fish was not significantly correlated with diet breadth (204050 125p 020): Discussion Assessing the proportion of prey categories in otter diet on the basis of faeces analysis must be regarded with caution. Carss and PArKInsoN (1996) and JACoBSEN and HANSEN (1996) pointed out some potential difficulties. The severest problem is the lack of statisti- cal independence of the data following errors in estimation of the number of fish re- corded per spraint and the number of droppings containing the remains of individual items. The applied recording procedure was chosen as a compromise between these de- mands. Identifying several different prey structures increases the likelihood of scoring a prey category (COTTRELL et al. 1995), while the probability of recording a single item re- peatedly, as may be common when using scales and vertebrae for quantifying prey num- ber (JaAcoBsEn and HAnSEN 1996), was minimized. The present results show that cultured stocks constitute an important food resource for otters. However, since carp fraction in the otter diet declined with the increasing pro- portion of ice cover at ponds, the advantages of using the additional food resource in the period of an ecological ‘bottleneck’ were reduced. Presumably better access to water under ice was provided at rivers, while water inlets were occasionally the only unfrozen places in the ponds. Farm fish was not taken in proportion to abundance. In April 41 % of the estimated otter diet consisted of carp, but only 26 % in August, despite the stock enhancement by a factor of 3.9. These considerations lack comparisons with the relative abundance of alter- native prey, as neither good density estimates nor data on seasonal changes in otter ranges were available. Still, information provided by the local fishermen indicate that the Otter predation on cyprinids 207 fish productivity of the Tysmienica at its upper and middle reaches was poor and varia- tions in wild fish abundance did not balance the changes in artificial food resources cre- ated by the vast pond complexes. This view gains some support from the fact that wild fish prey was markedly smaller than carps retrieved from spraints. A series of carp vulnerability periods related to fish culture regime may be respon- sible for these seasonal disproportions. Predation at carp farms may increase during stock translocations, as cultured fishes are more vulnerable when under stress asso- ciated with transport (OLLA et al. 1994). Fishes are also left unprotected due to lower- ing the water table during the few days of drawing down the ponds at harvests or re- stocking. In the study area these operations took place in March-April and September-October, and the proportion of carp in otter diet was relatively high in these months. Bad condition of carp after wintering may contribute to the peak of its exploitation in February-March. Another possible reason for the shifts in carp use by otters is meeting of nutrient and energy requirements at different seasons. Even 20% carp proportion in the diet can evolve clinical signs of thiamine deficiency in river otters Lutra canadensis after a long period (AurLerıich et al. 1995). Consumption of thiaminase-containing food by Tysmienica otters exceeded the above value by far throughout the year. Seasonal changes in the use of energy-rich prey coincided with inverse shifts in consumption of nutritionally valuable (thiaminase-free) food. In summer water temperatures are rela- tively high and some thiaminase-free prey types become available, whereas otters face hardship in winter as food requirements increase with decreasing water temperature (Kruuk 1995) and fish accessibility is restricted by ice cover. Shifts in prey choice, goVv- erned by a combined effect of seasonal food availability and its nutritional value, result in a balanced diet when it is averaged over long periods, but not in particular feeding periods (WestoßyY 1978). Although otters fed raw carp reject this food after a few days (S. SIKORA, pers. comm.), clinical symptoms of thiamine defficiency in captive animals are considerably delayed in time, dependent on the degree of thiamine deprivation and fat content in the diet (GErAcI 1974; AULERICH et al. 1995). Thus, under natural conditions switching to foods higher in thiamine may occur at longer intervals and thus produce a seasonal pattern. Still, spraint analysis allows no conclusions on the feeding behaviour of individual otters. Despite the presence of densely stocked ponds, a partial shift to non-fish food was ap- parent in summer. The same predation pattern was stated on the basis of a relatively small spraint sample (n = 74) in the study area a year later, in June-July 1995. In July 1995 crayfish constituted over half of the estimated otter diet by weight (J. KLoskowski, unpubl.). Similar dietary shifts were demonstrated in other eutrophic habitats of tempe- rate Europe (ErLINGE 1967; Wise et al. 1981) and interpreted as response to changes in fish dispersion and motility. However, carp stocks were regulated by transfers indepen- dent of natural seasonal population dynamics. Dispersion in densely stocked ponds was restricted. Moreover, in summer many carps were observed being torpid close to the water surface, probably due to temporal oxygen drops and were an easy prey for otters. By contrast, crayfish was rare in the canalised and polluted waterflows of the region (RADwan and GIRSZTOWTT 1994). In addition to the mentioned possible nutritional bene- fits, crayfish importance in the otter summer diet may be attributed to its increased activ- ity in this period (ErLINGE 1967). Preference for some alternative prey may have impor- tant implications for integrative management of aquatic habitats adjoining the fish farms, but its reasons have to be clearly determined. More information is needed on diet composition at individual levels and long-term numerical responses of otters to changes in prey availability in conditions of patchily dis- tributed monocultured fish supply. Considering the magnitude of cyprinid culture in cen- tral and eastern Europe and thriving otter populations in some parts of this region 208 J. KLosKowsKI (BRZEZINSKI et al. 1996), further progress to elucide the factors influencing otter depreda- tion of farm stocks may contribute to mitigation of large scale conflicts with aquacultural policy. Acknowledgements I gratefully acknowledge Ken IsHıı for assistance in both the field and laboratory, the managers E. LAGowskI and A. GAJDA for the permission to work on the fish-farms and the information on fish stocks, and J. IGSNnAToWIcz for his help in analysing the data on fish supply. Zusammenfassung Beuteerwerb des Fischotters Lutra lutra in von Weißfischen dominierten Habitaten Die Nahrung des Fischotters in einem Teichgebiet in Südostpolen wurde quantitativ untersucht, in- dem insgesamt 2048 Losungen, zweimal monatlich parallel zum alljährlichen Nutzungsmuster einer Karpfenzucht gesammelt und analysiert wurden. Die Zusammensetzung der Nahrung wurde mit Da- ten über den Fischbesatz verglichen. Die Präsenz der Karpfenbestände beinflußte die Zusammenset- zung der Nahrung mit Karpfen als dem wichtigsten Bestandteil. Der Biomasseanteil von Karpfen an der Gesamtnahrung variierte von 15.7 % im Juni bis 78.0 % im März. Die Änderungen des Angebotes an Karpfen konnten die jährliche Variation der Prädation nicht erklären. Die für die jahreszeitlichen Verlagerungen in der Ressourcennutzung verantwortlichen Faktoren werden diskutiert. References ÄULERICH, R. J.; Davıs, H. G.; BURSIAN, S. J.; SIKARSKIE, J. G.; STUHT, J. N. (1995): Suspected thiamine deficiency (Chastek’s paralysis) in northern river otter (Lutra canadensis). Scientifur 19, 297-304. BEJA, P.R. (1996): An analysis of otter Lutra lutra predation on introduced American crayfish Procam- barus clarkii in Iberian streams. J. Appl. Ecol. 33, 1156-1170. BEKKER, D. L.; NOLET, B. A. (1990): The diet of otters Lutra lutra in the Netherlands in winter and early spring. Lutra 33, 134-144. Brom, T. G. (1986): Microscopic identification of feathers and feather fragments of Palearctic birds. Bijdr. Dierk. 56, 181-204. BRZEZINSKI, M.; ROMANOWSKI, J.; CYGANn, J. P.; PABIn, B. (1996): Otter Lutra lutra distribution in Poland. Acta Theriol. 41, 113-126. Carss, D. N.; PARKINSON, S. G. (1996): Errors associated with otter Lutra lutra faecal analysis. 1. Asses- sing general diet from spraints. J. Zool. (London) 238, 301-317. COTTRELL, P. E.; TRITES, A. W.; MiILLER, E. H. (1995): Assessing the use of hard parts in faeces to identi- fy harbour seal prey: Results of captive-feeding trials. Can. J. Zool. 74, 875-880. CRAMB, S.; SIMMONS, K. E.L. (ed.) (1977): The Birds of the Western Palearctic. Vol. 1. Oxford: Oxford Univ. Press. Day, M. G. (1966): Identification of hair and feather remains in the gut of stoat and weasels. J. Zool. (London) 148, 201-217. DoBROWOLSKI, K. A.; BUKACINSKA, M.; BUKACINSKI, D.; CYGan, J. P; KACZMAREK, W. (1995): Environ- mental-Economic Evaluation of Fish Ponds in Poland. Warszawa: IUCN Poland. ENDER, F. (1966): Occurrence, properties, and physiological importance of thiaminase and other thia- mine antimetabolites in nutrition. In: B-Vitamine. Ed. by H.F. v. Kress and K. U. Brum. Stuttgart: F.K. Schattauer-Verlag. Pp. 149-164. ERLINGE, S. (1967): Food studies of the fish-otter Lutra lutraL. in south Swedish habitats. Viltrevy 4, 371-443. FEUNTEUN, E.; MARION, L. (1994): Assessment of grey heron predation on fish communities: The case of the largest European colony. Hydrobiologia 279/280, 327-344. GERACI, J. R. (1974): Thiamine deficiency in seals and recommendations for its prevention. J. Amer. Vet. Med. Ass. 165, 801-803. Otter predation on cyprinids 209 GRIFFITHS, D. (1977): Caloric variation in Crustacea and other animals. J. Anim. Ecol. 46, 593-605. HARRIS, R. S. (1951): Thiaminase. In: The Enzymes. Ed. by J. B. SUMNER. New York: Academic Press. Pp. 1186-1206. JACOBSEN, L.; HANSEN, H.-M. (1996): Analysis of otter (Lutra lutra) spraints: Part 1: Comparison of methods to estimate prey proportions; Part 2: Estimation of the size of prey fish. J. Zool. (London) 238, 167-180. KLEJMENOY, 1. J. (1962): Chimiceskij i vesovoj sostav ryb vodoemov SSSR i zarub£znych stran. Moskva: VNIRO. KRrUUkK, H. (1995): Wild Otters: Predation and Populations. Oxford: Oxford Univ. Press. KRUUK, H.; CARss, D. N.; CoNRoy, J. W. H.; Durgin, L. (1993): Otter (Lutra lutra L.) numbers and fish productivity in rivers in north-east Scotland. Symp. Zool. Soc. London 65, 171-191. Levins, R. (1968): Evolution in Changing Environment. Princeton: Princeton Univ. Press. Lisors, R. M.; HALLET-Lißois, C.; Rosoux, R. (1987): Elements pour l’identification des restes cräniens des poissons dulcaquicoles de Belgique et du Nord de la France. 1. Anguilliformes, GastE£rostei- formes, Cyprinodontiformes et Perciformes. Fiches Ost&eologie Animale Arch£ologie (ser. A) 3, 1-15. MACDONALD, S. M. (1995): Otter distribution in Europe. Cahiers d’Ethologie 15, 143-148. Mason, C. F. (1989): Water pollution and otter distribution: a review. Lutra 32, 97-131. Massıas, A.; BECKER, P. H. (1990): Nutritive value of food and growth in common tern Sterna hirundo chicks. Ornis scand. 21, 187-194. MYRCHA, A. (1969): Seasonal changes in caloric value, body water, and fat in some shrews. Acta Ther- iol. 16, 211-227. OLLA, B. L.; Davıs, M. W.; RyeEr, C. H. (1994): Behavioural deficits in hatchery-reared fish: Potential ef- fects on survival following release. Aquacult. Fish. Manage. 25, Suppl. 1, 19-34. PoDsIADto, E.; OLEcH, W. (1994): Variability of some body proportions in noble crayfish — Astacus asta- cus (L.) depending on the size of individuals and their sex. Ann. Warsaw Agricult. Univ.-SGGW, Anım. Sci. 30, 3-14. RADWAN, S.; GIRSZTOWTT, Z. (1994): Aktualny stan i problemy ochrony fauny wodnej woj. lubelskiego. In: Diagnoza stanu srodowiska przyrodniczego w wojewödztwie lubelskim. Ed. by T. WıLGAT and T. J. CHMIELEwSsKI. Lublin: TWWP. Pp. 125-151. SCHERZ, H.; SENSER, F. (1989): Food Composition and Nutrition Tables 1989/90. Stuttgart: Wiss. Ver- lagsges. STYPINSKA, M. (1978): Individual variabilities in absolute fertility of crayfish occurring in the waters of the Mazurian Lake district. Rocz. Nauk Roln. 98H, 177-203. TATARSKAYA, R. I.; KUBRYASHOV, I. B.; Fan, F.S. (1954): Raspostranienie i svojstva tiaminazy raznych Zivotnych form. Biokhimiya 19, 229-235. Wesrtoßy, M. (1978): What are the biological bases of varied diets? Am. Natur. 112, 627-631. Wise, M.H.; Linn, 1. J.;, KENNEDY, C.R. (1981): A comparison of the feeding biology of mink Mustela vison and otter Lutra lutra. J. Zool. (London) 195, 181-213. Author’s address: JANUSZ KrLosKkowskı, Department of Nature Conservation — Zaklad Ochrony Przyrody, Institute of Biology, M. Curie-Sktodowska University, Akademicka 19, 20-033 Lublin, Poland. & a: : m Z. Säugetierkunde 64 (1999) 210-219 ZEITSCHRIFT "FÜR © 1999 Urban & Fischer Verlag SÄUGETI ERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Contributions to the karyology and taxonomy of the genus Spalax Güldenstaedt, 1770 (Mammalia: Rodentia) in Turkey By M. SÖZEn, E. CoLAK, N. YiGit, S. ÖZKURT, and REYHAN VERIMLI Department of Biology, University of Ankara, Ankara, Turkey Receipt of Ms. 08. 02. 1999 Acceptance of Ms. 20. 04. 1999 Abstract The karyotypes of 17 specimens in 10 localities belonging to Spalax leucodon (Nordmann, 1840), and 2specimens of Spalax ehrenbergi Nehring, 1898 from Kilis in Turkey were analysed. It was deter- mined that $. leucodon has 2n=36 and NF=70 in the Bayındır population; 2n = 60, NF=82 and NFa = 78 ın Ankara (centrum, 15 km N, and 35 km S), Afyon 10 km E and Afyon 95 km SW popula- tions; 2n = 60, NF = 84 and NFa = 80 in the Burdur (centrum and 10 km W) population, and 2n = 60, NF = 76 and NFa = 72 in the Aksehir 10 km SE population. S. ehrenbergi from 15km E of Kilis has also 2n -S22 NE 7ArandıNEaA 70! According to these karyological findings the diploid chromosome number of the Bayındır popula- tion, and the NF and NFa values of Burdur, Aksehir, and Kilis populations are new for the genus Spa- lax ın Turkey. Key words: Spalax leucodon, Spalax ehrenbergi, karyology, taxonomy, Turkey Introduction The subterranean mole rats belonging to the family Spalacidae are distributed throughout southeastern Europe, Asıa Minor, Caucasus, Transcaucasus, Ukraine, Armenia, Syria, Pa- lestine, Iraq, Israel, Jordan, and northeastern Africa (OGneEv 1947; OnDrIAS 1966; Lay and NADLER 1972; CoRBET 1978; GIAGA et al. 1982; Nevo 1991: HARRISON and BATES 1991). To date, over 40 chromosomal forms of Spalax have been reported in the literature from these areas. According to the most recent morphological studies there are two species (S. leucodon and S.ehrenbergi) and nine subspecies (S. !. nehringi, S. I. armeniacus, S. 1. cilicicus, S. I. anatolicus, S. 1. turcicus, S. I. tuncelicus, S. e. intermedius, S. e. kirgisorum, and $. e. nevoi) of blind mole rats in Turkey (Kıvanc 1988; Coskun 1996, b). However, the results from karyological studies revealed nine karyological forms of S. leucodon (2n=38, 40, 50, 52, 54, 56, 58, 60, and 62) and four karyological forms of S. ehrenbergi (2n=52, 54, 56, and 58) in Turkey, and the number of chromosome arms (NF) for S. leucodon and S$. ehrenbergi varied from 70 to 82 and from 72 to 90, respectively (SoL- DATOVIc and Savıc 1978; Savıc and SoLDATovIıc 1979; YÜKSEL 1984; GÜLKAC and YÜKSEL 1989; YüRseL and GÜüLkAc 1992, 1995; Nevo et al. 1994, 1995; IvanıTsKAyA et al. 1997; Sö- ZEN and Kıvang 1998 a,b). NEvo et al. (1994, 1995) stated that each of the chromosomal forms is a separate biological species. They examined the populations using Nei’s genetic distance between populations obtained by allozyme electrophoresis and claimed that 0044-3468/99/64/04-210 $ 12.00/0 Karyology and taxonomy of Spalax in Turkey 2149 some populations having identical diploid chromosome numbers are different biological species, presumably representing about 20 such species in Turkey. They also showed that in Turkish Spalax, speciation and adaptation positively correlate with aridity stress and climatic unpredictability. 2n values and heterozygosity, 7, increase toward the ecologi- cally harsh, arid, and climatically unpredictable and geologically young central Anatolian Plateau from the west, north, south, and east (NEvo et al. 1994, 1995). The number of bio- logical species determined by a combination of chromosome number, genetic distances, and ecogeography of Spalax is tend to increase by new studies (SÖZEN and Kıvanc 1998 a, b). The results of these studies demonstrate the necessity to reexamine the species and subspecies specified morphologically, and to determine the borders of chromosomal forms. The aim of this present study is to give the karyologic characteristics of the blind mole rats collected from given localities, and thereby to contribute to karyology, taxon- omy, and speciation of the genus Spalax in Turkey. Material and methods The karyotypes of 1 female specimen from Bayındır (Izmir), 6 specimens (5 males, 1 female) from three localities in Ankara, 2 male specimens from 10 km SW of Afyon, 1 female specimen from 95 km SW of Afyon, 5 specimens (2 males, 3 females) from two localities in Burdur, and one female speci- men from 10 km SW of Aksehir belonging to Spalax leucodon (Nordmann, 1840), and 2 male speci- mens of Spalax ehrenbergi Nehring, 1898 from 15km E of Kilis in Turkey were analysed (Fig. 1, Tab. 1). Bayındır, Ankara, Afyon, Burdur, and Aksehir specimens were determined as S. leucodon, and specimens from Kilis as S. ehrenbergi. Karyotypes were prepared from bone marrow according to Forp and HAMERToN (1956), and about 25-30 metaphase cells, which were well-stained, and whose chromosomes were separate and distinct, were examined from each anımal. The diploid number of chromosomes (2n), the number of autosomal arms (NFa), and the total number of chromosomal arms (NF) were determined together with metacentric (m), acrocentric (a), subtelocentric (st), and subme- tacentric (sm) according to centromere positions, and sex chromosomes. The karyotype preparations and the animals examined were deposited in Department of Biology, Faculty of Science, University of Ankara. Fig. 1. Map of Turkey with localities of the analysed populations. 1. Bayındır, 2. Ankara, 3. Afyon 10 km W, 4. Afyon 95 km SW, 5. Burdur, 6. Aksehir, 7. Kilis 15 kmE. 212 M. Sözen etal. Table 1. The location and the number of anımals examined. Spalax leucodon (Nordmann, 1840) Locality Male Female Total Bayındır Ankara (Centrum) Ankara 1I5SkmN Ankara 35 km S Afyon 10kmE Afyon 95 km SW Burdur5SkmS Burdur 10 km W Aksehir 10 km SE | 3 2. D | DA WrHAamnND m Fa Spalax ehrenbergi Nehring, 1898 Kilis 15 km E 2 Results Spalax leucodon (Nordmann, 1840) Bayındır (Izmir) population: The karyotype of one female specimen from Bayındır in western Turkey was analysed. According to this analysis the Bayındır population has a karyotype of 2n=36, NF= 70. The karyotype contains 5 pairs of metacentric, 10 pairs of submetacentric, 2 pairs of subtelocentric, and 1 pair of acrocentric chromosomes (Fig. 2a). Ankara (centrum, 15 km N, and 35 km S), Afyon (10 km E, Afyon 95 km SW) popula- tions: We determined the karyotypes of these five populations as 2n=60, NF=82, and NFa = 78. The X chromosome is a medium-sized submetacentric, and the Y chromosome is the smallest subtelocentric. The autosomal set contains 10 pairs of subtelocentric, and 19 pairs of acrocentric chromosomes (Fig. 2b). Burdur 5km S and 10km W populations: The karyotype of Burdur populations has 2n=60, NF=84, and NFa = 80. The X chromosome is a medium-sized submetacentric, and the Y chromosome is the smallest subtelocentric. The autosomal set contains 11 pairs of subtelocentric and 18 pairs of acrocentric (Fig. 3a). Aksehir 19 km SE population: The Aksehir populations have a karyotype of 2n = 60, NF = 76, and NFa = 72. The X chromosome is a medium-size submetacentric, and the Y chromosome is the smallest subtelocentric. The autosomal set contains 7 pairs of subte- locentric and 22 pairs of acrocentric chromosoes (Fig. 3b). Spalax ehrenbergi Nehring, 1898 Kilis 15 km E population: The karyotype of 2 male specimens from 15 km east of Kilıis was examined. The karyotype contains of 2n=52chromosomes, NF=74, and NFa = 70. The X chromosome is a medium sized submetacentric and the Y chromo- some is acrocentric. The autosomal set of this population has 4 pairs of metacentric, 3 pairs of submetacentric, 3 pairs of submetacentric, and 15 pairs of acrocentric chromo- somes (Fig. 4). Karyology and taxonomy of Spalax in Turkey 213 HI KR AR aK an Ah na AR KR RR AK AN Lu TE oh Ah a AA Dh AR SA HD an a AN AN xx IN (1 nn AN an Nn ar nn nn nn un nn \y b SEN Fig. 2. The karyotype of a female Spalax leucodon from Bayındır (a), amale from Ankara (b). Discussion The first karyological analysis of the blind mole rat Spalax leucodon in Turkey was intro- duced by SoLDAToVIc and Savıc (1978) from the Thrace region of Turkey (Corlu and Ka- raevli), and corresponding investigations on the Asian part of Turkey (Havran and Selcuk) were performed by the same authors (Savıc and SoLpATovIc 1979). Later, new karyotypes were determined from the territory of Malatya by Yükseı (1984); from Malatya, Yazıhan, and Arguvan by GÜLkAc and YükseL (1989); from Kırsehir, Nevsehir, Kayseri, and Yoz- gat by YüksEL and GürkAc (1995); from Balıkesir, Izmir, Beysehir, Aydın, Erzurum, Sarıkamıs, Bolu, Bingöl, Denizli, Pınarbası, Malatya, Kütahya, Afyon, Konya, Sivas, An- kara, Kayseri, Havza, and Susehri by Nevo et al. (1994); from Malatya by IvAanıTsKAYA et al. (1997); from Sebil, Gülek by Sözen and Kıvanc (1998 a), and from Madenköy by Sö- zEN and Kıvanc (1998b). 214 M. Sözen etal. Ad 59 Kr ah Ar an an a AN Kr Ar ON AR an an na na an na aan an aa fan nn on eo.» gg nn Ww,) aa KA Ah hr NND MD nn an na an aa na MN na AR a0 00 no an nn an AN AR an 08° UX,\ b 6. Fig. 3. The karyotype of a female Spalax leucodon from Burdur (a), a male from Aksehir (b). The karyotype of $. ehrenbergi was first given by YükseL (1984) from Elazıg, then by YÜKsEL and GÜLkAg (1992) from Adıyaman, Hilvan, Suruc, and Gaziantep; by NEvo et al. (1994) from Diyarbakır, Urfa, Gaziantep, and Tarsus; by IvanıtskAyA et al. (1997) from Tarsus, Gaziantep, Elazıg and Urfa (Tab. 2). According to these studies the diploid karyotypes of Spalax leucodon in Turkey vary between 2n = 38 and 62, NF = 70 and 82, and NFa = 68 and 78. Spalax ehrenbergi also has a karyotype varying between 2n = 52 and 58, NF = 72 and 90, and NFa = 68 and 86. The karyotype of S.leucodon determined in this study from Bayındir, 2n=36, NF = 70 is a first record for $. leucodon populations in Turkey, and also for all Spalax po- Karyology and taxonomy of Spalax in Turkey 213 HH AR KK xx AN KR“ anid a“ aa aa na nrn nn nr ng» en NN Mn UI AN an nr nn a An Fig. 4. The karyotype of a male Spalax ehrenbergi from Kilıs. pulations. This karyotype pattern is smaller than in all the other populations in the distri- bution area of the Spalacidae. The karyotypes given by Savıc and SoLpATovIc (1979) from Havran and Selcuk, and by NEvo et al. (1994, 1995) from Balıkesir and Izmir are close to this karyotype. The diploid karyotypes of Ankara and Afyon populations determined here are identi- cal with the Arguvan population given by GÜLKAc and YÜksEL (1989) on the basis of chromosomal arm size and the chromosome morphology. These populations have 2n=60, NF=82 and NFa=78 containing 10 pairs of subtelocentric, 19 pairs of acro- centric autosomal chromomes, and a submetacentrice X chromosome. In contrast to our findings, Nevo et al. (1994, 1995) specified the diploid chromosome number as being 2n=62 from 30kmS of Ankara and 35 kmE of Afyon. This shows that two different chromosomal forms of Spalax are distributed in Ankara, and Afyon provinces. The diploid karyotype of Afyon, Ankara, Burdur, and Aksehir determined here is 2n=60, but the NF and NFa values are different (Tab. 2). NF = 84, NFa = 80 of the Bur- dur population, and NF = 76, NFa = 72 of the Beysehir population represents the first re- cords for Turkish Spalax. The diploid karyotype of S. ehrenbergi given by us is similar to the diploid chromo- some number but different, on the basis of chromosomal arm size and the chromosome morphology, from karyotypes given by YÜükseL (1984) from Elazıg by YÜKSEL and Gürkag (1992) from Adıyaman and Hilvan, by Nevo et al. (1994, 1995) from Diyarbakır and Urfa, and by Ivanıtskaya et al. (1997) from Birecik, Siverek, Diyarbakır and Elazıg, and from Urfa (Tab. 2). 216 M. Sözen etal. Table 2. Chromosomal records of Spalax leucodon (Nordmann, 1840) and Spalax ehrenbergi Nehring, 1898 from Turkey. *m: metacentric, sm: submetacentric, st: subtelocentric, a: acrocentric Spalax leucodon (Nordmann, 1840) Locality 2n NF NFa X Y Reference Corlu and Karaevli (in Thrace) SOLDATOVIC and Savıc (1978) Havran and Selcuk Savıc and SoLDAToVIc (1979) Malatya IvanITsKAYA et al. (1997) Malatya Yü&sEL (1984) Malatya and Yazıhan GÜLKAG and YÜRsEL (1989) Arguvan GÜLKAG and YÜRsEL (1989) Kırsehir, Nevsehir and Kayseri YÜKSEL and GÜLkAc (1995) Yozgat YÜKSEL and GÜLkAg (1995) Balıkesir and Izmir NEvo et al. (1994, 1995) Beysehir Nevo et al. (1994, 1995) Aydın, Erzurum Nevo et al. (1994, 1995) Sarıkamıs Nevo et al. (1994, 1995) Bolu and Bingöl NEvo et al. (1994, 1995) Denizli, Pınarbası Nevo et al. (1994, 1995) Malatya Nevo et al. (1994, 1995) Kütahya, Afyon, Konya, Sivas, Nevo et al. (1994, 1995) Ankara, Kayseri, Havza, Susehri Gülek SÖZEN and Kıvanc (1998 a) Sebil SÖZEN and Kıvanc (1998 a) Madenköy Sözen and Kıvanc (1998 b) Bayındır This study Ankara centrum, 15 km N, and This study 35kmS Afyon 95 km SW and 10 km E This study Burdur 5 km S and 10 km W This study Aksehir 10 km SE This study Spalax ehrenbergi Nehring, 1898 Locality 2n NF NFa X Y Reference Elazıg YükseL (1984) Adıyaman and Hilvan YÜKSEL and GÜLkAc (1992) Suruc YÜKSEL and GÜLkAc (1992) Gaziantep YÜKSEL and GÜLkAc (1992) Diyarbakır and Urfa NEvo et al. (1994, 1995) Gaziantep Nevo et al. (1994, 1995) Tarsus Nevo et al. (1994, 1995) Tarsus IvanıTsKAYA et al. (1997) Gaziantep IvanıtskAyA et al. (1997) Birecik, Siverek, Diyarbakır, Elazıg IvanıTsKayA et al. (1997) Urfa IvanıTskAYA et al. (1997) Kilis 15 km E This study The sex chromosomes are variable in both S$. leucodon and $. ehrenbergi. In most po- pulations of S. leucodon in Turkey, the X chromosome was described as being submeta- centric (SOLDAToVIc and Savıc 1978; YÜKSEL 1984; GÜLKAG and YÜKSEL 1989; YÜKSEL and GüÜLkAg 1995; IvanıTskAYA et al. 1997; SÖZEN and Kıvanc 1998, b), subtelocentric in two populations in western Turkey (Savıc and SoLpATovIc 1979), and metacentric only in the Gülek population (SöZEn and Kıvanc 1998 a). The Y chromosome is acrocentric (SOLDA- Karyology and taxonomy of Spalax in Turkey 217, tovıc and Savıc 1978; Savıc and SoLpATovIc 1979; SÖZEN and Kıvancg 1998 a, b), or subte- locentric (YÜKsEL and GÜLkAc 1995; YÜKSEL 1984; GÜLKAG and YÜKSEL 1989; SÖZEN and Kıvang 1998 a,b). We found that the X chromosome is also submetacentric in all popula- tions studied, and that the Y chromosome is subtelocentric. In $. ehrenbergi populations, the X chromosome is submetacentric (YÜKSEL 1984; IvanıtskAyA et al. 1997) or meta- centric (YÜKsEL and GÜLkAc 1992; IvanıTskAyA et al. 1997), and the Y chromosome is subtelocentric (YÜKSEL 1984; YÜKsEL and GÜLKAG 1992; IvanITskAyA et al. 1997). We de- termined that the X chromosome is submetacentric, and the Y chromosome is acrocentric in the Kilis population. The subterranean Spalacidae probably originated from a muroid-cricetoid stock in Asia Minor or vicinity, during Oligocene times, about 30-40 mya, and radiated adaptively underground in the Balkans, steppic Russia and Middle East, extending into North Africa (Savıc and NEvo 1990). The major important evolutionary feature was karyotypic evolution, mainly based on Robertsonian changes (Savıc and Nevo 1990). More than 40 karyotypes (2n = 38-62, NF = 72-124) occur across the eastern Mediterranean range of the family. Most karyo- types represent biospecies adapted to their different ecologies at multiple organizational levels. Three major chromosomal trends (2n = 38-62) occur across the entire Eurasian and east Mediterranean range of Spalacidae, all starting in Western Turkey. These in- volve: (1) the Near East and North Africa (2?n =38— 62); (2) the Balkans (?2n = 38 > 62), and (3) the Ukrainian and Russian steppes, 2n = 38 > 62 (NEvo 1991; NEvo et al. 1995). This trend has also been revealed in Turkey itself supporting the idea of ecologi- cal speciation via chromosome speciation (NEvo et. al 1995). The chromosome number of Spalax tends to increase during adaptive radiation from humid areas toward the ecologi- cally harsh, arid, and climatically unpredictable and geologically young central Anatolian Plateau from all directions (Nevo et al. 1994, 1995). The Bayındır population (2n = 36, NF = 70) determined in this study is now acceptable to be the speciation center of Spala- cidae. But this concept should be supported by new molecular and fosil findings. Nevo et al. (1995) determined 10 karyotypes and probably more than 20 new species, based on karyotypes and genetic distances, to the two superspecies leucodon and ehren- bergi, in Turkey. Later Sözen and Kıvanc (1998 a, b) determined three new karyotypes in the superspecies leucodon. In the present study, we have determined four extra karyolo- gic forms for the two superspecies leucodon and ehrenbergi, one of them has a new 2n value (Bayındır population), and three of them have new NF and NFa values (Burdur, Aksehire, and Kilis populations). These results bring the total number of karyological forms or biospecies of Turkish Spalax to approximately 30. According to the findings mentioned above, the borders of the areas of all chromoso- mal forms described from Turkey are not definite because of the possibility of the exis- tence of new localities and chromosomal forms. It will certainly be necessary to analyse greater numbers of populations of the blind mole rats to determine the borders of the areas of all the described chromosomal variations from Turkey and to find possible new karyological forms and thereby to explain precisely speciation, phylogeny, systematics, and the evolutionary history of Spalacidae in Turkey. Zusammenfassung Zur Karyologie und Taxonomie der Gattung Spalax Güldenstaedt, 1770 (Mammalia: Rodentia) in der Türkei Die Karyotypen von 17 Individuen der Art Spalax leucodon (Nordmann, 1840) aus 10 Probengebieten in der Türkei sowie von 2 Individuen der Art Spalax ehrenbergi Nehring, 1898 aus Kilis, Türkei, wur- 218 M. Sözen etal. den analysiert. 5. leucodon zeigte 2n = 36 und NF = 70 in Bayindir; 2n = 60, NF = 82 und NFa = 78 in Ankara (Zentrum, 15kmN und 35kmS), Ayfon (10km O und 95 km SW); 2n = 60, NF= 84 und NFa=80 in Budur (Zentrum und 10kmW) und 2n=60, NF=76 und NFa=72 in Aksehir (10km SO). S. ehrenbergi aus Kilis (15km OÖ) zeigte 2n =52, NF= 74 und NFa = 70. Nach unseren karyologischen Ergebnissen weisen die diploide Chromosomenzahl von Individuen aus Bayindir so- wie die NF und NFa bei Individuen aus Budur, Aksehir und Kilis bei der Gattung Spalax in der Tür- kei bisher nicht beobachtete Werte auf. References CoRBET, G. B. (1978): The Mammals of the Palaearctic region: A taxonomic Review. London: British Museum (Nat. Hist.). CosKun, Y. (1996): A new subspecies of Spalax nehringi (Satunin, 1898) (Rodentia: Spalacidae) from Turkey. Säugetierkdl. Mitt. 37, 103-109. CosKun, Y. (1996 b): Spalax nehringi nevoi, a new mole rat from southeast Anatolia, Turkey (Rodentia: Spalacidae). Säugetierkdl. Mitt. 38, 135-142. FORD, C. E.; HAMERTON, J. L. (1956): A colchicine hypotonic citrate, squash for mammalian chromo- somes. Stain Technol. 31, 247-251. G1AGA, E.; Savıc, I.; SoLDAToVIc, B. (1982): Chromosomal forms of the mole rat Microspalax from Greece and Turkey, Z. Säugetierkunde 47, 231-236. GÜLKAG, M. D.; YüksEL, E. (1989): Malatya yöresi kör fareleri (Rodentia: Spalacidae) üzerine sitogen- etik bir inceleme. Doga Tu. Biyol. D. 13, 63-71. HARRISON, D. L.; BATES, P. J. J. (1991): The Mammals of Arabia. Kent, England: Harr. Zool. Mus. IvANITSKAYA, E.; CosKUn, Y.; NEVo, E. (1997): Banded karyotypes of mole rats (Spalax, Spalacidae, Ro- dentia) from Turkey: a comparative analysis. J. Zool. Syst. Evol. Research 35, 171-177. Kıvang, E. (1988): Türkiye Spalax larının Cografik Varyasyonları. Ankara: 72, Teksir-Daktilo-Fotokopi Press. Lay, D. M.; NADLER, C. F. (1972): Cytogenetics and origin of North African Spalax (Rodentia: Spalaci- dae). Cytogenetics 11, 279-285. NEvo, E. (1991): Evolutionary theory and processes of active speciation and adaptive radiation in sub- terranean mole rats, Spalax ehrenbergi superspecies in Israel. Evol. Biol. 25, 1-125. NEVvo, E.; FiLıppuccı, M. G.; REDI, C.; KoRoOL, A.; BEILES, A. (1994): Chromosomal speciation and adap- tive radiation of mole rats in Asia Minor correiated with increased ecological stress. Proc. Natl. Acad. Sci. USA 91, 8160-8164. NEvo, E.; FıLıppuccı, M. G.; REDI, C.; Simson, S.; HETH, G., BEILES, A. (1995): Karyotype and genetic evolution in speciation of subterranean mole rats of the genus Spalax in Turkey. Biol. J. Linnean Soc. 54, 203-229. OGney, S. I. (1947): Mammals of the U.S.S.R. and Adjacent Countries. Vol. V. Rodents. Moskova: Izda- tel’stvo Akademii Nauk. ONDRIAS, J. C. (1966): The taxonomy and geographical distribution of the rodents of Greece. Säuge- tierkdl. Mitt. 14, 1-136. Savıc, 1.; NEvo, E. (1990): The Spalacidae: Evolutionary history, speciation, and population biology. In: Evolution of Subterranean Mammals at the Organismal and Molecular Levels Ed. by E. NEvo and O. A. Reıc. New York: Alan R. Liss. Pp. 129-143. Savıc, 1.; SOLDAToVIc, B. (1979): Contribution to the knowledge of the genus Spalax (Microspalax) Kar- yotype from Asia minor. Arh. biol. nauka. (Beograd) 31, 1-2. SOLDATOVIC, B.; Savıc, I. (1978): Karyotypes in some populations of the genus Spalax (Mesospalax) in Bulgaria and Turkey. Säugetierkdl. Mitt. 26, 252-256. SÖZEN, M.; Kıvanc, E. (1998a): Two new karyotypic forms of Spalax leucodon (Nordmann, 1840) (Mammalia: Rodentia) from Turkey. Z. Säugetierkunde 63, 307-310. SÖZEN, M.; Kıvang, E. (1998 b): A new karyotype of Spalax leucodon cilicicus Mehely, 1909 (Mamma- lıa: Rodentia) from type locality in Turkey. Isr. J. Zool. 44, 53-56. YükSEL, E. (1984): Cytogenetic study in Spalax (Rodentia: Spalacidae) from Turkey. Communications, C: Biologie 2, 1-12. Karyology and taxonomy of Spalax in Turkey 219 YÜKSEL, E.; GÜLKAG, M.D. (1992): On the karyotypes in some populations of the subterranean Mole Rats in the Lower Euphrates Basin, Turkey. Caryologia 45, 175-190. YÜKSEL, E.; GÜLKAG, M.D. (1995): Kızılırmak Havzası Kayseri-Kırsehir-Nevsehir- Yozgat Kesimi Spa- lax populasyonları üzerine Sitolojik Incelemeler. Ankara: TBAG-904 Tübitak press. Authors’ address: MUSTAFA SÖZEN, ERCÜMENT COLAK, NURi YiGit, and REYHAN VERIMLi, Department of Biology, University of Ankara, 06100 Besevler Ankara, Turkey; SAKiR OZKURT, Department of Biology, University of Gazi, 40100 Kırsehir, Turkey. Z. Säugetierkunde 64 (1999) 220-229 ZEITSCH RIFT® u © 1999 Urban & Fischer Verlag SÄUG ET|I ERKUN DE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Karyotype relationship among four species of Spiny mice (Acomys, Rodentia) By BÄRBEL KUNZE, F. DIETERLEN, W. TRAUT, and H. WINKING Institut für Biologie, Medizinische Universität zu Lübeck, Lübeck; Staatliches Museum für Naturkunde Stuttgart, Stuttgart, Germany Receipt of Ms. 02. 11. 1998 Acceptance of Ms. 24. 03. 1999 Abstract G-banded karyotypes of four species of the genus Acomys from different localities are described: Acomys cahirinus from Egypt, A. cineraceus from the Sudan, A. dimidiatus from Israel and A. minous from Crete. Diploid chromosome numbers vary considerably among these species, ranging from 2n =36 to 2n=50. The variation is due to Robertsonian fusions and/or fissions. We constructed a maximum parsimony tree using the common Robertsonian metacentrics as characters. Key words: Acomys, karyotype, Robertsonian chromosome, phylogeny Introduction The genus Acomys (spiny mice) has a wide geographical distribution from South Africa to southwest Asıa and as far north as Crete and Turkey. Preferring rocky habitats, Ac- omys displays a patchy distribution with many populations isolated on cliffs and rocky hills (VoLOBoUEV et al. 1991). Currently, 52 taxa are assigned to the genus Acomys. They have been assembled into a variety of subgenera, species groups, and species (BATES 1994). Due to the absence of unequivocal diagnostic markers, the taxonomic classification changed several times. The proposed number of species ranges from 14 (Musser and CARLETON 1993) to 38 (ELLERMANN 1941). We follow the classification of Denys et al. (1994) who gave diagnoses based on skull morphology, dental pattern, allozyme patterns, and karyotypes. VoLOBOUEV et al. (1991, 1996a, b) applied cytotaxonomy to clarify the confusing sys- tematics of this group. Karyotypes display a wide variation among species. Diploid chro- mosome numbers from 2n = 36 to 2n = 68 have been found. The variation is presumed to be mainly due to centric fusion and fission events (Robertsonian translocations). In this study we describe the G-band karyotypes of four species of Acomys and draw phylogenetic conclusions from the presence of common Robertsonian (Rb) chromo- somes. Material and methods Animals Our specimen of A. minous (Bate, 1905) was from Crete. Specimens from A. cahirinus (Desmar- est, 1819) were kindly provided by the Zoological Garden of Cairo. Animals originally derived from Abu Rawash (Egypt) had yellow coat colour, those from Kardasa (Egypt) were grey. A. cineraceus 0044-3468/99/64/04-220 $ 12.00/0 Karyotype relationship among four species of Acomys ZA (Fitzinger and Heuglin, 1866) was from two sources: the Blue Nile province (Sudan) and the province of Kordofan (Sudan). The Kordofan specimens were kindly provided by the late Dr. JocHEN NIETHAM- MER (Bonn). A. dimidiatus (Cretzschmar, 1826) was trapped in Jerusalem by the late Dr. ALFRED Gropp (Lübeck) and by Dr. JacoB WAHRMAN (Jerusalem). The A. minous (?) x A. dimidiatus (3) hy- brid had been bred and kindly supplied by Dr. JACOB WAHRMANN. Mitotic chromosomes Metaphase chromosomes were G-banded according to SEABRIGHT (1971). At least 10 metaphases from each species were analysed. Meiotic chromosomes The ovaries of the hybrid animal were dissected out and placed into cell culture medium (TC199). Oocytes were released by puncturing mature follicles with fine needles. The oocytes were collected and transferred into a petri dish with fetal calf serum. They were incubated at 36°C for about six hours, then in 1% Na-citrate for about 15 minutes. After transfer to clean slides in a drop of hypotonic solution, spreading and fixation were achieved by dropping methanol:acetic acid (3:1) onto the cells. Chromosomes were stained with orcein. Results and discussion Karyotypes The A. cahirinus specimens had 18 chromosome pairs (2n = 36), 16 of which were meta- centric (Fig. 1). Only the smallest pair of autosomes (no. 17) was acrocentric. Karyotypes of individuals from Abu Rawash and Kardasa were identical. The karyotypes are in keep- ing with the R-band karyotype published by VoLOBoUEV (1996 b). The A. minous complement consisted of 19 chromosome pairs (2n = 38). Of the auto- some pairs, 15 were metacentric and three acrocentric (Fig. 2). According to MATTHEY (1963), the A. minous karyotype is polymorphic. It contains 14 pairs of metacentrics and either four or five pairs of acrocentric autosomes. Our A. minous karyotype may have been derived from the latter one by a Robertsonian fusion (or the latter one from our karyotype by a fission event). A. cineraceus from the Blue Nile province (Sudan) was polymorphic. The chromo- some complement consisted of 25 pairs (2n = 50) with nine metacentric ones (Fig. 3) or of 24 chromosome pairs (2n = 48), among which ten were metacentric. One of the latter me- tacentrics was related to the acrocentric chromosomes no. 16 and no. 18 (indicated with asterisks in Fig. 3) by a Robertsonian fission/fusion event. The chromosome complement of specimen from Kordofan was identical to the 2n = 48 karyotype. The A. dimidiatus specimen from Israel had 19 chromosome pairs (?2n = 38), 16 of which were metacentric (Fig. 4). The same number of chromosomes was found in A. cf. dimidiatus from Saudi Arabia by VOLOBOUEV et al. (1991). The sex chromosomes of all four Acomys species were acrocentric. Common Robertsonian metacentrics To investigate the phylogenetic relationship, we compared the mitotic karyotypes of the four species. Single arm comparisons proved unsatisfactory as the low number of bands did not allow safe identification. We were in a better position with whole Robertsonian metacentrics (Rbs). They afforded the combined banding pattern of the two arms and the DR BÄRBEL Kunze etal. ne # 11 12 13 14 15 An a F 16 17 XY Fig. 1. G-band karyotype of Acomys cahirinus $ from Abu Rawash. Chromosomes were arranged ac- cording to size within each of three groups, metacentric autosomes, acrocentric autosomes and sex chro- mOosomes. centromere position as characters for identification. We focussed, therefore, on the recog- nition of common Rbs among the four species. Several Rbs common to two or more species have been identified (Tab. 1). The karyo- types of A. cahirinus and A. minous had 15 Rbs in common (Tab. 1: A-O). The arms of A. cahirinus Rb chromosome no. 1 had their counterparts in the A. minous acrocentrics no. 16 and no. 17. Three Rbs (D, L, O; Fig. 5, Tab. 1) from A. cahirinus were found in A. minous and A. cineraceus, another Rb (G; Fig.5, Tab. 1) in A. cahirinus, A. minous, and A. dimidiatus. Our identification of one common Rb in A. minous and A. dimidiatus was supported by diakinesis figures from an A. minous (2) x A. dimidiatus (3) hybrid. The ooyctes of Karyotype relationship among four species of Acomys 223 16 uf 18 AY Fig. 2. G-band karyotype of A. minous 3. Arrangement of chromosomes as in Fig. 1. the hybrid displayed three bivalents and two multivalent chains (Fig. 6). Most probably, the smallest bivalent (Fig. 6, arrowhead) consists of the paired acrocentrics no. 18 from A. minous and no. 18 from A. dimidiatus while the larger ring bivalents (Fig. 6, arrows) are the X chromosome bivalent and the bivalent of the only common Rb (G, Tab. 1). The two multivalent chains comprise all remaining chromosomes. Such synaptic chains are well known from hybrids between different Rb chromosome races, e.g., from Mus muscu- lus (JoHANNISSoN and WinkInG 1994). They are characteristic for pairing of Rbs with al- ternating arm composition. The two chains in the Acomys hybrid are, therefore, evidence for the presence of Rbs with non-homologous arm composition in the parent species. The chains are terminated at their ends by 3 acrocentrics (two from A. minous and one from A. dimidiatus) and one small A. dimidiatus metacentric chromosome which, according to VOLOBOUEV et al. (1996a) originated from an acrocentric one by an inversion. Hence, the pairing figures are fully accounted for. MATTHEY (1963) published a meiosis figure from a hybrid which was designated A. cahirinus (2) x A. minous (3) but which according to the classification of AL-SALEH (1988) is to be considered an A. dimidiatus (2) x A. minous ($) hybrid. Thus, this is in 224 BÄRBEL Kunze etal. 1 3 4 5 vd 6 7% 8 9 eu 2 > 34 Br Mr ar eu: ce: IE 2 N; %8 Bu Si “a 20 21 22 23 24 xı Fig. 3. G-band karyotype of A. cineraceus 3 from Blue Nile. Arrangement of chromosomes as in Fig. 1. fact the reciprocal hybrid to the one described by us and it confirms the conclusions drawn. The only apparent difference is the formation of three instead of two multivalent chains. This agrees with the presence of five instead of three acrocentric autosomes in the A. minous parent karyotype. In contrast to us, VOLOBOUEV et al. (1996 b) did not find a common Rb in the comple- ments of A. cahirinus and A. dimidiatus. This may have been due to a difference between populations. The animal investigated by VoLOBOUEV et al. (1996b) came from Saudi Ara- bıa, while our A. dimidiatus specimen was from Israel. Karyotype relationship among four species of Acomys 225 ‚Miu.g, Ye 94 u‘ Tome? L IIZ 12 [ 172.35 - D 0%) PN 10] ri 4 .“ ix $ 1 1 12 13 14 15 * 2. 16 Hi 28 m‘ 3 17 18 XXX Fig. 4. G-band karyotype of A. dimidiatus $. Arrangement of chromosomes as in Fig. 1. Karyotype evolution The presence of common Rbs (Tab. 1) was translated to a maximum parsimony tree (Fig. 7). The only inconsistant character with respect to the tree topology is Rb G. This Rb was either present at the root of the tree and lost in the A. cineraceus branch (Fig. 7a) or it was independently created by fusion in the A. dimidiatus and in the A. cahirinus/A. minous branch (Fig. 7b). We favour the interpretation represented in fig- ure 7a. The presence of a high number of chromosomes considered, the probability to create the same Rb in independent translocation events is low. According to our dendrogram A. cahirinus and A. minous are the closest relatives among the four species. According to morphological parameters A. minous was treated as a separate species which belongs to the A. cahirinus-A. dimidiatus group (CoRBET and Hırr 1991; Denys et al. 1994; DiETERLEN 1978). Morphologically A. cahirinus and 226 BÄRBEL Kunze etal. “7 » 53 x m Rs = D — DD 0) RbL Rb O Fig. 5. Common Rb chromosomes in three Acomys species. Designation of Rbs according to table 1. Chromosomes from A. cahirinus (1), A. minous (2), A. cineraceus (3), and A. dimidiatus (4). Table 1. Rb chromosomes common to at least two Acomys species (numbering of chromosomes ac- cording to figures 1-4). A. cahirinus no. A. minous no. A. cineraceusno. A. dimidiatus no. Designation N OOo own puma Hm ES m A B C D BE F G H I J K L M N 10) PrapRpj nr whN A. dimidiatus appear to be the closest relatives, they are nearly indistinguishable (VoLo- BOUEV et al. 1996b). The distance of A. cahirinus and A. dimidiatus is confirmed, how- ever, by a comparison of satellite DNA sequences (Kunze et al. 1999). The evolutionary history of the genus Acomys is not well documented, but the start- ing point appears to be tropical Africa with the oldest fossil records being about 7-8 my old (Denys et al. 1994). Acomys species with different Rbs were assumed by VOLOBOUEV et al. (1996a) to be derived from a common ancestor with a karyotye of 68-70 acro- centric chromosomes. DEnYs et al. (1994) assumed that this hypothetic common ancestor spread slowly over the continent and reached North Africa about 120000 years ago and Karyotype relationship among four species of Acomys 22], Fig. 6. Diakinesis figure of an A. minous (2) x A. dimidiatus (3) female hybrid. Arrowhead: bivalent of two acrocentric autosomes; arrows: ring bivalents of X/X and Rb G; m: short multivalent chain; M: long multivalent chain. A. cahirinus A. minous A. cineraceus A. dimidiatus A. cahirinus A. minous A. cineraceus A. dimidiatus Fig. 7. Maximum parsimony tree, based on common Rbs. a) and b) differ with respect to the interpreta- tion of G. Fusion events are indicated by resulting Rbs (A, B...O). Fission of Rb G is indicated by G”. appeared in Israel about 40000 years ago (TCHERNOV 1968). The Cretan population was assumed to be founded by an ancestor with an all acrocentric karyotype (DeEnys et al. 1994). From the chromosome tree of the four species we infer a different scenario of karyo- type diversification. A common ancestor with a nearly all acrocentric karyotype - it prob- ably contained Rb G - invaded Egypt about 120000 and Israel about 40 000 years ago. Waves of Robertsonian fusion started independently in Northern Africa and Israel. They led to a nearly all-metacentric state in the two species A. cahirinus and A. dimidia- 228 BÄRBEL KUNZE etal. tus. Since the karyotypes from the Cretan species A. minous is very similar to that of the Egyptian species A. cahirinus we assume that Crete was populated by Egyptian Acomys after fixation of most Rb chromosomes. Acomys may have reached Crete by ship with hu- mans. This assumption is in accordance with the lack of fossil Acomys records in Crete (DIETERLEN 1978). A corollary of this scenario is that Rb formation and fixation was a rapid process, at least in the Israelian spiny mice. The rate of Rb fixation (15 Rbs in 40000 years) is, how- ever, in accordance with estimates by FErrıs et al. (1983). They estimated the rate of Rb fixation in Mus musculus to be 1 per 1000 years. Acknowledgements The excellent technical assistance of C. REUTER is gratefully acknowledged. Zusammenfassung Karyotypische Verwandtschaftsbeziehungen zwischen vier Arten der Gattung Acomys (Rodentia) In dieser Arbeit werden G-Banden Karyotypen von vier verschiedenen Acomys Arten aus verschie- denen Regionen beschrieben: Acomys cahirinus aus Ägypten, A.cineraceus aus dem Sudan, A. dimidiatus aus Israel und A. minous von Kreta. Die diploide Chromosomenzahl variiert erheblich zwischen diesen Arten und reicht von 2n = 36 bis 2n = 50. Die Unterschiede werden im Wesentlichen durch Robertson-Translokationen verursacht. Wir benutzen das Auftreten von gemeinsamen Robert- son-Chromosomen in verschiedenen Arten als Merkmale zur Konstruktion eines Stammbaums. References AL-SALEH, A. A. (1988): Cytological studies of certain desert mammals of Saudi Arabia. 6. First report on chromosome number and karyotype of Acomys dimidiatus. Genetica 76, 3-5. BATE D. M. A. (1905): On the mammals of Crete. Proc. Zool. Soc. (London) 2, 315-323. BATES, P. J. J. (1994): The distribution of Acomys (Rodentia: Muridae) in Africa and Asia. Israel J. Zool. 40, 199-214. CoRBET, G. B.; Hırı, J. E. (1991): A World List of Mammalian Species. 3rd ed. Oxford: Oxford Univ. Press. CRETZSCHMAR, J. (1826): Säugethiere. In: Atlas zu der Reise im nördlichen Afrika von Edouard Rüp- pell. Erste Abteilung - Zoologie. Frankfurt: Brönner. Pp. 1-78. DENnYS, C.; GAUTUN, J.-C.; TRANIER, M.; VOLOBOUEY, V. (1994): Evolution of the genus Acomys (Roden- tia, Muridae) from dental and chromosomal patterns. Israel J. Zool. 40, 215-246. DIETERLEN, F. (1978): Acomys minous (Bate, 1905) -— Kreta-Stachelmaus. In: Handbuch der Säugetiere Europas. Vol. 1/I Ed. by J. NIETHAMMER and F. Krapp. Wiesbaden: Akad. Verlagsges. Pp. 452-461. ELLERMANN, J. R. (1941): The Families and Genera of Living Rodents. Vol. 2. Muridae. London: British Museum (Natural History). FITZINGER, L. J.,; HEUGLIN, T. V. (1866): Systematische Übersicht der Säugethiere Nordost-Afrika’s mit Einschluss der arabischen Küste, des rothen Meeres, der Somalie- und der Nilquellen-Länder, süd- wärts bis zum vierten Grade nördlicher Breite. Sitzungsber. K. Akad. Wiss. Wien. 54, 537-611. FERRIS, S. D.; SAGE, R. D.; PRAGER, E. M.; RıITTE, U.; WıLson, A. C. (1983): Mitochondrial DNA evolu- tion in mice. Genetics 105, 681-721. JOHANNISSON, R.; WinkInG, H. (1994): Synaptonemal complexes of chains and rings in mice heterozy- gous for multiple Robertsonian translocations. Chromosome Res. 2, 137-145. Kunze, B.; TRAUT, W.; GARAGNA, S.;, WEICHENHAN, D.; REDI, C. A.; WiınkiIng, H. (1999): Pericentric sa- tellite DNA and molecular phylogeny in Acomys (Rodentia). Chromosome Res. 7, 131-141. MATTHEY, R. (1963): Polymorphisme chromosomique intraspecifique et intraindividuel chez Acomys minous Bate (Mammalia-Rodentia-Muridae). Chromosoma 14, 468-497. Karyotype relationship among four species of Acomys 229 MUusseER, G. G.; CARLETON, M. D. (1993): Rodentia: Sciurognathi: Muridae: Murinae. In: Mammalian Species of the World. 2nd ed. Ed. by D.E. Wırson and D. A. M. REEDER. Washington, London: Smithsonian Institution Press. Pp. 501-755. SEABRIGHT, M. (1971): A rapid banding technique for human chromosomes. Lancet 2, 971-972. TCHERNOV, E. (1968): Succession of Rodent Faunas during the upper Pleistocene of Israel. Hamburg: Verlag Paul Parey. VOLOBOUENY, V. T.; TRAINIER, M.; DUTRILLAUX, B. (1991): Chromosome evolution in the genus Acomys: Chromosome banding analysis of Acomys cf. dimidiatus (Rodentia, Muridae). Bonn. zool. Beitr. 42, 253-260. VOLOBOUENY, V.; GAUTUN, J.-C.; SICARD, B.; TRANIER, M. (1996a): The chromosome complement of Acomys spp. (Rodentia, Muridae) from Oursi, Burkina Faso - the ancestral karyotype of the cahiri- nus-dimidiatus group? Chromosome Res. 4, 526-530. VOLOBOUEY, V. T.; GAUTUN, J.-C.; TRANIER, M. (1996 b): Chromosome evolution in the genus Acomys (Rodentia, Muridae): chromosome banding analysis of Acomys cahirinus. Mammalia 60, 217-222. Author’s addresses: Dr. BÄRBEL KUNZE, Prof. Dr. WALTHER TRAUT, Prof. Dr. HEINZ WinKINGg, Institut für Biologie, Medizinische Universität zu Lübeck, Ratzeburger Allee 160, D-23538 Lübeck; Dr. FrıTZ DIETERLEN, Staatliches Museum für Naturkunde Stuttgart, Rosenstein 1, D-70191 Stuttgart, Germany Z. Säugetierkunde 64 (1999) 230-238 ZEITSCHRIFT *% FL © 1999 Urban & Fischer Verlag SÄUG ET] ERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Mating behaviour of Ctenomys mendocinus (Rodentia, Ctenomyidae) By S. CAMiNn Unidad de Zoologia y Ecologia Animal, Instituto Argentino de Investigaciones de Zonas Aridas, Consejo Nacional de Investigaciones Cientificas y Tecnolögicas, Mendoza, Argentina Receipt of Ms. 04. 09. 1998 Acceptance of Ms. 01. 04. 1999 Abstract This is the first description of mating behaviour in Ctenomys mendocinus. Observations were con- ducted in a transparent acrylic pipe, under dim red light. The description of the courtship and copula- tion stages was based on 18 trials (7 males, 9 females, 17 couples). The mating behaviour of Ctenomys mendocinus was characterized by vocalizations, long court- ship, long bouts of precopulatory interactions, lengthy intromissions, a brief copulation stage, aggres- sive copulatory postures, and mutual indifference after ejaculation. The mating and copulation pat- terns in Ctenomys mendocinus are similar to those of other solitary subterranean, phylogenetically unrelated, forms (e.g., Spalax and geomyids). This suggests that the copulatory behaviour of C. mendocinus is closely related to its social structure. On the other hand, the structure of precopula- tory interactions in C. mendocinus is very much like that of some solitary and social bathyergids, pos- sibly reflecting phylogenetic affinities. Key words: Ctenomys mendocinus, copulatory behaviour, social biology Introduction The “tuco-tuco” Ctenomys mendocinus is a hystricomorph, fossorial, solitary rodent, and a seasonal breeder (Puic et al. 1992; Rosı et al. 1992), widely distributed throughout Ar- gentina (CABRERA 1961; Honackt et al. 1982). In Mendoza province this species is found in different mountain environments and occurs in isolated groups of low density (Rosı et al. 1992). Our knowledge of the reproductive behaviour of the genus in general and of Ctenomys mendocinus in particular is limited. The only data on copulatory behaviour of the genus Crenomys are those of ALrtuna et al. (1991), corresponding to Cienomys pear- soni of Uruguay. As far as C. mendocinus is concerned, its mating and copulatory pat- terns are unknown. In her review on the patterns of behaviour of hystricomorph rodents, KLEIMAN (1974) regarded brief copulations as one of the typical behaviours in this sub-or- der. HıckMAn (1982) compared copulatory behaviour in Cryptomys hottentotus to that de- scribed for other subterranean rodents: Spalax (NEvo 1969) and geomyids (ANDERSEN 1978; SCHRAMM 1961). The differences (spontaneous nature of copulation and short intro- missions in Cryptomys hottentotus vs. long courtship and long duration of intromissions in Spalax and geomyids) were explained by differences in social biology and burrow systems (social vs. solitary; copulations not restricted to particular areas vs. mating in specially constructed widened areas) (HıckMAn 1982). In contrast, BURDA (1989) suggested that the basic characteristics of copulatory behaviour in Cryptomys hottentotus may relate to the common mating patterns exhibited by most hystricomorphs, and hence the differ- 0044-3468/99/64/04-230 $ 12.00/0 Mating behaviour of Ctenomys mendocinus 231] ences may also be phylogenetic. The hystricomorph condition of Ctenomys mendocinus, as well as its fossorial and solitary habits, led to the investigation whether its copulatory behaviour reflects its social system or its phylogenetic affinities. If this behaviour were linked with the social structure, C. mendocinus would differ from social subterranean hystricomorphs by exhibiting longer courtship, aggressive copulatory postures, lengthy in- tromissions, and aggressiveness or indifference toward the partner after mating. On the contrary, no differences would be apparent between C. mendocinus, and social hystrico- morphs if its behaviour were mostly determined by the common, phylogenetically deter- mined, mating pattern of hystricomorphs. DEwsBURY (1972) established four attributes for a comparative study of copulation in mammals: First, the male and female may or may not become firmly locked or tied together by a strong mechanical connection during copulation. Second, pelvic thrusting may or may not occur during insertion. Third, multi- ple insertions with no sperm transfer may or may not be prerequisite to the occurrence of ejaculation. Fourth, either a single ejaculation or multiple ejaculations might be attained (Dewssury 1972). As each of these four attributes has two alternatives, patterns of copu- latory behaviour elaborated in this scheme can take any of 16 forms (DEwsBUurY 1972). Therefore, the main goals of this study were: a) to describe the mating behaviour of C. mendocinus, b) to assess whether its copulatory behaviour is consistent with HicKMAN’S (1982) hypothesis, and c) to define the DEwsBURIAN (1972) copulatory pattern in C. mendocinus. Material and methods A total of 29 adults (16 females: 130-197 g; 13 males: 156-328 g) was live-trapped in Cacheuta (Pro- vince of Mendoza, Argentina, 1330 m a.s.l) during March for five successive years (1992-1996). The animals were housed in a basement which received sunlight through a long narrow window above ground level. They were individually kept in plastic containers (50cm wide x 40cm high x 70 cm long) filled with alluvial soil to a depth of 3cm. A 30-cm long, 9-cm diameter plastic pipe was provided for shelter. Temperatures ranged between 14°C and 20°C. Alfalfa, carrots, and lettuce were supplied ad libitum. No free water was provided. During June-August, males and females were peri- odically placed together in a transparent acrylic pipe (200x11x16 cm). Trials were conducted in dim red light between 10:00 and 14:00 h. The female was placed in the pipe 45 minutes before introducing the male. Trials started when the male was introduced into the pipe, and pairs were permitted ap- proximately 1-1.5 h to initiate copulation. If copulation was initiated, observations were continued un- til attainment of the standard satiety criterion of 30 min with no intromissions (DEwSBURY 1975). All trials were video-taped. Trials observed totalled 174. Only trials with occurrence of copulations were considered in the analysis of mating behaviour. Therefore, the description of courtship and copulation was based only on the 18 trials in which copulation was attained (7 males, 9 females, 17 couples). Data are expressed as mean values (#SD) and/or ranges (mini.-max. values recorded). Results Mating patterns recorded in Ctenomys mendocinus are given in table 1. Both males and females were active from the start, seeking and slowly approaching each other, growling in a low-pitched voice. Some males and one female alternated growls with whines. Some males rubbed their perineal region against the pipe walls, and sometimes urinated (Fig. 1). On finding the urine of a female, males touched it with their nose for a few sec- onds. Upon encounters, partners sniffed at each other’s genitalia, parted, and met again to engage in precopulatory interactions: with their mouths in right angles and their inci- sors locked together they swayed from side to side. On occasions this swaying movement made one of the animals fall on its back thus exposing abdomen and genitalia to the part- DD S. CAMfn Fig. 1. Scent-marking behaviour of male of C. mendocinus. /L | SS a ä nNIm——— Fig. 2. Mating foreplay of C. mendocinus. ner (Fig. 2). The animals also stood on their hindfeet grasping each other’s cheek with forefeet and teeth. This playful behaviour ended when one of the partners moved away. Some soliciting females mounted the males in much the same way as males mount fe- males. A few females were initially inactive, reluctant to initiate courtship. They re- mained crouched, and exposed their incisors, ready to bite, when males came too close. Males responded to this threatening behaviour by presenting the side of the neck, half- closing the eyes and running away. In those cases where a male insisted on mounting a re- luctant female, the female drove him back after a short fight. After fights like these, a few males emitted the guttural “tuc-tuc” sound typical of tuco-tucos before resuming courtship. The courtship stage was long (maximum duration: 3900 sec, Tab. 1) and not continuous: partners played, parted, met again, and upon re-encounter courtship was re- sumed. First mounts occurred less than 8 min after the trial was started (Tab. 1), and were usually unsuccessful because of female resistance. In successful mounts the male mounted the female from behind, clasping her shoulders or lumbar region with his forefeet and bit- ing her neck (Fig. 3). Occasionally the male held the female’s hindfeet with his own. The female crawled away to escape, sometimes even carrying the male on her back. At first the male performed rapid shallow pelvic thrusts (PT), mostly without intromission, at a rate of 4 to 5 thrusts/sec. When the male achieved intromission, thrusts were deeper (DT) and slower (1 thrust/sec). Ejaculation, reached at the end of a number of deep thrusts, was obvious: the male grasped the female more strongly and a long lasting deeper thrust of 14 sec was discernible (SD = 10.01, range = 5-41), after which he dismounted slowly, and groomed his genitalia with forefeet and mouth. Mating behaviour of Cienomys mendocinus 233 By“ Fig. 3. Copulatory posture of C. mendocinus. Table 1. Mean duration (X), standard deviation (SD), range, frequency (f) and number of observa- tions (N) of mating and copulation patterns in Ctenomys mendocinus. Mount latency 894 10-3900 Intromission latency 905 14-3900 Courtship 1056 60-3 900 Precopulatory interactions 74 2-437 Pelvic Thrusts (PT) 33 3-154 Deep Thrusts (DT) Sl 5-130 Copulation Te 25 3277, Copulation stage 200 40-777 Only two males whined during copulation, and one female uttered a high-pitched squeak as the male exercised his last thrust. Courtship was feebly resumed for a few min- utes after copulation (Tab. 1), the animals walking up and down the pipe, growling and only occasionally locking their incisors until partners lost interest, and either crouched several centimeters apart or tried to escape by scratching the pipe walls until the end of the trial. A multiple-ejaculation pattern was absent in every trial. All episodes were ended by a single ejaculation. On nine of the trials ejaculation was achieved by the first mount with intromission. The remaining nine trials involved more than one mount but ejaculation was always attained on the last one. Mounting frequency was 2 (range 1-5), and the copulation stage lasted, at the most, 777 sec (Tab. 1). Discussion The mating behaviour of C. mendocinus was characterized by vocalizations, long court- ship, long bouts of precopulatory interactions, a lengthy intromission, a brief copulation stage, aggressive copulatory postures, and mutual indifference after ejaculation. Tuco-tuco females did not show any typical soliciting behaviour (e.g. lordosis), yet they were active, frequently starting the precopulatory interactions and occasionally mounting the male. As in other subterranean mammals, chemical (NEvo et al. 1976; GORMAN and STONE 1989, 1990; BURDA et al. 1990; HETH et al. 1992, MENZIESs et al. 1992) and chemosensory perception (HETH and ToDRANK 1995) can be very important in Ctenomys mendocinus. 234 S. CAMIN Table 2. Mating and copulation patterns in hystricomorph and other subterranean rodents of differ- ent social structures. Court- ship Length (sec) Social Sub-order References structure Post- mating behaviour of partners Species Ejacula- Copulatory tory Posture Intro- mission Length (sec) Geomys 15 bursarius Tho- momys talpoides Spalax 480-3 300 aggressive Solitary Sciuromorpha SCHRAMM (1961) 30-900 male rebuffed aggressive Solitary Sciuromorpha ANDERSEN (1978) aggressive male Solitary Myomorpha NEvo (1969) ehrenbergi Georychus capensis Cryptomys hottentotus 1-5 clumsy indifferent male chased soliciting female Solitary Social Hystrico- morpha Hystrico- BENNETT and JARVIS (1988 a) HıcKMAN (1982) morpha BENNETT and JARVIS (1988 b); BENNETT (1990) BENNETT and AGUILAR (1995) male not Socıal chased Cryptomys 30 damarensis Hystrico- morpha female Social pulling away from the male Hystrico- morpha <15 clumsy soliciting female Highly Social Hystrico- morpha Jarvıs (1991) cephalus glaber Ctenomys 60-3900 25-327 mendo- cinus aggressive mutual in- differerence Solitary Hystrico- morpha This study Tuco-tuco males become informed about the reproductive status of females by contacting the urine with their noses. The posture taken by males while rubbing their anogenital re- gion allows them to mark the walls of their burrows, resembling the behaviour of Spala- copus i.e. urinating on vertical surfaces using a leg lift (KLEımAn 1974). Vocal communi- cation is involved in mating behaviour of solitary species, e.g. Spalax (NEvo et al. 1987; HETH et al. 1987). In Ctenomys pearsoni only females whine, which indicates their full re- ceptivity to courtship (FrAncEscoLı 1995). Growls and whines in Ctenomys mendocinus are a signal of non-aggressiveness, and indicative of the animals’ willingness to copulate. Their long bouts of precopulatory interactions help to coordinate the mating behaviour, disminishing the partners’ fear and hesitation. Notwithstanding, precopulatory interac- tions might also play an important role in mating assessment, not just synchronisation of motivational states. The courtship of the solitary hystricomorph rodent Ctenomys mendocinus was longer than that of the social hystricomorph Cryptomys hottentotus (HıckMAn 1982), whereas it was similar in length to the courtship of the solitary myomorph Spalax ehrenbergi (NEvo 1969). Likewise, the duration of ejaculatory intromissions of C. mendocinus was similar to that of the solitary sciuromorph Thomomys talpoides (ANDERSEN 1978) but longer than Mating behaviour of Ctenomys mendocinus 235 Time (sec) em EM ji ii i Hg Od Mp Cd Species Fig. 4. Maximum duration of ejaculatory mounts in Crenomys mendocinus and other hystricomorph ro- dents expressed in seconds. Abbreviations: Ch= Cryptomys hottentotus (HıckMAn 1982), Cl= Chinchilla laniger (BıGnamı and BEACH 1968), Ps = Pediolagus salinicola (in KLEIMAN 1974), Hg = Heterocephalus glaber (Jarvıs 1991), Od = Octodon degus (KLEIMAN 1974), Mp = Myoprocta prat- ti (KLEIMAn 1971), Cd = Cryptomys damarensis (BENNETT and Jarvıs 1988b), Og= Octodontomys gliroides (KLEIMAN 1974), Cp = Ctenomys pearsoni (ALtunaA et al. 1991), Ed = Erethizon dorsatum (SHADLE 1946), Ps = Proechimys semi-spinosus (MALINIAK and EISENBERG 1971), Cm = Ctenomys men- docinus (present study). the duration in social hystricomorphs, such as Cryptomys hottentotus (HıckMAn 1982), Cryptomys damarensis (BENNETT and Jarvıs 1988b; BENNETT 1990), Heterocephalus gla- ber (Jarvıs 1991), and also the solitary sciuromorph Geomys bursarius (SCHRAMM 1961) (Tab. 2). The long ejaculatory intromission of some solitary hystricomorphs like Ctenomys mendocinus, Erethizon dorsatum (SHADLE 1946), and Proechimys semispinosus (MALINIAK and EISENBERG 1971) (Fig. 4) would be exceptions to the basic pattern of hystricomorph rodents reported by KLEIMAn (1974), which involves brief copulations 5-10 sec long. On the other hand, mating and copulation patterns of C. mendocinus are consistent with Hıckman’s (1982) predictions according to which solitary subterranean rodents show long courtship, aggressiveness and lengthy intromissions during copulatory behaviour. This evi- dence suggests that the copulatory behaviour of C. mendocinus is more closely related to social structure (solitary life) than to phylogeny. 236 S. CAMfN On the other hand, the structure of precopulatory interactions in C. mendocinus (e.g. locked incisors) is very much like the precopulatory interactions in Georychus capensis (Jarvıs and BENNETT 1991), the playful behaviour of Cryptomys hottentotus (BURDA 1989), the incisor fencing between young of Heterocephalus glaber (Laczy et al. 1991) and sparring between young of Cryptomys damarensis (BENNETT and Jarvıs 1988b), pos- sibly reflecting phylogenetic affinities. As previously stated, DEwsBury (1972) established four attributes for a comparative study of copulation in mammals based in the presence or absence of lock, intravaginal thrusting, multiple intromissions, and multiple ejaculations. C. mendocinus failed to exhibit lock, but it did show intravaginal thrusts. Although more than one mount with intromission was required in four of 18 tests, this appears to be attributable to the resistance offered by females in such occasions. As to whether an episode may either be terminated or not by a single ejaculation, it is likely that there will be more overlap between these two alternatives than for the first three criteria (DEwSs- BURY 1972). Although C. mendocinus is likely to resume copulation after the first ejacula- tion, in none of the 18 tests did second ejaculations occur within 30 min of the first one. The capacity to attain ejaculation on the first vaginal penetration and the absence of mul- tiple ejaculations suggest that C. mendocinus conforms to pattern twelve (absence of lock, presence of intravaginal thrusting, absence of a multiple-intromission pattern, absence of a multiple-ejaculation pattern) of Dewssury’s (1972) scheme unlike Ctenomys pearsoni (Aıtuna et al. 1991), Spalax ehrenbergi (NEevo 1969), Cryptomys hottentotus (HıcKMAN 1982), and Thomomys talpoides (ANDERSEN 1978) that conform to pattern nine (absence of lock, presence of intravaginal thrusting, presence of a multiple-intromission pattern, presence of a multiple-ejaculation pattern). Acknowledgements I would like to thank Ing. V. Roıs who helped to make this work possible. The manuscript was im- proved by the critical comments of Dr. L. MArone and Dr. P. BrLack. I also thank NeLıy HorAk for translating the manuscript, PATrıcıA DiAZ for writing the German summary and SonIA DEMARIA for the ıllustrations. This work is in partial fulfillment of the requirements for a doctoral degree for S. CAmin to be presented at the University of San Luis, Argentina. Zusammenfassung Paarungsverhalten von Ctenomys mendocinus (Rodentia: Ctenomyidae) Dieses ist die erste Beschreibung des Paarungsverhaltens von Ctenomys mendocinus. Die Beobach- tungen wurden in einem durchsichtigen Rohr aus Plastik durchgeführt. Die Beschreibungen von Brunst und Kopulation basieren auf 18 Tests (7 Männchen, 9 Weibchen, 17 Begattungen). Das Paarbil- dungsverhalten von Ctenomys mendocinus war durch Beschwichtigungslaute, lange Brunst, lange Spielsequenzen, eine lange Intromissionszeit, kurze Kopulation, aggressives Kopulationsverhalten und gegenseitige Gleichgültigkeit nach der Ejakulation gekennzeichnet. Das Paarbildungs- und Kopula- tionsmuster von C. mendocinus ähnelt dem solitär und subterran lebender Gattungen, die nicht näher verwandt sind, wie Spalax und Geomyiden. Dieses läßt vermuten, daß das Kopulationsverhalten von C. mendocinus eng mit der Sozialstruktur verbunden ist. Auf der anderen Seite, ist die Spielstruktur bei C. mendocinus, der einiger Gattungen der Familie Bathyergidae sehr ähnlich. Dieses spiegelt sehr wahrscheinlich hystricomorphe Verwandtschaft wider. Mating behaviour of Cienomys mendocinus 237 References ALTUNA, C. A.; FRANCESCOLI, G.; IZQUIERDO, G. (1991): Copulatory pattern of Ctenomys pearsoni (Ro- dentia, Octodontidae) from Balneario Solis, Uruguay. Mammalia 55, 316-318. ANDERSEN, D.C. (1978): Observations on reproduction, growth, and behavior of the northern pocket gopher (Thomomys talpoides). J. Mammalogy 59, 418-422. BENNEIT, N. C.; JARvIS, J. U. M. (1988 a): The reproductive biology of the Cape mole-rat Georychus ca- pensis (Rodentia, Bathyergidae). J. Zool. (London) 214, 95-106. BENNETT, N. C.; JARVIS, J. U.M. (1988b): The Social Structure and reproductive biology of colonies of the mole-rat, Cryptomys damarensis (Rodentia, Bathyergidae). J. Mammalogy 69, 293-302. BENNETT, N. € (1990): Behavior and social organization in a colony of the Damaraland mole-rat Cryp- tomys damarensis. J. Zool. (London) 220, 225-228. BENNETT, N. C.; AGUILAR, G. H. (1995): The reproductive biology of the giant Zambian mole-rat, Cryp- tomys mechowi (Rodentia: Bathyergidae). S. Afr. J. Zool. 30, 144. BIGNAMI, G.; BEACH, F. A (1968): Mating behaviour in the chinchilla. Anim. Behav. 16, 45-53. BURDA, H. (1989): Reproductive biology (behaviour, breeding, and postnatal development) in subterra- nean mole-rats, Cryptomys hottentotus (Bathyergidae). Z. Säugetierkunde 54, 360-379. BURDA, H., Bruns, V.; MÜLLER, M. (1990): Sensory adaptations in subterranean mammals. In: Evolu- tion in Subterranean Mammals at the Organısmal and Molecular Levels. Ed. by E.NEvo and O. A. Reıc. New York: Alan R. Liss. Inc. Pp. 269-293. CABRERA, A. (1961): Catälogo de los mamiferos de America del Sur. Rev. Mus. Cs. Nat. Bernardino Ri- vadavia Cs. Zool. 4, 309-732. DEWSBURY, D. A. (1972): Patterns of copulatory behavior in male mammals. The Quat. Rev. Biol. 47, 1- 32: DEwSsBURY, D. A. (1975): Diversity and adaption in rodent copulatory behavior. Science 190, 947-954. FRANcEScOLI, G. (1995): Las senales acüsticas en Cienomys (Octodontidae) de Uruguay. Actas X® Jor- nadas Argentinas de Mastozoologia. Pp. 94-95. GORMAN, M. L.; STONE, R. D. (1989): Repelling moles. In: Mammals as Pests. Ed. by. R. Purtman. Lon- don: Chapman and Hall. Pp. 79-81. GORMAN, M.L.; STONE, R.D. (1990): Mutual avoidance by European moles (Talpa europaea). In: Chemical Signals in Vertebrates V. Ed. by D. W. MAcDoNALDS, D. MÜLLER-SCHWARTZE, and S. E. NATYNCZUK. Oxford: Univ. Press. Pp. 367-377. HETH, G.; FRANKENBERG, E.; RAZ, A.; NEvo, E. (1987): Vibrational communication in subterranean mole-rats (Spalax ehrenbergi). Behav. Ecol. Sociobiol. 21, 31-33. HETH, G.; NEvo, E.; IKAN, R.; WEINSTEIN, V.; RAvıD, U.; Duncan, H. (1992): Differential olfactory per- ception of enantiomeric compounds by blind subterranean mole rats (Spalax ehrenbergi). Experien- tia 48, 897-902. HETH, G.; TODRANK, J. (1995): Assessing chemosensory perception in subterranean mole rats: different responses to smelling versus touching odorous stimuli. Anim. Behav. 49, 1009-1015. HıcKMan, G. C. (1982): Copulation of Cryptomys hottentotus (Bathyergidae) a fossorial rodent. Mam- malia 46, 293-298. HonackI1, H. J.; KInMAN, K. E.; KoEPPL, J. W. (1982): Mammal Species of the World. Lawrence, Kansas: Association System. Coll. Pp. 694. JARVIS, J. U. M.; BENNETT, N. C. (1991): Ecology and behavior of the family Bathyergidae. In: The Biol- ogy of the Naked Mole-Rat. Ed. by P. W. SHERMAN, J. U. M. Jarvıs, and R. D. ALEXANDER. Prince- ton: Univ. Press. Pp. 66-96. JARVIS, J. U. M. (1991): Reproduction of naked mole-rats. In: The Biology of the Naked Mole-Rat. Ed by P. W. SHERMAN; J. U. M. Jarvıs, and R. D. ALEXANDER. Princeton: Univ. Press. Pp. 384-425. KLEIMAN, D.G. (1971): The courtship and copulatory behaviour of the Green acouchi, Myoprocta pratti. Z. Tierpsychol. 29, 259-278. KLEIMAN, D. G. (1974): Patterns of behaviour in hystricomorph rodents. Symp. zool. Soc. Lond. 3, 171- 209. LAc&y, E. A.; ALEXANDER, R. D.; BRAUDE, S. H.; SHERMAN, P. W.; Jarvıs, J. U. M. (1991): An ethogram for the naked mole-rat: Nonvocal behaviors. In: The Biology of the Naked Mole-Rat Ed. by P. W. SHERMAN, J. U. M Jarvıs, and R. D. ALEXANDER. Princeton: Univ. Press. Pp. 209-242. MALINIAK, E.; EISENBERG, J. (1971): Breeding spiny rats. Proechimys semispinosus, in captivity. Int. Zoo. Yb. 11, 93-98. MENZIES, R. A.; HETH, G.; IKAnN, R.; WEINSTEIN, V.; NEvo, E. (1992): Sexual pheromone activity in urin- 238 S. CAMfin ary lipids and other fractions from urine of the male mole rat, Spalax ehrenbergi. Physiol. Behav. 52, 741-747. NEvo, E. (1969): Mole rat Spalax ehrenbergi: Mating behavior and its evolutionary significance. Science 163, 484-486. NEvo, E.; BODMER, M.; HETH, G. (1976): Olfactory discrimination as an isolating mechanism in speciat- ing mole rats. Experientia 32, 1511-1512. NEvo, E.; HETH, G.; BEILES, A.; FRANKENBERG, E. (1987): Geographic dialects in blind mole rats. Role of vocal communication in active speciation. Proc. Natl. Acad. Sci. USA 84, 3312-3315. Puic, S.; Rosı, M. 1.; VIDELA, F.; Roıc, V. G. (1992): Estudio ecolögico del roedor subterräneo Ctenomys mendocinus en la precordillera de Mendoza, Argentina: densidad poblacional y uso del espacio. Rev. Chil. Hist. Nat. 65, 247-254. Rosı, M. I.; Puic, S.; VIDELA, F.; MADOERY, L.; RoIG, V. G. (1992): Estudio ecolögico del roedor subterrä- neo Ctenomys mendocinus en la precordillera de Mendoza, Argentina: ciclo reproductivo y estruc- tura etaria Rev. Chil. Hist. Nat. 65, 221-233. SCHRAMM, P. (1961): Copulation and gestation in the pocket gopher. J. Mammalogy 42, 167-170. SHADLE, A.R. (1946): Copulation in the porcupine. J. Wildl. Mgmt. 10, 159-162. Author’s address: SERGIO CAMin, Unidad de Zoologia y Ecologia Animal, Instituto Argentino de Investigaciones de Zonas Aridas, CC. 507, 5500, Mendoza, Argentina Z. Säugetierkunde 64 (1999) 239-240 ZEITSCHRI TE © 1999 Urban & Fischer Verlag SÄUGETI se http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY WISSENSCHAFTLICHE KURZMITTEILUNGEN Free-ranging Vampire bats (Desmodus rotundus, Phyllostomidae) survive 15 years in the wild By M. TscHaApkA and G. S. WILKINSON Department of Zoology II, University of Erlangen-Nürnberg, Erlangen, Germany and Department of Zoology, University of Maryland, College Park, USA Receipt of Ms. 12. 06. 1998 Acceptance of Ms. 12. 10. 1998 Key words: Desmodus rotundus, vampire, Chiroptera, Phyllostomidae, longevity While there is considerable information on the life-span of bats of temperate zones, few data have been published about the life-span of tropical bats (see review by TUTTLE and STEVENSoN 1982). The Phyllostomidae or American leaf nosed bats of the Neotropics are in general much less extensively studied, their habitats are less accessible and they do not have periodic seasonal inactivity, which facilitates collection of survival data for tempe- rate zone bats (Davis and HırcHcock 1995). Reports exist only for some well-studied lo- cations and a few abundant species. For example, Artibeus jamaicensis have been studied extensively on Barro Colorado Island in Panama. One animal was reported reaching 7 years (Wırson and Tyson 1970) while two others were at least nine years old (GARDNER et al. 1991). Vampire bats (Desmodus rotundus: Phyllostomidae) have been studied longer than most the other neotropical bat species because of their potential economic impact. Du- ring his studies on social behavior in Desmodus rotundus one of the authors (GSW) banded over 600 vampire bats at various places in Costa Rica, primarily in Guanacaste province, but also in Atlantic rainforest at the La Selva Biological Station. On November 21st, 1994 MT caught a female Desmodus rotundus carrying band no. 507 at the Sendero Cave at Santa Rosa National Park (Guanacaste Province). The an- imal was pregnant and appeared to be in good condition. This bat was initially banded May 16th, 1980 at precisely the same location by GSW. During a visit to this cave in July 1994, GSW failed to observe any of the more than 300 vampire bats banded there be- tween 1980 and 1982. Subsequent frequent netting in the area between June of 1996 and June of 1997 revealed no more banded vampire bats (M. T. FERNANDEZ, Universidad Na- cional de Costa Rica, pers. comm). A second noteworthy record occurred at La Selva Biological Station (Heredia Pro- vince) on January 6th, 1996 when we recaptured another female vampire bat with an or- ange-green plastic band. This bat had been marked initially by GSW on September 2nd, 1981. Judging from the condition of the nipples this animal was lactating or postlactating. The distance between place of banding and recapture is approximately 2 km. While the animal was first caught at a roost in a big hollow Dipteryx panamensis located in primary forest near abandoned plantations, the recapture location is located at the edge of the La Selva property and serves probably as a flyway for animals roosting in old trees to the cattle pastures outside La Selva, as indicated by frequent captures of the species at that 0044-3468/99/64/04-239 $ 12.00/0 240 M. TscHAPkA and G. S. WILKINSON spot. GSW banded 50 vampire bats in 1981 at La Selva, and the reported animal is the first recapture known to us at the rather well-netted field station. It is noteworthy that both of our recaptures were females. WıLkınson (1985) reported that female vampire bats form roost associations which remain stable over long periods. At the time of first capture, neither female was pregnant or lactating but both had average adult forearm lengths and weights, indicating that they were probably near one year of age when captured. Our observations indicate, therefore, that Desmodus rotundus females can remain reproductively active until at least 15 years of age. While there is indirect dental evidence for 18 year old vampire bats (LorD et al. 1976), the animals reported here repre- sent, together with similar aged animals (13/22) reported by DELPIETRo et al. (1992), at an age of 15 years, the oldest recaptured vampire bats from the wild. Female vampire bats ap- pear, therefore, to be able to live longer than the fruit bat Artibeus jamaicensis, which is about the same body size and occurs in similar locations in the New World tropics. While we cannot determine if this difference in life history is due to a difference in diet or beha- vior, we suspect that the tendency of female vampire bats to share food under duress (WiLKınson 1984) represents an important behavioral adaptation that can increase longev- ity. In general terms our observations emphasize the remarkable status of bats as small mammals with small offspring-numbers per birth and a comparably long life-span. Acknowledgements MT’s work in Costa Rica was supported by a grant from the German Academic Exchange Service (DAAD), HSPIVAUFE and the DFG. For support in the field we want to thank K. STONER (Univer- sidad de Costa Rica) as well as C. Voıgr and M. HoLDERIED (University of Erlangen, Germany). MA- RIA TERESA FERNANDEZ (Programa Regional en Manejo de Vida Silvestre, UNA, Costa Rica) shared her unpublished data on Santa Rosa bats with us. References Davıs, W. H.; HıtcHcock, H.B. (1995): A new longevity record for the bat Myotis lucifugus. Bat Re- search News 36, 6. DELPIETRO, H. A.; MARCHEVSKY, N.; SIMONETTI, E. (1992): Relative population densities and predation of the common vampire bat (Desmodus rotundus) in natural and cattle-raising areas in north-east Argentina. Preventive Vet. Med. 14, 13-20. GARDER, A. L.; HANDLEY, C. O., Jr.; Wırson, D. E. (1991): Survival and relative abundance. In: Demo- graphy and Natural History of the Common Fruit Bat Artibeus jamaicensis, on Barro Colorado Is- land, Panamä. Ed. by C. O. HAnDLEy, Jr., D. E. Wırson, and A.L. GARDNER. Smithsonian Contr. Zool. No. 511. LorD, R. D.; MURADALI, F.; LAZARo, I. (1976): Age composition of vampire bats (Desmodus rotundus) in northern Argentina and southern Brazil. J. Mammalogy 57, 573-575. TUTTLE, M. D.; STEvENSoN, D. (1982): Growth and survival of bats. In: Ecology of Bats. Ed. by T. H. Kunz: New York, London: Plenum Press. WILKINSON, G. S. (1984): Reciprocal food sharing in vampire bats. Nature 308, 181-184. WILKINSON, G. S. (1985): The social organization of the common vamypire bat. I. Pattern and cause of as- sociation. Behav. Ecol. Sociobiol. 17, 111-121. Wiırson, D. E.; Tyson, E. L. (1970): Longevity records for Artibeus jamaicensis and Myotis nigricans J. Mammalogy 51, 203. Authors’ addresses: M. TscHAPKA, Department of Zoology II, University of Erlangen-Nürnberg, Staudtstraße 5, D-91058 Erlangen, Germany; G.S. WıLkınson, Department of Zoology, University of Maryland, College Park, MD 20742, USA BR N ee: N . Z. Säugetierkunde 64 (1999) 241-245 ZEITSCHRIFT "FÜR © 1999 Urban & Fischer Verlag SÄUGETIERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY The use of day roosts and foraging grounds by Natterer’s bats (Myotis nattereri Kuhl, 1818) from a colony in southern Germany By B.M. SIEMERS, InGRID KaAıpr, and H.-U. SCHNITZLER Lehrstuhl Tierphysiologie, Universität Tübingen, Tübingen, Germany Receipt of Ms. 23. 11. 1998 Acceptance of Ms. 15. 04. 1999 Key words: Myotis nattereri, telemetry, foraging, roosting, activity Natterer’s bat (Myotis nattereri Kuhl, 1818) occurs from SW-Europe and N-Africa through W-Asia (HorAcEK and HAnAK 1983). Little is known about its ecology, though fecal analysis indicates that prey is caught close to or from vegetation (GREGOR and BAUEROVA 1987; SHIEL et al. 1991; Beck 1991, 1995; TAAKE 1992; Swirt 1997). In beha- vioral experiments Natterer’s bat is capable of detecting arthropods close to vegetation by echolocation, using signals of broad band-width (SIEMERS and SCHNITZLER unpubl. data). In the course of a field study on echolocation and foraging behavior, we fitted three M. nattereri with radiotransmitters in order to locate their hunting areas. Here we present data on activity pattern, home range, and hunting area. Additionally, data on use and types of day roosts are given. The study was conducted in the vicinity of Mössingen, Baden-Württemberg, Germany on the foothills of the Swabian Alb (48°23’N, 9°01’E) from May through August 1996 (radiotracking between July 24th and August 15th). The study area is situated between 470 and 700 m above sea level and is characterized by fruit tree orchards, beech-domi- nated deciduous forests, and monocultures of spruce (Picea abies). Villages and roads lie interspersed. A colony of M. nattereri, comprising 50-60 animals distributed over several day roosts, was studied. Three adult non-lactating females (animal Al through A3) were fitted with erystal-pulsed transmitters (BD-2A and BD-2B supplied by Holohil Systems Ltd., Onta- rio, Canada) weighing 0.6-0.7g (6-7% of the animals’ body weight). The bats were tracked using FT 290-receivers (Andreas Wagner Telemetrieanlagen, Cologne, Germany) connected with 5-element Yagi antennae. Whenever we had radio contact, bearings were taken at 5 min intervals. With two observers at hand, two bearings were taken at the same time from different locations and an animal’s position was determined by triangula- tion using the computer-program Tracker (Camponotus AB, Solna, Sweden). Home ranges and core hunting areas were determined from fixes obtained by triangulation. The home range was defined as the smallest convex polygon comprising all fixes of an animal. The core convex polygon comprising 50 % of the fixes was considered as the core hunting area. Additionally, with only one observer at hand, we monitored the temporal use of the core hunting areas. We obtained telemetric data for animal Al from 8 nights, for A2 from 6 nights, and for A3 from 3 nights. Radio contact was maintained for about 70% of the time spent monitoring Al and for approximately 80% and 55 % for A2 and A3, respec- tively. 0044-3468/99/64/04-241 $ 12.00/0 242 B. M. SIEMERS, INGRID Karpr and H.-U. SCHNITZLER = roosting- area >" forest ®Ü core hunting areas W:]) settlement = home range Fig. 1. Home ranges (minimum convex polygon containing all radio fixes) and core hunting areas (core convex polygon containing 50 % of fixes) of three adult female Myotis nattereri (Al, A2, A3). The roosting area encompasses 13 artıfıcıal and natural roosts used by the colony under study. The colony under study roosted in bird nest boxes and artificial bat roosts, hung in the freestanding trees of an orchard belt. Between 3 and 30 individuals occupied a single roost at a time. By radio-telemetry we found a day roost in a hollow branch of a beech- tree (Fagus sylvatica) at 7 m height on the slope of a forested hill. During a hot period, anımal Al roosted therein for two consecutive days, on the second of which seven Nat- terer’s bats were counted leaving the roost. We discovered a total of 13 day roosts within a minimum convex polygon of 24.3 ha; about 90 % of it being orchards and 10 % hilly mixed forest (Fig. 1). As established by inspection of roost sites, the colony changed roosts at least 12 times in 11 weeks (May 1“ to July 23"). On one occasion, we discov- ered about 20 Natterer’s bats in an artificial bat roost together with a hornet queen (Ves- pa crabo) on its newly built nest. The first bats emerged from a roost 31.6 +# 9.6 min (mean + standard deviation, n = 24) after local sunset and the last ones returned to the roost 39.6 # 9.9 min (n = 19) before lo- cal sunrise, as monitored visually. Time between sunset and emergence and return and sunrise, respectively, remained fairly constant from May through August; thus the ani- mals’ active period was more than 1.5 h shorter in mid-summer than in spring and fall (Figz2). The home ranges, determined from radio fixes, measured 523 ha in Al, 123 ha in A2, and 80 ha in A3 (Fig. 1). The home range of A2 would have extended to approx. 580 ha if we had included one night (7./8. August) during which we followed A2 without being able to triangulate. Core hunting areas within those home ranges covered 2.8 ha for Al and 18.6 ha for A2. Due to a lack of sufficient fixes for A3, we could not determine the size but only the rough position of the bat’s core hunting area (Fig. 1). The bats’ presence in their core hunting area could be confirmed on each of 9 nights of inspection with only one observer at hand. Out of a total of 24h 15 min of tracking-time between 22:45 pm and 4:48 am across those 9 nights, individuals were present on average 56.6 % of the time in their core hunting area (Al: 86.2 %, A2: 46.8%, A3: 51.2%). Centers of the core hunting areas were located at a distance of 3.1+0.3km (n=3) from the roosting area. The animals were found up to a distance of 3.7+0.7km (n=3) from the roosting area. With the exception of the immediate surroundings of the roosting area, the home range of Al did not overlap with those of A2 and A3, whereas all fixes Day roosts and foraging grounds of Myotis nattereri 243 6:00 sunrise return 5:00 4:00 time 23:00 22:00 21:00 emergence sunset 01.Jun 01.Jul 01.Aug Fig. 2. Emergence of first (n = 24) and return of last (n = 19) Myotis nattereri in comparison with local sunset and sunrise between May 1“ and September 10°” 1996. On June 20", the animals delayed emer- gence about 25 min, waiting for a bout of heavy rain to pass (arrow). Note break in time axis. obtained for A3 lay within the home range of A2 (Fig. 1). In 3 nights we recorded the si- multaneous presence of Al and A2 in the same area. The core hunting areas of Al and A2 were 3.9 km apart; those of A2 and A3 were adjacent. The core hunting area of Al comprised mixed deciduous forest, a monoculture of coniferous forest (P. abies), an area that had been deforested by a storm and recently re- planted with oak (Ouercus spp.) and margins of pasture; hence, an area rich in edge struc- tures. The hunting areas of A2 and A3 were situated at the edge of a coniferous forest (P. abies) and included a fresh clearing and orchards with trees planted at distances be- tween 10 and 30 m apart. We conclude that animals were continuously on the wing, as signal direction kept changing most of the time. On one occasion only, it remained constant for 75 min while it was raining heavily, and the animal presumably hung in a sheltered place within a conifer- ous forest. We first detected the animals in the core hunting area 84.3 +25.8min (n=9) after they had left the roost in the evening. When the bats were heading back from the hunting areas to the roosts in the morning they covered the distance with 5.7+0.2 km/h 244 B. M. SIEMERS, InGRıD Karpr and H.-U. SCHNITZLER (mean + sd, n = 3). From photographs under stroboscopic illumination, the flight speed of M. nattereri was determined to be 15.5 +3.2 km/h (n =10). Thus the animals could have reached their roosting areas nearly three times faster than they actually did. We conclude that the anımals were hunting on their way to and from the core hunting area. Our findings confirm that the activity period of M. nattereri depends on sunset and sunrise, and thus on light intensity, as well as on weather conditions (see ENGLÄNDER and LAUFENS 1968; LAUFENS 1973; SwIFT 1997). We found that Natterer’s bats used individual core hunting areas at least during the study period; i.e. they showed site-fidelity. The existence of core hunting areas visited night after night is also reported for other European bat species, e.g. Myotis myotis (AU- DET 1990), Myotis emarginatus (KruLL et al. 1991), and Myotis daubentonii (ARNOLD pers. comm.). By fidelity to individual, exclusive hunting grounds the bats could avoid in- traspecific competition for resources (e.g. VON HELVERSEN 1989). From our data we can- not answer the question as to what degree core hunting areas overlap, but the simulta- neous presence of A2 and A3 in the same area might indicate that some overlap occurs. Another advantage of small and hence well known core hunting areas could be that the bats establish a detailed cognitive map, improving orientation in space and the repeated use of rewarding feeding sites. Concerning the habitat type used by M. nattereri, it is striking that coniferous forest was present in all of the three determined core hunting areas, whereas the study area is dominated by mixed deciduous forest. Extensive orchards were present in two of the core hunting areas. All core hunting areas were rich in horizontal and vertical edges. The hy- pothesis that M. nattereri hunts close to edges of vegetation is supported indirectly by fe- cal analysis (GREGOR and BAUERoOVA 1987; SHIEL et al. 1991; BEcK 1991, 1995; TAAKE 1992; Swirt 1997), predictions from wing morphometry (NORBERG 1981) as well as behavioral experiments on detection ability (SIEMERS and SCHNITZLER unpubl. data) and directly by visual observation in the field (ArLETTAZ 1996; SwıFt 1997; SIEMERS and SCHNITZLER Un- publ. data). The data presented here do not conflict with this view of M. nattereri’s for- aging ecology, but the spatial resolution of telemetry is too coarse for explicit confirma- tion. In our study we found a distance between roosting area and core hunting areas of about 3km for M. nattereri. Myotis blythii, M. daubentonii, and M. myotis travel about 4 km, 6-8 km, about 9 km and even up to 25 km between roosting and core hunting areas, respectively (AuDET 1990; ArLETTAZ 1995; ARNOLD pers. comm.). From these consider- able distances it may be concluded that intraspecific competiton forces individuals to hunt at some distance from roosts (VON HELVERSEN 1989), or that core hunting and roost- ing areas are chosen according to different criteria. Hunting grounds should yield abun- dant and accessible prey, while roosting areas should provide roosts protecting the bats from predators, providing a favorable micro-climate (Lewıs 1995) and enough space for conspecifics, especially in nursing colonies. The roosting area of the colony under study is characterized by a high density of bird nest boxes and artificial bat roosts, most of which are well exposed to the sun in spring, when the crowns of the free-standing fruit trees are still leafless. We presume that the abundance of possible roosts and their warm tempera- ture compared to hollow trees in the middle of a forest make the roosting area favorable. The localization of one day roost within a forest neighboring the orchard during a hot period had led us to speculate whether the bats might choose roosts in the cooler forest during hot summer days. The Natterer’s bats changed roosts often and are, according to Lewis (1995), to be categorized as low roost-fidelity species. As the bats changed fre- quently from one roost to another in the immediate vicinity, climatic differences are un- likely to play a major role and we consider the avoidance of parasites to be an important factor for those changes (e.g., LAUFENS 1973; Lewis 1995). Day roosts and foraging grounds of Myotis nattereri | 245 Acknowledgements We thank ELISABETH KALKO, ESTHER LANGEHEINECKE, GISELA LOTTER, and WIBKE THIES for discussion and valuable suggestions, the forestry authority of Mössingen for logistic support, ANDREAS BEIG, DIETMAR NILL, KAI SIEMERS, and ALBRECHT Rırk for help in the field and NıamH Ni BHLEITHIN and ROBERT PAXTon for correcting the English. Licenses: Naturschutzrechtliche Ausnahmegenehmigung 73-8/8852.21 / RP Tübingen; Versuchs- funkgenehmigung 77 55 9985 / BAPT Schwäbisch Hall References ARLETTAZ, R. (1995): Ecology of the sibbling mouse-eared bats (Myotis myotis, Myotis blythii): zoogeo- graphy, niche, competiton, and foraging. PhD Thesis. Martigny: Horus Publ. ARLETTAZ, R. ( 1996): Foraging behavior of the gleaning bat Myotis nattereri (Chiroptera, Vespertilioni- dae) in the Swiss Alps. Mammalia 60, 181-186. AUDET, D. (1990): Foraging behavior and habitat use by a gleaning bat, Myotis myotis (Chiroptera: Ves- pertilionidae). J. Mammalogy 71, 420-427. BEcK, A. (1991): Nahrungsuntersuchungen an der Fransenfledermaus, Myotis nattereri (Kuhl 1818). Myotis 29, 67-70. Beck, A. (1995): Fecal analysis of European bat species. Myotis 32/33, 109-119. ENGLÄNDER, H.; LAUFENS, G. (1968): Aktivitätsuntersuchungen an Fransenfledermäusen (Myotis natte- reri, Kuhl 1818). Experientia 24, 618-619. GREGOR, F.; BAUEROVA, Z. (1987): The role of diptera in the diet of Natterer’s bat, Myotis nattereri. Folia Zool. 36, 13-19. HORACEK, 1.; HANAK, V. (1983) Comments on the systematics and phylogeny of Myotis nattereri. Myotis 21/22, 20-29. HELVERSEN, O. von (1989): Schutzrelevante Aspekte der Ökologie einheimischer Fledermäuse. Schrif- tenreihe Bayer. Landesamt Umweltschutz 92, 7-17. KRULL, D.; SCHUMM, A.; METZNER, W.; NEUWEILER, G. (1991): Foraging areas and foraging behavior in the notch-eared bat, Myotis emarginatus (Vespertilionidae). Behav. Ecol. Sociobiol. 28, 247-253. LAUFENS, G. (1973): Beiträge zur Biologie der Fransenfledermäuse (Myotis nattereri Kuhl, 1818). Z. Säugetierkunde 38, 1-14. Lewis, S. E. (1995): Roost fidelity of bats: a review. J. Mammology 76, 481-496. NORBERG, U. M. (1981): Allometry of bat wings and legs and comparison with bird wings. Phil. Transact. Royal Soc. London B 292, 359-398. SHIEL, C. B.; MCANEYy, €. M.; FAIRLEY, J. S. (1991): Analysis of the diet of Natterer’s bat Myotis nattereri and the common long-eared bat Plecotus auritus in the West of Ireland. J. Zool. (London) 223, 299- 309; SWIET, S.M. (1997): Roosting and foraging behavior of Natterer’s bat (Myotis nattereri) close to the northern border of their distribution. J. Zool. (London) 242, 375-384. TAAKE, K.-H. (1992): Strategien der Ressourcennutzung an Waldgewässern jagender Fledermäuse (Chiroptera; Vespertilionidae). Myotis 30, 7-74. Authors’ address: BJÖRN M. SIEMERS, INGRID KAIPF, HANS-ULRICH SCHNITZLER, Department of Ani- mal Physiology, Zoological Institute, University of Tübingen, Auf der Morgen- stelle 28, D-72076 Tübingen, Germany. Email: bjoern.siemers@uni-tuebingen.de Z. Säugetierkunde 64 (1999) 246-250 ZEITSCHRIFT * Se © 1999 Urban & Fischer Verlag SÄUGETI au a http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Genetic relatedness in two Southern sea lion (Otaria flavescens) rookeries in the southwestern Atlantic By VALERIA B. SZAPKIEVICH, H. L. CAPpP0ZZ0, E. A. CRESPo, R. OÖ. BERNABEU, CECILIA CoMAs, and MARTA D. MuDRY Departamento de Biologta, Universidad de Buenos Aires, Buenos Aires; Museo Argentino de Ciencias Naturales, Buenos Aires; Centro Nacional Patagönico, Universidad Nacional de la Patagonia and Fundaciön Patagonia Natural, Chubut; Instituto de Biologia Celular, Universidad de Buenos Aires, Buenos Aires, Argentina Receipt of Ms. 23. 06. 1998 Acceptance of Ms. 25. 03. 1999 Key words: Otaria flavescens, protein electrophoresis, genetic distance The southern sea lion, Otaria flavescens (SHAw, 1800), is distributed along the coast of South America, from Torres (29°20’S 49°43’W) in southern Brazil in the Atlantic Ocean (Rosas et al. 1994) to Cape Horn in the extreme south, and from Cape Horn to Zorritos (4°S) in northern Perü in the Pacific Ocean (Riedman 1990). The northernmost breeding grounds are along the coasts of Uruguay (Isla de Lobos, Cabo Polonio, and La Coroni- lla). A total of 15,000 individuals was estimated for this area. To the north of the Urugua- yan breeding grounds, in southern Brazil, there are only two non-breeding rookeries where subadult males predominate. Seasonal movements have been documented for Rio Grande do Sul coast (Rosas et al. 1994). In addition, erratic records for the species have been reported from Rio de Janeiro (23°S) and even to 13°S (Castello 1984) but always by solitary individuals. In Argentina, they breed along the Patagonian coast, from Punta Bermeja at 41°08°S to 55°S on Tierra del Fuego Island. Presently, there are 54 breeding and non-breeding rookeries with a total of 51,000 individuals up to 47°05’S, 66°16’W (Revyes et al. 1999). In- dividual movements showing seasonal patterns have been demonstrated (CrEspo 1988; Crespo and PEDRAZA 1991). To the north of the Patagonian grounds, there are only two subadult male rookeries in Buenos Aires Province at the Mar del Plata (35°S) and Oue- quen harbours 38°30’S), as well as a breeding rookery at Isla Trinidad (39° S). Biochemical-genetic data have been used as a powerful tool for studying population biology; moreover, they can provide much information about the population structure of a species. Taking into account the reduced information available with respect to migra- tions and interchange of individuals, the aim of this study was to analyse the genetic variability in the southern sea lion in two rookeries: Isla de Lobos, Uruguay and Punta Norte, Peninsula Valdes. Collection of samples: Blood samples were collected from 70 southern sea lion pups (less than 30 days old), during the 1992/93 breeding season, from two rookeries 1,300 km distant from each other: Isla de Lobos, Maldonado, Uruguay (35 blood samples) (35°02’S, 52°55’ W) and Punta Norte, Peninsula Valdes, Chubut, Argentina (35 blood samples) (42°04'S, 63°47 W) (Fig. 1). All pups sampled were selected at random from breeding harems at both localities. About 2 to 5 ml of heparinized peripheral blood was collected 0044-3468/99/64/04-246 $ 12.00/0 Genetic relatedness in Otaria flavescens 247 30 ARGENTINA 32 34 ISLA DE LOBOS 36 Maldonado Buenos Aires 38 Mar del Plata Quequen > Isla Trinidad 40 | Punta Bermeja 42 PUNTA NORTE "Punta Leön 44 Isa Escondida ATLANTIC OCEAN sia Quintano 2 68 66 64 62 60 58 56 54 52 Fig. 1. Location of the study areas and of the geographical names mentioned in the text. @ Reference rookeries, 4 Sites of sample collection. from the extradural vein from the lumbar region. After blood extraction, pups were re- turned to their mothers. Genetic analysis: Blood samples were processed in situ, in the field. Plasma and ery- throcytes stored in liquid nitrogen were subjected to horizontal 5 %-7 % polyacrylamide gel electrophoresis (PAGE). Only 64 samples from the 70 collected were included for the electrophoretic interpretation. Nine protein systems representing 10 putative loci were analysed: lactate dehydrogenase A and B (LDHA, LDHB), malate dehydrogenase (MDH), phosphogluconate dehydrogenase (PGD), isocitrate dehydrogenase (IDH), superoxide dismutase (SOD), glutamate oxalacetate transaminase (GOT), and esterase D (ESD) from erythrocytes, and transferrin (TF) and albumin (ALB) from plasma. The pro- teins studied, the electrophoretic methods and staining procedures are given in table 1. To locate TF patterns, electrophoresis gels were run with human standards (HSA) and the genotypes were interpreted by direct comparison with them. Genotypic frequencies were tested by maximum likelihood for Hardy-Weinberg equi- librium. Differences in gene frequencies between the two rookeries were tested by means of a two-tailed binomial test (Zar 1996). Observed and expected mean gene diversity va- lues at population level (H, and H.) and NEı (1978) genetic distance (D) were calculated by using the Genetic Data Analysis (GDA) computer program (Lewis and Zaykın 1997). From the 10 examined loci, nine were monomorphic and six of them as shown in GALES et al. (1989): GOT, LDHA, LDHB, MDH, PGD, SOD. Only TF was polymorphic for both sea lion rookeries, and curiously, it was monomorphic in southern elephant seals (GALES et al. 1989). 248 VALERIA B. SZAPKIEVICH et al. Table 1. Screened proteins in Otaria flavescens. Optimal buffer systems: 1) Tris-maleic--EDTA-magne- sium chloride, pH 7,4 (Shaw and PrasAaD 1970); 2) Tris-EDTA-citric acid, pH 7.0 (SHAw and PRASAD 1970); 3) Phosphate-EDTA-citrate, pH 6.9 (SCHNEIDER 1988); 4) Phosphate-citric acid, pH 5.9 (HARRIS and Hopkınson 1976); 5) Lithium hydroxide-boric acid, pH 8.3 (BoumAan and BEARN 1965). Staining methods references: a) HARRIS and Horkınson (1976); b) Hırrıs and MorITz (1990); c) SCHNEIDER (1988). Abbreviature and Optimal buffer Staining enzyme commission system method number reference 1. Erythrocyte enzymes Lactate dehydrogenase TED EIEIE27 Malate dehydrogenase MDH 1.1.1.37 Phosphogluconate dehydrogenase PGD 1.1.1.44 Isocitrate dehydrogenase IDH 1.1.1.42 Superoxide dismutase SOD#F1571-1 Glutamate oxalacetate transaminase KONZERT Esterase D ESDS-RRl 2. Serum proteins Transferrin TF Albumin The same alleles of TF were found at PN and IL. Two alleles and three different elec- trophoretic TF phenotypes were observed. Both TF1 and TF2 bands showed slower electrophoretic mobility than that of human TFC. The heterozygotes TF 1-2 showed a two-banded pattern, consistent with the monomeric molecular structure of this protein. Under the assumption that genotypes are determined by two different alleles, the genotype distribution in both populations was in Hardy-Weinberg equilibrium according to the maximum likelihood estimation (Tab. 2). The allele Tf? was the most frequent in both sample sites. The two tailed binomial test showed that the gene frequencies were sig- nificantly different for the stocks (e = 2.5897, p< 0.01). The sample size could explain in part the observed differences in the gene frequencies. TF is usually a good molecular marker that is polymorphic in many mammals (GooDman et al. 1965; SHAUGHNESSY 1969; HERZocG et al. 1991; HARTL and FERRAND 1993), and as it showed polymorphism in the present study it could be used as a molecular marker according to the different frequen- cies found. Observed genetic heterozygosity for southern sea lions at Isla de Lobos was H, = 0.027 and H, = 0.003 for Punta Norte. These H values are in the extremes of pre- vious values given for other pinniped species (GALES et al. 1989). It has been stated that Table 2. Observed transferrin phenotypes, allele frequencies, expected and observed mean heterozyg- osity (He and Ho) and an estimate of the fixation index (f) in the two studied rookeries of Otaria fla- vescens. Rookery Number ofindividuals Gene frequencies of each phenotype 11 1 22 uR ne PUNTA 33 0.0147 0.9853 0.002941 0.002941 0.00000 NORTE ISLA DE 19 0.2167 0.7833 0.032542 0.026667 0.18309 LOBOS Genetic relatedness in Otaria flavescens 249 marine mammals as a group are low in genetic variation (SAUGHNESSY 1975; MCDERMID and BonnEr 1975; LIDICKER et al. 1981; GALES et al. 1989); however the H value for PN is extremely low, probably as a consequence of a bottleneck effect. Between the 30’s and the 50’s, the southern sea lion population in northern Patagonia was reduced to about 10 % of its original size remaining in a stable condition between 1975 and 1991. They were heavily exploited, mainly from Peninsula Valdes and Tierra del Fuego, by national permissioners for leather and oil. In Peninsula Valdes, the population was reduced during this period from 200,000 to less than 15,000 individuals (CrEspo and PEDRAZA 1991). Pre- sently, there are 35,000 individuals and the population is increasing at a rate of 3.5 % per year (CRESPO pers. comm.). On the other hand, the H value of IL is one of the highest found in pinnipeds accord- ing to GALES et al. (1989). One reason may be that another pinniped species, the south- ern fur seal Arctocephalus australis, has been the main exploited species in Uruguay in- stead of O. flavescens. However, VAZ-FERREIRA (1979) stated 30,000 individuals of O. flavescens for 1954, and between 1986 and 1990 LımA and BATALLES (pers. comm.) esti- mated 15,000 individuals, reaching the lowest number ever known for this stock and with a decreasing trend. The Nei genetic distance obtained between Punta Norte and Isla de Lobos was D = 0.003. This is a low value compared with intraspecific values found from populations of other mammalian species (SELANDER and JoHNSoN 1973). There are no previous re- ports of genetic distances for conspecific pinnipeds except for M. leonina from Macquarie and Heard Islands (D = 0.007, GALES et al. 1989). Taking into account that individual movements between Patagonian and Uruguayan stocks have been recorded from marked animals (LoRENZANI and LORENZANI, pers. comm.), this suggests a high mobility of individuals between rookeries, especially consid- ering the low number of marked animals from sites with several hundreds or thousands individuals. In this opportunity, the presence of the same fixed allele at each of the nine monomorphic loci and the low value of genetic distance leads to the conclusion that both rookeries belong to the same population in which gene flow is currently occurring. Acknowledgements The authors are indebted to many people who helped in the field sampling. We thank Centro Nacio- nal Patagönico (CONICET, ARGENTINA) and Instituto Nacional de Pesca (Uruguay) for their lo- gistical support in the field, P. SCHWARZBAUM and R. WERNER for their critical review of the manu- script and the anonymous referees for their comments and suggestions. We also thank financial support through the Binational Project (1992-93) for Uruguay and Argentina E.C. and H.L.C. grant from Marine Mammal Action Plan (UNEP) and UBACYT EX 228/95-97 and CONICET 211/97-99 M.M grants. References Bouman, B. H.; BEARN, A. G. (1965): The presence of subunits in the inherited group-specific protein of human serum. Proc. Natl. Acad. Sci. USA 53, 722-729. CASTELLO, H. P. (1984): Registros del elefante marino, Mirounga leonina (Carnivora, Phocidae) en las costas del Atläntico S.O., fuera del ärea de cria. Rev. Mus. Arg. Cs. Nat. “Bernardino Rivadavia”e Inst. Nac. Inv. Cs. Naturales 3, 235-243. CRESPo, E. A. (1988): Dinämica poblacional del lobo marino del sur Otaria, flavescens (SHAw, 1800), en el norte del litoral patagönico. Diss. thesis, Univ. Buenos Aires. CRESPO, E. A.; PEDRAZA, S. (1991): Estado actual y tendencia de la poblaciön de lobos marinos de un pelo (Otaria flavescens) en el litoral norpatagönico. Ecol. Austral 1, 87-96. 250 VALERIA B. SZAPKIEVICH et al. GALES, N. J.; ADAMS, M.; BURTON, H. R. (1989): Genetic relatedness of two populations of the southern elephant seal, Mirounga leonina. Mar. Mammal Scı. 5, 57-67. GOODMAN, M.; KULKARNI, A.; POULIK, E.; REKLYS, E. (1965): Species and geographic differences in the transferrin polymorphism of macaques. Science 147, 884-886. HARTL, G. B.; FERRAND, N. (1993): Genetic polymorphism of transferrin (Tf) and the haemoglobin al- pha chain (HBA) in the brown hare (Lepus europaeus): Anim. Genetics 24, 439-440. Harris, H., Hopkinson, D. A. (1976): Handbook of Enzyme Electrophoresis in Human Genetics. Am- sterdam: Nord Holland. HERZOG, S.; MUSHOVEL, C.; HATTEMER, H.; HERZOG, A. (1991): Transferrin polymorphism and genetic differentiation in Cervus elaphus L. (European red deer) populations. Heredity 67, 231-239. Hıruıs, D. M.; Moritz, C. (1990): Molecular systematics. Massachusetts: Sinauer Associates. Lewis, P. O.; ZAvkın, D. (1997): Genetic Data Analysis: Computer program for the analysis of allelic data. Version 1.0. Free program distributed by the authors over the internet from the GDA Home Page at http://chee.unm.edu/gda/ LIDICKER, W.; SAGE, R. D.; CALkıns, D. G. (1981): Biochemical variation in northern sea lions from Alaska. In: Mammalian Population Genetics. Ed. by M. H. SmitH and J. JoULE. Athens: Univ. Geor- gia Press. McDeERMID, E. M.; BoNNER, W.N. (1975): Red cell and serum protein systems of Grey seals and Har- bour seals. Comp. Biochem. Physiol. 50 B, 97-101. NeEı, M. (1978): Estimation of average heterozygosity and genetic distance from a small number of indi- viduals. Genetics 89, 583-5. Reyes, L. M.; Crespo, E. A.; Szapkievich, V.(1999): Distribution and population size of the southern sea lion (Otaria flavescens) in central and southern Chubut, Patagonia, Argentina. Mar. Mammal Sci. (in press). Riedman, M. (1990): The Pinnipeds; Seals, Sea Lions, and Walruses. Berkley: Univ. California Press. Rosas, F. C. W.; PINEDO, M.; MARMONTEL, M.; Haımovicı, M. (1994): Seasonal movements of the South American sea Lion (Otaria flavescens, Shaw, 1800) off the Rio Grande do Sul coast, Brazil. Mam- malıa 58, 51-59. SCHNEIDER, M.P.C. (1988): Variacao proteica em primatas de Amazonia e seu significado evolutivo. Diss. thesis. Univ. Federale Rio Grande do Sul, Porto Alegre, Brasil. SELANDER, R. K.; JOHNson, W. E. (1973): Genetic variation among vertebrate species. Ann. Rev. Eco- log. System 4, 5-91. SHAUGHNESSY, P. D. (1969): Transferrin polymorphism and population structure of the Weddell seal Leptonychotes weddelli (Lesson): Austr. J. Biol. Science 22, 1581-1584. SHAUGHNESSY, P.D. (1975): Biochemical comparison of the harbor seal Phoca vitulina richardi and P. v. largha. Ed. by K. RonALD and A. W. MANSFIELD. Biol. Seal 169, 70-73. SHAw, C.R.; PrAsAD, R. (1970): Starch gel electrophoresis of enzymes. A compilation of recipes. Bio- chem. Genetics 4, 279-320. VAZ-FERREIRA, R. (1979): South American sea lion. Mammals in the seas. FAO Fish. Ser. 2, 9-12. ZAR, J. H. (1996): Biostatistical Analysis. New Jersey: Prentice Hall. Authors’ addresses: VALERIA SZAPKIEVICH, MARTA MUDRY, CECILIA CoMAs, Grupo de Investigaciön en Biologia Evolutiva, Dpto. Biologia, Universidad de Buenos Aires, Ciudad Univer- sitaria, Pabellön II, 4° piso, 1428 Buenos Aires; H. Luis CAppoZZo, Laboratorio de Ecologia del Comportamiento y Mamiferos Marinos, Museo Argentino de Cien- cias Naturales, “B. Rivadavia”, Av. A. Gallardo 470, 1405 Buenos Aires; ENRI- QUE A. CREsPo, Centro Nacional Patagönico, Universidad Nacional de la Patagonia and Fundaciön Patagonia Natural, 9120 Puerto Madryn, Chubut; R. O. BERNABEU, Instituto de Biologia Celular, Universidad de Buenos Aires, Ar- gentina. Z. Säugetierkunde 64 (1999) 251-255 ZEITSCHRIFT % > $ FFÜR © 1999 Urban & Fischer Verlag SÄUG ETI ERKUN DE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Seasonal occurrence of killer whales (Orcinus orca) in waters of Rio de Janeiro, Brazil By S. SICILIANO, J. LAILSON BRITO Jr., and A. DE F. AZEVEDO Departamento de Vertebrados, Museu Nacional, Universidade Federal do Rio de Janeiro and Departa- mento de Oceanografia, Universidade do Estado do Rio de Janeiro, Rio de Janeiro, Brazil Receipt of Ms. 16. 02. 1999 Acceptance of Ms. 21. 04. 1999 Key words: Orcinus orca, occurrence, Rio de Janeiro coast, Brazil Killer whales (Orcinus orca) are found in all oceans and seas, from the polar regions to the equator, in both hemispheres. However, they appear to be more common near shores in cold temperate to subpolar waters (JEFFERSON et al. 1993). Records of killer whales are few or nearly absent for most parts of the Brazilian coast. The first reported Brazilian record is a stranding 128 km north of Rio Grande (31°57’S), southern Brazil (CASTELLO 1977). Additional records have remained anecdotal (e.g. CASTELLO and PınEDo 1986; DA- NIEL et al. 1992; Santos and SıcıLıano 1994) and only one stranding record is known for the Rio de Janeiro coast (GEIsE and BoroBIA 1988). More recently, a study on killer whale interactions with the swordfish and tuna fishery was conducted south and south- eastern of Brazil (DArLLA RosA 1995). A collection of recent sightings of killer whales for the Rio de Janeiro State coast (21°37’S-23°10’S) ıs presented and provides an opportu- nity for discussion of their presence and ecological requirements in an area previously un- covered. Information on killer whale presence was provided through the implementation and development of a sighting network along the Rio de Janeiro coast. The data set was ana- lysed to study individual occurrence and location patterns, based on characteristics that can be used to identify individuals uniquely (i.e. dorsal fin, saddle patch; Bıcc et al. 1987; O’SuLLivan and Muruin 1997). A total of 29 records of killer whale groups was confirmed on the coast of Rio de Ja- neiro for the period between October 1993 and November 1997 (Tab. 1, Fig. 1). One addi- tional record was obtained for November 1983. The period between August 1996 and February 1997 represents 76.7 % of all sightings. Killer whale sightings were concentrated on spring and summer months, accounting for 93 % of all confirmed records. Group size ranged from one to 15 individuals, but averaged 3.9 animals. Four groups included at least one adult male, and four groups at least one calf. Four individual killer whales have been photoidentified to date on the coast of Rio de Janeiro. While not every individual was photographed from each group, catalogued photographs showed that at least one individ- ual was resighted at a time interval of 37 days. All groups sighted were found in shallow coastal waters, well located inside the continental slope. The furthest offshore sighting was 38.9n. miles off Atafona (21°35’S), the northern coast of the state. The fact that most records are known for the area between Buüzios (22°44'S) and the city of Rio de Janeiro (22°56'S) may suggest higher chances for opportunistic sightings due to the increased hu- man recreational activity during the warmer months. However, this observed seasonality 0044-3468/99/64/04-251 $ 12.00/0 S. SICILIANO, J. L. BRITO JR. and A. DE F. AZEVEDO 2 JUOZ JINS au} 0} 9SOJD P3JE90J !esıe Junysıs ayJ Apngs sIyL ur sAeıI Jo uONJeIJUSIU09 9Z.Ie] Qyse 0z> ol1I oge) ‘J2ndo4 Op ereIg 966T I2qua9aq SI vI Apns SIQL JUOZ JINS 9UJ 0] 9509 P2JE9O] @ 07> ewoısenbes ‘9uo9erf 9661 12qwue9ad SI EI JUOZ JIns ay} 0} 9s0[9 P2Je90] ‘10Joq Aep ay] Apnıs sıy]L pa1y3ıs dno.1s awes ayy Afqıssod € 0 sorzng eqLıang 966 2qua9aq TO I Apnıs sIyL JUOZ JINS 9UJ 0} 9SOJ2 P>JE9O] 5 07> sozng 9661 1OqwaAoN 0E II UOZ JINS IUJ 0} 9SOJI POYeIO] ‘sojtusAnl [e.IoAas Apngs sIyL pue afeu J[npe 9uo }seo] Je a BR) 0 OLLJ Oge)7 ‘3J2n80 op eleld 9661 12qwaAoN EZ 0I "wo9 'sIod ‘sSowvy A pue OLLIAANAg IQ d'V Wd 07: Tl I vi riedınbj jo u Y‘9 9661 TIIOPO TO 6 "w1o9 'sıod ‘sowvy X pue oLLIAaNdg IA dV Wd 0E:Z0 I 07 euoJe}V JO U6T 9661 Joquusıdas 8 "109 'sIod ‘SOoWvA A pue oLLIaaNdg Id dV Wd 0€:C0 € DE euojeIvy Jo U6gE 9661 Isnanvy 07 /b "109 'sıad ‘sowvy N pue OLLIAANAg IA d'V WNd 0€E:70 v 0€> euoJe}V JJ0 S661 Amf ZI 9 SIANVNAag I p>110d31 ® 0 soIzng ‘OowBdeuniy ep eIeld 66T 2IORO IZ g sojrusanl SIAanvNaag A pue ofewpf Jjnpe ‘ofeuu JInpe r ZI-0L soIzng ‘oeBdeuniy ep eIeld Y661 12I0RO IZ 1m oprusanl Aeg apueig eyjf Apngs SIyL e pue 9fewsJ J[npe ‘ofeur Ipnpe € SI-0I “sp oIWM pue ourg ened r66L IeN € ZUIUIUIIMS JseJ !ofrusAnl o9NVIg d e pue Jfe9-IoyJou ‘ofeuu JInpe y sI-OI Aeg apueıng eyJf JoJuJ seuufeg E66 I290PO II 7 ollouef ap oIy Apnys sıy] UOZ JINS 9YJ 0} 9SOJ9 P9JEIO] +7 0 ‘ope.IUO) OS >p eleld E86T IOqwaAoN I 1NOS syıewoy azıs dnoIy (u) ydoq uore9o] -[IZeIg ‘OISUEL IP OIY JO 1se09 ayJ uO (DI10 SNUNAO) SOJEUM IO]JIN JO SPIOISI ZuNysIS "T JIqeL 253 Killer whales in coastal waters of Rio de Janeiro Apnys sıyL Apns sIqyL Apnys sıyL Apnys sIqyL Apns sIqyL Apnys sıqyL Apnys SIQqL Apnys sıqL Apngs sıyL Apngs sIqL Apnys sıqyL Apnys sıyL Apngs SIyL Apnys sıqyL Apnys sıyL p>110d931 P2110d91 Zurydesig ‘Aeı e Junpepe IfeyM 19]]1 Juo p>110da1 9UOZ JANS 9YJ 0} 9509 P9JEIO] Y9UOZ JINS 9YJ 0} 9SOJ9 P9YEIO] JUOZ Jıns UI 0) 9SO[ P>Ye90] ‘Furjre) qo]J 9UOZ JINS 9UJ 0] 9SO]JI P9JEIO] 9UOZ JINS OU] 0] 9SOJI P9JEIOL 9UOZ JINS 9UJ 0] 9SOJI P9JEIOT Sumuwmms poods MoJs JUOZ Jıns 9UJ 0} 9SOJI PSJEIOL :Nd 00:£0 yeoq durysij oyJ Poysdeonıde ssfeym Io] Zurmuuwmms poods Mo] ‘soprusAanl 0m] pue Jfed-IoyJou UOZ JANS 9UJ 0} 9SOJI POYe9O] song oge) Op [ere1iy ‘ereig ereig o19uerf op ory ‘eonlij ep eıseg ep ereig 191911 “ndendrejj ewosenbes ‘ejıy ep ereig eWenIelVv ‘BIOS eleld ewoıenbes ‘eJı\ ep eleid ewoıenbes ‘eJIA ep ereig o1DUef Hp oIy "ewouedj op ereig eursıenbes ‘ejıA ep ereId eWenIeIV 'BI9S eIeId eWenIeIV ‘eI9S eIeld ewozenbes Jo ‘soyen sop 9 s9JeQ sop o1tonbsad eWEeNIEIV ‘BIOS eIeld ewroıenbes ‘ejıA ep ereig L661 ‚9wuıns L661 Arenıqaq 9I L661 Arenıqoq pl L661 Arenıq9q L661 Arenueg /7 L661 Atenugg (7 L661 Arenugg p] L661 Atenugg ZI L6617 Arenueg 80 L661 Arenueg [0 9661 PQqUSIIA 0E 9661 LOW 67 9661 PqWHAA FZ 9661 1Oquıasadl IT 9667 quad SI 254 S. SICILIANO, J. L. BRITO JR. and A. DE F. AZEVEDO 215 nm? 30‘ oO & yt e; ’ STB \ Io N R DE JANEIRO STATE @ 43.10.11, 12, 14, 28, 29 207% 7 ar ichh) 23° Ss za 15,:16,217,,18 219 pe aa8> 24,25 Pr | TLANTIC i\ n OCEAN son nn un Zt 45’ W 44 W 43:W Jay 41’ W Fig. 1. Sighting locations of killer whales (Orcinus orca) along the Rio de Janeiro state coast. Map key refers to table 1. may also reflect a true killer whale occurrence pattern at the Rio de Janeiro coast. Up- dwelling conditions present along the Rio de Janeiro coast in summer months lower sur- face water temperature to 18°C, or even less. The resighting interval of 37 days for an individual whale poses some questions on group composition, habitat preferences and distribution patterns. BAırD et al. (1992) listed some of the behavioural and ecological differences between transient and resident killer whale populations in the Pacific northwest. According to these authors, the most important differences relate to diet and habitat use. Transient killer whales show small groups sizes (1-15); unpredictable seasonal occurrence and foraging area generally in coastal waters. Our observations suggest some degree of ecological requirements listed for transient killer whales in BAıkD et al. (1992). Killer whales may visit shallow waters of Rio de Janeiro in search of favorable prey that could include small as well as large whales. Some potential prey are the marine tucuxi (Sotalia fluviatilis), the franciscana (Pontoporia blainvillei), the rough-toothed dolphin (Steno bredanensis), the Atlantic spotted dolphin (Stenella frontalis), the bottlenose dolphin (Tursiops truncatus), the com- mon dolphins (Delphinus spp.) and Bryde’s whale (Balaenoptera edeni;). However, there is no evidence of feeding by killer whales on marine mammals in our observations. On the other hand, such potential predation pressure on small dolphins could explain why Sotalia groups are virtually confined to coastal shallow bays and/or riv- er mouths associated with turbid waters. More recently, Ort and DAnıLewicz (1996) re- ported the presence of three franciscanas in the stomach of a stranded female killer whale in southern Brazil. It is also possible that killer whales take advantage of the up- welling conditions on the coast of Rio de Janeiro and may forage on a variety of season- ally abundant sharks, rays, large fish (e.g. Euthynnus alletteratus “bonito”, Scombero- morus spp. “cavala”, and Coryphaena hippurus “dourado”) and cetaceans. At least two Killer whales in coastal waters of Rio de Janeiro 255 observations were conducted on the presence of rays: it was noted in one case a whale at- tacking a ray (unidentified species). These sightings indicate that killer whales, once thought to be rare in shallow coastal waters of southeastern Brazil, may use this habitat seasonally as a foraging ground. Acknowledgements We thank A.P.Dı BEneDITTo, R. Ramos, and the photographers E. FERNANDES (Petrobras) and A. Branco (Jornal O Globo) for the access to their data. D. E. SERGEANT, P. H. OTT, D. DANILEWICZ, and L. DArLLA Rosa kindly commented on the manuscript. References Biss, M. A.; ErLis, G. M.; FoRD, J. K. B. ; BALcoMB, K. (1987): Killer whales: a study of their identifica- tion, genealogy and natural history in British Columbia and Washington state. Nanaimo, B. C.: Phantom Press. BAIRD, R. W.; ABRAMS, P. A; Dirr, L. M. (1992): Possible indirect interactions between transient and re- sident killer whales: implications for the evolution of foraging specializations in the genus Orcinus. Oecologia 89, 125-132. CASTELLO, H.P. (1977): Food of a killer whale: eagle stingray, Myliobatis, found in the stomach of a stranded Orcinus orca. Sci. Rep. Whales Res. Inst. 29, 107-111. CASTELLO, H. P.; PINEDo, M.C. (1986): Sobre unos avistajes en el mar de distintas especies de cetaceos en el sur del Brasil. In: Actas Primera Reuniön de Trabajo de Expertos en Mamiferos Acuäticos de America del Sur. Ed. by H. P. CAsTELLo and I. R. Waıs. Buenos Aires, Argentina: Museo Argentino de Ciencias Naturales and Fundaciön Vida Silvestre Argentina. Vol. I. Pp. 61-68. DALLA Rosa, L. (1995): Interagöes com a pesca de espinhel e informagöes sobre a dieta alimentar de orca, Orcinus orca, no sul e sudeste do Brasil. Fundagäo Universidade do Rio Grande: Monografia de Bacharelado. DANIEL, M. C.; METZLER, P. M.; Nunes, V. A.; ROCHA, A. R.; TALASKA, A. (1992): Nota sobre o primeiro registro de Orcinus orca em UÜbatuba, litoral norte do Estado de Säo Paulo. In: Anales III Reuniön de Trabajo de Especialistas en Mamiferos Acuäticos de America del Sur. Ed. by J. A. OPoRrTOo, L. M. BrıEvA, and R. PrADerı. Valdivia, Chile: Centro de Investigaciön y Manejo de Mamiferos Marinos. Vol. 3. Pp. 23-25. GEISE, L.; BOROBIA, M. (1988): Sobre a ocorröncia de cetäceos no estado do Rio de Janeiro, entre 1968 e 1984. Rev. Bras. Zool. (Säo Paulo) 4, 341-346. JEFFERSON, T. A.; LEATHERWOOD, S.; WEBBER, M. A. (1993): Marine Mammals of the World. Rome: FAO Species ıdentification guide. O’SULLIVAN, S.; MuLLin, K.D. (1997): Killer whales (Orcinus orca) in the northern Gulf of Mexico. Mar. Mam. Scı. 13, 141-147. OTT, P. H.; DAnILEwIcCZ, D. (1996): Presence of franciscanas (Pontoporia blainvillei) in the stomach of a killer whale (Orcinus orca) stranded in southern Brazil. Whalewatcher 30, 27. SANTOS, M.C. DE O.; SICILIANO, S. (1994): Novos registros de cetäceos para o litoral do estado de Säo Paulo, Brasil. In: Anais VI Reuniäo de Trabalho de Especialistas em Mamiferos Aquäticos da America do Sul. Ed. by A. XımEnEZ and P.C. Sımözs-Lopes. Florianöpolis, Brasil: Universidade Federal de Santa Catarina. Vol. 6. Pp. 58. Authors’ adress: SALVATORE SICILIANO, Departamento de Vertebrados, Museu Nacional, Universi- dade Federal do Rio de Janeiro, Rio de Janeiro, RJ 20940-040 Brazil, siciliano@openlink.com.br, JosE LAıLson BRITO Jr. and ALEXANDRE DE FREITAS AZEVEDO, Departamento de Oceanografia, Universidade do Estado do Rio de Janeiro, Rua Säo Francisco Xavier, 524 - sala 4018E Rio de Janeiro, RJ 20550-013 Brazil Z. Säugetierkunde 64 (1999) 256 ZEITSCHRIFT ® FÜ © 1999 Urban & Fischer Verlag SÄUGETI ERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Buchbesprechung QunmSIYEH, M. B.: Mammals of the Holy Land. Lubbock: Texas Tech University Press 1996. 389 pp., 72 black and white photographs, 43 maps, hard cover. Price $ 35,-. ISBN 0-89672-364-X. This book deals with the mammals of Israel and the western part of Jordan, i.e., an area between lati- tudes 29° and 34° N and longitudes 34° and 38° E. The author, born in the Holy Land and presently working at Duke University in the United States, gives a concise introduction into the mammalian fauna of an area in the border zone between Asia and Africa. After a short introduction and an account of the historical development of mammal research in the Holy Land, the author makes short remarks on the study of mammals in general, deals with “Mammalian evolution and human history”, discusses different mammalian adaptations and the inter- relationship between mammalian parasites and human health and introduces the reader to the geogra- phy and ecology of the Holy Land. After remarks on the zoogeography of mammals and aspects of conservation, the most extensive part of the book from page 59 through 316 deals with a synopsis of the mammals of the Holy Land. For eight mammalian orders a short introduction is given and a dichotomous key allows the deter- mination of genera within mammalian families. Subsequently, the mammalıan species that live in the considered geographical area are described. First the original descriptions and synonyms are cited, fol- lowed by a diagnosis. An account of the geographical range of the different species is illustrated by a distribution map. The local status is characterized and information on the biology of the respective species is given. Under the heading “Genetics” information on the karyotype is supplied. The final section “Human interactions” supplies information on names given to species by the local inhabitants, as well as the significance of the mammalian species in rural medicine and folklore. The description of many of the considered species includes black and white photos. The quality of printing of these half- tone illustrations is generally poor, i.e., in most cases very dark. This reduces the information drasti- cally that was originally intended by inclusion of the photos. Following the major part of the book, it continues with short notes on introduced and domesti- cated mammals, a compilation of scientific terms in a glossary, thirty pages of references and an ap- pendix listing localities with their geographical coordinates. Because of the diffculties in transcription of both Arabic and Hebrew names into English, many localities are represented by more than one name or spelling. Finally, a detailed index of more than 12 pages concludes this book, which will cer- tainly be of great help to those visitors of the Holy Land who are interested in the mammalian fauna of this area. P. LANGER, Gießen Instructions to Authors Submission and Acceptance of Manuscripts: Manuscripts for publication should be sent to the managing editor, Prof. Dr. D. Kruska, Institut für Haustierkunde, Christian-Albrechts-Universität, Olshausenstr. 40-60, D-24118 Kiel, Germany, e-mail: dkruska@ifh.uni-kiel.de. Acceptance of the manuscript follows the bylaws of the German Society for Mammalogy (Deutsche Gesellschaft für Säugetierkunde). Receipt of the manuscript will be confirmed immediately by mail, and as soon as the peer reviews are received the authors will be informed concern- ing the decision for acceptance. All correspondence dealing with details of production should be sent to: Zeitschrift für Säugetierkunde, Urban & Fischer Verlag, Löbdergraben 14a, D-07743 Jena, Germany. Form of the Manuscript: Each manuscript should be submitted in triplicate, one original and two copies, in English or German in DIN A4 format either from a typewriter or legible computer print with a left-hand margin of 4cm and true double spacing. The first page of the manuscript should contain the following information: title name(s) of author(s): full first name for female authors, first initials only for male authors department and university affiliation In case of long titles, a running title not exceeding 72 characters should be provided. Script type should be uniform throughout the manuscript. Use of bold print, italics and spaced-letters must be avoided. Page footnotes and appendices at the end of the text are not allowed. Names of authors cited in the text are to be underlined in pencil. Botanical and zoological generic and species names should be underlined in pencil with a wavy line, whereby the name of the scientist who first described the species is not underlined. Authors should indicate in the margins the approximate location in the text for illustrations and tables. At the end of the manuscript following References, the exact postal address(es) of the author(s) should be given. Content of the Manuscript: Manuscripts can be published as original investigations or short communications. a) Original investigations: In addition to the text, original investigations should include illustrations, tables and references, and must not exceed 30 typewritten, double-spaced pages. The text should be divided into: Abstract (in English), Introduction, Material and methods, Results, Discussion (or together as Results and Discussion), Acknowledgements, Zusammenfassung (in German), References. If written in German, an English title should precede the English abstract; if written in English, a German title should precede a German summary. b) Short communications: Short communications must not exceed 5 typewritten, double-spaced pages and do not require either an Abstract or Summary. If written in German, an English title should be given on page 1; if written in English, please give a German title on page 1. Short communications need not be headed with subtitles into In- troduction, Materials and methods, Results, and Discussion but should be organized according to this form. Key words: Please supply five key words following the Abstract. Hlustrations: Number and size of illustrations must be kept to an absolute minimum. Only well focussed and read- ily reproducible original prints showing optimal contrast can be accepted; colour plates cannot be processed. La- belling on figures should be uniform, and large and clear enough for eventual reduction in size. On the back of each illustration the following information should be given, using a soft pencil: figure number, name of the journal, first author. Captions to figures should be written on a separate sheet of paper. Tables: The number and size of tables must be reduced to a minimum. Each table should appear on a separate page including heading and running number. Comprehensive tables not fitting onto one page should follow on consecutive DIN A4 format sheets of paper. References: Each paper cited in the text must appear in alphabetical order at the end of the text. Only internation- ally available literature should be quoted. The names of the journals should be given in abbreviated form in conformity to international use. Examples a) From journals: Karko, E. K. V; HANDLEY, C. O. Jr. (1994): Evolution, biogeography, and description of a new species of Fruit-eat- ing bat, genus Artibeus Leach (1821), from Panamä. Z. Säugetierkunde 59, 257-273. b) From books: HERRE, W.; Röhrs, M. (1990): Haustiere - zoologisch gesehen. 2. Aufl. Stuttgart, New York: Gustav Fischer. c) From reference books: NIETHAMMER, J. (1982): Cricetus cricetus (Linnaeus, 1758) -— Hamster (Feldhamster). In: Handbuch der Säugetiere Europas. Ed. by J. NIETHAMMER und F. Krapp. Wiesbaden: Akad. Verlagsges. Vol. 2/1, 7-28. Printing: The principal author will receive page proofs of the manuscript as well as a set of the illustrations and the original manuscript for proof reading. Only type-setting errors should be corrected. Any major revision undertaken by the author at this stage will result in costs to the authors themselves. The conventional proof-readers’ symbols should be used. The corrected page proofs should be signed by the author to indicate that the paper is “ready for print” and be sent to Prof. Dr. Peter Langer, Institut für Anatomie und Zellbiologie, Justus-Liebig-Universität Gies- sen, Aulweg 123, D-35385 Giessen, Germany, e-mail: peter.langer@anatomie.med.uni-giessen.de Reprints: 50 reprints of each paper are supplied free of charge. When returning the page proofs additional reprints can be ordered at the cost of the author(s). Vorbehalt aller Rechte Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, ins- besondere 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 gesetz- lichen Bestimmungen des Urheberrechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmun- gen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Ver- vielfältiger 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 in- ternal 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 CCC. Publisher: Urban & Fischer Verlag GmbH & Co. KG, Branch Office Jena, P. 0. Box No. 100537, D-07705 Jena, Germany. Phone ++49(0)36 41/6 26-3; Fax ++49(0)36 41/62 65 00; e-mail: journals@urbanfischer.de Type setting, printing and binding: druckhaus köthen GmbH Advertisement bookings: Enquiries should be directed to Urban & Fischer Verlag, Advertising Manager: Sabine Schröter, P.O. Box 1005 37, D-07705 Jena, Germany. Phone ++49(0)36 41/62 64.45, 9 Fax ++49(0)36 41/62 64 21 Advertising rates: The price list from January 1, 1999 is effective at present. Distribution and subscription agency: For customers in the German speaking countries: Please turn to your bookshop or to SFG-Servicecenter Fachverlage, Zeitschriftenvertrieb: Barbara Dressler, Löbdergraben 14 a, D-07743 Jena, Germany. Phone ++49(0)36 41/62 64.44, Fax ++49(0)36 41/62 64.43; For customers in all other countries: Please contact your local bookshop or Stockton Press., c/o Debbie Floyd, The Sub- scription Dept., Houndmills, Basingstoke, Hants RG 21 6XS, UK. Phone ++44(0)12 56/32 92 42. Fax ++44/1256/812358. Subscribers in the USA can call toll free on ++1/800/ 747 3187. Terms of delivery (1999): 1 volume consisting of 6 issues. Subscription rate (1999): Price per volume: DM 478.00*; UK Sterling 192.00 Single issue price: DM 95.00*; UK Sterling 39.00 * Suggested list prices. Prices are subject to change without notice. Reduced membership rates are available on request. Please turn to our publ. house. All prices mentioned above are exclusive postage. USdollar prices are converted at the current exchange rate. We accept the following credit cards: VISA Euro-/Mastercard/American Express, (Please note card-No. and expiry date). Banking connections: For customers in the German speaking countries: Deutsche Bank AG Jena, Konto-No. 6 284 707, BLZ 820 700 00; Postbank Stuttgart, Konto-No. 92 372 704, BLZ 600 100 70. For costumers in the other countries: UK Post Office Giro Account No. 5 192 455. Full details must accompany the payment. Subsciption information: If not terminated the subscription will be effected until recalled. If no cancellation is given until October, 31st the delivery of the journal will be continued. Copyright: The articles published in this journal are protected by copyright. All rights are reserved. No part of the journal may be reproduced in any form without the written permission of the publisher. Authorization to photocopy items for internal or personal use of specific clients, is granted by Urban & Fischer, for libraries and other users registered with the Copyright Clearance Center (CCC) Transaction Reporting Service, provided that the base fee of $12.00 per copy is paid directly to CCC, 222 Rosewood Drive, Danvers, Ma 01923, USA, 0044-3468/99/064/ $ 12.00/0 This consent does not extend to copying for general distribution, for advertising or promotional purposes, for creating new collective works or for resale and other enquiries. In the cases, specific written permission must be obtained from the pub- lisher Urban & Fischer. Printed in Germany PA STOCKTON Printed on acid-free paper effective with vol. 61, no. 1, 1996. © 1999 Urban & Fischer Verlag New: For detailed journal information see our home page: http://www.urbanfischer.de/journals Abstracted/Indexed in Animal Breeding Abstracts; Current Contents Agriculture, Biology Environmental Sciences; Biological Abstracts; BIOSIS database; Current Advances in Ecological and Environmental Sciences; Dairy Science Abstracts; Elsevier BIOBASE/Cur- rent Awareness in Biological Sciences; Fisheries Review; Helminthological Abstracts; Index Veterinarius; South Pacific Periodicals Index; Veterinary Bulletin; Key Word Index to Wildlife Research; Wild Review (Fort Collins); Zoological Record &L 700 Z4S7 MAMM ZEITSCHRIFT FÜR SÄUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Barlow, Kate E.; Jones, G.: Roosts, echolocation calls and wing morphology of two phonic types of Pipistrellus pipistrel- lus. - Quartiere, Echoortungslaute und Flügelmorphologie von zwei akustischen Typen von Pipistrellus pipistrellus.. 257 Revilla, E.; Delibes, M.; Travaini, A.; Palomares, F.: Physical and population parameters of Eurasian badgers (Meles meles L.) from Mediterranean Spain. - Physische und demographische Parameter des Eurasischen Dachses (Meles ne lage ee ee ER: 269 Antinuchi, C. D.; Busch, Cristina: Intrageneric comparisons in urine-concentrating capacity and renal morphology among three species of Akodon from different geographic rainfall regimens. - Intragenerische Vergleiche zwischen der Fähigkeit, Urin zu konzentrieren und der Morphologie der Niere bei drei Akodon-Arten aus geographischen Re- Bionen mie Unterschiedlichen Niederschlagsmustern .-.........2202cnsansanesannandsennernensesentennsneessnnnenneneeanene 277 Borkowska, Anetta: Genetic and morphological variation among populations of the bank vole Clethrionomys glareolus from north-eastern Poland: the seasonal aspect. - Genetische und morphologische Differenzierung zwischen Rötel- mauspopulationen (Clethrionomys glareolus) aus dem Nordosten Polens: Saisonale Aspekte ........zz2u@22eeeeeeeeeen- 285 Granjon, L.; Bonnet, Amelie; Hamdine, W.; Volobouev, V.: Reevaluation of the taxonomic status of North African gerbils usually referred to as Gerbillus pyramidum (Gerbillinae, Rodentia): Chromosomal and biometrical data. - Neu- bewertung der taxonomischen Stellung nordafrikanischer Rennmäuse, die gewöhnlich Gerbillus pyramidum (Gerbil- linae, Rodentia) zugeschrieben werden: Chromosomale und biometrische Daten ............22222202022202eee nennen nenn 298 Suchentrunk, F.; Polster, Karin; Giacometti, M.; Ratti, P.; Thulin, C.-G.; Ruhle, C.; Vasilev, A. G.; Slotta-Bachmayr, L.: Spatial partitioning of allozyme variability in European mountain hares (Lepus timidus): gene pool divergence across a disjunct distributional range? - Räumliche Verteilung der Allozymvariabilität bei europäischen Schnee- hasen (Lepus timidus): Genpool-Divergenz in einem disjunkten Verbreitungsgebiet? ...........22222222022see nennen 308 TE ee a ee 319 ISSN 0044-3468 - Z. Säugetierkunde - 64(1999)5 - S. 257-320 - Oktober 1999 URBAN & FISCHER 1999 GES Bes SE ET ZEITSCHRIFT FÜR Ar: SÄUGETIERKUNDE INTERNATIONAL JOURNAL USE OF MAMMALIAN BIOLOGY Herausgeber/Editor Deutsche Gesellschaft für Säugetierkunde Schriftleitung/Editorial Office D. Kruska, Kiel - P. Langer, Giessen Wissenschaftlicher Beirat/Advisory Board W. Fiedler, Wien - S. Halle, Jena - G. B. Hartl, Kiel - R. Hutterer, Bonn -H.-G. Klös, Berlin - E. Kulzer, Tübingen - P. Lüps, Bern - W. Maier, Tübingen - O. Anne E. Rasa, Bonn -H. Reichstein, Kiel - M. Röhrs, Hannover - H. Schliemann, Hamburg - G. Storch, Frankfurt Deutsche Gesellschaft für Säugetierkunde Altvorsitzende/Living Past Presidents D. Starck, Frankfurt (1957-1961, 1967-1971) - H. Frick, München (1972-1976) -— M. Röhrs, Hannover (1977-1981) —- H.-J. Kuhn, Göttingen (1982-1986) - E. Kulzer, Tübingen (1987-1991) — U. Schmidt, Bonn (1992-1996) Amntierender Vorstand/Managing Committee Vorsitzender/President: H. G. Erkert, Tübingen Mitglieder/Board Members: H. Frädrich, Berlin - R. Hutterer, Bonn - D. Kruska, Kiel — Marialuise Kühnrich, Hamburg - R. Schröpfer, Osnabrück - Angela Wöhrmann-Repenning, Kassel Z. Säugetierkunde 64 (1999) 257-268 ZEITSCHRIFT FÜ © 1999 Urban & Fischer Verlag SÄUGETI ERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Roosts, echolocation calls and wing morphology of two phonic types of Pipistrellus pipistrellus By Kate E. BARLow and G. JONES School of Biological Sciences, University of Bristol, Bristol, U.K. Receipt of Ms. 27. 08. 1997 Acceptance of Ms. 05. 05. 1999 Abstract Variation in the number of bats in maternity roosts of two phonic types of P. pipistrellus was investi- gated. Also, bats of the two phonic types were caught at maternity roosts, and their wing morphology and echolocation calls studied. 45 kHz P. pipistrellus maternity roosts contained significantly fewer bats than 55 KHz P. pipistrellus roosts. There was significant variation in mean frequency of maximum energy (FMAXE) of echolocation calls used by bats among roosts of 55 KHz P. pipistrellus, but not among roosts of 45 kHz P. pipistrellus. However, within each phonic type differences among roosts only accounted for a small proportion of the variation in echolocation call frequency; a much larger proportion was due to differences among individuals. Forearm length, an indicator of body size, was larger in 45 kHz P. pipistrellus than in 55 kHz P. pipistrellus, but there was no relationship between body size and geographic roost location in either phonic type. Variation in echolocation call frequency was not correlated with body size in either phonic type. Variation in echolocation call frequency among individuals may allow roost members to identify others in their group, but it is more likely to have evolved as a result of other influencing factors. Some variables of wing morphology differed be- tween the two phonic types, but it is not clear how these differences relate to flight performance. Key words: Pipistrellus pipistrellus, cryptic species, ultrasound, body sıze, ecomorphology Introduction Maternity roosts of the vespertilionid bat Pipistrellus pipistrellus (SCHREBER, 1774) are formed from May to July in the British Isles. These maternity roosts are aggregations of mainly adult female bats and their pups (STEBBINGS 1968; SPEAKMAN et al. 1991) and are usually found in buildings (CorBET and Harrıs 1991). Adult females may occupy a num- ber of different roosts during the year, but are often loyal to the same set of roosts for several years (THomrson 1992). The number of females in roosts of P. pipistrellus in the British Isles varies widely from a few bats to over a thousand in some cases (SPEAKMAN et al. 1991); up to double that number emerge from roosts when young bats are flying, usually during July. In this study, we investigated the roosting ecology and wing morphology of P. pipistrellus in the British Isles. P pipistrellus exists as two phonic types over much of Europe (Jones and van Parıss 1993). Search-phase echolocation calls (GRIFFIN et al. 1960) of these phonic types have a frequency of maximum energy (FMAXE) at around 55 KHz in one type, and at around 45 kHz in the other. We will refer to the phonic types as 45 kHz P. pipistrellus and 55 kHz P. pipistrellus throughout this study, though there is now unequivocal evidence that they are cryptic species (BARRATT et al. 1997; BARLOW 0044-3468/99/64/05 - 257 $ 12.00/0 258 KAtE E. BARLow and G. JoNES 1997; BARLOw and Jones 1997 a, b; BARLoWw et al. 1997; Jones 1997). The nomenclature of P. pipistrellus is currently being amended accordingly by the International Commission on Zoological Nomenclature. There are several benefits to animals living in groups, which may include increased ac- cess to resources, information transfer, decreased risk of predation, and increased repro- ductive success (HAMILTON 1971; WArD and ZAHavı 1973; PULLIAM and CArAco 1984; Brown 1988; WILKINSON, 1992; SPEAKMAN et al. 1992; SPEAKMAN et al. 1995; FENTON et al. 1994). Roosting communally may also have energetic benefits (TRUNE and SLOBODCHI- KOFF 1976; ROVERUD and CHAPPELL 1991). There are costs, however, of coloniality, includ- ing for example increased parasite loads (Brown and Brown 1986; BARcLAY 1988; LEwIs 1996). Optimal colony size will differ according to ecological circumstances. We predicted that the two phonic types of P. pipistrellus might have different colony sizes since they show differences in diet (BARLOw 1997) and in habitat use (VAUGHAN et al. 1997 a). Group cohesion may be achieved by bats if individuals produce individually identifi- able communication calls specifically to maintain group coherence or to identify their re- latives (e.g. BALCoMBE 1990; RAasmuson and BArcLAY 1992; SCHERRER and WILKINSON 1993) or their group mates (e.g. CHENEY and SEYFARTH 1982; FoRD 1989; WILKINSON and BoUGHMAN 1998). Bat echolocation calls may function in communication (FENToN 1985, 1994). PEARL and FEnTon (1996) suggest that echolocation call structure may be colony- specific and used in group recognition, and therefore in the maintenance of group cohe- sion. There is variation in echolocation call frequency among individual P. pipistrellus (MıLLEr and Decn 1981), which could allow individual or colony identification, although individual variation may be caused by sex, or body size effects (Jones 1995). Bats of the two phonic types of P. pipistrellus use separate maternity roosts (JONES and van Parıss 1993). First, we counted and compared the numbers of bats in maternity roosts of the two phonic types. Second, we measured body size, indicated by forearm length, and variables of wing morphology of the phonic types. We also investigated varia- tion in body size with geographical roost location in the two phonic types. Third, we in- vestigated whether variation in echolocation call frequency could be explained at the in- dividual level by correlating with body size, or at the roost level by varying among roosts. Material and methods Roost counts The number of adult bats in maternity roosts of the two phonic types were counted at evening emer- gence between late May and early July 1992-6. In most cases, time-expanded recordings of echoloca- tion calls were recorded as bats emerged from the roosts, and a Sona-Graph was used to determine the phonic type of the bats. Overlap in the frequency of maximum energy in echolocation calls be- tween phonic types is small (<5%, JonEs and van Parıss 1993), and roosts can be ascribed to phonic type unambiguously when large numbers of bats are recorded. For some roosts the heterodyne output of a bat detector (S-25; Ultra Sound Advice, London, UK), tuned first to 45 KHz and then to 55 KHz, was used to determine phonic type. Roost counts were transformed with the square root transforma- tion to achieve normality (ZAr 1984). The number of bats in roosts of each phonic type was compared with a t-test. Bat capture at roosts Adult female bats were caught with a hand-net during evening emergence at 16 roosts of each of the two phonic types during June 1993-1996. The length of the left forearm was measured to the nearest 0.1 mm with dial callipers, as an index of body size, and a wing tracing was made of the left wing of each captured bat. A magnetic tablet (SummasSketch III, Summagraphics, Fairfield, USA) and soft- ware written by Professor J. M. V. RAYNER (School of Biological Sciences, University of Bristol) were Roosts of two phonic types of pipistrelle 299 used to digitise the wing tracings and morphological variables were measured from them (NORBERG and RAYNnER 1987). Variables measured were wingspan (B), total wing area (S), hand-wing area (HWA), hand-wing length (HWL), arm-wing area (AWA), and arm-wing length (AWL); variables cal- culated were aspect ratio (AR), tip length ratio (TL), tip area ratio (TS), and tip shape index (I). Each bat was released from the hand in open habitat, and its echolocation call sequence was re- corded via the high frequency output of a bat detector (S-25) to a Portable Ultrasound Processor (PUSP; Ultra Sound Advice, London, UK). A 2.2 s sequence of digitised signal (sampled at 448 kHz) was stored in the PUSP and replayed to a Walkman (WM-D6C; Sony, Tokyo, Japan) at one tenth of the original speed. The bat detector (S-25) microphone had a response of #3 dB from 20-120 kHz; the Walkman had a response of #3 dB from 40 Hz to 15 kHz. The recordings were analysed by using a Digital Signal Processing Sona-Graph (5500; Kay Elemetrics, Pine Brook, New Jersey, USA; 512 point fast Fourier transform with Hamming window, 400 Hz frequency resolution). The mean fre- quency containing most energy (FMAXE) of calls produced by each bat was calculated from power spectra of 3-6 echolocation calls. Each roost was considered to be composed of either 45 kHz P. pipistrellus or 55 kHz P. pipistrellus, on the basis of the mean FMAXE of all bats caught from that roost. Roosts were assigned to 45 KHz P. pipistrellus if the roost mean FMAXE was less than 49 KHz, and to 55 kHz P. pipistrellus if the roost mean FMAXE was greater than 52 kHz (Jones and VAN PA- russ 1993). This categorisation allowed unambiguous separation of the phonic types, with each phonic type corresponding to the two different genotypes with a sequence divergence of >11% in the cyto- chrome 5 gene of mitochondrial DNA identified by BARRATT et al. (1997). Variation in FMAXE of echolocation calls among roosts of each phonic type was investigated by using analysis of variance (ANOVA). Variance component estimates were calculated to determine how much variation in FMAXE was explained by differences among roosts, and how much by differ- ences among individuals (SoKAL and RoHLr 1995). Variables of wing morphology were compared be- tween phonic types with t-tests or Mann Whitney tests. Geographical variation in forearm length, ac- cording to roost location, was investigated by using multiple least squares regression analysis on roost latitude and longitude for each phonic type. The relationship between individual forearm length and FMAXE was investigated in the two phonic types. Results Roost counts The number of bats in 33 roosts of 45 kHz P. pipistrellus ranged from 20 to 223, with a median of 76 bats. The number of bats in 40 roosts of 55 kHz P. pipistrellus ranged from 30 to 650, with a median of 203 bats. There were significantly more bats in 55 kHz P. pipistrellus roosts than in 45 kHz P. pipistrellus roosts (t7ı = 6.15, P< 0.001; Fig. 1). Echolocation calls The 16 roosts of each of the two phonic types at which bats were caught are shown in fig- ure 2; between 6 and 20 adult female bats were caught at each roost. Figure 3 shows the distribution of individual FMAXE in the two phonic types. A comparison of FMAXE of echolocation calls found in this study and in previous studies of the two phonic types of P. pipistrellus is shown in table 1. In 45 kHz P. pipistrellus, there was no significant differ- ence in FMAXE among roosts (Fıs 165 = 1.66, NS; Tab. 2). Variance component estimates showed that only 5.5% of the variation in FMAXE was explained by differences among roosts, whereas 94.5% was explained by differences among individuals. In 55 kHz P. pipistrellus, there was a significant difference in FMAXE among roosts (F13.204 = 3-45, P<0.00, Tab. 2). Variance component estimates showed that 15.2% of the variation in FMAXE was explained by differences among roosts, and 84.3% by differences among in- dividuals. Three bats (of 401 recorded) which were assigned to 45 kHz P. pipistrellus on the basis of roost mean FMAXE, had FMAXE in the range 52-54 KHz (Fig. 3). These three individuals were therefore not included in further analysis. 260 Kate E. BARLow and G. JONES 14 Number of roosts = rn je») D > Er = = 2 EEE ED Zr WEBER BERN BT ZI ZI] EEREEER TE EEE TEE DE DER TEE ET Fin EIER FIE ZI I TE] TEEEELEEEEEEEEERIIEEEEEEEREEEEEEREREREEFERTEEEREE [_ 22 u DE EEEEEREREEEEEEREEEPEEER Beer] KEREEEEEEEEEEEEEEEEEEERESEEEEREEEEEEEEEEEEEEEEREER 9 Be er ae EL ESDEEER 7 z EEE EEESESEERSESSEEER EEE, Zr F7 - RER 2 >) en) oO >) >) >) (>) oO &) >) Si en Sl a Se ET >) em) >) >) >) em) =) (>) >) =) >) un >) un >) un >) un =) un >) @) —! — oa [a] an an xy xy un un Number of bats Fig. 1. Histogram showing the frequency distribution of the number of bats in 45 KHz P pipistrellus roosts (white bars) and in 55 kHz P pipistrellus roosts (hatched bars). There were significantly more bats in 55 kHz P. pipistrellus roosts than in 45 KHz P. pipistrellus roosts. Wing morphology There was much overlap in forearm length (mm) between the two phonic types (45 KHz P. pipistrellus: mean = 32.0, sd= 0.82, range 29.9-33.9, n=178; 55 KHz P. pipistrellus: mean = 31.7, sd = 0.77, range 29.9-33.7, n = 220; Fig. 4). However, forearm length was sig- nificantly longer in 45 kHz P pipistrellus than in 55 kHz P pipistrellus (t396 = 3.87, P < 0.001). Multiple regression analysis of forearm length on two measures of geographi- cal roost location, latitude and longitude, showed that there was no relationship between forearm length and roost location in either 45 kHz P pipistrellus (m = 0.024, F} 175 = 2.12, NS) or 55 kHz P pipistrellus (r* = 0.012, F5>17 = 1.28, NS). There was also no correlation between forearm length and FMAXE of echolocation calls in either 45 kHz P pipistrellus (T176 = 0.08, NS), or 55 kHz P. pipistrellus (r>18 =-0.05, NS). The variables B, S, HWA, HWL, AWL, TS, and I were all significantly larger in 45 kHz P pipistrellus than in 55 KHz P. pipistrellus (Tab. 3). However, there was much overlap in all these variables be- tween the two phonic types. Roosts of two phonic types of pipistrelle 261 Os [e\®) Fig.2. A map of mainland Britain showing the roosts at which bats of the two phonic types of P. pipistrellus were caught. Open circles represent 45 KHz P. pipistrellus roosts (n = 16); closed circles represent 55 kHz P. pipistrellus roosts (n = 16). Discussion Wing morphology and echolocation calls The FMAXE of echolocation calls of the two phonic types of P. pipistrellus recorded in this study was similar to that found in previous studies (Jones and van Parıss 1993; VAUGHAN et al. 1997b), the two types differing by 8-9 kHz on average. It is unclear whether the three bats (0.75% of total) that were classified as 45 kHz P. pipistrellus, but whose FMAXE fell within the range of 55 kHz P. pipistrellus were in fact individuals of 45 kHz P.pipistrellus with unusually high FMAXE, or were individuals of 55 kHz P. pipistrellus in a 45 kHz P. pipistrellus roost. 262 KATE E. BARLow and G. JoNES 23 20 10 5 0 Se Percentage of bats — N LEELLLLE ELLE ZELLE ELLE LTE DSL LED EL LEE LLLLL LEHE LEHE LLELE SL LEE L LEE LESE HL LLEH LS EHHF LH SSLHFFTHHFLTFLHE EHELLLL ZELL LE LLL LED ZELLE LLL DELL DELL LLL ELLE LLL LEE LLL LEST LEE ZZ LLLZLLSL EZ SLEEHLSSLLFLR LED EL LLLLLSL EL L LS LLLLT TTS LLLTLTLLE GL L ZELLE DL LT LEBE LL LTD LH ZZ ELLE ZELL LLLLL LEHE LH LT LT TEL TH HL ELLE ZELL ZELL LLLLLLEEL LT LLLLFLLLLZL LS 5 4-5 5 BELLE ZELL ELLE LLLL LS LEZ ELLE LT ZELL LLLL ZELLE LS LS HL HH HL ST LE LEHE ELLE LLC LH L ELF FL ELLE LE L LE, N | l | a SI | N [en] an [ee] EN >) - aN an SH \oO [0 (6,0) ON Da N | Er A ARD NDR ne Ra NEN — [eo] an St un \oO = 0,0) ON ) — oa [Sg] nn \O 5 [0,0] x x x x S Ay 23 Z. Säugetierkunde 64 (1999) 277-284 ZEITSCHRIFT SS FÜR © 1999 Urban & Fischer Verlag SÄUG ETIERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Intrageneric comparisons in urine-concentrating capacity and renal morphology among three species of Akodon from different geographic rainfall regimens By €. D. Antinuchı and CRısTINA BUSCH Departamento de Biologia, Universidad Nacional de Mar del Plata, Mar del Plata, Argentina Receipt of Ms. 09. 11. 1998 Acceptance of Ms. 18. 03. 1999 Abstract Urine-concentrating capacity and renal morphology were examined for Akodon azarae, A. iniscatus and A. cursor from different geographic rainfall regimens. A. azarae and A. cursor rejected sodium chloride solution above 0.45 M, whereas A. iniscatus did not. A. azarae and A. iniscatus may concen- trate urine to similar osmolarity values. These values were higher than those observed for A. cursor. Percent medullary thickness of the kidney for Akodon was related to both geographic rainfall regi- mens and urine osmolarity. Key words: Akodon species, urine osmolarity, renal morphology Introduction Water is the primary constituent of anımals. Mammals, particularly those adapted to arid habitats, can minimise their urinary water loss due to the fact that their kidneys can pro- duce urine that is significantly more concentrated than their plasma (BROOKER and WITH- ERS 1994). In mammals, adaptive radiation into new environments may involve both phy- siological and structural adaptations of the kidney. Hence, interspecific variation in these characteristics may represent adaptive responses to habitat differences in water availabil- ity. Urine-concentrating ability, and consequently the ability to conserve water have been evaluated in many morphological and functional studies on the mammalian kidney. These data have been correlated, in many instances, wıth water availability and habitat distribu- tion. One example is given by an earlier study performed by SPERBER (1944) after examin- ing species representing most mammalian orders, highlighting the relationship between mammal distribution related to climatic factors and their ability to concentrate urine throughout renal morphology. Later on, SCHMIDT-NIELSEN and O’Derr (1961), by studying several mammalian species, found a positive correlation between the relative medullary thickness (RMT; SPERBER 1944) and maximum urine osmolality. These studies have been followed by countless others, exploring the relationship between urine-concentrating abil- ity and RMT in hundreds of species of mammals (e.g. Hewırt 1981; Dunson and LAZELL 1982; LAWLER and GELUSO 1986; BEUCHAT 1990 b; BEUCHAT 1996). Other structural indices have been used to show the relationship between the relative size of the renal medulla and the capacity to concentrate urine: ratio of medullary to cortical thickness (M/C; GELUSO 1978), percent medullary thickness (PMT; HEISINGER and BREITENBACH 1969), relative me- 0044-3468/99/64/05 — 277 $ 12.00/0 278 C. D. AntinucHı and CRISTINA BUSCH dullary area (RMA; BROWwNFIELD and WUNDER 1976; HEwITT 1981) and percent medullary area (PMA; SchmiD 1972). Nevertheless, the above-mentioned indices must be carefully employed since recent studies on renal morphology have established some limitations in regards to their utilisation. Thus, studies performed by BROOKER and WITHERS (1994), have concluded that for marsupials, only those indices representing relative medullary lenght of the kidneys were correlated to climatic factors, whereas those representing medullary area were not. Moreover, BEUCHAT (1996) found that the relationship between thickness of the medulla and concentrating ability is neither proportional nor direct. Sigmodontinae rodents are abundant in many dry areas of South America, but few studies on their physiological adaptations have been published (Mares 1975; 1977a; 1977b; 1977 c; CorTES et al. 1988). Akodon is a polytypic genus that had a rapid adaptive radiation in the earliest Plio- cene (5.67 M. y. B. P.; ArpFELBAUM and Reıc 1989) and was distrubuted in South America from humid to semiarid regions with less than 200 mm precipitation per year. Hence, it is an interesting model for studying intrageneric differences in kidney function and struc- ture that could arise as a consequence of selection pressures imposed by environmental constraints. The aim of this study was to analyse the intrageneric pattern in urine-concentrating capacity and renal morphology of A. azarae, A. cursor, and A. iniscatus that could be the result of adaptive responses to habitats with different mean yearly rainfall. Material and methods Experimental animals Animals of both sexes were collected using Sherman live traps. Thirty four individuals of Akodon azarae were captured at Necochea, Buenos Aires Province (38°29 S, 58°50’W; rainfall average: 830 mm.year'; Pampeana biogeographic province; CABRERA and WiLLINK 1973). A. azarae is a mouse of moderate size (25 g body mass), strongly associated to natural grassland, particularly to open vege- tation formations (BONAVENTURA 1992; REDFORD and EISENBERG 1992). This species, which is the more representative rodent species of the pampa’s grasslands, is found from southern Brazil to central Ar- gentina (REDFORD and EISENBERG 1992). Thirty six individuals of Akodon iniscatus, which is a small size mouse (20 g body mass), were cap- tured at Puerto Madryn, Chubut Province (42°77'S, 65°82’ W; rainfall average 198 mm.year'; Patag6- nica biogeographic province). Captured mice were associated with vegetated coastal dunes with scarce availability to water. Distribution of this species also includes the xeric Monte and southern Espinal biogeographic provinces (CABRERA and WILLınK 1973; REDFORD and EISENBERG 1992). Thirty five individuals of Akodon cursor were captured at Posadas, Misiones Province (27°22’S, 55°58’ W; rainfall average 1604 mm.year '; Paranense biogeographic province; CABRERA and WILLINK 1973; REDFORD and EISENBERG 1992). This species, which is a medium size mouse (40 g body mass), is distributed in southern and central Brazil, Uruguay, Paraguay, and northeastern Argentina. At Mis- iones province it is found in most habitats but prefers flat and less moist areas (REDFORD and EISEN- BERG 1992). Climatic data records were obtained from the Meteorological Service of the Argentine Air Force. Only adult animals that maintained or gained body weight in the laboratry were used. Laboratory conditions Captured mice were carried to the laboratory and housed individually in anımal cages (30x22x15 cm). Wood shavings and cotton for nesting material were placed on cage floors. All anı- mals were maintained under a natural photoperiod (10L:14D). Temperature ranged from 18 to 25°C and relative humidity ranged from 50 to 80%. Animals were fed with dehydrated pellets ad lıbitum (composition: minimum protein = 21 %; maximum fiber = 4.5 %; minimum fat = 8%; average calcium = 1.8 %;, phosphorous = 1.1 %; maximum ashes = 8%). Tap water was provided ad libitum. Waterhandling in three species of Akodon 29 Salt-water regimen Tap water and sodium chloride, prepared in distilled water at different concentrations (0.05, 0.15, 0.25, 0.35, 0.45, 1 and 1.5 M), was offered as a drinking source. To measure consumption, a saline solu- tion was provided during 24h in inverted graduated glass Erlenmeyer, L-shaped drinking tubes, from which animals rapidly learned to drink. An additional inverted Erlenmeyer was placed near the cage to measure evaporation, which proved to be negligible. Once the Erlenmeyer with saline solution was removed, individuals were deprived of food and water and placed in a urine collection apparatus. Urine collection Urine was collected in glass vials containing mineral oil by using a collecting apparatus similar to that described by Drozpz (1975). Urine samples were collected for a total period of 14h, after which sal- ine solutions were removed and frozen at -20°C for analysis. Urine samples were discarded when contaminated by fecal material. Urinalyses Urine was analysed for total osmolarity measured with an Advanced 3MO osmometer. For evaluating urine-concentration capacity, urine data were analysed as follows: (1) for the whole rank of saline so- lutions offered (tap water to 1.5 M NaCl) and (2) grouping the data into two categories: low salt con- centration (tap water to 0.25 M Na Cl) and high salt concentration (0.35 to 1.5M NaCl). The latter procedure was performed to mitigate the potential effect of the low number of individuals due to ani- mal supply limitations. ANOVA was used to test the null hypothesis of no effect of saline solution treatments over urine osmolarity. Tukey’s test a posteriori was used to ıdentify differences between treatments. T test was used to test the null hypothesis of no effect of saline solution treatments grouped in low or high salt concentration categories over urine osmolarity. Renal index (PMT) Data on renal structure (relative size of the renal medulla) was collected as an ancillary measure of urine-concentrating capacity. Within 2 weeks following urine collection, animals were sacrificed by ether inhalation. Fresh kidneys were removed and fixed in 10 % formaline. The kidney tissue was de- hydrated and embedded in paraffin wax using standard histological techniques. Serial sagittal sections were cut at 10 um, and stained with heamatoxylin and eosin. The thicknesses of the cortical and me- dullary regions were measured to the nearest 0.1 mm with the aid of an ocular micrometer. For each kidney, percent medullary thickness (PMT) was determined by dividing the medullary thickness by the combined thickness of both regions (HEISINGER and BREITENBACH 1969). A mean of 10 measured PMT values was determined for each specimen. ANOVA was used to test the null hypothesis of no differences in relative medullary thickness (PMT) of the kidney among species. Tukey’s test a poster- iori was used to identify differences between species. Results Fluid consumption Above 0.25 M of saline solution Akodon iniscatus and A. cursor decreased their fluid con- sumption as saline concentration increased, whereas in A. azarae it was above 0.35 M. Furthermore, A. iniscatus drank saline solutions up to 1.5M, whereas A. azarae and A. cursor rejected saline solutions above 0.45 M (Fig. 1). Urinalyses Akodon iniscatus and A. azarae reached a similar urine osmolarity value (=3500 mOs- mol, ANOVA, F = 147.9, n=9, d.f.=2, Tukey’s test, P > 0.05). These values were higher 280 C. D. ANTINUCHI and CRISTINA BUSCH 120 100 [o.) oO Solution consumed (mi.day ”) [e)} [e>] 40 20 0 tapwater 0.05 0.15 0.25 0.35 0.45 1 165 Solution concentration (M) Fig. 1. Fluid consumption of Akodon azarae (MW), A. cursor ($), and A. iniscatus (4) drinking various concentrations of sodium chloride solution. Vertical lines show +1SD. low solution high solution 5000 d concentration concentation __» pooled data pooled data => 4000 E il O E 3000 = S E 2000 an oO ® 'E 0 tap water 0.05 0:15..2.025 0.35 0.45 1 8 Solution concentration (M) Fig. 2. Total urine concentration for Akodon azarae (MW), A. cursor (@), and A. iniscatus (4) on fluid regime. Vertical lines show + or - 1SD. than those observed for A. cursor (1000 mOsmo, Tukey’s test, P<0.05). However, A. iniscatus reached this value at 1.5M of ingested solution, whereas A. azarae and A. cursor reached it at a value of 0.45 M of ingested solution. The intraspecific comparison of pooled data between low and high solution concentra- tions (Fig. 2) showed statistical differences in urine osmolarity for both Akodon azarae (t- test, n (low) = 12, n (high) = 8, t= 4.2, df = 18, P< 0.001) and A. iniscatus (t-test n (low) = 18, n (high) = 7, t=-5.1, df = 23, P< 0.001). On the other hand, these differences were not significant for A. cursor (t-test n (low) = 18, n (high) = 7, t=-1.8, df = 23, P > 0.089). Waterhandling in three species of Akodon 281 100 60 er [sp] \ = u Rn Se650 I S R I N 40 8 © u 2) O = S S S 30 S & n E s i 20 < E S < S x 10 < 198 830 1604 Mean annual rainfall (mm) Fig. 3. Comparison of the percentage medullary thickness (PMT) of the kidneys of the selected Ako- don species from three different rainfall regime areas. Vertical lines show + or - 1SD. The interspecific comparison of pooled data at low and high solution concentration treatments showed that Akodon azarae exhibited the highest urine osmolarity followed by A. iniscatus and A. cursor (ANOVA, low: F = 24.74, n=46, df=2, p<0.001; high: 2 277528407 26, dt -2,P=0.03). Renal index (PMT) The three species showed significant differences in percent of medullary thickness (PMT; ANOVA, F = 71.87, n = 37, df = 2; Tukey’s test, P< 0.001). Akodon iniscatus showed the highest mean value (PMT = 76.5 + 3.4 %, n = 13), A. azarae showed the medium mean va- lue (PMT = 69.4+3.6%, n=12) and A. cursor showed the lowest mean value (PMT = (U E92, NE ae Discussion Small mammals vary dramatically in their ability to concentrate urine osmotically. Mesic mammals generally have a low capacity to concentrate urine and the rely on drinking as an avenue of water gain as was reported for many families of rodents (WHITFORD and ConLEyY 1971; GREENE and FERTIG 1972; MAREs 1977 a). On the other hand, small desert mammals cannot rely on drinking and their water requirements must be met by pre- formed water and/or metabolic water and by reducing the urinary water loss to a minimal level (BROoKER and WITHERsS 1994). These adaptations are well documented for Dipod- omys spectabilis (SCHMIDT-NIELSEN 1964) and D. merriami (Nagy and GRUCHACZ 1994). Unlike the last rodent species cited above, the three species of Akodon studied pre- sently, drink water to survive. However, at high concentrations of water solutions, these species do not drink and a progressive deterioration of body condition was observed. 282 C. D. ANTINUCHI and CRISTINA BUSCH With respect to urine concentration, urine osmolarity in species of Akodon differed after ingesting relatively low or high NaCl drinking solutions. Furthermore, there seems to be different renal handling of excess salt intake among the three species: A. cursor, which lives in higher rainfall areas, favor drinking relatively low concentrations of NaCl solution, and choose not to drink high NaCl solutions. In contrast, A. iniscatus and A. azarae, which live in dryer areas, did not drink substantial amounts of NaCl solutions, but showed good tolerance to high NaCl solutions, by forming a concentrated urine and conserving water. Therefore, the general pattern of water-salt balance found in the three analysed spe- cies of Akodon was similar to that reported for Rattus rattus, Phillotis osilae, Oryzomys longicaudatus, Akodon varius, and some phillotine rodents (NORMAN and BAUDINETTE 1969; Dunson and LAZELL 1982; MAREs 1977 a; 1977 cc). Renal index has been regarded by some researchers as a good indicator of urine-con- centrating ability in mammals (SCHMIDT-NIELSEN and O’Deır 1961). Significant relation- ships between urine concentrating ability, renal index, and the availability of water have been reported for the genera Sylvilagus, Microtus, and Peromyscus (HEISINGER and BREI- TENBACH 1969; HEISINGER et al. 1973, MAcMiıLLEn 1983). On the other hand, BEUCHAT (1993) has found the renal index to be less reliable. BEUCHAT (1996) also found a signifi- cant relationship between the thickness of the inner medulla only in species from mesic environments, reflected in total medullary thickness, and the concentrating ability of the kidney. However, the percent of medullary thickness of the kidney (PMT) for Akodon species from xeric, mesic, and humid environments, was related to both the geographic rainfall regime and to its renal-concentrating ability. Thus, intrageneric differences in urine concentrating capacity and renal morphology would be a consequence of the adaptive radiation of this genus in the earliest Pliocene (5.67 M. y. B. P. ApFELBAUM and Reıc 1989), related to different selection pressures. Other influences such as diet, environmental conditions and behaviour affect the over- all urine-concentrating ability (BEucHAT 1990 a; 1990 b). Studies in natural conditions of these factors would be important to elucidate the physiological adaptations of these spe- cies. Acknowledgements The authors wish to express their gratitude to Boßy TAYLoR and Dra. GEORGINA BARRANTES for provid- ing specimens of Akodon iniscatus and Akodon cursor used in this study. Thanks also to the late Dr. OsvaLpo REIG and members of the Laboratorio de Histologia for suggestions related to the plan- ning of this work. We also thank Dr. Oscar IRIBARNE, Dra. ALEJANDRA LOPEZ MANANES, Lic. ROXANA ZENUTO, Dra. Ana MALIZIA, and Dr. MARCELO KITTLEin for improving the manuscript with their com- ments and suggestions. This work was granted by Universidad Nacional de Mar del Plata. Zusammenfassung Intragenerische Vergleiche zwischen der Fähigkeit, Urin zu konzentrieren und der Morphologie der Niere bei drei Akodon-Arten aus geographischen Regionen mit unterschiedlichen Niederschlags- mustern Die Fähigkeiten zur Konzentration des Urins und die Morphologie der Nieren von Akodon azarae, A. iniscatus und A. cursor aus Regionen mit unterschiedlichen Niederschlagsmustern wurden unter- sucht. Im Gegensatz zu A. iniscatus verweigerten A. azarae und A. cursor NaCl-Lösungen über 0,45M. A. azarae und A. insicatus können ihren Urin bis zu etwa gleichhohen Werten konzentrieren. Diese Werte lagen höher als bei A. cursor. Die Prozentwerte der Nierenmarklänge zeigen eine Bezie- hung zwischen dem Niederschlagsmuster der Region und der Urinmolarität. Waterhandling in three species of Akodon 283 References APFELBAUM, L. 1.; REIG, ©. A. (1989): Allozyme genetic distances and evolutionary relationships in spe- cies of akodontine rodents (Cricetidae: Siıgmodontinae). Biol. J. Linn. Soc. 38, 257-280. BEUCHAT, C. A. (1990 a): Metabolism and the scaling of urine concentrating ability in mammals: resolu- tion a paradox? J. Theor. Biol. 143, 113-122. BEUCHAT, C. A. (1990 b): Body size, medullary thickness, and urinary concentrating ability in mammals. Am. J. Physiol. 238R, 298-308. BEUCHAT, C. A. (1993): The scaling of concentrating ability in mammals. In: New Insights in Vertebrate Kidney Function. Ed. by J. A. Brown, R. J. BALMENT, and J. C. Rankın. Cambridge: Cambridge Univ. Press. Pp. 259-279. BEUCHAT, C. A. (1996): Structure and concentrating ability of the mammalian kidney: correlations with habitat. Am. J. Physiol. 40R, 157-179. BONAVENTURA, S. M.; KRAVETZ, F. O.; SUAREZ, O. V. (1992): The relationship between food availability, space use and territoriality in Akodon azarae (Rodentia, cricetidae). Mammalia 56, 407-416. BROOKER, B.; WITHERS, P. (1994): Kidney structure and renal indices of dasyurid marsupials. Aust. )J. Zool. 42, 163-176. BROWNFIELD, M. S.; WUNDER, B. A. (1976): Relative medullary area: A new structural index for estimat- ing urinary concentrating capacity of mammals. Comp. Biochem. Physiol. A. 58, 413-419. CABRERA, A. L.; WILLINK, A. (1973): Biogeografia de America Latina. Washington D. C.: OEA.E. V. Chesneau. CoRTES, A.; ZULETA, C.; ROSENMANN, M. (1988): Comparative water economy of sympatric rodents in a Chilean semi-arid habitat. Comp. Biochem. Physiol. A 91, 711-714. DrozDpz, A. (1975): Feeding and nutrition. Metabolic cages for small rodents. In: Methods for Ecologi- cal Bioenergetic. IBP Handbook. Vol. 24. Ed. by W. GRropxZzinskKI, R.Z. KLEKOWSKI, and A. Duncan. Oxford: Blackwell Sci. Publ. Pp. 325-351. Dunson, W. A.; LAZELL, J. D. Jr. (1982): Urinary concentrating capacity of Rattus rattus and other mammals from the lower Floida Key. Comp. Biochem. Physiol. A 71, 17-21. GELUSo, K.N. (1978): Urine concentrating ability and renal structure of insectivorous batts. J. Mammal- ogy 59, 312-323. GREENE, J. R.; FERTIG, D. S. (1972): Water sources for house mice living in salt marshes. Physiol. Zool. 45, 125-129. HEISINGER, J. F.; BREITENBACH, R. (1969): Renal structural characteristics as indexes of renal adaptatıon for water conservation in the genus Sy/vilagus. Physiol. Zool. 42, 160-172. HEISINGER, J. F.; Kıng, T. S.; HALLing, H. W.; FieLds, B. L. (1973): Renal adaptation to macro- and mi- cro-habitats in the family Cricetadae. Comp. Biochem. Physiol. A 44, 767-774. HEWwITT, S. (1989): Plasticity of renal function in the Australian desert rodent Notomys alexis. Comp. Biochem. Physiol. A 69, 297-304. LAWLER, R. M.; GELUSo, K.N. (1986): Renal structure and body size in heteromyid rodents. J. Mammal- ogy 67, 367-372. MACMILLEN, R. E. (1983): Water regulation in Peromyscus. J. Mammalogy 64, 38-47. MARES, M. A. (1975): South American mammal zoogeography: evidence from convergent evolution in desert rodents. Proc. Nat. Sciences 72, 1702-1706. MARES, M. A. (1977 a): Aspects of the water balance of Oryzomys longicaudatus from northwest Ar- gentina. Comp. Biochem. Physiol. A 57, 237-238. MARES, M. A. (1977b): Water economy and salt balance in a South American desert rodent Eligmodon- fia typus. Comp. Biochem. Physiol. A 56, 325-332. MARES, M. A. (1977c): Water balance and other ecological observations on three species of Phyllotis in Northwestern Argentina. J. Mammology 58, 514-520. Nagy, K. A.; GRUCHACZ, M. J. (1994): Seasonal water and energy metabolism of the desert-dwelling kangaroo rat (Dipodomys merriami). Physiol. Zool. 67, 1461-1478. NORMAN, F. Y.; BAUDINETTE, R. V. (1969): Water economy and salt balance of an insular population of Rattus rattus Linnaeus. J. Mammalogy 50, 487-493. REDFORD, K. H.; EISENBERG, J. F. (1992): Mammals of Neotropics. The South Cone. Chicago: Univ. Chi- cago Press. SPERBER, I. (1944): Studies on the mammalian kidney. Zool. Bridr. Uppsala 22, 249-431. SCHMID, W. D. (1972): Nocturalism and variance in ambient vapor pressure of water. Physiol. Zool. 45, 302-309. 284 C. D. ANTINUCHI and CRISTINA BUSCH SCHMIDT-NIELSEN, K. (1964): Desert Animals: Physiological Problems of Heat and Water. New York: Oxford Univ. Press. SCHMIDT-NIELSEN, B.; O’DELL, R. (1961): Structure and concentrating mechanism in the mammalian Kidney. Am. J. Physiol. 200, 1119-1124. WHITFORD, W. D.; ConLEy, M. I. (1971): Oxygen consumption and water metabolism in carnivorous mouse. Comp. Biochem. Physiol. A 40, 797-803. Author’s address: C. D. ANTINUCHI and CRISTINA BUSCH, Departamento de Bioligia, FCEyN, UNMAP. €. €. 1245 (7600) Mar del Plata, Argentina. 67 IS Z. Säugetierkunde 64 (1999) 285-297 ZEITSCHRIFT ®® FÜR © 1999 Urban & Fischer Verlag SÄUGETI ERKU NDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Genetic and morphological variation among populations of the bank vole Clethrionomys glareolus from north-eastern Poland: the seasonal aspect By ANETTA BORKOWSKA Institute of Biology, University of Bialystok, Bialystok, Poland Reicept of Ms. 10. 03. 1999 Acceptance of Ms. 11. 05. 1999 Abstract The relationship between genetic and morphological variation was studied in five populations of the bank vole (Clethrionomys glareolus) from NE Poland. For the assessment of genetic polymorphism 37 enzyme loci were investigated. The proportion of polymorphic loci (P) ranged from 0.162 to 0.270 and the average observed heterozygosity (Ho) from 0.074 to 0.095 in populations studied. F-statistics indicated the existence of seasonal variation in allele frequency in four populations (Fsr = 0.011- 0.018, p<0.05). The differentiation among vole populations was greater in autumn (Fsr = 0.032, p < 0.0001) than in spring (Fsr = 0.023, p < 0.0001). Morphometric variation in 3 groups of parameters (I - body size, II - size of internal organs, III — cranial and mandibular traits) was subjected to princi- pal component analysis (PCA). There was a very clear separation of season in all three groups of parameters in each population. PCA analysis revealed differences among vole populations in body size, organ size and cranıum size and shape in spring and only in body size and cranıum shape in au- tumn. No correlation was found between Rogers’ genetic distances and Mahalanobis distances, as cal- culated from 3 groups of morphological characters in both seasons (r = |0.01-0.43|, NS; Mantel’s test). The data suggest that there is no equivalent degree of divergence on these two levels of integration in C. glareolus. Key words: Clethrionomys glareolus, allozymes, morphometry, differentiation Introduction There is abundant geographic variation in both morphology and gene frequency in most species. The extent of geographic variation results from balance of forces tending to pro- duce local genetic differentiation and forces tending to produce genetic homogeneity (SLATKINn 1987). Protein electrophoresis has been widely used to describe genetic differ- ences among populations of rodent species (LEITNER and HARTL 1988; GALLARDO et al. 1992; GEBCZYNSKI et al. 1993; FEDORoOV et al. 1995). The degree of genetic differentiation among populations over wide geographical distances is higher than within a narrow geo- graphic area (GEBCZYNSsKI et al. 1993; FEDoRovV et al. 1995). It has also become obvious that there was considerable morphological variation among local populations of the spe- cies. Morphological differences between animals from various regions have hitherto mainly been related to the environmental conditions or, as in genetic instances, to the geographic distance and isolation (HAITLINGER 1965, 1970; Hansson 1985; BAKER 1992; SARA and CASAMENTO 1995). 0044-3468/99/64/05 - 285 $ 12.00/0 286 ANETTA BORKOWSKA Genetic and morphological observations are very rarely carried out on the same mate- rial. Thus, there were only a few estimations of the relationships between genetic and morphometric divergence among populations in mammals (HArTL et al. 1993; KITCHENER et al. 1994). On the other hand, there are numerous studies dealing with this problem in other groups of animals (LAZARIDOU-DIMITRIADOU et al. 1994; KYRIAKOPOULOU-SKLAVONOU et al. 1991; BAKER 1992; LoBo 1995). The intrapopulation variability may have an effect on differences among populations. There is considerable evidence that body mass, organ weights, and gut morphology of ro- dents can change seasonally in response to changes in feeding habits, reproductive state, ambient temperature or photoperiod (HAmMmonD 1993; NAGy and NEGus 1993; HAMMOND and DıamonD 1994; BORKOWSKA 1995; CAMPBELL and MACARTHUR 1996). Likewise, a spring generation of rodents differs in allele frequency from an autumn generation (FE- DYK and GEBCZYNSKkI 1980). Thus, seasonal variation may also be reflected in morphologi- cal and genetic differentiation among bank vole populations. The bank vole, Clethrionomys glareolus (Schreber, 1780) is one of the most common Palearctic rodent species. It has a wide geographical distribution from the British Isles to Lake Baikal, and from Kola Peninsula to Asia Minor (RAczynskiI 1983). The genetic dif- ferences among bank vole populations have been determined in Austria (LEITNER and HARTL 1988) and Poland (GeEBcZyNskI et al. 1993). Morphological differences were found between eastern and western (HAITLINGER 1965) as well as between mountain and low- land vole populations in Poland (HAITLINGER 1970). The purpose of this study was to determine variability in allozymes and morphology (body size, size of internal organs, and skull dimension) among the bank vole populations over short geographic distances. Next, the seasonal aspects of the interpopulation diver- gence will be taken into account. Material and methods A total of 391 individuals of C. glareolus was collected from 5 populations in the vicinity of Bialystok (NE Poland 23°07E, 53°18’N, Tab. 1). The minimum distance between two sites was 10 km, maximum 50 km. Animals were caught in live-traps during two seasons: spring (May-June) and autumn (Octo- ber-November) in 1995-97. The voles were brought into the laboratory and dissected. A spring gen- eration of the bank vole consisted of over-wintered individuals, while in autumn the populations were only made up of current-year animals. Samples of blood plasma, kidney, liver, and salivary gland were taken from each vole and stored at -85°C until used for electrophoresis. Tissues were homogenized in phosphate buffer (0.01M, pH 7.5) and then centrifuged at 12000 rpm for 15 min at 4°C. Protein electrophoresis was performed on (1) starch gel following the running and staining conditions given by SELANDER et al. (1971), HAR- rıs and Hopkınson (1976), Quavı and Kır (1980), and (2) cellulose acetate plates (SEARLE 1985), and (3) agar gel (NIELSENn 1977). Gene products for the following 37 presumptive enzyme loci were ana- lysed. Locus were the following (E.C. number are given in parentheses): «@Gpd-1, «Gpd-2, BGpd-1, and BGpd-2 (1.1.1.8), Sdh (1.1.1.14), Ldh-1 and Ldh-2 (1.1.1.23), Mdh-1 and Mdh-2 (1.1.1.37), Me-1 and Me-2 (1.1.1.40), Idh-1 and Idh-2 (1.1.1.42), Pgd (1.1.1.44), Dia (1.6.2.2), Cat (1.11.1.6), Sod-1 and Sod-2 (1.15.1.1), Aat-1 and Aat-2 (2.6.1.1), Pgm-1, Pgm-2, and Pgm-3 (2.7.5.1), EstB3 and EstD (3.1.1.1), Amyl1-2 (3.2.1.1), Pep-2 and Pep-3 (3.4.11), Acy (3.5.1.14), Ald-1 and Ald-2 (4.1.2.13), Acon- 1 and Acon-2 (4.2.1.3), Mpi (5.3.1.8), Pgi (5.3.1.9), Alb, and Prot. A locus was considered poly- morphic if the frequency of the most common allele did not exceed 0.95. Alleles at polymorphic locı were designated alphabetically with increasing anodal migration of the corresponding allozymes. BIOSYS-1 (SwoFFORD and SELANDER 1989) was used to calculate observed average heterozygosity (Ho) and x” contingency test for homogenity. If significant deviation from Hardy-Weinberg equili- brium was found the fixation index Fıs was calculated. The deviation of Fıs from zero was tested by X = NF%, where N is the total sample size (NEı 1977). Wright’s Fsr was calculated to quantify the Genetic and morphological variation in Clethrionomys glareolus 287 amount of genetic differentiation between seasons in each population and among populations in spring and autumn, separately. An F-statistic value was considered to be significantly different from zero if statistically significant heterogeneity among samples at the same hierarchical level was found (x? test). Pairwise Rogers’ genetic distances (Dx) were estimated from allozyme data and clustered using the UPGMA method (SWwOFFORD and SELANDER 1989). Morphometric variation in the 5 populations of the bank vole was assessed by taking 39 mea- surements on 349 individuals older than 3 months (age was determined according to PUcEK and ZEIDA 1968). Morphological characters were divided into three groups: I - body size (5 external measurements: total body mass TBM to the nearest 0.1 g, and head and body length HBL, tail length TL, hind foot length HFL, ear height EH to the nearest 0.01 mm; PUcEK 1981); II - organ size (13 internal measurements: dry mass of stomach STM, small intestine SIM, large intestine LIM, caecum CM, liver LM, kidneys KM, spleen SM, heart HM, lungs LUM, scraped muco- sa SMM to the nearest 0.001 g, and length of small intestine SIL, large intestine LIL, and caecum CL to the nearest 0.01 cm; MyrcHA 1964; DıamonD and KArAsov 1984; HAMMoND and DiıAaMmonD 1994); III — skull dimension (21 cranial measurements: condylobasal length CBL, total cranium length TCL, basal length BL, rostral width RW, interorbital width IW, zygomatic width ZW, mastoid width MW, palatal height PH, skull height per auditory bullae SPB, skull height between auditory bullae SBB, upper diastema length UDL, mandibule length ML, mandibular ramus height MH, recorded by dial caliper to the nearest 0.01 mm, and length of nasal bones NBL, length of frontal bones FBL, length of sagital crest SCL, length of interparietal bone IBL, incisive foramen length IFL, palatal length PL, upper molar series length from the alveoles UMSL, lower molar series length from the al- veoles LMSL, recorded by binocular microscope with micrometer ocular to the nearest 0.01 mm; HAITLINGER 1965; Vıro and NIETHAMMER 1982). Morphological data were quantitatively compared in three groups separately by the use of princi- pal component analysis (PCA, STATISTICA, StatSoft. Inc. 1995). The scores of the first principal component were used to calculate Mahalanobis distance (D°) in a pairwise fashion between all popu- lations in both seasons. To test differences among populations studied and seasons, ANOVA test and Fisher’s least significant difference tests (L.S.D.) were performed on the first three factors of PCA (PC1, PC2, PC3). MANTEL’s (1967) test was used to test relationships between Rogers’ genetic dis- tances and Mahalanobis morphological distances among vole populations in both seasons. The analy- sis was performed using the TFPGA computer programme (MiLLEr 1997). Results Genetic analysis Fourteen of the 37 loci examined were found to be polymorphic in the bank vole from NE Poland, as defined using the 0.95 common allele frequency: Ldh-2, Me-2, Dia, Cat, Aat-2, Pgm-1, Pgm-2, Pgm-3, EstB3, EstD, Amyl-2, Pep-2, Acy, Mpi. But only five loci: Me-2, Dia, Pgm-3, EstB3, and Amyl-2 were polymorphic in every population studied and in both seasons. The percentage of polymorphic loci (P) ranged from 0.162 to 0.270 and the mean observed heterozygosity (Ho) from 0.070 to 0.095 in the bank vole populations (Tab. 1). There were no significant differences in observed heterozygosity H, between seasons in each population and among populations either in spring or in autumn (p > 0.05; Kruskal-Wallis test). Two populations in spring (BIA, ZED) and two others in autumn (PRZ, SZE) showed significant departures from Hardy-Weinberg equilibrium in a few loci (Tab. 2). There was a significant deficit of heterozygotes in all these popula- tions. To measure genetic differences between spring and autumn generations Fsr values were calculated (Tab. 3). In four populations mean Fsr values were statistically significant indicating that above 1 per cent of genetic variation in a C. glareolus population was attri- butable to differences between seasons. There was a low but significant genetic differen- 288 ANETTA BORKOWSKA Table 1. Sample size (N), percentage of polymorphic locı (P) and observed heterozygosity (Ho) in spring and autumn generations of C. glareolus. Abbreviations of the populations are in paren- theses. Population Spring generation Autumn generation Bialystok (BIA) Suprasl (SUP) Przewalanka (PRZ) Zednia (ZED) Szelagowka (SZE) Table 2. Allele frequencies at 14 polymorphic loci in spring (SG) and autumn (AG) generations of Clethrionomys glareolus from NE Poland. Asterisks indicated deviation from Hardy-Weinberg equilibrium. Locus Bialystok Suprasl Przewalanka Zednia Szelagowka Allele Genetic and morphological variation in Clethrionomys glareolus 289 Table 2. (Continued) Bialystok Suprasl Przewalanka Zednia Szelagowka a 0.761 OT NEE TEE 0 EIIZEZE0 72006023 0.718 0.690 0.598 b 0.239 020952031722 023727 103067720280752.0308 0282702937 20378 c 0.000 0.000 0.083 0.013 0.000 0.000 0.019 0.000 0.017 0.024 EstD a 0.875 0.857 0862 091 0.806 0.793 0.808* 0.962 0.948 0.878 b 0.125 0401B.0238777.0:07927 1041947720417 1782. 0419227010387 20.052 70122 € 0.000 0.000 0.000 0.000 0.000 0.037 0.000 0.000 0.000 0.000 Amyl-2 a 0.000 0.000 0.0221 0.000 0.000 0.000 0.000 0.000 0.000 0.000 b 01607.2.0:8435 2210750709087 2.077337 20:59 7.08 0.784 0.727 0.868* c 05332 4.043577 30229E7E 0 02702074 2091415103188 0:2116577.02737 2.0.1322 Pep-2 a 0.102 0.080622.0:065522.0:0132220:.0167°=70:06377:0:000 0:000,29.0:07222041595 b 0.898 VELI URS EEE VE EN BI 120007210:8752.2.0:8:17> € 0.000 0.014 0.000 0.013 0.000 0.024 0.019 0.000 0.054 0.024* Acy a 0.000 0.000 0.033 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.045 01028-2740:00072.0415872.0:0002720.03722:0:000 01077222 0:000920:012 c 0.955 0ITPEEEN:ICTTEEE NV: 81271. 000720.96377 31000 0:923725521:00052:0.988 Mpi 0.000 0.000 0.000 0.000 0.016 0.013 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.065 0.074 0.019 0.000 0.034 0.000 0.989 0195 055,.0:968.200: 977127209 05252.0.9132720981 1.000 0.949 1.000 0.011 0.064 0.032 0.029 0.016 0.000 0.000 0.000 0.017 0.000 no om Table 3. Mean F-statistic values at all lociı for seasonal generations in five populations of Clethrio- nomys glareolus from NE Poland. Asterisks denote statistical significance for Fsr as determined by the x? test (see text); *p < 0.05; **p < 0.01; NS = nonsignificant. Site abbreviations as in Tab. 1. Population 0.012** 0.0182 0.011* 0.011* 0.012 NS tiation among the bank vole populations in spring (mean Fsr = 0.023; p < 0.0001) in eight loci (Dia, Aat-1, Pgm-1, Pgm-2, Pgm-3, EstB3, Pep-2, Acy). However, in autumn the mean Fsr value was higher (Fsr = 0.032; p< 0.0001) and 16 single locus Fsr values were statistically significant (Ldh-1, Ldh-2, Me-2, Idh-1, Pgd, Dia, Cat, Aat-1, Aat-2, Pgm-l, Pgm-2, Pgm-3, EstD, Pep-2, Acy, Mpi). Rogers’ genetic distance estimates based on pair- wise comparisons of the five sites ranged from 0.018 to 0.029 in spring and from 0.023 to 0.039 in autumn, indicating that populations were very similar to one another. The link- age distances were so low that any groupings in the clusters could be considered to be random associations. However, two geographical nearest populations (SUP and ZED) 290 ANETTA BORKOWSKA 0 0.005 0.010 0.015 0.020 0.025 0.030 0.035 Rogers’ genetic distance Fig. 1. Phenogram generated by UPGMA cluster analysis based on Rogers’ genetic distances among five populations of C. glareolus in spring (solid lines) and autumn (broken lines). Site abbreviations as in Tab. 1. clustered together in both seasons (Fig. 1). Correlation of two genetic distance matrices from various seasons revealed that genetic differentiation among the bank vole popula- tions in spring was not covered by interpopulation differentiation in autumn (r = 0.06, Z = 0.01; NS; Mantel’s test). Morphological variation Seasonal variation was conspicuous in all the populations studied. The first principal com- ponent (PC1) explained over 50% of the total variation in body size (I) and about 40% in organ size (II) and cranıum dimension (III) in each population. Nevertheless, the be- tween-season variation in three groups of morphological parameters described by PC1 was significant in all populations of the bank vole (Tab. 4). Principal component analyses of C. glareolus revealed significant differences in body size (group I) among populations both in spring (ANOVA on PC1: F = 7.24; p< 0.0001) and autumn (ANOVA on PC1: F = 10.12; p< 0.0001). However, the first PC based on group II of morphological parameters divided the bank vole populations only in spring (F = 2.78; p< 0.05) but not in autumn (F = 1.01; NS). The first component (PC1) based on craniometric data (group III) explained 36% of total amount of phenotypic variability in spring and about 32% in autumn (Tab. 5). The differences observed among the bank vole populations also referred to skull dimension. The PC1s were of the size-type with all coefficients positive in both seasons, which suggested an influence of overall skull size on group separation. The second factor (PC2) based on craniometric parameters ac- counted for 8.35% of the variation in spring. In this season it loaded positively on mea- surements of skull and mandibule length (CBL, TCL, BL, UDL, ML) and negatively on parameters of skull width (RW, IW, ZW) and height (PH, SPB, SBB; Tab. 5). Thus, PC2 could be interpreted as a shape-type component in spring. In autumn, however, the sec- ond (PC2) and the third (PC3) factors which explained 24.3% of the total variance to- gether represented ‘shape’ variation (Tab. 5). Analyses of variance confirmed that in spring the bank vole populations differed significantly in cranium size (PC1: F = 6.45; p< 0.0001) and cranium shape also (PC2: F = 4.18; p< 0.01, PC3: F = 13.59; p< 0.0001). Genetic and morphological variation in Clethrionomys glareolus 291 Table 4. Seasonal variation in body size (I), organ size (II), and skull dimension (III) in five popula- tions of Clethrionomys glareolus from NE Poland revealed by one-way ANOVA on PC1. EV = eigenvalue, V = per cent of total variance, p = significance level for between-season comparison in ANOVA. Only factor loading values greater than absolute 0.60 are presented. Site abbreviations as in Tab. 1. Acronyms are described in the text. Population EV Factor loadıngs BA 967 5344 HBL (0.91), TBM (0.85), TL (0.74) 0.0000 II 461 35.43 LM (0.84), KM (0.78), SIM (0.77), SMM (0.70) 0.0000 III 656 31.23 BL (-0.96), CBL (-0.95), TCL (-0.92), UDL (-0.81), ZW (-0.78), 0.0003 MW (-0.76) 3.18 63.68 HBL (0.92), TBM (0.88), TL (0.82) 0.0000 6.68 5142 _STM (0.91), KM (0.87), LIM (0.83), SIM (0.82), HM (0.80), 0.0000 CM (0.78), LUM (0.74) 8.45 40.26 TCL (0.97), CBL (0.95), BL (0.94), NBL (0.85), ZW (0.83), 0.0009 ML (0.83), UDL (0.82), IFL (0.72) 3.13 62.66 TBM (-0.94), HFL (-0.92), EH (-0.82), TL(-0.77) 0.0000 5.89 45.29 SIM (0.89), STM (0.83), SMM (0.82), LUM (0.82), CM (0.80), 0.0000 LIM (0.75) 6.99 33.30 _CBL (0.88), BL (0.86), MW (0.77), ZW (0.73), FBL (0.72), 0.0000 UDL (0.71) 276 5522 TBM (-0.93), HBL (-0.93), TL (-0.84) 0.0000 6.11 47.01 STM (0.91), CM (0.84), KM (0.83), SIM (0.82), LIL (0.70) 0.0000 9.04 43.07 TCL (-0.94), CBL (-0.93), BL (-0.93), ZW (-0.84), NBL (-0.84), 0.0000 MW (-0.80), UDL (-0.77), FBL (-0.76), IFL (-0.74) 3.24 64.84 HBL (0.93), TBM (0.85), EH (0.84), TL (0.78) 0.0000 4.59 35.30 STM (0.90), CM (0.85), KM (0.82), HM (0.75) 0.0000 793 3778 BL (0.93), CBL (0.92), TCL (0.92), UDL (0.88), ML (0.88), 0.0106 ZW (0.80), NBL (0.80), MW (0.72), IFL (0.71) However, two ‘shape’ components (PC2 and PC 3) only separated vole populations in au- InDEENNONVA on P@I7 7ER =71.46; NS, PC: 7E=19.717 p<0:.001, PC3:’ E- 11.76; p < 0.0001). Mahalanobis distances (D°) calculated from the first principal component based on group I of morphological parameters ranged from 0.046 to 1.456 among populations in spring and from 0.001 to 1.618 in autumn. The differences in organ and cranıum size (PC1 based on group II and III) were not found among populations in autumn. Thus, in this season the Mahalanobis distances among populations based on these two groups of variables ranged only from 0.0-0.587. Likewise, D* index also reached the low values when we analysed differences among populations in size of voles internal organs (group II; D” range 0.023-0.426) and cranium size (group III, D* range 0.022-1.317) in spring. Cluster diagrams generated on the basis of Mahalanobis distances from group I of morphological parameters were similar in spring and autumn (r = 0.71; Mantel’s test) but the correlation coefficient was not significant (Z = 2.16; P = 0.0560, Fig. 2 A). There was no significant correlation in Mantel’s test between matrices of Mahalanobis distances in spring and autumn based on group II (r=-0.34, Z= 0.10; NS, Fig. 2B) and group III of morphological parameters (r = -0.28, Z = 0.75; NS, Fig. 2C). 292 ANETTA BORKOWSKA Table 5. Component loadings for the first three principal components (PC1, PC2, PC3) analysed in spring and autumn generations of Clethrionomys glareolus. Five populations are included in the analy- sis and PCA is based on 21 craniometric variables. EV = eigenvalue, V = per cent of total variance. Acronyms are described in the text. Variables Spring generations Autumn generation 0.164 0.200 0.151 0.101 U 0.419 0.316 -0.516 0.090 0.014 0.090 0.108 0.406 0.275 0.705 0.605 -0.073 -0.587 -0.071 -0.671 -0.808 Sr 15275 Table 6. Correlation between Rogers’ genetic distances Dx and Mahalanobis distances D? calculated from PC1 of groups I, II, and III of morphological parameters in spring (SG) and autumn (AG) gen- erations of Clethrionomys glareolus. r = correlation coefficient, Z = statistic value, p = probability level after randomly permuting the values in one of the distance matrices 1 000 times, Mantel test. Comparison Generation DE. DI) Dr xD’) Dr x D’ (I) Correlation between genetic and morphological divergence Cluster diagrams generated on the basis of Rogers’ distances and Mahalanobis distances (in three groups of parameters separately) were not similar either in spring or in autumn, suggesting no equivalent degree of divergence on the two levels of integration. To quantify the lack of this relationship correlation coeffceients between Dr matrices and D° matrices were calculated, with regard to season and group of morphological parameters (Tab. 6). The correlation coefficients (r) were low and not significant in all cases of Mantel’s test. Genetic and morphological variation in Clethrionomys glareolus 293 BIA A SUP PRZ (EEEERTERETEETEREEEUERTERTEREREUERTEERER| 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Mahalanobis distance (D2 I) 0 0.05 0.10 0.15 0.20 0.25 0.30 Mahalanobis distance (D2 II) BIA C Lisrslsso loss los lo nn 0 01 02 0.3 04 05 0.6 0.7 0.8 Mahalanobis distance (D2 I) Fig. 2. UPGMA cluster analysis using Mahalanobis distances based on PC1 from morphological data of (a) group I (D’I), (b) group II (D’II) and (c) group III (D’III) among C. glareolus populations in spring (solid lines) and autumn (broken lines). Site abbreviations as in Tab. 1. 294 ANETTA BORKOWSKA Discussion There are three general areas where the present data are noteworthy. Firstly, the bank vole populations differed from each other over short geographical distances both geneti- cally and morphologically. Secondly, the genetic and morphological divergence among populations varied between seasons. Finally, there was no correlation between genetic and morphological differentiation of the populations in both seasons. The results of the analyses on allozyme variation in C. glareolus indicated that genetic polymorphism is not relatively high in this species. Percent of polymorphic loci in the populations studied approximated the values obtained for other populations of the bank vole in Poland (FEpyKk and GEBCZYNSKI 1980; GEBCZYNSKI et al. 1993; GEBCZYNSKI and RATKIEWwIcZ 1998). However, the observed heterozygosity (Ho) slightly exceeded Ho va- lues obtained both in Poland (Ho = 0.073; GEBCZYNsKI and RATKIEwIcZ 1998) and in Aus- tria (Ho = 0.075; LEITNER and HARTL 1988). Morphological analysis supported that the bank vole populations exhibited body size, organ, and cranıum size variation within its distribution, even over short geographical dis- tances. It is known that C. glareolus from various geographical regions differs consider- ably in intestinal morphology and body size (HAnsson 1985). Comparison of Western and Eastern Polish bank voles revealed the presence of several skull characters differing among the populations (HAITLINGER 1965). Likewise, mountain populations of C. glareolus differed from lowland populations in respect to dimensions and proportions of the body and skull (HAıTLinGEr 1970). According to Hansson (1985), all geographical differences in C. glareolus are related to ecogeographical rules, to possible demographic patterns, and to various adaptations following different modes of feeding (1. e., more gran- ivores or more foliovores anımals). However, morphological differences observed among the bank vole populations over short geographical distances seem to be due to adapta- tions to various local habitats. Genetic structure of the population changes between seasons in four out of five bank vole populations. Differences in allele frequency between spring and autumn generations of C. glareolus have been already noted (FEpyK and GEBCZYNSKI 1980; GEBCZYNSKI and RATKIEwIczZ 1998). It is interesting that seasonal intrapopulation variability strongly af- fected divergence among vole populations. Thus, the differentiation among autumn popu- lations was greater than among those in spring. Furthermore, correlation of two genetic distance matrices from spring and autumn revealed that genetic differentiation among the bank vole populations was not equivalent in various seasons. Smaller differences among vole populations in spring than in autumn suggested elimination of rare heterozygotes from the populations. Winter mortality can reach 77% of autumn numbers in €. glareolus (Pucer et al. 1993). However, it did not cause between-season changes in Ho value, as GEBCZYNSKI and RATKIEWICZ (1998) noted, but decreased the Fgr value, the measure of genetic differences among populations. Next, throughout breeding season the intrapopu- lation genetic variability increased. It seems that two processes, dispersion and mating system, can be responsible for increasing the intrapopulation variability, and consequently increasing Fsr value among populations in autumn. The dispersal rates of C. glareolus in- dividuals vary significantly with seasons, being the highest in early summer and in au- tumn (Guiwicz 1988). Additionally, C. glareolus was characterized by a promiscuous mat- ing system, and multiple paternity was common in natural populations (RATKIEwICZ and BORKOWSKA 1999). Hence, there was a deficit of homozygotes at all loci, showing signifi- cant departures from Hardy-Weinberg equilibrium. Seasonal variation in morphological parameters of the bank vole occurred in each population studied. The dynamic aspect of body mass and gut size of small herbivores was widely noted and explained as a physiological response to fluctuating environmental conditions (HAmMonD 1993; BORKOWSKA 1995; CAMPBELL and MACARTHUR 1996). Con- Genetic and morphological variation in Clethrionomys glareolus 2.95 trary to the genetic data, the differentiation in body size among bank vole populations was similar in spring and autumn. Therefore, it seems that body size could be an indicator of morphological divergence among populations of the bank vole. However, variation in size of internal organs strongly depended on a state of sexual activity (HAMMoNnD and DıamonD 1994) or food availability (CAMPBELL and MACARTHUR 1996). This is why the differences in group II in morphological parameters appeared among populations only in spring. The divergence in cranium dimension demonstrated seasonal variation also. Thus, cranıum size and shape divided the populations in spring. However, individual variation related to age in craniometric parameters (HAITLINGER 1965) is high in autumn popula- tions, which consisted of relatively young animals. Thus, the ‘shape’ components (PC2 and PC3) only separated vole populations in this season. Genetic and morphological differentiation patterns were discordant in C. glareolus. The analyses revealed that in both seasons allozymic variation did not correspond to mor- phological variation specified either by body size, or by size of internal organs, or by cra- nium dimension. Among the studies that have combined electrophoretic and adequate morphometric data to examine intraspecific population differentiation patterns, discor- dant genetic and morphological differentiation was found in amphibians (KyRIAKOPOU- LOU-SKLAVOUNOU et al. 1991) and birds (BAER 1992). The authors suggested that genetic patterns were haphazard, while the morphological differences were due to either climatic adaptation or random divergence through founder effects. On the other hand, changes in environmental conditions, either temporarily or permanently, strongly influence both ge- netic and morphological variance among populations of small mammals. Thus, it seems that a temporal variation in the genetic and morphological constitution of a population has the potential for revealing the agents responsible for microevolutionary change. Acknowledgements I would like to thank Prof. M. GEBcZYNskK1 for his attention to this work and critical reading of the manuscript. I am also grateful to Dr. M. A. RATKIEWwIcZ for his advice and help with electrophoresis. Zusammenfassung Genetische und morphologische Differenzierung zwischen Rötelmauspopulationen (Clethrionomys glareolus) aus dem Nordosten Polens: Saisonale Aspekte Bei 5 Rötelmauspopulationen (Clethrionomys glareolus) aus dem Nordosten Polens wurde die gene- tische und morphologische Variabilität untersucht. Zur Abschätzung der genetischen Variabilität wur- den 37 Enzymloci mittels Proteinelektrophorese untersucht. Der Polymorphiegrad (P) reichte von 0,162 bis 0,270 und der durchschnittliche beobachtete Heterozygotiegrad (Ho) von 0,047 bis 0,095. Be- rechnungen zur F-Statistik ergaben das Auftreten von saisonalen Unterschieden in den Allelfrequen- zen von vier Populationen (Fsr = 0,011-0,018, p< 0,05). Die Differenzierung zwischen den Rötel- mauspopulationen war im Herbst (Fsr = 0,032, p<0,0001) größer als im Frühling (Fsr = 0,023, p < 0,0001). Die morphometrische Variation in drei Gruppen von Parametern (I -— Körpergröße, II — Größe innerer Organe, III — Schädelmaße) wurde mittels Hauptkomponentenanalyse (PCA) unter- sucht. In allen drei Parametergruppen zeigten sich klare saisonale Unterschiede. Die PCA ergab Un- terschiede zwischen den Rötelmauspopulationen hinsichtlich der Körpergröße, der Größe innerer Or- gane sowie der Schädelgröße und -form für den Frühling, während sie sich im Herbst nur in bezug auf die Körpergröße und die Schädelform unterschieden. Genetische Distanzen nach Rogers und Mahala- nobis-Distanzen, berechnet für 3 Gruppen morphologischer Merkmale in beiden Jahreszeiten, waren nicht miteinander korreliert (r = |0.01-0.043|, NS; Mantel-Test). Nach unseren Daten zeigen bei der Rötelmaus die beiden untersuchten Merkmalssysteme keinen vergleichbaren Grad an Differenzie- rung zwischen Populationen. 296 ANETTA BORKOWSKA References BAKER, A. J. (1992): Genetic and morphometriec divergence in ancestral European and descendent New Zealand populations of chaffinches (Fringilla coelebs). Evolution 46, 1784-1800. BORKOWSKA, A. (1995): Seasonal changes in gut morphology of the striped field mouse (Apodemus agrarius). Can. J. Zool. 73, 1095-1099. CAMPBELL, K. L.; MACARTHUR, R. A. (1996): Seasonal changes in gut mass, forage digestibility, and nu- trient selection of wild muskrats (Ondatra zibethicus). Physiol. Zool. 69, 1215-1231. DIAMOND, J. M.; KARASov, W. H. (1984): Effect of dietary carbohydrate on monosaccharide uptake by mouse small intestine in vitro. J. Physiol. 349, 419-440. FEDOROV, V. B.; FREDRIKSSON, R.; FREDGA,K. (1995): Genetic differentiation among populations of Myopus schisticolor (the wood lemming)-isozyme variation. Heredity 74, 267-273. FEDYK, A.; GEBCZYNSKI, M. (1980): Genetic changes in seasonal generations of the bank vole. Acta Theriol. 25, 475-485. GALLARDO, M. H.; ARANEDA, C.; KÖHLER, N. (1992): Genic divergence in Spalacopus cyanus (Rodentia, Octodontide). Z. Säugetierkunde 57, 231-237. GEBCZYNSKI, M.; RATKIEWICZ, M. (1998): Does biotope diversity promote an increase of genetic varia- tion in the bank vole population? Acta Theriol. 43, 163-173. GEBCZYNSKI, M.; RATKIEWICZ, M.; DRYDEN, G. L. (1993): Electrophoretic variation among five Polish populations of the bank vole. Biochem. Syst. Ecol. 21, 825-831. GLiwicz, J. (1988): Seasonal dispersal in non-cyclic populations of Clethrionomys glareolus and Apode- mus flavicollis. Acta Theriol. 33, 263-272. HAITLINGER, R. (1965): Morphological analysis of the Wroclaw population of Clethrionomys glareolus (Schreber, 1780). Acta Theriol. 10, 243-272. HAITLINGER, R. (1970): Morphological characters of mountain populations of Clethrionomys glareolus (Schreber, 1780) and Apodemus agrarius (Pallas, 1773). Acta Theriol. 15, 269-282. HAMMOND, K. A. (1993): Seasonal changes in gut size of the wild prairie vole (Microtus ochrogaster). Can. J. Zool. 71, 820-827. HAMMOND, K. A.; DIAMOND, J. M. (1994): Limits to dietary nutrient intake and intestinal nutrient up- take in lactating mice. Physiol. Zool. 67, 282-303. Hansson, L. (1985): Geographic differences in bank vole Clethrionomys glareolus in relation to ecogeo- graphical rules and possible demographic and nutritive strategies. Ann. Zool. Fennici 22, 319-328. HARRIS, H.; Hopkinson, D. A. (1976): Handbook of Enzyme Electrophoresis in Human Genetics. Ams- terdam: North-Holland. HARTL, G. B.; SUCHENTRUNK, F.; NADLINGER, K.; WiILLInG, R. (1993): An integrative analysis of genetic differentiation in the brown hare Lepus europaeus based on morphology, allozymes, and mitochon- drial DNA. Acta Theriol. 38, Suppl. 2, 33-57. KITCHENER, D. J.; SCHMITT, L. H.; MAHARADATUNKAMSI (1994): Morphological and genetic variation in Suncus murinus (Soricidae: Crocidurinae) from Java, Lesser Sunda islands, Muluku and Sulawesi, Indonesia. Mammalia 58, 433-451. KYRIAKOPOULOU-SKLAVOUNOU, P.; SJÖGREN, P.; JANSSoN, H. (1991): Discordant genetic and morphologi- cal differentiation between three Rana ridibunda populations from northern Greece. Hereditas 155, 51-61. LAZARIDOU-DIMITRIADOU, M.; KARAKOUSIS, Y.; STAIKOU, A. (1994): Geographical variation in shell mor- phology and isoenzymes of Helix aspersa Müller, 1774 (Gastropoda, Pulmonata), the edible land snail, from Greece and Cyprus. Heredity 72, 23-35. LEITNER, M.; HARTL, G. B. (1988): Genetic variation in the bank vole Clethrionomys glareolus: biochem- ical differentiation among populations over short geographic distances. Acta Theriol. 33, 231-245. Loß0o, J. A. (1995): Morphometric, isozymic and mitochondrial variability of Africanized honeybees in Costa Rica. Heredity 75, 133-141. MANTEL,N. (1967): The detection of disease clustering and generalized regression approach. Cancer Res. 27, 209-220. MILLER, M. P. (1997): TFPGA. Tools for population genetic analysis. A windows program for the analysis of allozyme and molecular population genetic data. Version 1.3. Northern Arizona Univ., Flagstaff. MYRCHA, A. (1964): Variations in the length and weight of the alimentary tract of Clethrionomys glareo- lus (Schreber, 1780). Acta Theriol. 9, 139-148. Nagy, T. R.; NEGus, N. C. (1993): Energy acquisition and allocation in male collared lemmings (Dicrosto- nyx groenlandicus): effects of photoperiod, temperature, and diet quality. Physiol. Zool. 66, 537-560. Genetic and morphological variation in Clethrionomys glareolus 2 Neı, M. (1977): F-statistics and analysis of gene diversity in subdivided populations. Ann. Hum Genet. (London) 41, 225-233. NIELSEN, J. T. (1977): Polymorphism for the number of genes coding for salivary amylase in the bank vole, Clethrionomys glareolus. Genetics 85, 155-169. Pucek, Z. (1981): Keys to Vertebrates of Poland Mammals. Warszaw: Polish Scientific Publ. Pucek, Z.; ZEIDA, Z. (1968): Technique of determining age in the red backed vole, Clethrionomys glar- eolus (Schreber, 1780). Small Mammal. Newsl. 2, 51-60. PUCEK, Z.; JEDRZEJEWSKI, W.; JEDRZEJEWSKA, B.; PUCEK, M. (1993): Rodent population dynamics in a primeval deciduous forest (Bialowieza National Park) in relation to weather, seed crop, and preda- tion. Acta Theriol. 38, 199-232. Quavı, H.; Kır, S. (1980): Electrophoretic patterns of aminoacylase-1 (Acy-l1) activity. Biochemical Ge- netics 18, 669-679. RACZYNSKI, J. (1983): Taxonomic position, geographic range, and ecology of distribution. In: Ecology of the Bank Vole. Ed. by K. PETRUSEwICZ. Acta Theriol. 28, Suppl. 1, 3-10. RATKIEWICZ, M.; BORKOWSKA, A. (1999): Multiple paternity in natural populations of the bank vole (Clethrionomys glareolus). Z. Säugetierkunde (in press) SARA, M.; CASAMENTO, G. (1995): Morphometrics of the wood mouse (Apodemus sylvaticus, Mammalia, Rodentia) in the Mediterranean. Boll. Zool. 62, 313-320. SEARLE, J. B. (1985): Isoenzyme variation in the common shrew (Sorex araneus) in Britain, in relation to karyotype. Heredity 55, 175-180. SELANDER, R.K.; SMITH, M. H.; YAndg, S. Y.; JOHNSON, W. E.; GENTRY, J. B. (1971): Biochemical poly- morphism and systematics in the genus Peromyscus. I. Variation in the old-field mouse (Peromyscus polionotus). Studies in Genetics VI. Univ. Texas Publ. 7103, 49-90. SLATKIN, M. (1987): Gene flow and the geographic structure of natural populations. Science 236, 787- PA STATISTICA StatSoft. Inc. (1995): STATISTICA user’s guide, version 5, Tulsa: StatSoft Inc., Polish Ed. by StatSoft Polska, Kraköw. SWOFFORD, D. L.; SELANDER, R. B. (1989): BIOSYS-1: a computer program for the analysis of allelic var- iation in population genetics and systematics. Release 1.7. Illinois: Illinois Natural History Survey. VIro, P.; NIETHAMMER, J. (1982): Clethrionomys glareolus (Schreber, 1780) — Rötelmaus. In: Handbuch der Säugetiere Europas. Vol. 2/I. Ed. by J. NIETHAMMER und F. KrAapp. Wiesbaden: Akad. Ver- lagsges. Pp. 109-146. Author’s address: AnNETTA BORKOWSKA, Institute of Biology, University of Bialystok, ul. Swierko- wa 20 B, 15-950 Bialystok, Poland. e-mail: abork@noc.uwb.edu.pl Z8 DIE _ -- Z. Säugetierkunde 64 (1999) 298-307 ZEITSCHRIFT Sr FUR © 1999 Urban & Fischer Verlag SAUGETIERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Reevaluation of the taxonomic status of North African gerbils usually referred to as Gerbillus pyramidum (Gerbillinae, Rodentia): Chromosomal and biometrical data By L. GRANJON, AMELIE BONNET, W. HAMDINE, and V. VOLOBOUEV Museum National d’Histoire Naturelle, Paris, France, Universite Mouloud Mammer, Tizi-Ouzou, Algeria and Institut Curie, Paris, France Receipt of Ms. 14. 12. 1998 Acceptance of Ms. 05. 05. 1999 Abstract The chromosomal and biometrical attributes of large-sized, hairy-footed gerbils from North Africa usually referred to as Gerbillus pyramidum were studied. High-resolution banding techniques as well as external and skull biometry were used to compare specimens from Mauritania and Algeria. All specimens studied were characterized by the same karyotype, comprising 40 chromosomes and 74 autosomal arms. Gerbils from Algeria were found to be larger than those from Mauritania for most of the skull measurements, as well as for some external measurements. Comparisons with pub- lished data from other North African countries (Senegal, Morocco, Tunisia) suggest that all the pre- viously figured 2n = 40 karyotypes do represent the same species, chromosomally significantly distinct from 2n = 38 chromosome individuals found in Egypt and Sudan that correspond to true Gerbillus pyramidum. Based on our results and awaiting contrary evidence, we propose that the 2n =40 chro- mosome specimens found from Senegal to Libya correspond to a unique species, for which the name Gerbillus tarabuli should be applied. This species of wide distribution in northern Africa shows an ap- parently important biometrical variability, to be related with eco-climatological varıations of the en- vironment in which these populations live. Key words: Gerbillus, North Africa, chromosomes, biometry, systematics Introduction Rodents of the genus Gerbillus constitute a significant part of the arid and semiarid com- munities of mammals, from North Africa to India through the Arabian Peninsula and the Middle East. From a taxonomic point of view, their diversity is established, but, as stated by Musser and CARLEToN (1993), ... “[t]his genus has never been adequately revised”. Asa result, the number of species recognized has varied considerably according to var- ious authors (see review in LAy 1983), until Lay (1983) produced a list of 62 tentative spe- cies which was nearly entirely adopted by Musser and CARLEToON (1993). Clearly identi- fied in these lists, a number of taxonomic questions remained. Part of this problem undoubtly lies in the great number of ancient, often superficial, descriptions of new taxa that were only based on crude comparisons of colour and other morphological (external and a few skull) characteristics. The use of new morphological characters (see LAY 1983) and, more importantly, the development of cytogenetical investigations (starting from MAaTTHEY 1952) have improved our knowledge of the systematics in this group, without providing, however, significant clarification in its taxonomic arrangement to date. 0044-3468/99/64/05 — 298 $ 12.00/0 Taxonomy of North African gerbils 299 Among the taxonomical problems identified in Gerbillus, the status of large-sized, hairy-footed gerbils from North Africa, the Sinai, and Israel that have been referred to as Gerbillus pyramidum Geoffroy, 1825 is still a matter of debate. Often considered to include populations of individuals characterized by diploid numbers of chromosomes (2n) ranging from 38 (Wassır et al. 1969) to 66 (WAHRMAN and ZaHavI 1955), the name G. pyramidum was restricted by LAy (1983, following Lay et al. 1975) to 2n = 38 specimens from Egypt and Sudan. According to the latter author, the populations from the Sinai and coastal areas of Israel characterized by high diploid numbers should be referred to a species yet to be identified, but probably different from G. pyramidum. On the other hand, populations from North Africa west of Egypt and Sudan have been studied by various authors, under different species names. When performed, standard chromosomal analyses regularly yielded 2n = 40 chromosomes for individuals from these populations: Tunisia (JORDAN et al. 1974; CHiBanI and LAMINE-CHENITI 1982), Morocco (Lay et al. 1975), Algeria (MATTHEY 1952), Mauritania (Kreın et al. 1975), Senegal (HUBERT and BÖHME 1978; GRANJoN et al. 1992). In this region, two species have been described, which are proposed by Lay (1983) as potentially valid, and possibly characterized by this diploid number of 40 chromosomes. These are G. tarabuli described by THomas (1902) from Libya (initially as a subspecies of G. pyramidum), and G. riggenbachi, described by the same author (THomAs 1903) from Western Sahara (and then said to be... “[a] representative of G. pyramidum”). In this study, we investigate gerbils from Mauritania and Algeria using high resolution chromosome banding techniques, and also bring together all the available biometrical in- formation (including original one) on these large-sized gerbils from North Africa, with the aim of discussing the systematic implications of these data, and making some sugges- tions about the nomenclature in this group. Material and methods The skulls and skins of the specimens studied are deposited in the collection of the Laboratoire de Zoologie, Mammiferes et Oiseaux, at the Museum National d’Histoire Naturelle. The tissue explants and a portion of the cells of the karyotyped specimens are routinely kept in liquid nitrogen in the cell and tissue collection of the same laboratory. Chromosomal analyses Six specimens (3 females and 3 males) of large-sized, hairy-footed gerbils from coastal areas of Mauri- tania, as well as two specimens (one female and one male) from Beni Abbes (Algeria) have been kar- yotyped. Chromosome analysis was performed on preparations obtained from fibroblast cultures es- tablished after tail biopsy. Mitotic chromosomes were studied by G-banding (GTG; SEABRIGHT 1971), R-banding (RBG; VIEGAS-PEQUIGNOT and DUTRILLAUX 1978) and C-banding (CBG; SUMNER 1972; see ISCN 1995). For each specimen, at least 20 metaphases were analysed. The origin, sex, and number of the specimens studied are as follows: Mauritania: from Ivik, Banc d’Arguin (male n? 88-006); near Nouamghar (female n° 95-002; male n? 95-007); surroundings of Nouakchott (male n° 95-035; female n° 95-081); Tamzakt (male n° 95-082). Algeria: from Beni-Abbes (male n? 97-013; female n? 97-044). Morphometric analyses Preliminary analyses have revealed the presence of at least 4 species of Gerbillus ın coastal areas of Mau- ritania (GRANJoN et al. 1997). Among them, a sample of 27 adult specimens of large-sized, hairy-footed gerbils referable to the species with 2n = 40 chromosomes has been isolated, on the basis of the confron- tation of cytogenetical, morphological, and biometrical results (unpubl. data). The sample from Algeria is composed of 15 adult specimens from Beni Abbes and El Golea. On all these specimens, classical exter- nal measurements (weight, head and body, tail, hind foot, and ear length) have been taken, as well as Sskull measurements: GLS (greatest length of skull), ZYW (greatest zygomatic width), IOB (least 300 L. GRANION et al. breadth of interorbital constriction), DIA (length of diastema), LPF (length of palatal foramina), UTR (crown length of upper tooth row), BBC (breadth of braincase) and BUL (greatest length of bullae). These measurements have been taken as described in CHıMIMBA and DiPPpENAAR (1995). They were se- lected for comparative purpose with other studies, being the more regularly presented measurements on Gerbillus skulls. The origin, sex, and number of the specimens studied are as follows: Mauritania: from the surroundings of Nouakchott, Trarza (females n° 1994.1273 and 1997.1495, male n? 1994.1272); from Nbeika, Tagant (male n® 1997.1479); from Agneitir, Inchiri (female n° 1997.1481); from Sei-Rakna, Trarza (females n° 1997.1484 and BLM386, male n° BLM388); from Amatlich EI Gleitat, Adrar (male n° 1997.1480); from Tiguent, Trarza (female n° 1997.1486, males n° 1997.1485, and BLM39]); from Tam- zakt, Trarza (female n° 1997.1494, males n? 1997.1482, 1997.1483, 1997.1491, BLM448, and BLM455); from Chott Boul, Trarza (females n° 1997.1490, 1997.1496, 1997.1497, BLM441, and BLM459A, male n° 1997.1489); from Akchar, Inchiri (female n° 1997.1474, male n° BLM423); from Hassı Tifouggag, Bra- kna (male n° 1997.1493). Algeria: from El Golea (females n° 1997.503, 1997.504, 1997.505, and 1997.506, males n° 1997.507 and 1997.508); from the surroundings of Beni-Abbes (males n° 1997.509, 1997.510, 1997.511, 1997.513, 1997.514, 1997.515, and 1997.516, females n? 1997.512 and 1997.517). Bi Zeuıe w* u a en 2 mem “ ee m u N nen be } >. 0a = mei a wer ER ze jeu=d an juc a je = Fig. 1. R-banded (RBG) karyotype of a male G. tarabuli. Note that in addition to the late replicating long arm of the Y chromosome, the long arm of pair 14, an intercalar segment of the long arm of pair 1 and distal segments of the pairs 17 and 18 are also late replicating. Taxonomy of North African gerbils 301 Results Chromosomal analysis After R-banding, the karyotypes of the six specimens from Mauritania and the two speci- mens from Algeria were found to be identical. They consist of 40 chromosomes compris- ing 13 pairs of metacentric, 5 pairs of submeta- to subtelocentric and one pair of acro- centric chromosomes, resulting in a number of autosomal arms (NFa) of 74. Both sex chromosomes are large submetacentrics, the X chromosome being the largest of the set and the Y chromosome being equal in size to the second pair of autosomes. Their short arms are identical and result from an autosome-gonosome translocation (Fig. 1). In spite of the important quantity of C heterochromatin revealed in all centromeric and some in- tercalar regions (Fig. 2), no well-defined heteromorphism for C-band positive heterochro- matin was observed. . © “_ . & 6 & & 1. # 5 L be “a 1 2 3 4 S MM; w: ww en w ° 3 R .« R . a ; 3 . 3 6 7 8 9 a u € 12 13 15 ri - u» E; 18 19 Fig. 2. C-banded (CBG) karyotype of male G. tarabuli. Note that the late replicating segments of the Fig. 1 are C-band positive. 302 L. GrANJON et al. Morphometric data The ranges of variation of the measurements taken on the samples from Mauritania and Algeria are given in tables 1 (body measurements) and 2 (skull measurements), together with the values extracted from various other studies. No statistical test could be per- formed between our data and those from other authors, due to the fact that individual data (and/or standard deviation values) were rarely available in the latter. Between the series of Algeria and Mauritania that we measured, significant differences were found for all skull measurements, except diastema length (Mann-Whitney U tests, p < 0.05), the spe- cimens from Algeria being consistently larger. As far as body measurements are con- cerned, Algerian specimens were found to be larger for hindfoot (U test, p = 0.0003) and weight (U test, p = 0.014) values, whereas Mauritanian specimens were found to be larger for tail length (U test, p = 0.0001) and ear length (p = 0.0314) values. For Mauritania, our data mostly fall in the range of the values given by Kıeın et al. (1975) for a sample of Gerbillus sp. “agag group” characterized by a 40 chromosome kar- yotype. However, our sample appears characterized by a somewhat longer tail and, possi- bly, a longer hindfoot (but Krein et al. [1975] did not indicate precisely whether they in- cluded the claw in their measurement). Concerning the Algerian specimens, our results also correspond to the range of variation indicated by KowAuskı and RZEBIK-KOWALSKA (1991), the latter being so large that one may suspect that young individuals were in- cluded in the sample considered by these authors. For all these measurements, the sample of G. pyramidum from Egypt (OsBorn and Heımy 1980) reaches the highest values, the specimens of G. p. pyramidum from the Nile Valley being the largest of all. Table 1. Body measurements (in mm) of samples of large-sized, hairy-footed gerbils from Northern and Saharo-Sahelian Africa, often referred to as G. pyramidum. TL = Total length; HB = Head and body length; Hf = Hindfoot length; E = Ear length. The sample from Egypt includes G. p. pyramidum, G. p. gedeedus and G. p. elbaensis. Reference G. p. tarabuli type 149 30 (s. u.) 15 THoMmAas (1902) G. riggenbachi type 132 30 (s. u.) 13 Thomas (1903) G. pyramidum Senegal 156 34 14 HUBERT and BÖHME (al) (1978) Gerbillus sp. (“agag” 123-148 26-29 12-16 Kein etal. (1975) group) Mauritania (n = 46) Gerbillus sp. Maurita- 134-161 13-16 This study nia(n= 27) G. pyramidum Algeria 200-274 110-172 13-17.5 Kowauskı and RZE- (n = 83-87) BIK-KOWALSKA (1991) Gerbillus sp. Algeria 130-150 12-15 This study (n=14) G. pyramidum Tunisia 222-267 124-149 27-35 13-17 JorDan etal. (1974) (n= 30) G. p. tarabuli Lybie 246-289 132-165 30-35 14-17 Ranck (1968) (n='31) G. p. tibesti Tchad 258-300 146-176 32-37 15-18 SETZER and RANCK (m=15) (ER) G. pyramidum, Egypte 102-135 128-180 30.5-39 14-20 OÖsBorn and HELMY (n = 60-70) (1980) G. pyramidum, Sudan 97-121 125-149 28.3-30 11.9-16.1 TAwıLL and NIETHAM- (m 5) MER (1989) 303 Taxonomy of North African gerbils (=u) 0'9T-0'ST III EREESIIE Five zae uepnS "unpnuvıdd 'O (EL-09 = U) Sıech SUSI SEO EC E3i0T8i9T RI o7dAaq on 2 Sp-u ATS era FAN SEE peyaL nsoqu d’O (TE=U) ae! Eh bie, SSL. LI] 9’SE-8°TE eIgAT ı7ngvo uunpnunadd 5 (0€E = U) IArEOE Va VOL IST 760805 Asiye56z eistun,, unpnunıdd 9 (ZI-01 = U) 07-56 O'ST-LEI BOSSE DE ÜE erlosjy ds snp1q100 (18-£8 = U) CAT Ce eLIsd]y wunpnunidd 'Q (LZ=U) vrI-gel SEES NUT "NE eruejLnew ds sn7p1q409 (9 = U) rz£-9'87 eiueyuınew (dno1s „Sege,,) ‘ds snyp1g429 (Zu) spE je3auas uunpnuvıdd 9 (DIS) gE TE odA} 1yopquossı OD ST 128 adA} ıyngnAavı OD 'SO9U2I9JOI IOJ [ 9Jge) pue ‘uorreuejdx> ‘suomerAdIgge 10F IX9J 99S "uınpnunıdd OD se 0) P9119J91 USJJO ‘“eILIFW uelfsyeS-oleyes pur UIOYLION WOAJ S[IQIS3 P2I00F-Altey ‘pazis-odıe]j Jo sojdures Jo (ww ur) sJuswoınsesw [nyS "7 'OIgeL 304 L. GRANION et al. Discussion A karyotype with 2n=40, NFa = 74, comprising 18 pairs of submetacentric to meta- centric and one small pair of acrocentric autosomes has already been presented for large- sized, hairy-footed gerbils from Senegal (HuUBERT and BÖHME 1978; GRANJON et al. 1992), Morocco (Lay et al. 1975), and Tunisia (JoRrDAN et al. 1974; CHIBANI and LAMINE-CHENITI 1982). Earlier drawings of the chromosomes of Algerian specimens by MATTHEY (1952) are more difficult to interpret, but most probably represent the same pattern. The X chro- mosome appears as a relatively large submetacentric and the Y chromosome as a middle sized submetacentric in Lay et al. (1975), whereas both sex chromosomes appear as meta- centrics slightly different in size in JORDAN et al. (1974). In all instances, two relatively large pairs of submetacentric chromosomes characterized by very small short arms are clearly visible, but the number of other pairs identified as submetacentrics varies from 5 (JoRDAN et al. 1974) to 9 (Lay et al. 1975). However, the absence of chromosome banding in these studies makes it difficult and somewhat arbitrary to identify chromosome mor- phology unambiguously (metacentric vs submetacentric, or submetacentric vs subtelo- centric). Slightly different arrangements and interpretations of these karyotypes can be done, resulting in figures very similar to the ones we obtained in the specimens from Mauritania and Algeria that we studied. Awaiting further data, these similarities suggest a more or less complete homology between the 2n =40 karyotypes reported from all these countries, a fact that was proven here between the specimens from Mauritania and Algeria by virtue of high resolution banding techniques. On the other hand, the karyotype of specimens from Egypt as shown in Wassır et al. (1969) and Lay et al. (1975) is characterized by 38, all meta- to submetacentric chromo- somes. Four pairs are considered to be submetacentric by Lay et al. (1975), among which not one appears to have particularly short arms. The same may be true in the specimens from Sudan studied by TAawıL and NIETHAMMER (1989), who described the chromosomes to be all metacentric or submetacentric, but no photograph of the karyotype was pro- vided in this study. Moreover, many other chromosomes are clearly distinct in size and morphology from those figured in Karyotypes with 2n = 40. This means that at least some of them are rearranged differently between the two karyotypes (2n=38 and 2n = 40). For instance, the absence of acrocentric pairs in 2n=38 chromosome gerbils is most probably the result of a tandem translocation, a chromosomal rearrangement known for its strongly negative heterotic effect in the heterozygous state (WRIGHT 1982). Finally, the comparison of G-, R- and C-banded sex chromosomes of G. pyramidum (see Fig.2 in WAHRMAN et al. 1983, where unfortunately chromosome banding data were not presented for autosomes) with those of the 2n = 40 individuals presented here, shows clear differ- ences. It appears finally that there is probably more than a simple difference of one pair of small acrocentric chromosomes (that could be explained by one single event) between the 2n=38 and 2n =40 karyotypes. This difference needs to be more accurately docu- mented, which will be possible once 2n = 38 specimens are studied by high resolution banding techniques. Nevertheless, the rearrangements suggested by the comparison made above constitute a strong argument for considering the specimens characterized by these two karyotypes as belonging to two distinct species. The eastern, 2n =38, one corre- sponds to G. pyramidum, the type specimen of which was collected in the surroundings of the Great Pyramids in Egypt, and which would range along the Nile Valley, in Egypt, and Sudan, and in the oases of the region as proposed by Musser and CARLEToN (1993). On the other hand, the apparent homogeneity of the 2n =40 chromosome karyotype from Northern Senegal in the West eastwards to Tunisia suggests that the corresponding speci- mens may belong to only one species. G. riggenbachi was described by THomAs (1903) from Rio de Oro, a coastal site on the Tropic of Cancer which is some 350-400 km distant Taxonomy of North African gerbils 305 Sr. oO G. riggenbachi type „Tunisia %* G. tarabuli type + Morocco + 2n=40 specimens Algeria m 2n=38 specimens Fig. 3. Map of northern Africa, showing localities where 2n = 40 and 2n = 38 chromosome specimens of Gerbillus have been recorded, and type-localities of G. riggenbachi and G. tarabuli. from the site where the specimens from Northern Mauritania chromosomally analysed here originated (Fig. 3). G. pyramidum tarabuli was described by the same author one year earlier, on the basis of specimens caught in various localities of central West and the coastal North of Libya (ThomAs 1902). A number of these localities, including Sebha, from which the type specimen was caught, are situated 550 to 600 km east from south- eastern Tunisia (Fort Saint), where JoRDAN et al. (1974) reported specimens with 40 chromosomes (Fig. 3). Although the whole region can be environmentally subdivided in a number of ways (see, for instance the climatic and phytogeographic subdivisions pre- sented by LE HouErou (1992), an arid to semi-arid nucleus has persisted at least through- out the Pleistocene and Holocene in the lowlands along the Tropic of Cancer, which ex- tension has varied according to climatic changes (LE Houerou 1992). The absence of a significant north-south barrier between the western coast and the Libyan desert is an- other argument to support the existence of one wide-ranging species of large-sized, hairy- footed gerbil characterized by the 2n = 40, NFa = 74 karyotype described here. Contrary to Lay (1983, followed by Musser and CARLETON 1993), we thus propose to abandon G. riggenbachi as a valid species and, for reasons of priority, to consider only G. tarabuli as being present in this region. Its distribution would range, more or less continuously, from northern Senegal in the west (DUPLANTIER et al. 1991) to the Cyrenaican Plateau of Libya (RAnck 1968) and the Tibesti Mountains of Chad (SETZER and RAnck 1971) in the east. The hiatus in G. pyramidum distribution mentioned by RAnck (1968) in eastern Li- bya and western Egypt, corresponding to the Cyrenaican Plateau and the northern part of the Libyan desert may well represent the barrier that has been at the origin of the dif- ferentiation between G. pyramidum and G. tarabuli. The morphological and biometrical data appear of relatively little value for a priori characterization of these species. Specimens of G. pyramidum from Egypt can reach a lar- ger size than that of specimens from the other origins but the biometrical characteristics of the series from Sudan fall in the range of the values obtained for other samples. Con- versely, significant differences can be evidenced between samples of G. tarabuli from var- ious origins (as here between the samples from Mauritania and Algeria). This apparently important morphometrical variability within both G. pyramidum and G. tarabuli is prob- ably related to eco-climatological variations of the environment in which these popula- tions live (see PETTER 1961), and would deserve further analyses. The proposal of Lay et al. (1975) to consider G. riggenbachi as a distinct species, following comparison of mu- seum specimens, has never been substantiated, and would need to be critically examined on the basis of large-scale analyses, taking into account this environmentally and geo- 306 L. GRANJON etal. graphically determined variability. Meanwhile, it seems important to gather as much chro- mosomal information as possible on these gerbils from all over North Africa, in an at- tempt to map more precisely their distribution. The establishment of diagnostic morpho- logical characteristics could then be envisaged, to determine whether splitting of what we propose to call G. tarabuli ıs justified. Acknowledgements We are grateful t0o M. LoMBARD for technical assistance. We also thank F. CoLas and A.T. Dia, heads of the project “Biodiversite du littoral mauritanien”, financed by the European Commission, during which the specimens from Mauritania were caught, with the help of J.-F. Cosson, J. Cuisin, and M. TRANIER. Thanks go also to J.-F. Voısın for the German version of the abstract. This is contribution no. 99-2 from EA 2586. Zusammenfassung Neubewertung der taxonomischen Stellung nordafrikanischer Rennmäuse, die gewöhnlich Gerbillus pyramidum (Gerbillinae, Rodentia) zugeschrieben werden: Chromosomale und biometrische Daten Es wurden die chromosomalen und biometrischen Characteristika großer, nordafrikanischer Renn- mäuse untersucht, die gewöhnlich dem Taxon Gerbillus pyramidum zugeschrieben werden. Hoch- auflösende Bänderungstechniken zur Chromosomenanalyse wie auch verschiedene Körper- und Schädelmaße wurden zum Vergleich von Tieren aus Mauretanien und Algerien herangezogen. Alle untersuchten Rennmäuse zeigten den gleichen, durch 40 Chromosomen und 74 autosomale Arme ge- kennzeichneten Karyotyp. Rennmäuse aus Algerien waren generell größer als jene aus Mauritanien. Vergleiche mit den publizierten Angaben über Rennmäuse aus anderen nordafrikanischen Ländern (Senegal, Marokko, Tunesien) legen nahe, daß alle zuvor beschriebenen 2n = 40 Karyotypen dieselbe Spezies repräsentieren, welche sich chromosomal deutlich von den Individuen mit 2n = 38 abhebt, die in Ägypten und im Sudan zu finden sind und dem echten Gerbillus pyramidum entsprechen. Nach un- seren Ergebnissen schlagen wir vor, daß die von Senegal bis Lybien gefundenen Exemplare mit 2n=40 einer einzigen Art entsprechen, welche den Namen Gerbillus tarabuli tragen sollte. Diese, im nördlichen Afrika weitverbreitete Art zeigt ein anscheinend breites Spektrum an biometrischer Varia- bilität, die mit der Variabilität des Lebensraums zu tun haben Könnte. References CHIBANI, J.; LAMINE-CHENITI, T. (1982): Etude caryologique des gerbilles (genre Gerbillus) de Tunisie. Rev. Fac. Sc. (Tunis) 2, 147-153. CHIMIMBA, C. T.;, DIPPENAAR, N. J. (1995): The selection of taxonomic characters for morphometric ana- lysis: a case study based on Southern African Aerhomys (Mammalia: Rodentia: Muridae). Ann. Carnegie Mus. 64, 197-217. DUPLANTIER, J.-M.; GRANJON, L.; BA, K. (1991): Decouverte de trois especes de rongeurs nouvelles pour le Senegal: un indicateur suppl&mentaire de la d&sertification dans le nord du pays. Mammalıa 55, 313-315. GRANJON, L.; Cosson, J.-F.; CUISIN, J.; TRANIER, M.; CoLas, F. (1997): Les Mammiferes du littoral mauri- tanıen: 2 - Biog&ographie et Ecologie. In : Environnement et littoral mauritanien. Ed. by F. CoLas. Montpellier, France: CIRAD (Collection Colloques). Pp. 77-89. GRANJON, L.; DUPLANTIER, J.-M, CATALAN, J.; BRITTON-DAVIDIAN, J. (1992): Karyotypic data on rodents from Senegal. Israel J. Zool. 38, 263-276. HUBERT, B.; BÖHME, W. (1978): Karyotype of Gerbillus pyramidum I. Geoffroy (Rodentia, Gerbillidae) from Senegal. Bull. Carnegie Mus. Nat. Hist. 6, 38-40. ISCN (1995): An international system for human cytogenetic nomenclature. Ed by F. MıTELMman. Basel: Karger. Taxonomy of North African gerbils 307 JORDAN, R. G.; Davis, B. L.; BAccAR, H. (1974): Karyotypic and morphometric studies of Tunisian Ger- billus. Mammalia 38, 667-680. KLEIN, J.-M.; POULET, A. R.; SIMONKOVICH, E. (1975): Observations Ecologiques dans une Zone enzoo- tique de peste en Mauritanie. 1. Les rongeurs, et en particulier Gerbillus gerbillus Olivier, 1801 (Rodentia, Gerbillinae). Cahiers de ’ORSTOM, Serie Entom. Medic. Parasitol. 13, 13-28. KowaLskI, K.; RZEBIK-KOWALSKA, B. (1991): Mammals of Algeria. Wroclaw: Ossolineum. Lay, D. M. (1983): Taxonomy of the genus Gerbillus (Rodentia: Gerbillinae) with comments on the ap- plications of generic and subgeneric names and an annotated list of species. Z. Säugetierkunde 48, 329-354. Lay, D. M.; AGERsoNn, K.; NADLER, C. F. (1975): Chromosomes of some species of Gerbillus (Mammalia, Rodentia). Z. Säugetierkunde 40, 141-150. LE Houerou, H.N. (1992): Outline of the biological history of the Sahara. J. Arid Environment 22, 3- 30. MATTHEYy, R. (1952): Chromosomes sexuels multiples chez un Rongeur (Gerbillus pyramidum Geof- froy). Arch. Klaus-Stiftung für Vereb.-forsch. 27, 163-166. MUSSER, G. G.; CARLETON, M. D. (1993): Family Muridae. In: Mammal Species of the World -— A Taxo- nomic and Geographic Reference. Ed. By D. E. Wırson and D. M. REEDER. Washington: Smithso- nian Inst. Press. Pp. 501-755. ÖSBORN, D. J.; HELMY, I. (1980): The contemporary land mammals of Egypt (including Sinai). Fieldiana Zoology, New Series N° 5. Chicago : Field Mus. Nat. Hist. PETTER, F. (1961): Repartition g&ographique et Ecologie des Rongeurs desertiques (du Sahara occiden- tal a l’Iran oriental). Mammalıia 25 (N° special), 1-222. RANncK, G.L. (1968): The rodents of Libya. Taxonomy, ecology and zoogeographical relationships. U.S. Nat. Mus. Bull. 275, 1-264. SEABRIGHT, M. (1971): A rapid banding technique for human chromosomes. Lancet 2, 971-972. SETZER, H. W.; RAncK, G.L. (1971): A new gerbil (genus Gerbillus) from the Chad. Proc. Biol. Soc. Washington 84, 55-58. SUMNER, A.T. (1972) : A simple technique for demonstrating centromeric heterochromatin. Exp. Cell Res. 75, 304-306. TAwıL, S. A.; NIETHAMMER, J. (1989): Über Gerbillus pyramidum (Rodentia, Gerbillidae) im Sudan. Z. Säugetierkunde 34, 57-59. THoMmas, ©. (1902): On the Mammals collected during the Whitaker Expedition to Tripoli. Proc. Zool. Soc. Lond. 2, 2-13. Thomas, ©. (1903): On a small collection of Mammals from the Rio de Oro, Western Sahara. Novit. Zool. 10, 300-302. VIEGAS-PEQUIGNOT, E.; DUTRILLAUX, B. (1978): Une methode simple pour obtenir des prophases et des prome£taphases. Ann. Genetique 21, 122-125. WAHRMAN, J.; ZAHAVI, A. (1955): Cytological contributions to the phylogeny and classification of the ro- dent genus Gerbillus. Nature 175, 600-602. WAHRMAN, J.; RICHLER, C.; NEUFELD, E.; FRIEDMANN, A. (1983): The origin of multiple sex chromo- somes in the gerbil Gerbillus gerbillus (Rodentia: Gerbillinae). Cytogenet. Cell Genet. 35, 161-180. WAsSIE, K.; LUTFY, R. G.; Wassır, S. (1969): Morphological, cytological and taxonomical studies of the rodents genera Gerbillus and Dipodillus from Egypt. Proc. Egypt. Acad. Sci. 31, 203-209. WRIGHT, S. (1982): Character change, speciation, and the higher taxa. Evolution 36, 427-443. Authors’ addresses: Dr. LAURENT GRANJON and Dr. VITALY VOLOBOUEN, Laboratoire de Zoologie, Mam- miferes et Oiseaux, Museum National d’Histoire Naturelle, 55 rue Buffon, F-75005 Paris, France; AMELIE BonNET, Laboratoire de Conservation des Especes Ani- males, Parc Zoologique de Paris, 53 avenue Saint-Maurice, F-75012 Paris, France; WATIK HAMDINE, Universitt Mouloud Mammeri de Tizi-Ouzou, Institut d’Agronomie, Laboratoire d’Ornithologie et d’Ecologie des Vetebres, 15000 Tizi- Ouzou, Alge£rie. SE SEUn Da Z. Säugetierkunde 64 (1999) 308-318 ZEITSCHRIFT SF FÜR © 1999 Urban & Fischer Verlag SAUGETIERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Spatial partitioning of allozyme variability in European mountain hares (Lepus timidus): gene pool divergence across a disjunct distributional range? By F. SUCHENTRUNK, KARIN POLSTER, M. GIACOMETTI, P. RATTI, C.-G. THULIN, C. RUHLE, A.G. VasırL Ev, and L. SLOTTA-BACHMAYR Research Institute of Wildlife Ecology, Vienna Veterinary University, Vienna, Austria, Institut für Tier- pathologie, University of Berne, Switzerland; Jagd- und Fischereiinspektorat Graubünden, Chur, Switzer- land; Department of Conservation Biology and Genetics, Uppsala University, Uppsala, Sweden; Jagd- und Fischereiverwaltung St. Gallen, St. Gallen, Switzerland, Institute of Plant and Animal Ecology, Yeka- terinburg, Russia; Nationalpark Hohe Tauern, Austria Receipt of Ms. 19. 04. 1999 Acceptance of Ms. 28. 07. 1999 Abstract To investigate if the postglacial dispersion of mountain hares (Lepus fimidus) into the present geographically separated ranges in Europe has produced marked gene pool differentiation, 209 indivi- duals from Scandinavia, Russia, the Alps, Scotland, and Ireland were screened for allozymic variabil- ity at 40 structural gene loci by horizontal starch gel electrophoresis. Polymorphisms were detected at 13 locı. Most alleles were identical with those of brown hares (Lepus europaeus) studied earlier in Europe. Average expected heterozygosity (2.0-5.0 %) and rates of polymorphism (8.8-29.4 %) in re- gions or subspecies were comparable to those of local samples of European brown hares studied ear- lier. Despite a high amount (31.3 %) of “private alleles”, genetic distances (NeEr’s 1978 D: 0.000-0.008 among subspecies, and 0.000-0.017 among regions) were similar to those found among local samples of central European brown hares. This indicates low genetic differentiation among gene pools of sub- species or regions. Also, relatively low mean Fsr values (0.157 for regions, 0.14 for subspecies) and low numbers of significantly differing allele frequencies indicated little genetic differentiation. WRIGHT’s (1978) hierarchical F-statistics revealed that less than 1% of the relative genetic variation was partitioned among subspecies but 13.6 % among regions within subspecies. All results conform to the hypothesis of a quite panmictic gene pool of late-glacial and postglacial mountain hares in Eu- rope. They also support the view that no severe drift has occurred in postglacial populations during the colonization of the present ranges. Key words: Lepus timidus, allozymes, colonization, genetic differentiation, disjunct distri- bution Introduction Mountain hares (Lepus timidus) have a disjunct distribution in Europe, with natural ranges in the subarctic/arctic regions of Russia and Fennoscandia, the Baltic region and Poland, the Alps, Scotland, and Ireland. Mountain hares from these regions are consid- ered separate subspecies (L.t. timidus, L.t. kozhevnikovi, L.t. sylvaticus, L.t. varronis, L.t. scoticus, L. t. hibernicus), mainly due to morphometric differences and pelage colora- tion (see ANGERBJÖRN and Frux 1995). According to late Pleistocene and early Holocene 0044-3468/99/64/05 - 308 $ 12.00/0 Spatial partitioning of allozyme variability in Zepus timidus 309 ecotopes, geography, and fossil records (e.g., Lang 1994; STUART 1982; Dörrpes 1997; see also CorBET 1986), however, this hare species was most probably continuously distributed over large parts of Europe between the northern and the Alpine ice sheets by the end of the last glaciation period (at 10.000-12.000 YBp). Apparently, they were hunted by Mag- dalenien Cro-Magnon people of central Europe (e.g., Döppes 1997). This study addresses the degree of cross gene pool differentiation among the currently spatially well separated subspecies of European mountain hares. Provided the late Pleis- tocene population of mountain hares did exhibit a panmictic gene pool across large parts of central and north-central Europe, rather than an already substructered one, and no se- vere or long lasting demographic bottlenecks (“founder effects” etc.) have occurred dur- ing the postglacial colonization of the present ranges, we should expect a low gene pool differentiation among the currently acknowleged subspecies in Europe. Alternatively, a possible structuring of the late glacial gene pool into regional populations and/or strong genetic drift during the post-glacial colonization period might have led to signifcant ge- netic differences among the subspecies in Europe. Material and methods A total of 209 mountain hares was collected at diverse localities in the Alps (Switzerland, France, Austria), Scandinavia, the Ural mountains (Russia), Scotland (U.K.), the Irish Republic, and North- ern Ireland (U.K.) between 1994 and 1996. Detailes of sampling localities and sample sizes are given in figure 1. These hares can be allocated to four nominal subspecies: L.t. fimidus (Scandinavia, Ural), L.t. varronis (Alps), L. t. scoticus (Scotland), and ZL.t. hibernicus (Ireland). The sample from Scotland (Mull) may also include the subspecies L.t. hibernicus, or L.t. scoticus x L.t. hibernicus hybrids be- cause of the introduction of this subspecies to Mull in the last century (CoRBET and SOUTHERN 1977; see also Frux 1970 for mountain hares from mainland Scotland). Among the presently studied moun- tain hares from Sweden (/. t. timidus) introgression of L.r. sylvaticus cannot be entirely excluded. The Swedish hares were shot in January/February but unfortunately the hunters did not make any special remarks as to blue/grey coat colour which is typical for L. f. sy/vaticus (BERGENGREEN 1969). Sexing of hares was carried out by inspection of their internal reproductive organs. Age (adult vs. juvenile/subadult) was estimated by body size, body weight, and by checking for the occurrence of the lateral epiphyseal protrusion of the ulna; the latter method separates juveniles/subadults (born in the last reproductive season) from older ones (WALHovD 1965). The following 25 isozymes/-systems encoded by 40 hypothetical structural gene loci were assayed for allozymic variation by horizontal starch gel electrophoresis (isozyme/-system, abbreviation, E.C. number, and corresponding structural gene loci in parentheses): alpha-glycerophosphate dehydrogen- ase (GDC, 1.1.1.8, Gdc), sorbitol dehydrogenase (SDH, 1.1.1.14, Sdh), lactate dehydrogenase (LDH, 1.1.1.27, Ldh -1, -2), malate dehydrogenase (MOR, 1.1.1.37, Mor -1, -2). malic enzyme (MOD, 1.1.1.40, Mod -1, -2), isocitrate dehydrogenase (IDH, 1.1.1.42, Idh -1, -2), 6-phospho-gluconate dehy- drogenase, (PGD, 1.1.1.44, Pgd), glutamate dehydrogenase (GLUD, 1.4.1.3, Glud), catalase (CAT, 1.11.1.6, Cat), superoxide dismutase (SOD, 1.15.1.1, Sod -1, -2), purine nucleoside phosphorylase (NP, 2.4.2.1, Np), aspartate amino-transferase (AAT, 2.6.1.1, Aat -1, -2), hexokinase (HK, 2.7.1.1, Hk -1, -2, -3), creatine kinase (CK, 2.7.3.2, Ck -1, -2), adenylate kinase (AK, 2.7.4.3, Ak -1, -2), phospho-gluco- mutase (PGM, 2.7.5.1, Pgm -2, -3), esterases (ES, 3.1.1.1, Es -1; ES-D, 4.2.1.1, Es-D), acıd phosphatase (ACB, 3.1.3.2, Acp -1), fructose-1,6-diphosphatase (FDP, 3.1.3.11, Fdp-1), £-galactosidase (#-GAL, 3.2.1.23, ß-Gal), peptidases (PEP, 3.4.11, Pep -1, -2), fumarate hydratase (FH, 4.2.1.2, Fh), aconitase (ACO, 4.2.1.3, Aco -1, -2), mannose phosphate isomerase (MPI, 5.3.1.8, Mpi), glucose phosphate iso- merase (GPI, 5.3.1.9. Gpi -1, -2). Tissue preparation, electrophoresis and protein-specific staining followed GRILLITSCH et al. (1992). Allelic variants were resolved by direct side-by-side comparison of migrating allozymes, including five brown hares (Lepus europaeus) on the same gels. For designation of alleles we used the nomenclature of GRILLITSCH et al. (1992). Genotypes at polymorphic loci were determined in each specimen accord- ing to the principles of enzyme electrophoresis (e. g., RICHARDSOoN et al. 1986; ROTHE 1994). In several individuals, however, genotypes could not be determined for the entire set of loci due to insufficient 310 F. SUCHENTRUNK etal. .t. timidus af a = & sr Fig. 1. Sampling locations of mountain hares (full circles) and associated subspecies names. Sample sizes in parentheses. Switzerland: canton Grisons, central and northern parts (49); Engadin (23); Val Mesolcina, Val Calanca, Val Bregaglia, Val Poschiavo (15); canton Glarus (16); Austria: Hohe Tauern (4); France: St. Veran (3); Abries (3); Chäteauroux (1); Aiguilles (1); Sweden: Jämtland (22); Växvik re- gion (6); Väster- and Norbotten (7); Uppland (7); Norway: Ringebu ( 19); South Norway (5); Telemark (5): Russia: Polevskoy, Ural (14); Scotland (U.K.): Mull (5); Northern Ireland (U.K.): Autrim (1); Ty- rone (1); Rep. Ireland: Mayo (1); Sligo (1). Open circles: Local populations of brown hares (Lepus euro- paeus) from Austria (cf., HARTL et al. 1993) used for comparison of genetic differentiation. quality of resolution producing ambiguous interpretations. All population genetic statistics regarding regional samples of L. timidus and the comparison of L. fimidus regional samples and the L[. euro- paeus local samples were based on 40 locı. For the comparison of L. timidus subspecies all analyses were based on 34 locı (omitting the Mor-2; Ck-1, -2; Pgi; Cat; Gdc loci) because of total lack of data for these loci in certain subspecies. Allele frequencies were calculated by using the BIOSYS-1 pc package 1.7 (SWoFFORD and SELAN- DER 1989). Allele frequencies of hares from Switzerland were tested for independence of age class (young of the year vs. older animals) or sex by Fisher’s exact tests (using SPSS). Association of geno- types between loci was also tested by Fisher’s exact tests for each pair of polymorphic loci with aggre- gated genotypes to check for linkage disequilibrium. Significance was based on sequential Bonferroni procedures (with a nominal alpha = 0.05) to account for multiple testing (Rıce 1989). Allele frequen- cies at single loci were tested for significant variation between pairs of regions and pairs of subspecies by Fisher’s exact tests of aggregated alleles in cases of more than two alleles per locus and sequential Bonferroni procedure. The BIOSYS-1 pc package, release 1.7 (SWOFFORD and SELANDER 1989) was also used to calculate the rate of polymorphism (P, 99 % criterion), the mean number of alleles per locus (A), and mean heterozygosities (H.-expected, H,-observed) for each regional, subspecies, and local sample. It was further employed to calculate WRIGHT’s (1978) non-hierarchical and hierarchical F-statistics. The latter was calculated to test for partitioning of genetic variability among subspecies relative to partitioning among regions within subspecies. NEI’s (1978) genetic distances, corrected for small sample sizes, Ro- GERS’ (1972) distances and modified RoGERS’ distances (WRIGHT 1978) between all pairs of regional and subspecies samples of mountain hares and local samples of brown hares were calculated. Regard- ing brown hares, eight local samples studied earlier in the same laboratory (HarTL et al. 1993) were used with adjusted numbers of loci (n = 40). Relationships of pairwise genetic distances were revealed by an unrooted Wagner dendrogram (FARRIS 1972). Spatial partitioning of allozyme variability in Zepus timidus Sal Results Polymorphism was revealed at 13 loci. The overall rate of polymorphism (99 % criterion, 40 locı considered) for European mountain hares amounted to 32.5 %. Polymorphic loci, al- leles, and associated allele frequencies are given in table 1 for four regions of Europe (disre- garding subspecific allocation), and in table 2 for subspecies. Allele frequencies of moun- tain hares from Switzerland did not vary significantly among age classes or sexes. Values of Table 1. Allele frequencies (%) at polymorphic loci of mountain hares from four regions of Europe based on 40loci. H,= average observed heterozygosity, H.= average expected heterozygosity, P = rate of polymorphism (99% criterion), A = mean number of alleles per locus. Significant devia- tions of genotype frequencies from expected Hardy-Weinberg frequencies are indicated for respective loci and regions (? p < 0.05; p < 0.01, significance tests using exact probabilities). Scandinavia Alps Ural NW Europe (n = 74) (m=3142)) (n= 14) (m=39)) Subspecies L. t. timidus L. t. varronis L. t. timidus L. t. scoticus and Locus Allele L. t. hibernicus a b a b € a d a d a b a b a b c e a b c a b c a b a b c a b a b 312 F. SUCHENTRUNK etal. Table 2. Allele frequencies (%) at polymorphic loci of four subspecies of mountain hares from Eu- rope based on 34 locı. H, = average observed heterozygosity, H. = average expected heterozygosity, P=rate of polymorphism (99 % criterion), A = mean number of alleles per locus. Significant devia- tions of genotype frequencies from expected Hardy-Weinberg frequencies are indicated with the “a” allele for respective loci and subspecies (? p<0.05; ” p<0.01, significance tests using exact probabil- ities). Subspecies L. t. timidus L. t. varronis L. t. scoticus L. t. hibernicus Locus Allele (n = 88) (= 12) IS) (n = 4) a b a b © a d a d a b a b a b c e a b c a b c a b a b c a b a b genetic variability for the regions (based on 40 loci) are listed in table 1, and for the subspe- cies (based on 34 loci) in table 2. The observed genotypic distributions differed significantly from Hardy-Weinberg expectations at six loci in three regional samples (Tab. 1) and at six loci in two subspecies samples (Tab. 2). Basically, all these significant genotype deviations were due to heterozygote deficiencies. Pairwise NeEr’s (1978) genetic distances in mountain hares, corrected for small sample sizes, ranged between 0.000-0.008 among subspecies, and between 0.000-0.017 among regions (Tab. 3). Modified Rogers’ distances (WRIGHT 1978) ranged between 0.037-0.117 among subspecies, and between 0.024-0.135 among regions (Tab. 3). Pairwise genetic distances between single regional samples of mountain hares and Spatial partitioning of allozyme variability in Zepus timidus 318 single local samples of central European brown hares ranged between 0.068-0.093 (NeEr’s 1978 D), and between 0.253-0.295 (Rogers’ modified distances; WRIGHT 1978). In mountain hares, locus-specific Fsr and Fıs values did not show any particular con- cordance across loci; for the regional samples mean Fsr = 0.157, mean Fıs = 0.17, and mean Fır = 0.3. For the subspecies samples the respective values were 0.14, - 0.02, and 0.12. The relative genetic differentiation (Fsr values) for pairs of subspecies are listed in table 4 along with associated significances of heterogeneity of allele frequencies. Only Table 3. Ner’s (1978) genetic distances for small sample sizes (above diagonal) and modified Rogers’ distances (below diagonal) among pairs of mountain hares from European regions (based on 40 loci), and subspecies (based on 34 loci). Regions: SCAN = Scandinavia, ALPS, URAL, NWE = Northwest Europe (Scotland, Ireland). SCAN ALPS URAL NWE L.ttimidus L.t.varronis L.t.scoticus L.t. hibernicus (1) (2) (3) (4) (5) (6) (7) (8) 0.010 0.001 0:013530:008 - 0.017 04135 - Table 4. Fs7r values for pairs of subspecies (above diagonal) and significance values for heterogeneity of allele frequencies (below diagonal); significance is based on exact Fisher’s test and sequential Bon- ferroni procedure (sig.: p< 0.05; n.s.: p > 0.05). Significance is given with significantly varying allele frequencies at least at one locus. L.t. timidus (1) L.t. varronis (2) L. t. scoticus (3) L.t. hibernicus (4) Table 5. Fsr values for pairs of sampling regions of L.r. varronis and L.t. timidus, respectively (above diagonal), and significance values for heterogeneity of allele frequencies (below diagonal), based on exact Fisher’s test and sequential Bonferroni procedures (sig.: p< 0.05; n.s.: p > 0.05). Significance is given if at least one locus shows significantly varying allele frequencies in a pairwise comparison. L.t. varronis Switzerland 0.146 0.048 0.173 0.014 (1) Austria ass 0.055 0.237 0.107 (2) France as HS: 0.171 0.024 3) L.t. timidus Ural (4) Scandinavia (5) 314 F. SUCHENTRUNK etal. 6.53% of all possible pairwise comparisons of region-specific allele frequencies at poly- morphic loci (59 tests) yielded significant differences. According to theory, significance of pairwise Fsr-values is given with allele frequencies varying significantly at least at one lo- cus studied (WRIGHT 1978). Fsr-values for pairs of sampling regions within the subspecies L.t. timidus and L.t. varronis, respectively, as well as associated significance values for heterogeneity of allele frequencies are given in table 5. Significant differences of allele frequencies at polymorphic loci were found only in 5.4 % of all possible pairwise compari- sons between regions (92 tests). Details of WrIGHT’s (1978) hierarchical F-statistics giving the proportions of genetic variation partitioned among the four subspecies studied, rela- tive to the regional effect on genetic partitioning, is presented in table 6. An unrooted Wagner dendrogram based on Rogers’ distances, depicting genetic relationships among mountain hares from the four regions studied in Europe and eight local samples of brown hares from central Europe (HarTL et al. 1993) is presented in figure 2. Table 6. Wright’s hierarchical F-statistics in European mountain hares, based on 34 allozyme loci. Var- iance components and F-statistics combined across loci. Comparison variance component X NG sampling regions'-subspecies sampling regions'-total variance subspecies-total variance ! Austria, France, Ireland, Scandinavia, Scotland, Switzerland, Ural. Lepus timidus Lepus europaeus URAL SCANDINAVIA ALPS NW-EUROPE Fig. 2. Unrooted Wagner dendrogram (midpoint rooting of longest path) depicting genetic relation- ships among mountain hares (Lepus timidus) from various regions of Europe and brown hares (L. euro- paeus) from eight local samples of central Europe. The dendrogram is based on Rocers’ (1972) dis- tances, calculated from allele frequencies at 40 loci. Total tree length = 0.192, distance between “Ural” and “Scandinavia” = 0.032, cophenetic correlation coefficient = 0.986 Spatial partitioning of allozyme variability in Lepus timidus 318 Discussion The level of gene pool variability of European mountain hares, as indicated by allozyme heterozygosity, rate of polymorphism, and mean number of alleles per locus in the diverse regions and subspecies is similar to that of brown hares from various continental Euro- pean regions (HARTL et al. 1989, 1990, 1992, 1993, 1995; SUCHENTRUNK et al. 1998, 1999). The presently found heterozygosity values are typical for undisturbed populatiens of ter- restrial mammalian species of diverse orders (NEvo 1978; TIEDEMANN et al. 1996). The somewhat reduced rates of polymorphism and mean numbers of alleles per locus in Scot- ish and Irish mountain hares are likely due to the low sample sizes. The H.:P rates within regions and subspecies ranged between 0.107-0.256. These values fall within the range of “undisturbed” populations (TIEDEMANN et al. 1996), indicating populations without genet- ic depletion e.g., due to severe bottlenecks or long-term low effective population size. Average numbers of alleles per locus (A) do not give any hint for depauperated gene pools in Scandinavia, the Ural or the Alps. The low A-values for mountain hares from Scotland and Ireland are most probably due to the low sample sizes for these regions. The overall rate of polymorphism of the mountain hares (32.5 %) appears to be some- what greater than in brown hares. Combining the data of Harrtı et al. (1989, 1990, 1992, 1993, 1994, 1995) and SUCHENTRUNK et al. (1998, 1999) for brown hares from various re- gions of Europe yields an overall rate of polymorphism of 25.9 %. Adjusting the set of loci analysed in brown hares to the presently studied set (40 loci) results in a value of 27.5 % for brown hares. The still somewhat higher value of mountain hares is due to three polymorphic loci (Me-2, Acp-l, Acon). But these three loci are only marginally poly- morphic with variant alleles occurring in one or two regions, respectively. Furthermore, the Acp-1” allele occasionally found in some mountain hares from Switzerland may result from rare cases of hybridization (cf. e.g., BALDENSTEIN 1863; FRAGUGLIONE 1966; SCHRÖ- DER et al. 1987; Tuurın et al. 1997 a). In general, most of the locı found polymorphic ın brown hares (Hartr et al. 1990, 1992, 1993; SUCHENTRUNK et al. 1998; 1999) are also poly- morphic in the presently studied mountain hares. Moreover, most of the locı with several alleles in brown hares ( Es-l, -Gal, Pep-2, Mpi) reveal several of these alleles in the mountain hares too. These very similar allele patterns hamper differential diagnosis by allozymes between mountain and brown hares. When comparing allozyme patterns of brown and mountain hares GRILLITSCH et al. (1992) screened only few mountain hares from one region in Aus- tria. With that small and regionally limited sample size they obviously have missed some polymorphisms in mountain hares. Their results suggested a differential diagnosis between these two species by three loci. However, the allelic differences at the 5-Gus and the Pgm-1 loci between the two species (GRILLITSCH et al. 1992) could not be proven presently be- cause of dubious zymograms. The present results suggest that, among the array of locı screened, only the Acp-1 locus has alleles alternately fixed in the two species, with occa- sional cases of introgressive hybridization in mountain hares from the Alps. However, at present no allozyme data of brown hares from regions of potential introgressive hybridiza- tion are available to substantiate this hypothesis. In Scandinavian mountain hares no hint of introgressive hybridization was found presently, although Thuuin et al. (1997 a) reported presence of mountain hare mtDNA in brown hares from the Upland region. Despite the relatively high amount of “private alleles” at several loci (31.3 % of all studied loci) in various regions or subspecies of mountain hares, overall genetic distances among gene pools of the diverse regions or subspecies are generally low in magnitude. Because of their generally low frequencies, “private alleles” do not greatly influence ge- netic differentiation. Ners (1978) genetic D-values among subspecies are similar to those found among local samples of brown hares within central Europe (e.g., HArTL et al. 1989, 1990, 1992, 1993). However, brown hares also exhibit low genetic differentiaton even 316 F. SUCHENTRUNK et al. across large geographic distances in Europe; this indicates a rather panmictic network and a lack of discernible populations (HARTL et al. 1990; SUCHENTRUNK et al. 1999). Only mountain hares from the Ural region and Ireland show a slightly increased genetic diver- gence to mountain hares from the other study regions or subspecies. In Ural mountain hares thıs slight separation is only based on significantly increased frequencies of the ß- Gal® and the Idh-2“ alleles. It might e.g., result from the large geographic distance be- tween the Ural and the other regions or indicate introgression of gene pool elements of L.t. kozhevnikovi. According to OGNEFF (1929) this subspecies occurs in the south Ural (near Miass) some 150 km south of the presently studied collection site. However, skull dimensions and the external features of the presently studied individuals tend to conform with those of L.r. timidus of various parts of European Russia, rather than with those of L.t. kozhevnikovi (albeit there is little morphometric differentiation between the two sub- species in Russia; see OGNEFF 1929). The slightly raised level of genetic separation of Irish mountain hares is particularly due to changes in allele frequencies at the Es-d locus and the presence of a “private al- lele” (Ldh-2°) with a frequency of 25 %. However, significant changes in allele frequen- cies between ZL.r. hibernicus and the other subspecies was only found at the Pep-2 locus. In view of the small sample size from NW Europe, and the fact that Scottish mountain hares were only screened from the isle of Mull, where Irish mountain hares had been lib- erated in the last century, no conclusions regarding the genetic differentiation between these two subspecies can be drawn. The generally low level of genetic differentiation between subspecies or regions of mountain hares is also indicated by the small proportion of relative genetic variability partitioned among subspecies or regions and particularly by the low proportions of signif- icantly varying allele frequencies between pairs of subspecies (6.3 %) or regions (5.4 %). Moreover, while 13.6 % of the relative genetic variation are partitioned among regions within subspecies, less than 1 % is partitioned between the subspecies studied. This means that gene pool divergence is greater among sampling regions within Z.r. timidus and L.t. varronis, respectively, than between all studied subspecies. Hence, no distinct gene pools of the studied subspecies can be identified. Also, a single mountain hare collected in the Primorje region of Far East Sıbiria analysed in our laboratory did not reveal any new allele. All results agree with sequence data demonstrating an admixture of mtDNA haplo- types in mountain hares from Scandinavia and other parts of Europe (ThuLin et al. 1997b) without separation in clear phylogeographic units. The present allozyme results also conform to the hypothesis of a rather panmictic gene pool of late- and postglacial po- pulations in central Europe without any specific phyletic blocks. They are also in agree- ment with the view that there were no severe demographic bottlenecks, founder effects, long-term low effective population sizes, multiple regional extinctions etc. in post-glacial populations during the colonization of the present ranges. Obviously, separation of post- glacial European mountain hares into several geographic ranges has not resulted in dis- tinct gene pools; and there is very little measureable evolutionary divergence between the pools of coding genes in the phenotypically specified subspecies. Acknowledgements We are grateful to the hunters from Grisons and Glarus (Switzerland), to the Swedish hunters, to Mr. K. Knott (forestry commission Oban, Scotland) and Dr. R. MAurizıo (Vicosoprano, Switzerland) for shooting the hares and/or providing organ samples. Mr A. KöRBER produced the figures. Dr. M. PAvLEnKoO (Russ. Acad. Sciences, Vladivostok) provided organ samples of a mountain hare from Far East Sibiria. Dr. M. Janovsky (Berne) transferred most of the Swiss samples and Dr. I. UpI- NA (Moscow) the Far East Sibirian sample to our laboratory. Spatial partitioning of allozyme variability in Zepus timidus 317 Zusammenfassung Räumliche Verteilung der Allozymvariabilität bei europäischen Schneehasen (Lepus timidus): Genpool-Divergenz in einem disjunkten Verbreitungsgebiet? Die Allozymvariabilität von 209 Schneehasen (Lepus timidus) von Skandinavien, Rußland, den Al- pen, Schottland und Irland wurde mittels horizontaler Stärkegelelektrophorese von 40 Strukturgenloci festgestellt, um zu prüfen, ob die postglaziale Besiedelung der heutigen disjunkten Schneehasen-Ver- breitung in Europa zu einer markanten genetische Differenzierung geführt hat. Die meisten Allele an den 13 polymorphen Loci waren identisch mit den schon früher bei europäischen Feldhasen (Z. euro- paeus) gefundenen. Durchschnittliche erwartete Heterozygotie-Werte pro Region bzw. Subspecies (2,0-5,0%) sowie die Polymorphieraten (8,8-29,4%) entsprachen denen bei Feldhasen aus früheren Untersuchungen. Trotz der hohen Rate (31,3%) an Allelen, die ausschließlich in einzelnen Regionen oder Subspecies vorkamen, lagen die genetischen Distanzen (Ner's D: 0,000-0,008 zwischen Subspe- zies, 0,000-0,017 zwischen Regionen) grundsätzlich im Bereich jener Werte, wie sie zwischen lokalen Feldhasen-Populationen in Mitteleuropa früher festgestellt wurden. Ebenso zeigten die relativ gerin- gen mittleren Fsr Werte (0,157 zwischen Regionen; 0,14 zwischen Subspecies) sowie die geringe Zahl an signifikanten paarweisen Unterschieden von Allelfrequenzen eine geringe genetische Differenzier- ung zwischen den Regionen bzw. Subspecies an. Während 13,6 % der relativen genetischen Variabili- tät zwischen Regionen innerhalb von Subspecies verteilt waren, lag der entsprechende Wert für die Verteilung zwischen den Subspecies unter einem Prozent. Alle Ergebnisse entsprechen der Hypothese einer panmiktischen spät- bzw. postglazialen Population in Mitteleuropa und der Annahme, daß bei der Kolonisation der heutigen disjunkten Verbreitungsgebiete in Europa keine starke genetische Drift erfolgt ist. References ÄNGERBJÖRN, A.; FLux, J. E. C. (1995): Lepus timidus. Mammalian Species 495, 1-11. BALDENSTEIN, T. C. zu (1863): III. Zoologische Mittheilungen. Der Bastardhase. Jahresber. d. Naturf. Gesellsch. Graubündens. NF. VIII. Jahrg. 98-101. BERGENGREN, A. (1969): On genetics, evolution and history of distribution of the heath hare, a distinct population of the arctic hare (Lepus timidus). Swed. Wildl. (Viltrevy) 6, 381-460. CoRBET, G. B. (1986): Relationships and origins of the European lagomorphs. Mammal Rev. 16, 105-110. CoRBET, G. B.; SOUTHERN, H.N. (1977): The Handbook of British Mammals. Sec. Ed. London: Black- well. Dörpes, D. (1997): Die jungpleistozäne Säugetierfauna der Gudenushöhle (Nieder-Österreich). Wiss. Mitt. Niederösterr. Landesmus. 10, 17-32. FARRITS, J. S. (1972): Estimating phylogenetic trees from distance matrices. Am. Nat. 106, 645-668. Frux, J. E. C. (1970): Colour change of Mountain hares (Lepus timidus scoticus) in north-east Scotland. J. Zool. (London) 162, 345-358. FRAGUGLIONE, D. (1966): L’hybridation du lievre commun et du lievre variable, Lepuse. europaeus x L. timidus varronis (Pallas, 1878). Säugetierkdl. Mitt. 14, 92-100. GRILLITSCH, M.; HARTL, G. B.; SUCHENTRUNK, F.; WILLING, R. (1992): Allozyme evolution and the mole- cular clock in the Lagomorpha. Acta Theriol. 37, 1-13. HARTL, G. B.; SUCHENTRUNK, F.; WILLING, R.; GRILLITSCH, M. (1989): Biochemisch-genetische Variabili- tät und Differenzierung beim Feldhasen (Lepus europaeus) in Niederösterreich. Wien. Tierärztl. Mschr. 76, 279-284. HARTL, G. B.; MARKOWSKI, J., Kovacs, G.; GRILLITSCH, M.; WirLinG, R. (1990): Biochemical variation and differentiation in the brown hare (Lepus europaeus) of Central Europe. Z. Säugetierkunde 55, 186-193. HARTL, G. B.; MARKOWSKI, J.; SWIATCKI, A.; JANISZEWSKI, T.; WILLING, R. (1992): Genetic diversity in the Polish brown hare Lepus europaeus Pallas, 1778: implications for conservation and management. Acta Theriol. 37, 15-25. HARTL, G. B.; SUCHENTRUNK, F.; NADLINGER, K.; WiLLing, R. (1993): An integrative analysis of genetic differentiation in the brown hare ZLepus europaeus based on morphology, allozymes, and mitochon- drial DNA. Acta Theriol. 38, suppl. 2, 33-57. HARTL, G. B.; WirLing, R.; NADLINGER, K. (1994): Allozymes in mammalian population genetics and 318 F. SUCHENTRUNK etal. systematics: Indicative function of a marker system reconsidered. In: Molecular Ecology and Evolu- tion: Approaches and Applications. Ed. by B. SCHIERWATER; B. STREIT; G. P. WAGNER; and R. DE- SALLE. Basel: Birkhäuser. Pp. 299-310. HARTL, G. B.; SUCHENTRUNK, F.; WILLING, R.; PETZNEK, R. (1995): Allozyme heterozygosity and fluctu- ating asymmetry in the brown hare (Lepus europaeus): a test of the developmental homeostasis hy- pothesis. Phil. Trans. R. Scoc. Lond. B 350, 313-323. Lang, G. (1994): Quartäre Vegetationsgeschichte Europas. Methoden und Ergebnisse. Jena: G. Fischer. NeEı, M. (1978): Estimation of average heterozygosity and genetic distance from a small number of indi- viduals. Genetics 89, 583-590. Nevo, E. (1978): Genetic variation in natural populations: patterns and theory. Theor. Pop. Biol. 13, 121-177. ÖGNEFF, S. J. (1929): Zur Systematik der russischen Hasen. Zool. Anz. 84, 69-83. Rice, W. S. (1989): Analyzing tables of statistical tests. Evolution 43, 223-225. RICHARDSON, B. J.; BAVERSTOCK, P. R.; ADAams, M. (1986): Allozyme Electrophoresis. A Handbook for Animal Systematics and Population Studies. San Diego: Academic Press Inc. ROGERS, J. S. (1972): Measures of genetic similarity and genetic distance. Studies in Genetics, Univ. Texas Publ. 7213, 145-153. ROTHE, G. M. (1994): Electrophoresis of Enzymes. Laboratory Methods. Springer LAB Manual. Berlin: Springer. SCHRÖDER, J.; SOVERI, T.; SUOMALAINEN; H. A., LINDBERG; L.-A.; VAN DER Loo, W. (1987): Hybrids be- tween Lepus timidus and Lepus europaeus are rare although fertile. Hereditas 107, 185-189. STUART, A. J. (1982): Pleistocene Vertebrates in the British Isles. London, New York: Longman. SUCHENTRUNK, F.; HARTL, G. B.; FLux, J. E. C.; PARKES, I.; HAIDEN, A.; TAPPER, S. (1998): Allozyme hetero- zygosity and fluctuating asymmetry in brown hares (Lepus europaeus) introduced to New Zealand: Developmental homeostasis in populations with a bottleneck history. Acta Theriol. Suppl. 5, 35-52. SUCHENTRUNK, F.; MICHAILOV, C.; MARKOV, G.; HAIDEN, A. (1999): Population genetics of Bulgarian brown hares Lepus europaeus: allozymic diversity at zoogeographical crossroads. Acta Theriol. (in press). SWOFFORD, D. L.; SELANDER, R. B. (1989): BIOSYS-1. A computer program for the analysis of allelic variation in genetics. Release 1.7. Champaign, IL: Illinois Natural History Survey. TIEDEMANN, R.; HAMMER, S.; SUCHENTRUNK, F.; HARTL, G. B. (1996): Allozyme variability in medium-sized and large mammals: determinants, estimators, and significance for conservation. Biodiv. Lett. 3, 81-91. THULIN, C.-G.; JAAROLA, M.; TEGELSTRÖM, H. (1997 a): The occurrence of mountain hare mitochondrial DNA in wild brown hares. Mol. Ecol. 6, 463-467. THULIN, C.-G.; IsSAKSsoN, M. ; TEGELSTRÖM, H. (1997 b): The origin of Scandinavian mountain hares (ZLe- pus timidus). Gibier Faune Sauvage, Game Wildl. 14, 463-475. WALHOVD, H. (1965): Age criteria of the mountain hare (Lepus timidus L.) — with analyses of age and sex rations, body weights, and growth in some Norwegian populations. Meddel. fra Statens Viltun- dersokelser 2. serie. 22, 1-57. WRIGHT, S. (1978): Evolution and genetics of populations. Vol. IV. Variability within and among natural populations. Chicago: Univ. Chicago Press. Authors addresses: DR. FRANZ SUCHENTRUNK, KARIN POLSTER, Research Institute of Wildlife Ecology, Vienna Veterinary University, Savoyenstraße 1, A-1160 Vienna, Austria; DR. MAR- co GIACOMETTI, Untersuchungsstelle für Wildtierkrankheiten, Institut für Tier- pathologie, Länggass-Strasse 122, CH-3012 Bern, Switzerland; Dr. PEIDER RATTI, Jagd- und Fischereiinspektorat Graubünden, Loestrasse 2, CH-7000 Chur, Switzer- land; CARL-GUSTAF THULIN, Department of Conservation Biology and Genetics, EBC, Uppsala University, Uppsala, Box 7003, S-75007 Uppsala, Sweden; DR. CHRISTIAN RUHLE, Jagd- und Fischereiverwaltung St. Gallen, Davidstrasse 35, CH-9001 St. Gallen, Switzerland; Dr. ALEXEY VAasıL’Ev, Institute of Plant and Ani- mal Ecology, Ural Div. Russian Academy of Sciences, 8 Marta Street 202, 620219 Yekaterinburg, Russia; DR. LEOPOLD SLOTTA-BACHMAYR Nationalpark-Institut des Hauses der Natur, Museumsplatz 5, A-5020 Salzburg, Austria. Z. Säugetierkunde 64 (1999) 319-320 ZEITSCHRIFT ®&&# FÜR © 1999 Urban & Fischer Verlag SAUG ETI ERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Buchbesprechungen WANDREY, R. (1997): Die Wale und Robben der Welt: Vorkommen, Gefährdung, Schutz. Kosmos Na- turführer. Stuttgart: Franckh-Kosmos Verlags-G.m.b.H. & Co. 285 pp., 93 Farb-, 24 s/w-Zeichnungen, 108 Farbfotos, 120 zweifarbige Verbreitungskarten. DM 49,80 / 6S 364,- / sFr 47.80. ISBN 3-440-07047-6. Dr. RÜDIGER WANDREY, der Direktor des „Zoos am Meer“ in Bremerhaven, hat einen schönen und instruktiven Naturführer über aquatische Säugetiere vertaßt. Der Titel seines Buches ist eher tief- staplerisch zu verstehen, denn es werden neben Walen (Cetacea) und Robben (Pinnipedia) auch die Seekühe (Sirenia), sowie der Eisbär (Ursus maritimus), der Meerotter (Enhydra lutris) sowie der Kü- stenotter (Lutra felina, nach W. C. WOZENCRAFT, 1993, korrekt: Lontra felina) behandelt. Dieses Buch ist eindeutig mehr als ein Bestimmungsbuch für den Säugetierliebhaber, da der Autor zusätzlich zu den Bestimmungshilfen Einführungen in die Biologie bietet. Er informiert über Schwimm- und Tauchvermögen, charakterisiert die Sinnesleistungen, behandelt ferner Probleme der Haltung und er- wähnt auch die Rolle von Walen und Sirenen in der Mythologie. Die problembeladenen Beziehungen zwischen dem Menschen und den aquatischen Säugern werden ebenfalls erörtert, so wird der Einfluß von Meeressäugetieren auf die Fischerei besprochen und es wird auf Schutzprogramme und Beobach- tungsmöglichkeiten hingewiesen. In den Abschnitten über die einzelnen Arten folgen auf eine Be- schreibung der Art Bemerkungen zur Verbreitung, zur Bestimmung, zur Ernährung, Lebensweise und Fortpflanzung und Angaben zum heutigen Bestand. Die Verbreitung der einzelnen Arten wird durch klare, übersichtliche und informative Verbreitungskärtchen illustriert, die selbsterklärend sind. Das Buch gewinnt nicht nur an Informations-, sondern auch an ästhetischem Wert durch seine Illustratio- nen; die Habitusbilder der Wale sind durch die Graphikerin MARIANNE GOLTE-BECHTLE zu Papier ge- bracht worden, die anderen behandelten Arten werden durch vorzügliche Farbphotos anschaulich ge- macht. Gesteigert wird der positive Eindruck, welchen dieser Naturführer macht, durch seine sorgfältige drucktechnische Gestaltung. Ein Glossar bietet Erklärungen zu Fachbegriffen, eine Liste macht den Leser mit Institutionen, Zoologischen Gärten und Verbänden, welche sich mit dem Schutz aquatischer Säugetiere beschäftigen oder deren Biologie erforschen, bekannt. Ein kurzes Literaturver- zeichnis und ein sechs Seiten langes Schlagwortregister schließen dieses erfreuliche Buch ab. Zwei Aussagen sollten korrigiert werden: Es ist nicht notwendig, für die Bartenwale neben dem gültigen Begriff „Mysticeti“ auch das obsolete Wort „Mystacoceti“ zu benutzen. Auf Seite 251 schreibt der Autor, daß die Geruchsorientierung über Geschmacksknospen auf der Zunge erfolge. Auch wenn Geruch und Geschmack eng miteinander verbunden sind, ist die Aussage so, wie sie im Text steht, vom Autor sicherlich nicht gemeint. P. LANGER, Gießen THIEDE, U.: Auf Haustierspuren zu den Ursprüngen der Japaner. Vor- und frühgeschichtliche Haus- tierhaltung in Japan. München: iudicium Verlag 1998. 152 pp., 55 Abb. ISBN 3-89129-429-8. DM 38,— Der vorliegende schmale, aber inhaltsreiche Band der Zoologin und Japanologin U. THIEDE befaßt sich mit Ergebnissen der japanischen Haustierforschung, die besonders in den 80er und 90er Jahren eine Fülle neuerer Erkenntnisse hervorgebracht hat. Das Ziel der auf breiter methodischer Grund- lage (Morphologie, Biochemie, Genetik) durchgeführten japanischen Forschungsarbeiten war (und ist) darauf ausgerichtet, durch Vergleiche von alten japanischen Haustierrassen mit Haustieren des benachbarten asiatischen Festlandes Herkunft und Einwanderungswege der auf den japanischen In- seln auftauchenden Haustiere zu rekonstruieren. Damit eröffnet sich die Möglichkeit, die Geschichte der menschlichen Besiedlung Japans einer Klärung näher zu bringen, weil die Wanderwege der Haus- tiere mit jenen der sie haltenden Menschen zwangsläufig identisch sind. Sprachbarrieren ist es wohl zuzuschreiben, daß die Ergebnisse japanischer Haustierforschung in der westlichen Welt bis heute durchweg unbekannt geblieben sind. Das vorliegende, durch zahlreiche Verbreitungskarten und Dia- 0044-3468/99/64/05 — 319 $ 12.00/0 320 Buchbesprechungen gramme reich illustrierte Buch, das auf eine Vielzahl japanischer haustierkundlicher wie auch anthro- pologischer Arbeiten zurückgreift, ist angetan, diese Lücke zu schließen. Auf zwei einleitende Ab- schnitte folgen 3. Forschungsgeschichte der Bevölkerungsentwicklung Japans (mit einer Darstellung zahlreicher Theorien), 4. Ergebnisse der neuesten anthropologischen Forschung in Japan, 5. Wegbe- gleitende Pflanzen und Haustiere, 6. Aktueller Stand der vergleichenden Haustierforschung in Japan, 7. Die Haustiere im einzelnen und 3. Von Mäusen, Menschen und Viren. Beschlossen wird der Band mit einer Zusammenfassung und Diskussion, die übersichtlich informiert, wann und von wo welche Haustiere den japanischen Archipel vermutlich besiedelten. Sieht man einmal von Schafen, Enten und Gänsen ab, kommen heute in Japan die gleichen sogenannten klassischen oder traditionellen Haustiere vor wie in Europa. Erwähnenswert ist allerdings, daß die Einwanderungen erst in den Jahr- hunderten um Chr. Geburt oder noch viel später erfolgten, wo doch in Europa schon vor mehr als 6000 Jahren Haustiere ein untrennbarer Bestandteil menschlichen Daseins waren. Das 224 Titel um- fassende, vorwiegend japanische Arbeiten aufführende Schriftenverzeichnis unterstreicht, daß U. THIEDE einen profunden Beitrag zur Haustierforschung in Japan vorgelegt hat, der nicht nur in je- der zoologisch-haustierkundlichen Fachbibliothek einen Platz beansprucht, sondern auch unter Prä- historikern, Anthropologen, Ethnologen und Kulturgeschichtlern Aufmerksamkeit verdient. Abschlie- ßend seien ein paar Randbemerkungen erlaubt. Bei der redaktionellen Überarbeitung des Textes ist übersehen worden, daß sich die Verbreitungskarte der Wachtel nicht in Abb. 52 (S. 132), sondern Abb. 53 befindet. Nicht einheitlich gehandhabt werden die lateinischen Haustiernamen, z.B. Equus przewalskii f. caballus, andererseits Felis silvestris f. catus. Und „rückverwildert“ sollte besser wohl „verwildert“ heißen. Schließlich ist auf S. 114 zu verbessern: „... 19% der untersuchten genetischen loci polimorph sind ....“. (nicht ist) H. REICHSTEIN, Flintbek Redaktionelle Hinweise Einsendung und Annahme von Manuskripten: Ms. zur Veröffentlichung sind einzusenden an den 1. Schriftleiter: Herrn Prof. Dr. Dieter Kruska, Institut für Haustierkunde, Christian-Albrechts-Universität, Olshausenstr. 40-60, D-24118 Kiel. e-mail: dkruska@ifh.uni-kiel.de Über die Annahme von Ms. zur Veröffentlichung wird gemäß Geschäftsordnung der Deutschen Gesellschaft für Säugetierkunde entschieden. Der Eingang von Ms. wird sofort bestätigt, und nach Erhalt der Gutachten wer- den die Autoren über die Entscheidung umgehend informiert. Schriftwechsel, der die technische Drucklegung betrifft, ist zu richten an: Zeitschrift für Säugetierkunde, Urban & Fischer Verlag, Löbdergraben 14a, D-07743 Jena. Äußere Form der Manuskripte: Die Ms. sind 3fach und vollständig als Original mit 2 Kopien in deutscher oder englischer Sprache einzureichen. Sie müssen mit Schreibmaschine oder als deutlich lesbare Computerausdrucke auf DIN A4-Blättern (21x 29,7 cm), mit einem linken Heftrand von 4cm und durchgehend mit einem doppelten Zeilenabstand geschrieben sein. Jedes Ms. soll auf der 1. Seite folgende Angaben enthalten: Titel der Studie; Name(n) der Autoren (männliche Vornamen nur in Initialen, weibliche ausgeschrieben); Institution(en), an der (denen) die Studie durchgeführt wurde. Ferner soll bei längeren Titeln ein kurzer Kolumnentitel (Seitenüber- schrift) von maximal 72 Anschlägen angegeben werden. Die Schrifttypen müssen einheitlich sein, ohne halbfette, gesperrte, kursive etc. Hervorhebungen. Fußnoten auf den Seiten oder Appendices am Ende sind nicht zugelassen. Autorennamen im Zusammenhang mit Literatur- zitaten sind im Text mit Bleistift zu unterstreichen. Wissenschaftliche zoologische und botanische Gattungs-, Art- und Unterartnamen sind mit einer Schlangenlinie zu unterstreichen. Der Name von Erstbeschreibern bei wis- senschaftlichen Namen wird nicht unterstrichen. Im Ms. ist am Rand mit Bleistift zu vermerken, an welcher Stelle die einzelnen Abbildungen und Tabellen ein- gefügt werden sollen. Am Ende der Studie, im Anschluß an das Literaturverzeichnis, sind die genauen postalischen Adressen des (der) Autoren anzugeben. Umfang und Aufbau der Manuskripte: Manuskripte können als Originalarbeiten oder als wissenschaftliche Kurz- mitteilungen veröffentlicht werden. a) Originalarbeiten: Originalarbeiten sollen einschließlich Abbildungen, Tabellen und Literaturverzeichnis einen Gesamtumfang von 30 Ms.-Seiten nicht überschreiten. Der Text soll eine klare Gliederung aufweisen in: Abstract (in englischer Sprache), Einleitung, Material und Methode, Ergebnisse, Diskussion (oder zusammengefaßt als Er- gebnisse und Diskussion), Danksagungen, Zusammenfassung und Literatur. Bei deutschsprachigen Studien ist über dem englischen Abstract der Titel in englischer Sprache anzugeben. Bei englischsprachigen Studien ist entsprechend über der deutschen Zusammenfassung der Titel in deutscher Sprache anzugeben. b) Wissenschaftliche Kurzmitteilungen: Wissenschaftliche Kurzmitteilungen sollen einen Gesamtumfang von 5 Ms.-Seiten nicht überschreiten. Sie enthalten kein Abstract und keine Zusammenfassung. Bei deutschsprachigen Ms. ist jedoch auf der 1. Seite zusätzlich eine englische Übersetzung des Titels anzugeben, bei englischsprachigen entsprechend eine deutsche Übersetzung. Wissenschaftliche Kurzmitteilungen sind äußerlich nicht in die Kapitel Einleitung, Material und Methode, Ergebnisse und Diskussion gegliedert, müssen aber in der textlichen Darstel- lung diese Strukturierung erkennen lassen. Key words: Bis zu fünf Key Words sollen dem Abstract folgen. Abbildungen: Anzahl und Größe der Abbildungen sind unbedingt auf das notwendige Mindestmaß zu beschrän- ken. Es können nur kontrastreiche, reproduktionsfähige Originale verwendet werden, farbige Abbildungen wer- den nicht angenommen. Beschriftungen müssen einheitlich und wegen späterer Verkleinerung groß und kräftig sein. Auf der Rückseite sind laufende Numerierung, Zeitschrift und Autorennamen anzugeben. Alle Bildunter- schriften sollen auf einem gesonderten Blatt gelistet werden. Tabellen: Tabellen müssen auf ein Mindestmaß beschränkt werden. Jede Tabelle muß auf einem gesonderten DIN A4-Blatt angelegt sowie mit Überschrift und laufender Numerierung versehen sein. Umfangreichere Tabel- len müssen auf Folgeblättern als Fortsetzung erscheinen. Literaturverzeichnis: Jede zitierte Arbeit ist in alphabetischer Reihenfolge aufzuführen. Es soll nur international erreichbare Literatur zitiert werden. Die Zeitschriften werden mit ihrem Kurztitel angegeben. Beispiele: a) Zeitschriftenbeiträge: Karko, E. K. V; HANDLEY, €. O. Jr. (1994): Evolution, biogeography, and description af a new species of Fruit-eat- ing bat, genus Artibeus Leach (1821), from Panama. Z. Säugetierkunde 59, 257-273. b) Bücher: HERRE, W.; RönHrs, M. (1990): Haustiere - zoologisch gesehen. 2. Aufl. Stuttgart, New York: Gustav Fischer. c) Handbuchbeiträge: NIETHAMMER, J. (1982): Cricetus cricetus (Linnaeus, 1758) — Hamster (Feldhamster). In: Handbuch der Säugetiere Europas. Ed. by J. NIETHAMMER und F. Krapp. Wiesbaden: Akad. Verlagsges. Vol. 2/I, 7-28. Drucklegung: Der jeweilige Hauptautor erhält sowohl die Andrucke der Bilder zur Prüfung wie auch den kom- pletten Umbruch seines Artikels zusammen mit dem Ms. Es dürfen nur tatsächliche Satzfehler korrigiert wer- den. Vom Ms. abweichende Änderungen werden dem Autor in Rechnung gestellt. Es sind die üblichen Korrekturzeichen zu verwenden. Ein als druckfertig bezeichneter Abzug ist zu senden an: Herrn Prof. Dr. Peter Langer, Institut für Anatomie und Zellbiologie, Justus-Liebig-Universität, Aulweg 123, D-35385 Giessen. e-mail: peter.langer@anatomie.med.uni-giessen.de Sonderdrucke: Verfasser von Originalbeiträgen und Wissenschaftlichen Kurzmitteilungen erhalten 50 Sonder- drucke gratis. Mehrbedarf kann gegen Berechnung bei Rücksendung des Umbruchs bestellt werden. Vorbehalt aller Rechte Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, ins- besondere 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 gesetz- lichen Bestimmungen des Urheberrechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmun- gen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Ver- vielfältiger 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 in- ternal 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 CCC. Publisher: Urban & Fischer Verlag GmbH & Co. KG, Branch Office Jena, P. 0. Box No. 1005 37, D-07705 Jena, Germany. Phone ++49(0)36 41/6 26-3; Fax ++49(0)36 41/62 65 00; e-mail: journals@urbanfischer.de Type setting, printing and binding: druckhaus köthen GmbH Advertisement bookings: Enquiries should be directed to Urban & Fischer Verlag, Advertising Manager: Sabine Schröter, P.O. Box 1005 37, D-07705 Jena, Germany. Phone ++49(0)36 41/62 64.45, 9 Fax ++49(0)36 41/62 64 21 Advertising rates: The price list from January 1, 1999 is effective at present. Distribution and subscription agency: For customers in the German speaking countries: Please turn to your bookshop or to SFG-Servicecenter Fachverlage, Zeitschriftenvertrieb: Barbara Dressler, Löbdergraben 14a, D-07743 Jena, Germany. Phone ++49(0)36 41/62 64.44, Fax ++49(0)36 41/62 64.43; For customers in all other countries: Please contact your local bookshop or Stockton Press., c/o Debbie Floyd, The Sub- scription Dept., Houndmills, Basingstoke, Hants RG 21 6XS, UK. Phone ++44(0)12 56/32 92 42. Fax ++44/1256/812358. Subscribers in the USA can call toll free on ++1/800/ 747 3187. Terms of delivery (1999): 1 volume consisting of 6 issues. Subscription rate (1999): Price per volume: DM 478.00*; UK Sterling 192.00 Single issue price: DM 95.00*; UK Sterling 39.00 * Suggested list prices. Prices are subject to change without notice. Reduced membership rates are available on request. Please turn to our publ. house. All prices mentioned above are exclusive postage. USdollar prices are converted at the current exchange rate. We accept the following credit cards: VISA Euro-/Mastercard/American Express, (Please note card-No. and expiry date). Banking connections: For customers in the German speaking countries: Deutsche Bank AG Jena, Konto-No. 6 284 707, BLZ 820 700 00; Postbank Stuttgart, Konto-No. 92 372 704, BLZ 600 100 70. For costumers in the other countries: UK Post Office Giro Account No. 5 192 455. Full details must accompany the payment. Subsciption information: If not terminated the subscription will be effected until recalled. If no cancellation is given until October, 31st the delivery of the journal will be continued. Copyright: The articles published in this journal are protected by copyright. All rights are reserved. No part of the journal may be reproduced in any form without the written permission of the publisher. Authorization to photocopy items for internal or personal use of specific clients, is granted by Urban & Fischer, for libraries and other users registered with the Copyright Clearance Center (CCC) Transaction Reporting Service, provided that the base fee of $12.00 per copy is paid directly to CCC, 222 Rosewood Drive, Danvers, Ma 01923, USA, 0044-3468/99/064/ $ 12.00/0 This consent does not extend to copying for general distribution, for advertising or promotional purposes, for creating new collective works or for resale and other enquiries. In the cases, specific written permission must be obtained from the pub- lisher Urban & Fischer. Printed in Germany AR STOCKTON Printed on acid-free paper effective with vol. 61, no. 1, 1996. © 1999 Urban & Fischer Verlag New: For detailed journal information see our home page: http://www.urbanfischer.de/journals Abstracted/Indexed in Animal Breeding Abstracts; Current Contents Agriculture, Biology Environmental Sciences; Biological Abstracts; BIOSIS database; Current Advances in Ecological and Environmental Sciences; Dairy Science Abstracts; Elsevier BIOBASE/Cur- rent Awareness in Biological Sciences; Fisheries Review; Helminthological Abstracts; Index Veterinarius; South Pacific Periodicals Index; Veterinary Bulletin; Key Word Index to Wildlife Research; Wild Review (Fort Collins); Zoological Record ZEITSCHRIFT FÜR SÄUGETIERKUNDE INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Goymann, W.; Leippert, D.; Hofer, H.: Parturition, parental behaviour, and pup development in Indian false vampire bats, Megaderma lyra. - Geburt und Aspekte der Jungenaufzucht und -entwicklung bei Indischen Falschen Vampi- en. Mengdenmn Bea =... 2ncaneeenn anne ee ee en 321 Garde, J. M.; Escala, Carmen: Coats and moults of the water vole Arvicola sapidus Miller, 1908 (Rodentia, Arvicolinae) in southern Navarra (Spain). - Felle und Fellwechsel bei Arvicola sapidus Miller, 1908 (Rodentia, Arvicolinae), aus ne Senn Er ee 332 Smorkatcheva, Antonina V.: The social organization of the Mandarine vole, Lasiopodomys mandarinus, during the reproductive period. - Soziale Organisation der Mandarin-Wühlmaus (Lasiopodomys mandarinus) während der Re- een 344 Cervantes, F.A.; Ramirez-Silva, J. P.; Marin, Ariadna; Portales, GloriaL.: Allozyme variation of Cottontail rabbits (Sylvilagus) from Mexico. - Allozym-Variation von Wollschwanz-Kaninchen (Sylvilagus) aus Mexiko ..........2.......- 356 Cassinello, J.: Allosuckling behaviour in Ammotragus. — Fremdsaugen bei Ammotragus..........222s2222200000eesnnenennen 363 Wissenschaftliche Kurzmitteilungen £ Kock, D.; Künzel, T.: The maned rat, Lophiomys imhaussii Milne-Edwards, 1867, in Djibouti, NE-Africa (Mammalia: Rodentia: Lophiomyinae). - Die Mähnenratte, Lophiomys imhaussii Milne-Edwards, 1867, in Djibouti, NO- Afrika ennaln 18 DES IN elchie ee 371 Gigirey, A.; Rey, J. M.: Faecal analysis of the edible dormouse (Glis glis) in the northwest Iberian Peninsula. - Kot- analyse von Siebenschläfern (Glis glis) von der nordwestlichen iberischen Halbinsel............uur-.uuussennssenneenn- 376 Mitteilung der Gesellschaft u 380 a nes in 383 Zn ISSN 0044-3468 - Z. Säugetierkunde - 64(1999)6 - 5. 321-384 - Dezember 1999 URBAN & FISCHER a” GES: RS> = rs ZEITSCHRIFT FÜR #&: SÄUGETIERKUNDE OF MAMMALIAN BIOLOGY INTERNATIONAL JOURNAL Sg Herausgeber/Editor Deutsche Gesellschaft für Säugetierkunde Schriftleitung/Editorial Office D. Kruska, Kiel - P. Langer, Giessen Wissenschaftlicher Beirat/Advisory Board W. Fiedler, Wien - S. Halle, Jena - G. B. Hartl, Kiel - R. Hutterer, Bonn - H.-G. Klös, Berlin - E. Kulzer, Tübingen - P. Lüps, Bern - W. Maier, Tübingen - O. Anne E. Rasa, Bonn -H. Reichstein, Kiel - M. Röhrs, Hannover - H. Schliemann, Hamburg - G. Storch, Frankfurt Deutsche Gesellschaft für Säugetierkunde Altvorsitzende/Living Past Presidents D. Starck, Frankfurt (1957-1961, 1967-1971) — H. Frick, München (1972-1976) -— M. Röhrs, Hannover (1977-1981) — H.-J. Kuhn, Göttingen (1982-1986) - E. Kulzer, Tübingen (1987-1991) — U. Schmidt, Bonn (1992-1996) Amntierender Vorstand/Managing Committee Vorsitzender/President: H. G. Erkert, Tübingen Mitglieder/Board Members: H. Frädrich, Berlin -— R. Hutterer, Bonn -— D. Kruska, Kiel — Marialuise Kühnrich, Hamburg - R. Schröpfer, Osnabrück - Angela Wöhrmann-Repenning, Kassel Z. Säugetierkunde 64 (1999) 321-331 FÜR © 1999 Urban & Fischer Verlag SÄUG EI ERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Parturition, parental behaviour, and pup development in Indian false vampire bats, Megaderma Iyra By W. GoyMAnNnN, D. LEIPPERT, and H. HOFER Max-Planck Institut für Verhaltensphysiologie, Seewiesen, Zoologisches Institut der Universität München, München, and Institut für Zoo- und Wildtierforschung, Berlin N die 30. 03. 1999 ging from 0.53 mm/d to 1.35 mm/d. Females started leaving pups behind (‘park’) either in the day roost or in special night roosts when pups were between 1 and 23 days old. Possible reasons for both the variability in growth and the onset of leaving pups on their own are discussed. Contact calls in this species consisted of repeated squeaking sounds and were not only emitted by mothers and pups, but also by other false vampires. They occurred most frequently in the morning and evening, during bursts of major activity in the colony. Key words: Megaderma lyra, parturition, postnatal growth, night roosts, contact calls Introduction Parturition has been observed only in few bat species (Wimsatt 1960; Kunz et al. 1994), including one report on captive Indian false vampires (Megaderma Iyra; GOPALAKRISHNA et al. 1976). To our knowledge, this is the first description on parturition in free-ranging Indian false vampires. Postnatal growth is an important life-history trait in mammals and represents an im- portant index of maternal input (ÖFTEDAL and GITTLEMAN 1989). Differences in this early development may affect fitness, such as offspring survival and future reproductive success (e.g. CLuTron-Brock 1991; HorFER and East 1996). Growth is influenced by environmen- tal variables such as food, temperature, maternal status, and condition. Young bats remain nutritionally dependent on their mothers for a prolonged period of time and female bats provide their young with milk until they achieve at least 90% of adult wing dimension and 70% of adult body mass (Kunz and STERN 1995). Hence, maternal input (Evans 1990) should play a crucial role in patterns of postnatal growth before pups achieve flight (BARcLAY 1994). We were interested to find out whether there are individual differences in postnatal growth and maternal attendance during early development of Indian false vampires. 0044-3468/99/64/06 - 321 $ 12.00/0 322 W. GoyMAnN etal. Materials and methods Caves are natural roosts of Indian false vampires but this species exploits also man-made structures (BrosseErt 1962; AuDer et al. 1991; BALASINGH et al. 1994; MARIMUTHU et al. 1995). We investigated a temple-dwelling colony of about 60 false vampires in a small village about 15 km south-east of Tiru- nelveli (Southern India). Observations were conducted at the beginning of the dry season (February- May 1995), when females are expected to give birth and rear their young (BALAsınGH et al. 1994). Maximum outside temperature was 39.1 +2.6°C (mean +sd; day, N = 38), minimum temperature was 25.9+1.4°C (night, N= 38). Average temperature inside the day roost was rather constant with 32.6 +0.9°C during day (N = 43) and 31.2 # 0.9 °C during night (N = 43). The day roost was continuously illuminated with dım red light (nine 15 W bulbs), which did not seem to disturb the false vampires, as bats often hung close (<20 cm) to the bulbs. False vampires were caught with mistnets during emergence from the roost and removed from the net immediately. Lactating females carrying a pup were released without further handling to avoid injuries to the pup. Other individuals were sexed and weighed. Forearm length was measured to the nearest millimetre. Seventy percent of caught individuals carried a collar and coloured plastic beads from a previous study (BALASINGH et al. 1992). The beads, however, were invisible in most of the cases and only a few bats could be identified from these tags. To improve individual identification we tagged 34 adults (-50% of the colony) with coloured wing bands (Museum Bonn, Germany, size E). Males were tagged on the left and females on the right forearm. “Parked” pups were tagged by painting either their claws with nail elamel or their ears with non-toxic dye (Marabu). When a pup was hanging alone at a roost- ing place we measured the rıght forearm with a pair of vernier callipers while standing on a chair be- side the pup. We used the NONLIN function implemented in SYSTAT 5.0 (WiLkınson et al. 1992) to derive an equation for the forearm growth of pups. The equation was calculated combining data from 2 captive (colony in Munich, Germany) and 5 free-ranging false vampires of known age. Total observation time in the day roost comprised 335 hours. Four other buildings in the vicinity of the temple, where false vampires perched during the night (night roosts) were observed during 105 hours. Night roosts were inspected daily to record whether any pup was ‘parked’ inside the roost. Observations in night roosts were done mainly with a night vision scope supported by stationary red- light torches. Pup vocalisations were recorded with an electrostatic microphone (Petterson), amplified with an Ultrasonic Detector (Petterson D940), and stored into a Racal Store 4 DS tape recorder with a tape speed of 30 in/s. The recordings were digitised and analysed using a custom-made sound analysis soft- ware (Sona, M. Knipschild, Dortmund). Statistical analyses were conducted with SYSTAT 5.0 (WiLkınson et al. 1992), following the proce- dures recommended by SokAL and RoHLr (1995) and LAMPRECHT (1992). The significance level was set as «= 0.05 (two-tailed). Data are presented as mean +SD or, when skewed, as median/interquar- tile range. Results Parturitions Female false vampires gave birth to a single pup. The first pup was born on March 17'*, the last one on April 18'®. After this date no more pregnant females were observed. The total number of pups was about 20. Five females were observed during parturition and another female was seen eating the placenta immediately after parturition. All of these observations occurred between 11.05 h and 12.39 h local time, suggesting that the time around noon could be preferred for parturition (binomial test, N=6,x=(0,p< 0.05). During parturition females generally produced a hollow space between abdomen and wings by slightly spreading the wings around the abdomen. One female did not cover her abdomen and we were able to observe the emergence of the pup. At 11.05 h we saw the head and upper breast of the pup emerging out of the vagina. The mother bowed forward and licked the young, especially its head and flanks. At 11.08 h the pup emerged from the vagina. However, its left forearm still stuck to the vagina. The pup revolved round its Parturition, parental behaviour, and pup development in Megaderma Iyra 928 forearm and thus got into the position to clasp to its mother in carrying position. Its head was now close to the inguinal nipples and its legs close to the breast of its mother. At 11.16 h the mother stopped licking the pup. Twenty minutes later she started licking again, especially the pup’s left flank. She repeatedly pushed her pup’s left flank. At 11.38 h the pup’s forearm suddenly emerged from the vagina, while the mother was still pushing the pup. The pup slipped slightly downwards and made a squeaking sound. It clasped the neck of the mother with its feet and the mother started to lick the pup’s left forearm and, by pushing the pup to one side, her own belly intensively. At 11.45 h the pla- centa began to emerge, while the mother was still licking her belly. Then she slightly wrapped her wings around her pup. One minute later the mother again began to lick her belly and the pup. Now the placenta left the vagina completely and hung down the umbil- ical cord. At 11.51 h the mother started to feed on the placenta, which took her 4 minutes. Then she again licked the pup and her own genitals. At 12.05 h the mother wrapped her wings around the pup and rested. After parturition all mothers licked their pup and their own genital region intensively, the latter may be a support to discharge the placenta. Mothers ate the placenta almost immediately after it had been discharged (N = 4, latency of feeding: 1.5/1-2 min, duration of feeding: 3.5/3-4 min). False vampire pups cling to their mother in the ‘carrying-position’, venter to venter, the head pointing towards the anus of the mother. With its mouth the pup sucks at one of the two inguinal nipples. These nipples are special attaching warts, no functional teats. The pup wraps its legs around the neck of the mother, crossing them behind her shoulders. The wings are folded at the mother’s flank. When false vampire pups tried to get into this carrying position after birth their mothers assisted them in different ways. Two bats used their muzzles to guide their pups to teats and inguinal nipples. Two other bats made stretching movements while their pups tried to get into the carrying position after birth. Three different pups struggled 10, 20, and 45 minutes until they succeeded in reaching the carrying position. It seemed to be dif- ficult and laborious for the pup, which is delivered head-first to revolve around itself and reach the carrying position. In the carrying position pups frequently lost the sucking grip to an inguinal nipple and the heads suddenly emerged out between the wings of the mothers. The mothers then pushed their pup back with their muzzle. Larger pups fre- quently hung pendant-like, grasping only the neck of their mother. In this position they stretched their wings, groomed themselves, and did wıng flapping. Sex ratio and growth rates The sex ratio of identified pups was balanced (6 males and 6 females). The sex of two other pups was not determined. To derive a model for the forearm growth of false vam- pire pups we combined data of 5 pups (3 females, 2 sex unknown) of known age with data of 2 female pups from a captive colony in Munich. We computed a logistic equation (fore- arm length = 64.382 [e%ase+0.97) , 11 ) which appropriately fitted to the data (cor- rected r° = 0.97, Fig. 1). The mean birth forearm length calculated with this equation was 34 mm. During the first 22 days of development the slope of the logistic equation almost equalled the slope of a linear curve (Fig. 1). We took data from 6 individuals for which appropriate data were available for this period and calculated their individual growth rates. Individuals showed considerable variation, ranging from 0.53 mm/d to 1.35 mm/d (Fig. 2). Two pups parked in night roosts (Fig. 2, individuals no. 2 and 3) had relatively low growth rates, but the sample size does not allow a meaningful statistical comparison. 324 W. GoYMAnN etal. 70 65 60 € e55 S 5 50 = D® 45 S 40 35 30 0 10 20 30 40 50 60 70 80 age (days) Fig. 1. Logistic growth model (forearm length = 64.382 [e"aee+0.697) | 1)-1), estimated using data of 2 captive (white symbols) and 5 free-ranging false vampires (black symbols). During days 1 to 22 growth almost follows a linear curve (forearm length = 1.008 age + 34.455). ‘Parking’ pups Females carried their pups during flight even when the young had almost reached adult size (own observations; BrossEr 1962). However, mothers also ‘parked’ their pup during night-time either in so-called night roosts (N =4) or in the day roost (N =4). Mothers leaving their pup in night roosts came to these roosts between 18.50h and 19.10h (N =4), probably immediately after they had left the day roost. They chose a roosting place and started to ‘park’ their pup. This procedure was complicated and took some time. First mothers landed at the ceiling and hung with the pup in carrying position. Then they made rhythmic and jerky movements with their muzzles towards the pup’s rump (muzzle-pushing’) and bowed themselves ventrally towards the ceiling, probably to give the pup the opportunity to grip the roost’s surface with its legs. During these bowing movements mothers spread their forearms. They had to bow up to 12 times until the pup clung to the ceiling. This behaviour included several bouts of jerky muzzle-pushing and Parturition, parental behaviour, and pup development in Megaderma Iyra 329 1.6 @& 1.4 12 1.0 6 = > 0.8 mean linear forearm growth (mm/d) (67) 0.6 Oel mu me oa or Rue Tine cn Bar al en in ae (0) 1 2 &) 4 5 6 model individual no. Fig. 2. Linear growth rates (+SD) of 6 individuals until day 22 compared with the linear proportion of the logistic growth curve from figure 1 (numbers refer to sample size on which the respective calcula- tion is based). bowing. The separation procedure lasted from 5 to 21 minutes (N =4, median 9.75/8 -10.5 min). Immediately after the pup had attached to the ceiling the mothers left the roost. Pups were left alone for 23% of the time, whereas 77% of the remaining time the mother joined it and the two hung together (Tab. 1). Mothers stayed in a night roost at least a minute and maximum 202 min (N =4) and pups were on their own for at least 3 and maximum 170 min (N =4). Detailed data were available for one mother in night roost 2: her median attendance interval was 83.5/18.5-151.5 min (N = 12) and her median absence interval was 39.5/30-59 min (N = 10). Age at ‘parking’ At what age did mothers ‘park’ their pups? If exact pup age was not known (5 cases) we estimated the age using the logistic forearm growth model. For three pups the exact age was known. Age of pups ‘parked’ for the first time inside the temple varied from 2 to 326 W. GoyManN etal. Table 1. Time (in minutes) mothers spent at and apart from the roosting site of their young (NR = night roost) total time mother present mother absent 173 (71%) 71. (29%) 1682 (77%) 514 (23%) 90 (100%) 0 (0%) 104 (81%) 25 (19%) 2.049 (77%) 610 (23%) Table 2. Forearm length (in mm) and age (in days) of pups when they were ‘parked’ for the first time (NR = night roost, DR = day roost, F = female pup, M = male pup, * = age calculated using logistic growth equation, see figure 1) location pup forearm length (mm) age (d) NR1 15 51 bl NR2 M3 45 10* NR 3 M4 56 232 NR4 jan - 16 19 days, median age was 10.5/2.5-18 days (N =4, Tab. 2). Pups ‘parked’ for the first time in night roosts were 10-23 days old, their median age was 13.5/10.5-19.5 days (N=4, Tab. 2). However, the age difference of pups ‘parked’ in temple and night roosts was not significant (Mann-Whitney U-test, U = 10, p = 0.564). First flight Twenty-seven days after the first birth had occurred, the first pup was fluttering inside the temple. From then on the number of flying pups increased to at least 4 until the end of the study period. However, so far no pup seemed to leave the day roost. One pup made its first flight attempts when its forearm was 58 mm long. Two days later its forearm length was 60 mm. This was the last time the pup could be measured, since its increasing flying ability made further measurements impossible. In contrast, another pup hanging in night roost 4 did not make any attempts to fly when its forearm was 62 mm long. A third pup in night roost 3 could be recorded until its forearm measured 58mm. The two captive bats from the laboratory colony (Munich, Germany) started to fly when their forearms were 60.5 mm and 62.6 mm long. Their respective ages were 35 and 30 days. Contact calls Contact calls of false vampires consisted of repeated squeaking sounds with a peak fre- quency of about 12 kHz (Fig. 3). A total of 107 contact calls was counted between Feb- ruary 9'® and April 14 (Fig. 4). Then contact calls occurred so frequently that we stopped 327 sdnd Jo ‘ou anıyeinwund 31.3. Parturition, parental behaviour, and pup development in Megaderma Iyra distributed throughout the day. A comparison of " and April 14" and the frequency of contact 1128, 22: 1:8) time (date) and hence 54 of all recorded contact calls (50.5%) occurred before a single pup was born counting them. Interestingly contact calls could be heard from the very onset of the study (Fig. 4). Contact calls were not equally total observation time between February 9' LO (es) [f9) oO eat aA A = ” to) (es) - = ku. ae VS Een nn re Jr re Tr Qu em = | SE | I AIISIISISIIIIIIIUUUUUIIIIIIIIIIIIIIIIIIIIIIIIIIIIIUIIIIIUIIIIIIUIUIUIIII > IIIÜÜÜÜÜÜÜIÜIÜÜIÜÜIUÜUIUUÜUUUÜUUUUUUUUÜUUÜÜUUÜUÜUÜÜÜÜUUUUUÜUUÜÜÜUUÜÜÜUUÜUUUUUUÜUUNS > ® IIIIIIIÜÜIÜIIIÜÜIÜIIIIIIIÜUÜUIÜIÜIÜÜÜÜÜÜUÜIÜIÜÜUÜUÜUUÜÜUUÜUÜÜÜUUUUUUUUUUUÜUUUUUUÜUN [(®) 77) SIT Y DS im) je [a»} ei #2 oO "Oo A ! 63 gen ° IISISISSSIUIIIIIIIIIIIIIIIIIIIIIIIIY = DISSSSSIIIIIIIIIIIIIIIIIIIIIIIIIIIISY {ae} IISSSISISSIIIIIIIIIIIIIIIIIIIIIIT oO IIIIIIIIIIIIIIIIIIIIIII =) IISISIIIIIIIIIIIIIIIIN o III S Ss GES (o) DO 9 — RS oO 8 mn -_ oO {ss} ee) gen - (@) TEN Be) m > „ ja) &) a a | un [77 > & (eb) [97] = 8 = fe} [>] 7 No) [s>) — ; [s»} - al je10) n ei = - -_ [®) | WM ersceguseg Tb; Su S Ren # ai oO le) oO oO (ee) eo) ® 5 = 2 = S : N >) oo (de) 9 < 0.0122 —p< 0.001. Ist 2nd 3rd irregular 11.9-15.9 11.9-18.9 14.6-20.5 15.4-20.8 WEIGHT Ist ; 125.3-198.2 2nd 144.7-248.2 Ird 159.3-260.0 irregular 147.8-207.1 Ist 2nd 3rd irregular lst 2nd 3rd 172-197 179-198 170-202 169-197 36.1-38.5 36.8-39.9 37.8-40.1 irregular 37.0-40.5 moulting phases were statistically significant in most cases. This was also the case with the other parameters of the third phase of the first moult and the first phase of the sec- ond moult. Nevertheless, the differences between the second coat specimens and those still undergoing the first phase of the second moult were not quite as significant, which re- inforces the hypothesis suggested earlier concerning the brief duration of the second coat. Specimens with irregular moults displayed intermediate biometric values (Tab. 3) be- tween the second and third phase of the moult and no significant differences were evi- dent. These data support the aforementioned theory of a possible similarıty between these designs and those typical for the phases. Among those specimens undergoing the second moult, and captured from autumn to the beginning of winter, 22% of the males and 30 % of the females were immature. This attests to the delay of the onset of sexual maturity for anımals born at the end of the re- productive period (GARDE and EscALA 1996). A. sapidus from southern Navarra reaches sexual maturity mainly during the second moult, like the water vole from the Ebro River Delta (VENTURA and GosALBEZ 1990) and several other species of rodents (MoREL 1981; Espana et al. 1985; SANS-CoMA et al. 1987). Most of the animals which were in the process of a second moult were captured be- tween the end of the summer or in mid-winter, coinciding with the capture of those ani- mals from classes II and III (GArDE and EscArLA 1996 a). Hence one might believe that the second moult also is mainly determined by age. Adult coats Once the second moult has finished, the animals acquire a third coat which is the first adult coat. Later on the specimens will go through a series of moults, influenced by envi- ronmental conditions, sex and age, which will lead to the acquisition of the adult coats. 340 J. M. GARDE and CARMEN ESCALA 56 specimens (22.8 % of the adults), with totally formed adult coats and without mela- nin prints, were studied. This small amount suggests that little time elapses between the two consecutive moults. Although it was not possible to distinguish between the adult coats themselves, two types of coats associated with the seasons were differentiated; one in the winter, which is thick and made up of much longer hair, and one in the summer. Such observations have been made by VENTURA and GoSALBEZ (1990) in A. sapidus from the Ebro River Delta. The pigmentation of the adult coat is lighter than that of the second coat, a character- istic which becomes more noticeable with age. In some older individuals, yellowish mark- ings are displayed which vary from one individual to another (GARDE 1992). The 56 sample specimens all belonged to age classes IV and V (Tab. 1), which confirms the existence of several coats following the first adult coat. All the specimens with third or later coats, wıth no evident signs of moult were sexually mature, except for two males corresponding to age class IV, captured in February and in December (GARDE and EscaA- A 1996b). The disproportion that existed between the males (16 specimens-11.6 % of the adult males) and the females (40 specimens-37.4 % of the adult females) was striking. These data suggest that the onset of a new moult in females might be influenced by phy- siological processes connected with reproduction. The distribution of adult specimens with the third coat throughout the year reaches a maximum in March (35%) and in September (36 %). These months coincide with the end of winter and summer and are periods with lowest moulting activity. However, it is likely that this monthly sequence is also influenced by the sexual activity of the female as previously mentioned. Adult moults In the analysed sample of the 245 specimens with totally formed adult coats, 189 (77.2 %) showed signs of moulting. All were sexually active, except 9 (6 males and 3 females), and were captured between November and February (GARDE and EscALA 1996b). Moulting adult specimens could be found every month of the year (64-87 %). The monthly frequencies of moult were higher between October and January and lower in February and March. The higher frequencies seem to correspond to a possible autum moult. The increase in percentage of moult beginning in April could be due to the com- mencement of spring moult and to the beginning of mating activity (GARDE and ESCALA 1996 b). The identification of two coat types in the adult specimens, one in winter and one in summer, suggests an autumn and spring moult. These two moults allow the animals to ac- quire the above-mentioned coats. Some authors (SAINT-GIRONS 1967; MorREL 1981) refer to such changes in adult coats as “seasonal moults”, owing to their relationship with the environment. Other authors, e.g. Espana et al. (1985) for M. spretus and SANS-CoMA et al. (1987) for R. rattus insist on there being no relationship between the changes of the adult coat and the time of year; both cases were observed in southern Spain. Only 20 % of the specimens (20 males and 18 females) showed uniform designs. The majority (28) was captured between September and January and all belonged to age class IV, which suggests that the first adult moults maintain a certain uniformity but, show variations with time. This group of young adults displaying a regular moult undoubtedly corresponded to samples that had just reached adult age at that time of the year and were going through the autumn moult. In fact, several animals that were going through the third phase of the second moult displayed overlying melanin prints typical for the third moult (Fig. 3: i-I), and thus corresponded to the autumn moult. As has already been stated, this moult nor- Coats and moults of Arvicola sapidus 341 mally takes place between September and January. Most of its sequence was established (Fig. 3: a-h), except for the final phases, either due to the lack of specimens displaying those markings or because the final reabsorption of pigment was irregular. The autumn moult started from the lateral glands. The remainder of the process was very similar to the subadult moult (Fig. 2), although its development was somewhat more variable and ir- regular. The characteristics of the autumn moult for A. sapidus coincide with MoRrEL’s (1981) observations in A. terrestris. It was impossible to determine the sequence of the spring moult since all the speci- mens captured during this season displayed irregular topographies. It is therefore prob- able that its development is more irregular, unlike the autumn moult, as noted by VEN- TURA (1988) for A. terrestris. The majority (80 %; n = 151) of specimens with adult moults displayed irregular mark- ings. Nevertheless, it should be pointed out that in several specimens there was a varia- tion in pigmentation as well as in the way the hairs were renewed in the lateral gland re- gion. This fact, which was observed by QuaAy (1968), has also been confirmed by VENTURA and GosSALBEZ (1990) in A. sapidus from the Ebro River Delta. In some females a similar variation was also observed near the mammary glands (Fig. 3: c, p), which also was men- tioned by VENTURA (1988) in A. terrestris. The appearance of small melanin prints distributed as a mosaic pattern over the whole reverse coat (Fig. 3: m-n) was also observed in specimens from age class V. These “scat- tered” designs are a clear sign of the absolute irregularity that characterises coat changes in older individuals and observed by More (1981) in A. terrestris. The existence of irregular designs of large melanin spots (Fig. 3: o-p) was also re- corded. This widespread pigmentation is a sign of a fast moulting process. 21 specimens, 3 males and 18females (12 of which were in gestation) displayed these characteristics. Most of the specimens (12) which displayed this design, did so between May and August, the months of greatest sexual activity. If the female sexual activity was an influencing-fac- tor in the onset of a new moult, these results now seem to strengthen the idea that sexual activity also affects the adult moulting process in A. sapidus from southern Navarra. These observations coincide with those made by More (1981) in A. terrestris, who pointed out that during the reproductive period the females in gestation displayed signifi- cant delays in their moults, which were later completed at the end of the gestation period or slowed down and took place between two pregnancies. This was noticed by the appear- ance of simultaneous hair growth. Similar effects have been found in other rodents, for example, in A. sylvaticus (SAINT GIRONS 1967). Zusammenfassung Felle und Fellwechsel bei Arvicola sapidus Miller, 1908 (Rodentia, Arvicolinae), aus dem Süden von Navarra (Spanien) Es werden Angaben zu den Besonderheiten von Fell- und Haarwechsel einer Population von Arvicola sapidus (n = 363) aus dem Süden von Navarra gemacht. Die Untersuchungen beziehen sich auf Pig- mentierungsflecke der Hautinnenseite. Die südwesteuropäische Schermaus zeigt ein 1. oder jugend- liches Haarkleid, das weniger dicht, kürzer und dunkler ist als die folgenden. Danach erfährt sie einen 1. oder juvenilen Haarwechsel, der in Zusammenhang mit dem Alter steht; er ist sehr regelmäßig und vom sublateralen Typ. Der größte Teil von in diesem Haarwechsel befindlichen Tieren ist sexuell un- reif (82-93 %). Danach bekommt die Schermaus ihr 2. oder subadultes Haarkleid, das dichter, länger und heller als das vorangehende ist. A. sapidus beginnt bald mit ihrem 2. oder subadulten Haarwechsel, der mit dem Alter korreliert und von sublateralem Typ ist. Im Unterschied zum 1. Haarwechsel verläuft er unregel- mäßiger. Während des Haarwechsels erreichen die Tiere allmählich ihre sexuelle Reife (78-70 %). 342 J. M. GARDE and CARMEN ESCALA Der subadulte Haarwechsel führt zum 3. Haarkleid, das länger und heller als das 2. Haarkleid ist. Diesem 3. Haarkleid entspricht die Gesamtheit der adulten Haarkleider, die ununterscheidbar sind; das Winterfell ist von größerer Länge und Haardichte als das Sommerfell. Nach der Bildung des adulten Haarkleids unterliegen die Tiere ständig einer Reihe von saisonalen Haarwechseln: Herbstlicher Haarwechsel, der ähnlich wie der subadulte Haarwechsel verläuft und auch eine gewisse Regelmäßigkeit zeigt, zumindest in den ersten Stufen, und der etwas weniger ausge- prägte Frühlingshaarwechsel, der Topographien von auffallend hoher Unregelmäßigkeit zeigt. Der Verlauf des Haarwechsels der adulten Tiere wird auch durch das Alter beeinflußt und im Falle der Weibchen durch die Fortpflanzungsprozesse. References BECKER, K. (1952): Haarwechselstudien an Wanderratten (Rattus norvegicus Erxl.). Biol. Zentralblatt 71, 626-640. CABRERA, A. (1914): Fauna Iberica: Mamiferos. Madrid: Mus. Nac. Ciencias Naturales. 441 pp. EspAaNa, M.; PALOMO, L. J.; ZAMORANO, E.; SANS-CoMA, V. (1985): Über Haarwechsel und Haarkleid von Mus spretus Lataste, 1883 aus Südspanien (Rodentia, Muridae). Spixiana 8, 1-16. GARDE, J.M. (1992): Biologia de la rata de agua Arvicola sapidus Miller, 1908 (Rodentia, Arvicolidae) en el Sur de Navarra (Espaha). Tesis Doctoral, Universidad de Navarra. GARDE, J. M.; EscALA, C. (1996 a): Estructura poblacional de Arvicola sapidus Miller, 1908 (Rodentia, Arvicolidae) en el Sur de Navarra (Espana). Real Sociedad Espanola Historia Natural 92, 189-194. GARDE, J. M.; EscALA, C. (1996b): Reproductive cycle of Arvicola sapidus (Rodentia, Arvicolidae) from southern Navarre, Spain. Acta Theriol. 41, 353-365. GARDE, J. M.; ESCALA, C.; VENTURA, J. (1993): Determinaciön de la edad relativa en la rata de agua me- rıdional, Arvicola sapidus Miller, 1908 (Rodentia, Arvicolidae). Donana, Acta Vert. 20, 266-276. GOSALBEZ, J. (1982): Contribuciön al conocimiento de los roedores de la Peninsula Iberica y su interes biolögico. Tesis Doctoral, Universidad de Barcelona. KAHMANN, H.; TIEFENBACHER, L. (1970): Über Haarwechsel und Haarkleid des Gartenschläfers Eliomys quercinus Linnaeus, 1776. Z. Säugetierkunde 35, 89-103. Le LoUARN, H.; SAINT-GIRONS, M. C. (1977): Les Rongeurs de la France. Faunistique et biologie. Paris: Inst. Nat. Recherche Agronomique. MARTIN, R. L. (1973): Molting in the rock vole, Microtus chrotorrhinus. Mammalıa 37, 342-352. MILLER, G. S. (1912): Catalogue of the Mammals of Western Europe. London: Brit. Mus. (Nat. Hist.). MOREL, J. (1981): Le campagnol terrestre, Arvicola terrestris (L.), en Suisse: biologie et syst&matique (Mammalia, Rodentia). Tesis Doctoral, Universite de Lausanne. PALOMO, L. J.; VARGAS, J. M. (1988): Deroulement topographique et temporel des mues regulieres de la souris a queue courte Mus spretus Lataste, 1883. Mammalıa 52, 75-83. Quvay, W. B. (1968): The specialized posterolateral sebaceous glandular regions in Microtine Rodents. J. Mammalogy 49, 427-445. REICHSTEIN, H. (1982): Gattung Arvicola Lac£pede, 1799. In: Handbuch der Säugetiere Europas. Band 2/1. Nagetiere 2. Ed. by J. NIETHAMMER und F. Krapp. Wiesbaden: Akad. Verlagsges. Pp. 209- 252. RoDe, P.; DiDIER, R. (1946): Les Mammiferes de France. Paris: N. Boube&e. SAINT-GIRONS, M. C. (1967): Etude du Genre Apodemus Kaup, 1829 en France. Mammalia 30, 547-600. SAINT-GIRONS, M.C. (1973): Les mammiferes de France et du Benelux. Faune marine exceptee. Paris: Doin. SANS-CoMA, V.; ROSADO, L. M.; GosSÄLBEZ, J. (1987): Un estudio de la morfometria y del comienzo de la actividad reproductora de Apodemus sylvaticus (L., 1758) en la Peninsula Iberica basado en la con- sideraciön de los pelajes y las mudas como indicadores de la edad. In: Mamiferos y Helmintos. Ed. by V. Sans-CoMA, S. Mas-CoMAa and J. GoSALBEZ. Barcelona: Ketres. Pp. 89-98. SANS-CoMA, V.; ZAMORANG, E.; VARGAS, J. M.; ANTÜNEZ, A. (1987): Über den Haarwechsel freilebender Hausratten, Rattus rattus (L., 1758), in Südspanien. Zool. Abh. Mus. Tierkd. Dresden 42, 125-136. STEIN, G. H. W. (1960): Zum Haarwechsel der Feldmaus, Microtus arvalis (Pallas, 1778), und weiterer Muroidea. Acta Theriol. 4, 27-43. VENTURA, J. (1988): Contribuciön al conocimiento del genero Arvicola Lac£pede, 1799, en el nordeste de la Peninsula Ib£rica. Tesis Doctoral, Universidad de Barcelona. Coats and moults of Arvicola sapidus 343 VENTURA, J. (1992): Coats and moults in Arvicola terrestris from the northeast of the Iberian Peninsula. (Mammalia, Rodentia: Arvicolidae). Zool. Abh. Mus. Tierkd. Dresden 47, 95-110. VENTURA, J.; GOSALBEZ, J. (1990): Caracteristicas de los pelajes y las mudas en Arvicola sapidus (Roden- tia, Arvicolidae). Donana, Acta Vert. 17, 3-15. ZABALA, J. (1983): Primera cita de Arvicola sapidus (Miller, 1908, Mammalia, Rodentia) para la provin- cia de Guipüzcoa. Munibe 35, 109-114. Authors’ address: JUAN MANUEL GARDE and CARMEN ESCALA, Department of Zoology and Ecology, University of Navarra, P. O. Box 177, E-31080 Pamplona, Spain ARE: Z. Säugetierkunde 64 (1999) 344-355 ZEITSCHRIFT ®=# FÜR © 1999 Urban & Fischer Verlag SAUG ETl ERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY The social organization of the Mandarine vole, Lasiopodomys mandarinus, during the reproductive period By ANTONINA V. SMORKATCHEVA Department of Vertebrate Zoology, St. Petersburg state University, St. Petersburg, Russia Receipt of Ms. 20. 04. 1999 Acceptance of Ms. 14. 07. 1999 Abstract The social system in a free-living population of the mandarine vole Lasiopodomys mandarinus was ex- amined in Selenginski District, Buryatia, by use of the mark-capture method. Mandarine voles lived in extended family groups. The members of the group occupied a common burrow and were strongly attached to it. The summer groups consisted of one breeding male,1-5 breeding females, and young of 1-3 generations with a mean of 8.7 (range 3-22) individuals per burrow. Most of the offspring re- mained within the natal territory at least up to 50 days. None of the 72 young males and only three of the 73 young females became reproductive while staying in natal burrows. The change of sire appears to be the necessary condition for the reproductive activation of philopatric daughters. Thus, L. mandarinus exhibits a high level of sociality based on communal breeding, prolonged pair-bonding and parent-young relationships. This set of traits is also reported in the literature for La- siopodomys brandti. It is suggested that sociality was characteristic of ancestral above ground form and represented the precondition to occupy the recent niche of the fossorial stenophage. Key words: Lasiopodomys mandarinus, mating system, philopatry, territoriality, subterra- nean rodents Introduction The mandarine vole Lasiopodomys mandarinus inhabits the grasslands of China, Korea, North Mongolia, and the borderland in the south of the Transbaical region of Russia. Very little is known about its habits. Meanwhile, scanty information from the works of former authors (Ferıssov 1955; Hyan-VAan-Dı 1960) and especially the recent studies in the Mongolia and Transbaical region (DMiTRiEvV 1980; SMORKATCHEVA et al. 1990) charac- terise Lasiopodomys mandarinus as an extremely interesting aberrant form. Apparently, throughout its range L. mandarinus leads an almost completely subterra- nean existence. Its burrows are extensive: the number of entrances is up to 50-70 and the tunnels are up to 95 m long. Most of the tunnels are close to the surface and serve to for- age underground for roots and tubers. In Mongolia and Buryatia the mandarine vole feeds almost entirely on Stellera chamaejasme using both its massive roots and green parts (DmMitRIEvV 1980; SMORKATCHEVA et al. 1990). Even when foraging for stems and leaves of Stellera, voles rarely go away more than 0.5-1 m from a burrow entrance which usually is near this plant (SMORKATCHENVA et al. 1990). The reproductive period lasts in the Transbaical region at least from early April through late August. Winter breeding is likely although there is an autumn break (SMoRr- KATCHEVA et al. 1990; SMORKATCHEVA 1993). Unlike most voles of grasslands — e.g. Lasio- 0044-3468/99/64/06 — 344 $ 12.00/0 The social organization of the Mandarine vole 345 podomys brandti, Microtus socialis, M. arvalis, M. gregalis, M. ochrogaster (SVIRIDENKO 1934; KHRUSTZELEVSKI 1954b; TAıtt and KreBs 1985; GETZ et al. 1987) — L. mandarinus does not display great outbreaks of numbers. Although it seems to undergo cyclic density fluctuations, their amplitude is comparatively low and the species is never very abundant. Even in optimal habitats the maximal density of burrows does not exceed 5-7 per hectare (DaiItkıEv 1980). The reproductive potential of the mandarine vole is low compared with most other microtines. Mean litter size determined by the number of scars in the uterus was 3.65 (DMiTkıEv 1980); the number of embryos in nature averaged 4.4; mean number of newborn in the laboratory was 3.3 (ZORENKO et al. 1994). Gestation is 22-24 days. Pups open their eyes only after 13-16 days and wean around 18-22 days. Females become sexu- ally mature at about 38-45 days, males at 55-60 days (ZoRENKO et al. 1994). Thus, this species demonstrates the set of characters associated with K-selection (Mc ARTHUR and Wiırson 1967). Some data on the spatial organisation and group composition obtained during two field seasons were briefly reported earlier (SMORKATCHENVA et al. 1990). The aim of this study is to summarise the information on the mandarine vole social system based on data collected throughout 1986-1993. Material and methods Study area This work was carried out in Selenginski District of Buryatia near Lake Torm, 16 km SWW Selendu- ma (50°53’N, 106°01’E). Data are presented for the periods June-July 1986; August-mid September 1990; mid April-late May 1991; June-July 1992, and early August-early September 1993. Additionally, some information on spacing and burrow dynamics was obtained in late September 1986 and in late July 1990. At the beginning of each trapping period one of four areas (8-16 ha) within the gentle slopes was selected. We had to change trapping areas because of asynchronous decreases in number of different local pop- ulations separated by steep rocky slopes or agricultural fields. When the density of inhabited burrows was low (0.1-0.2 per ha), it was practically impossible to obtain sufficient amount of data. Thus, the local population with the highest density of burrows was selected. Microrelief and vegetation of these four areas are similar. The steppe community is dominated by Festuca lenensis with dispersed Stellera chamaejasme, Artemisia frigida, Potentilla acaulis, P. tanacetifolia, Leontopodium sp., Thymus serpil- lum, Arenaria capillaris, Lilium tenuifolium, Youngia tenuifolia, Veronica incana, Rumex acetosella, Astragalus sp., Oxytropys sp., etc. Everywhere more or lesser overgrazing results in vegetation IMpoVv- erishment combined with increase of Stellera chamaejasme productivity (GORSHKoVA et al. 1977). In the areas under study its density varied from about 50 up to 150-200 individuals per 100 m’. Field studies All burrows within the chosen area were marked by stakes. Thereafter, the area was inspected daily (early in the morning or after rain) to reveal fresh mounds and plugs of soil. Every fresh mound was plotted (1:75 or 1:150). In this way the data on dynamics for at least 30 days were obtained for 29 burrows. In addition, four burrows were monitored from the very first mounds and were plotted and measured at different stages of development. All burrows in the area were live-trapped. The traps were slightly buried at the fresh mounded burrow entrances, from two to ten per one burrow and were moved according to the movements of the fresh mounds. Slices of stale bread with sunflower oil were used as bait. The live-traps were checked at 3-4 h intervals from 6a.m. to 11 p.m. We closed traps during very cold nights in autumn and in spring and during very hot afternoons in summer. When first captured, animals were marked by toe-clippings. At each capture voles were weighed, sexed, notes were taken on molt and sexual condition (vaginal smears were taken in females with open vulva). Most of the animals were released immediately. In the case of death of voles in live-traps they were autopsied in order to obtain additional information on the reproductive organs. 346 ANTONINA V. SMORKATCHEVA Data analysis Those voles that weighed more than 26 g and/or had adult fur were classified as adults, the remainder as young. Scrotal males and pregnant, nursing, or perforate females were classified as reproductive, the others considered to be nonreproductive. The age of young was determined by the moult develop- ment and weight. Since the composition of groups did not change significantly through a trapping period, our esti- mations of group size refer to a given session as a whole. Individuals that were not caught repeatedly were assumed to be residents of those group burrows where they were marked. In doing so it was re- cognised that for some anımals (especially subadult individuals) this assumption is not true, and there- fore the values of group size must be somewhat overestimated. When no unmarked voles were caught in a burrow for at least 10 days, it was presumed all residents older than 20 days to be marked and these burrows were considered as completely trapped. To verify this, 4 burrows were excavated at the end of the trapping session; of 9 animals caught by hand when excavating, 8 were marked. During spring (1991) and summer (1986 and 1992) sessions, most of the monitored burrows (22/27) were com- pletely trapped. During fall sessions all members of a family apparently were not marked in any of 16 live-trapped burrows, because after the vegetation of Stellera had finished, the trapability of manda- rıne voles decreased. Group territory sizes were determined on the basis of fresh mound plotting (25-48 points were obtained for each burrow) for a period of approximately one month. Demographic background At live-trapped areas the number of inhabited burrows per ha ranged from 0.7 up to 3.0, i.e. middle- high density in this species. Only during the spring session 1991, a low density (0.2 burrows per ha) population was under study. The mean number of adult females per adult male varied from 1 up to 5, with a mean value being 1.9. The sex ratio among young was near 1:1. Results As judged from the trapping data, mandarine voles live in extended family groups. The members of a group occupy the common burrow and are strongly attached to it. Of the 194 marked and repeatedly caught voles 92 % were caught at only one burrow. Among adults marked throughout the first week of each trapping session, 72% of males (13 of 18) and 64 % of females (19 of 28) were caught in the same burrow system up to the end of the given session. Mating system Of five burrows live-trapped during the spring session of April-May 1991 (the beginning of the reproductive period) three were inhabited by pairs of adults without offspring, one by one female with two young, and one by a single adult male. At the onset of the reproductive period both polygynous and monogamous groups were present in the population. Of a total of 17 groups completely caught during the sum- mer sessions, 7 (41%) contained one breeding female, 7 (41%) contained two breeding females, 2 (12%) contained four and 1 (6 %) contained five breeding females. The num- ber of breeding females averaged 2.4 (n= 9) in 1986 and 1.8 (n=8) in 1992. Each of two excavated burrows inhabited by several reproductive females had only one large nest. Thus, there is evidence of joint rearing of pups. From pooled data of 1986 and 1992, re- productive males were present in 17 burrows of 21 live-trapped and in 15 groups of 17 completely caught. In five burrows “additional” reproductive males were observed. In one case the female changed her mate after the former had died. In three burrows the “additional” males were caught only once and, apparently, were dispersers. Only in one The social organization of the Mandarine vole 347 case two adult males were present in the same burrow simultaneously for at least five days. Of 16 burrows live-trapped during the autumn sessions (pooled data for 1990 and 1993), in 2, 11, and 3 burrows two, one and no reproductive females were present, respec- tively. In 5 burrows one adult scrotal male was caught, 8 units did not contain scrotal males. Each of the remaining three groups included two scrotal males, one of which being an adult overwintered individual, and another being a young of this year. Parent-young relationships and natal dispersal During the spring trapping period only two young males were caught born approximately at the end of March; in mid-May they lived in the natal burrow with an adult female, ap- parently their mother. Already in June two cohorts of young were live-trapped in some of the burrows: indi- viduals born in late April-early May and those born in mid-late May. Simultaneously the newborn pups may be present in nests, as was shown by the subsequent trapping. Of a to- tal of 14 family groups completely trapped by July where young were present, 3 included at least three litters, 9 at least two litters and one at least one litter. For these groups, the mean number of the young known to be present was the same in 1986 and 1992 and con- sisted of 6.4 individuals. In August-September the young from the same group usually belonged to the same cohort born in July; the older young voles were caught in only three burrows of 16 live- trapped. The majority of juveniles marked during the first week of each trapping period at 20- 30 days of age were recorded staying at the natal site up to 50-60 days, and some of them even up to 70-80 days (Fig. 1). All young males (n = 72) and most young females (70 of 73) remained nonreproduc- tive while staying in their natal burrows. In late June-mid July 1992 three cases of repro- ductive activation in young philopatric females were registered: a pregnancy in a female 100% 80% 0 60% u A2 40% age, days Fig. 1. Proportions of marked juvenile mandarine voles known to remain at natal territory up to the ages of 40, 50, 60, and 70 days. 1-males (n = 30), 2-females (n = 22). 348 ANTONINA V. SMORKATCHEVA of about 60 days and the phase of late proestrous-estrous in two females of about 40 days. It is noteworthy that one of the estrous females and the pregnant female were from the same family group, where an old mature female (their mother?) was present and the change of the sire-male occurred. The composition of groups at the onset of the reproductive period allows to suppose dispersion of young males to be obligatory. Five males marked within natal territories were then captured outside at distances of 20-80 m. Two of them were found in the terri- tories of the adjacent groups, the others at single burrow entrances. They were at an age of 45-75 days when dispersing; all but one were registered as nonscrotal. None of them were caught later. The rather long distance of natal dispersal is likely in these cases be- cause all burrows within a radius of at least 100 m were live-trapped. Besides, 7 unmarked males at the age of 40-60 days were caught (each only once) near single burrow en- trances, outside permanent family burrows. These individuals also seem to be dispersers. Iwo to three voles of the same age were captured at the same entrance, suggesting that they could be littermates. Concerning the young females, two facts of their resettling in the adjacent burrows (distances 40 and 45 m) were recorded. These individuals were at least 70 days old when dispersing. Both were caught at the new sites as reproductive. Besides, five young females were initially marked in small recent underdeveloped burrows and were most likely among the founders of new breeding units. All dispersers except one male and one female were registered in late June-mid July. Size of groups In spring, at the very beginning of the mass breeding period, all families were small and included 2-3 individuals. According to the data of 1986 and 1992 for completely trapped burrows, in June-July the average social group contained 9.4 (n=9) and 8.0 (n = 8) individuals, respectively. By July 1986 the number in one of the families reached 22 individuals; afterwards this family and group burrow gradually divided into two. The former mature male remained in one part, the new unrelated male appeared in another. The formation of new families was ob- served in early July 1986 (2 groups) and mid-July 1992 (1 group). From the very begin- ning, each of them consisted of one male with two females. Among the founders were both young and adult individuals. During the late summer-fall the mean number of individuals caught per burrow was 3.47 and 3.72 in 1990 and 1993, respectively. Spacing and territorial relationships In L. mandarinus the density of inhabited burrows is comparatively low; the nearest ones usually are separated by a distance of several dozens of meters. Although this space is of- ten holed by the set of destroyed or left tunnels, it seems to prevent a direct contact be- tween the resident individuals from distant groups. Occasionally burrow-clusters were ob- served, several group territories being located close to each other (Fig. 2). In such cases burrow systems of different family groups were not connected by tunnels. Members of a certain group normally did not visit strange burrows. During 5 trapping sessions only 9 marked voles were observed to intrude. Among them, 4 individuals (2 young males and 2 young females) most likely were dispersers. Three.voles (one adult female and two adult males) lost their mates and were probably looking for new ones. The remaining two voles were a pair occupying an adjacent burrow, of which hosts had disappeared two weeks be- fore. The social organization of the Mandarine vole 349 4 ++ + + i + + a + + ++ $.t, Trr ® Bo serie + + + ® 2.0 MARS IEEN, °s 2 .2) 0900. !00 = 25 I ri A ++ es 3 + : ® +0 dt 9 ige, > +r ® 4 Pa + Ho . 0 + 21 +++ ”s : ee + Ht * 4 ++ ar 00 = .;0d + + 2 = f u Vet x n ++ Br tr Nas 2, rn ve + an 5 ++ 2 1 + rel [e®] + +. + oe, 9% at Bat De: oO + + \ ® .r E ® % S 0.009 + I; ® (e) © +® +ro O0 + Sie 00 + BO or I Soniztt a 00 o“ + „00 Oo is we +: No (6) art ©® o eo © [) . © ° IV ! ee (e) + eV + Ba et 10 m o ie) Fig. 2. Dynamics of five mandarine vole burrows (June-July, late September 1986) as revealed by soil mounds. Dashed lines indicate areas covered by old mounds in early June. Open circles, solid circles, and crosses indicate new mounds plotted in June, July, and September, respectively. 350 ANTONINA V. SMORKATCHEVA Group territory sizes and dynamics At the very beginning of the spring trapping session (mid-April 1991) most part of the area under study was covered with mounds of soil indicating that many burrows were in- habited in March. For 20 well-distinguishable burrows the mean size (+ SE) determined by the old mounds was 265 + 55.9 m’. However, the following observation revealed only 4 burrows where the fresh mounds had appeared; their areas determined by the mounds registered from 20 April up to 20 May were 110 m” (inhabited by a single female with young), 70, 15, and 25m” (each inhabited by a pair of adults). The fifth burrow was founded by a single male in early May and reached 25 m” by 20 May. During the summer sessions of 1986 and 1992, a total of 14 burrows were followed through at least one month, most of them from the first decade of June up to the first decade in July. For this period, the mean size (+SE) of the burrows consisted of 142 + 32.4 m” (range 50-370 m”, n=13, pooled data of 1986 and 1992). The fourteenth burrow was extremely large; by July it reached about 600 m? (this value is not included in the calculation) with the family consisting of 22 individuals. In late July this family had di- vided into two, which began to exploit the different parts of the enormous territory. By October these burrows were separated from one another by a distance of over 40 m (Fig. 2, burrow ]). In early-mid July the dispersal of young was attended by the emergence of new small burrows (e.g., burrow III, Fig. 2). Three burrows that were monitored from the first mounds grew very rapidly: within two weeks they reached 15, 15, and 54 m’. By late Sep- tember the soil mounds in smaller burrows covered areas of 112 and 140 m? (1986). In late summer-late autumn a considerable increase of the burrowing activity was ob- served. During fall daily 3-6 (up to 12) fresh mounds per burrow were recorded, e.g. equivalent to only 1-3 (up to 6) mounds per burrow in summer. An area exploited by a family group from mid-August up to mid-September 1993 averaged (+ SE) 177 + 32.0 m” (n= 11), range 80-330 m’. Discussion From the trapping data results, the basic type of spatial organisation for L. mandarinus is group territoriality. The mating system varies from polygyny to monogamy, the former predominating. Mateships are prolonged and appear to dissolute if some of the mates die. Young born in spring as well as later stay with their parents for a long time. As has been shown in the laboratory (ZORENKO et al. 1994; SMORKATCHEVA et al. 1997), the life cycle of L. mandarinus is characterized by rather long intervals between weaning (about 19- 22 days) and the earliest age of fertility (55-60 days for males, 36-38 days for females). In this study most of the young were known to remain in their natal burrows during this on- togenetic phase. Apparently, they help their mother rear subsequent 1-2 litters both by direct and indirect parental investment as it was observed in the laboratory (ZORENKO et al. 1994). For reproduction, sons apparently must leave the natal territory. According to trapping data young males seem to disperse at the age of 45-70 days, i.e. about the time of puberty. Daughters may stay in the natal territory as nonreproductive female, stay and reproduce, or disperse. In contrast to males no young female was recorded as migrant, although two individuals were known to settle at new sites. Several females were captured in their natal burrow as reproductive. L. mandarinus display strong incest-avoidance and contact with strange male is necessary for reproductive activation of young females (SMORKATCHEVA et al. 1997). Evidently females wait for a mate in their natal territory for some time after becoming physiologically fertile. Meanwhile, they increase their inclusive fitness by helping kin individuals and receive a chance to inherit the parental burrow and The social organization of the Mandarine vole 3a territory thereby, avoiding risks involved in dispersal. The change of sire appears to be the necessary condition to realise this chance. Most of polygynous groups are likely formed in such a way. The question remains open about the proximate causes of natal dispersal in L. man- darinus. From our preliminary laboratory data only amicable interactions occur between family members independent of their age. These observations do not support the hypoth- esis that aggression from adult members of the group forces the young to disperse (CHRIS- TIAN 1970; ANDERSoN 1980; BooNSsTRA et al. 1987). It is more probable that different fac- tors trigger the dispersal in the two sexes. Internal physiological cues might be sufficient to promote male dispersal. In female ontogeny not the “dispersal phase”, but the “mating or waiting for a mate phase” seems to be present. Natal dispersal of females was shown to be preceded by maturation and mating in M. arvalis (BoycE and Boyce 1988). Hormo- nal events induced by copulation probably account for strong female bonding to male in monogamous M. ochrogaster (CARTER and GETZ 1993). In the latter species, the important role of non-resident, non-paired males in reproductive activation of young females was clearly demonstrated (CARTER et al. 1980; McGuikre and GETZ 1991; McGuire et al. 1990; Lyons and GETZ 1993). For L. mandarinus, I hypothesise that not only non-resident males activate the reproduction, but also may promote the dispersal of young females if their mate does not remain with their family after copulation. If this hypothesis is true the following predictions should be realised: (i) the higher the number of non-paired males in a population, the greater the propor- tion of young females becoming reproductive (ii) the higher the mortality in fathers, the greater the proportion of daughters repro- ducing within the natal territory. Thus, (iii) if the survival of mated males is high and numerous unmated males are pre- sent in a population, then high levels of female natal dispersal and increasing numbers of new breeding units should be expected. These assumptions should be examined both in field demographic studies and in lab- oratory ethology experiments. Thus, in nature L. mandarinus exhibits group territoriality, prolonged pair-bonding, and parent-young relationships. Previously the species was reported to display the charac- ters of K-strategy (small litter size, slow development and sexual maturity, incest-taboo), paternal care activity, care by weaned young of pups, long latency, and low level of copu- latory stimulation (ZoRENKO et al. 1994; SMORKATCHEVA 1997). All of these traits are usually attributed to monogamy (KLEIMAn 1977; DEWSBURY 1990). However, L. mandari- nus combines a monogamous system of rearing with a polygynous system of grouping and mating. It is this combination of traits that underlies the high level of sociality in the man- darine vole. The social structure of this species appears to be similar to that of the prairie vole M. ochrogaster and the pine vole M. pinetorum. The latter two species are the classical ex- amples of monogamous microtines, although both display communal reproduction under certain conditions as well (GavisH et al. 1981; FITZGERALD and Mapıson 1983; GETZ et al. 1990). The tendency of the young to philopatry and cooperative breeding has been found in the prairie vole (McGuire et al. 1993) as well as in the pine vole (FiITZGERALD and MA- DISON 1983). These examples of monogamous or communal breeding units in voles are of- ten considered to be exceptional (e.g. ANDERSON 1980; WoLFF 1985; NELsoNn 1987). How- ever, the analysis of the available data dealing with the spatio-social structure for Old World microtines shows that this pattern is not at all that rare. Apparently, it is typical for Microtus socialis (SHCHIPANOV and Kasarkın 1996), Eolagurus luteus (LABUNETZ 1968; SHUBIN 1974), Prometheomys schaposchnikovi (GAMBARYAN et al. 1957; TurovV 1926), La- siopodomys brandti (KHRUSTZELEVSKI 1954a; GRosSE et al. 1984; DMITRIEV et al. 1992; Xın-Rong et al. 1998), and Ellobius talpinus (SHUBIN 1961; ZuBKoO and OSTRYAKOV 1961). 332 ANTONINA V. SMORKATCHEVA The mating system of M. ochrogaster ıs generally believed to be an adaptation to homogeneous, stable, low-food habitats where the females are widely dispersed and it is better for a male to guard a selected mate instead of searching for others (GETZ 1978; NeELson 1987; GETZ and CARTER 1996). However, the benefits associated with living in groups are unclear for this species (GETZ et al. 1990; McGuikeE and GETZ 1995). In M. pinetorum and L. mandarinus it is the fossorial habits that might explain prolonged pair-bonding as well as phylopatry and sociality if we presume that: (1) the risk of above-ground wandering is especially high for the fossorial animals; (ii) the construction of new tunnels is also associated with high costs (PowELL and FrıED 1992); (iii) the male and the philopatric offspring gain indirect fitness benefits by forage tun- nel building and maintenance and probably care for pups (PowELı and FrıED 1992). However, the statement of ErwoonD (1983) that direct care of offspring (huddling, grooming, retrieving) by a male is rather a result of his staying with the mother appears to be likely for voles. Thus, the set of traits mentioned above (prolonged strong pair-bonding, retention by mature offspring, care of the offspring by all group members) is not specific only for fos- sorial microtines but is expected to be typical for all of them. It seems to be true for a few species other than M. pinetorum and L. mandarinus, whose social pattern has been reported, Prometheomys schaposchnikovi (Turov 1926) and Ellobius talpinus (SHUBIN 1961; Zugko and OSTRYAKoV 1961). When the group composition of L. mandarinus ıs compared with that of other social microtines, the former appears to resemble most closely L. brandti by the complexity and large size of units. In both species the summer groups consist of 6-8 individuals (up to 20-25) and often include several reproductive females and 2-3 generations of youngs; the division of oversize families into several distinct ones has been discribed previously (Grosse et al. 1984; DmitrIEV et al. 1992; Xın-Rong et al. 1998; this study). In contrast to this, the social groups generally contain 2-3 reproductive anımals wıth only a few off- spring in Microtus socialis (SHCHIPANOV and Kasarkın 1996), M. ochrogaster (GETZ et al. 1993), M. pinetorum (FITZGERALD and Mapıson 1980), and Eolagurus luteus (SHUBIN 1974). The analysis of data available on sexual maturity in voles reveals a second trait that L. mandarinus and L. brandti have in common. This is the great obligatory delay of fertility with respect to weaning. In Lasiopodomys the minimal age of fertility in females has been reported as 35-40 days (L. mandarinus see ZORENKO et al. 1994; L. brandti see ZORENKO and JAKOBSONE 1986), males maturing even later. This phase of ontogeny may be considered as “period of helping”. In most studied microtines from grasslands, both the delay of maturation and dispersal are facultative with the earliest conceptions in fe- males occurring at about weaning or soon after, at 15-30 days (PoKrovskı 1967; SHUBIN 1974; NADEAU 1985; BoycE and Boyce 1988). The exceptions are M. pinetorum (SCHADLER and BUTTERSTEIN 1979) and probably, Ellobius talpinus (ZuBKo and OsTRryAKov 1961). Finally, the third trait of Lasiopodomys apparently associated with a high level of so- ciality, is the predominance of tactile contacts during interactions of these voles. Thus, the genus Lasiopodomys seems to be among the most social of voles. The genus now occurs in the open, highly seasonal grasslands of Central Asia. The colonial organisa- tion of L. brandti is thought to be associated with (1) construction of complex by ex- tended winter burrows, (2) storage of winter-food supplies in snow-free steppe, and (3) use of acoustic communication like ground squirrels to avoid predators (NAaumov 1955; Nıkorskı 1979; Grosse et al. 1984). The mandarine vole has solved both the problem of predators and that of winter storage by the transition to subterranean existence and for- aging. Moreover. the northern subspecies L. mandarinus vinogradovi presents an excep- tional example of the tendency towards monophagy, that is rare among voles (the only known monophagous species is M. breweri, ROTHSTEIN and TAMARIN 1977) as well as The social organization of the Mandarine vole 538 among the fossorial mammals (OrLov 1978; Nevo 1995). At the same time, neither its ap- pearance nor its skull and bones of the extremities seem to demonstrate significant mor- phological adaptations to a fossorial mode of life. It can be hypothesized that the high level of sociality was characteristic of the ancestral above the ground form and repre- sented the precondition to occupy the recent niche. Probably it is the collective tunnel construction that allows the mandarine voles to feed on the dispersed large roots, as is ac- cepted for mole-rats Bathyergidae (Jarvıs 1981; LovEGROVE and WisseEL 1988). Acknowledgements I thank EKATHERINE TSYTsULInA for her assistance in the field and am also grateful to Dr. NATALIA AMBRAMSOnN for reviewing the English version of the manuscript. Zusammenfassung Soziale Organisation der Mandarin-Wühlmaus (Lasiopodomys mandarinus) während der Reproduktionsperiode Das Sozialsystem in freilebenden Populationen von Lasiopodomys mandarinus wurde im Selenginski Distrikt in Burjatien mit Markierungs-Wiederfang-Methoden untersucht. Mandarin-Wühlmäuse lebt- en in umfangreichen Familiengruppen. Die Mitglieder einer Gruppe waren streng an einen gemeinsa- men Bau gebunden. Die Gruppen bestanden im Sommer aus einem reproduzierenden Männchen, 1- 5 reproduzierenden Weibchen und von 1-3 Generationen von Jungtieren; durchschnittlich fanden sich 8,7 Individuen pro Bau (Umfang 3 bis 22). Die meisten Nachkommen blieben im Elternterritorium wenigstens 50 Tage lang. Keines von 72 jungen Männchen und nur drei von 73 jungen Weibchen be- gannen sich im Geburtsbau zu vermehren. Der Wechsel des Vatertieres ist wahrscheinlich eine wich- tige Bedingung für die reproduktive Aktivierung von philopatrischen Töchtern. Folglich zeigt L. mandarinus einen hohen Grad an Sozialıtät, der auf gemeinsamer Jungenpflege und verlängerten Paarbindungen sowie auf verlängerten Bindungen zwischen Eltern und Jungtieren begründet ist. Diese Merkmale werden in der Literatur auch für die Art Lasiopodomys brandti be- schrieben. Vermutlich war diese soziale Organisation für die nicht subterrane Anzestralform der bei- den Arten typisch und die Voraussetzungen dafür, die Nische eines unterirdischen Stenophagen zu be- setzen. References ÄNDERSON, K. (1980): Evolutionary implications of microtine behavioral systems on the ecological stage. Biologist 62, 70-88. BoonsTRA, R.; KREBS, C. J; GAINES, M. S.; JOHNSON, M. L.; CRAINE, I. T. M. (1987): Natal philopatry and breeding systems in voles (Microtus sp.). J. Anim. Ecol. 56, 655-673. Boyce, C. C. K; Boyce, J. L. (1988): Population biology of Microtus arvalis. Il. Natal and breeding dis- persal of females. J. Anim. Ecol. 57, 723-736. CARTER, C. S.; GETZ, L. L. (1993): Monogamy and the prairie vole. Sci. Amer. 268, 100-106. CARTER, C. S.; GETZ, L. L.; GAVISH, J. L.; MCDERMOTT, A. (1980): Male-related pheromones and the acti- vation of female reproduction in the prairie vole (Microtus ochrogaster). Biol. Reprod. 23, 1038- 1045. CHRISTIAN, J. J. (1970): Social subordination, population density and mammalıan evolution. Science 168, 84-90. DEwSBurY, D. A. (1990): Individual attributes generate contrasting degrees of sociality in voles In: So- cial System and Population Cycles in Voles. Ed. by R. H. TAMARIN, R. S. OSTFELD, S. R. PuGH, and G. BuJALska. Basel, Boston, Berlin: Birkhäuser-Verlag. Pp. 1-10. 354 ANTONINA V. SMORKATCHEVA DMirkiev, P. P. (1980): Ecology of the mandarine vole Microtus mandarinus, and its role in pasture bio- coenoses of North Mongolia. Zool. Zh. 59, 1852-1861 (in Russian). DMITRIEV, P. P; SHVETZOV, J. U. G.; DULAMTZEREN, S. (1992): The Mammals of Khangain Mountains: Fauna, Ecology, Importance for Biogeocenoses. Moscow: Nauka. (in Russian). Eırwoop, R. W. (1983): Paternal care in rodents. In: Paternal Behavior in Rodents. Ed. by R. W. Er- wooD.N. Y. Pp. 235-257. FETISsov, A. S. (1955): On the distribution and ecology of the Vinogradov vole in the East Siberia. Dok- lady AN SSSR. 104, 326-328 (in Russian). FITZGERALD, R. W.; Mapıson, D.M. (1983): Social organization of a free-ranging population of pine vole, Microtus pinetorum. Behav. Ecol. Sociobiol. 13, 183-187. GAMBARYAN, P. P.; KARAPETYAN, V. S.; AIRUMYAN, K. L.; KAsARYAnN, K. T.; MEILUMYAN, S. K. (1957): On the ecology of the Prometheus’ mice. Materialy po izucheniyu fauny Armyanskoi SSR. Erevan. 3, 5-16 (in Russian). GAVISH, L.; CARTER, C. S.; GETZ, L. L. (1981): Further evidence for monogamy in the prairie vole. Anim. Behav. 29, 955-957. GETZ, L. L. (1978): Speculation on social structure and population cycles in Microtine rodents. Biologist 60, 134-147. GETZ, L. L.; CARTER, C. S. (1996): Prairie-vole partnerships. Amer. Sci. 84, 56-62. GETZ, L. L.; MCGUIRE, B.; HOFMANN, J. E.; PIZZUTO, T.; FRASE, B. (1990): Social organization and mating system of the prairie voles, Microtus ochrogaster. In: Social System and Population Cycles in Voles. Ed. by R.H. TAMmARIN, R. S. ÖSTFELD, S. R. PUGH, and G. BuJALsKA. Basel, Boston, Berlin: Bir- khäuser-Verlag. Pp. 69-79. GETZ, L. L.; MCGUlRe, B.; PızZuUTo, T.; HOFMANN, J. E.; FRASE, B. (1993). Social organization of the prairie voles, Microtus ochrogaster. J. Mammology 74, 44-58. GETZ, L. L.; HOFMANN, J. E.; KLATT, B. J.; VERNER, L.; COLE, F.R.; LINDROTH, R.L. (1987): Fourteen years of population fluctuations of Microtus ochrogaster and M. pennsylvanicus in east central Illi- nois. Can. J. Zool. 65, 1317-1325. GORSHKOVA, A.; GRINEVA, N. A.; JURAVLEVA, N. A. (1977): Ecology and Pasture Digression of Transbai- cal Steppe Societies. Novosibirsk: Nauka. (in Russian). GROSSE, W.-R.; STUBBE, A.; SCHUH, J.; DawAA, N. (1984): Zum Raum-Zeit-Verhalten von Microtus brandti und Alticola argentatus. Wiss. Beitr. M.-Luther-Univ. Halle-Wittenberg 21, 124-131. Hyan-VaAn-Dı (1960): Some faunistic finds in Yang-tze-basin and their zoogeographical importance. Zool. Zh. 39, 595-599 (in Russian). JARVIS, J. U. M. (1981): Eusociality in amammal: cooperative breeding naked mole-rat colonies. Science 212, 571-573. KHRUSTZELEVSKI, V. P. (1954 a): On the ecology of the brandt’s voles. Age regularities and reproductive rate. Izvestiya Irkutskogo gos. Protivochumnogo instituta Sibiri I Dalnego Vostoka 12, 156-170 (in Russian). KHRUSTZELEVSKI, V.P. (1954b): Regularities in Brande’s vole density fluctuations in south-east of Transbaical region. Izvestiya Irkutskogo gos. Protivochumnogo instituta Sibiri I Dalnego Vostoka 13, 81-105 (in Russian). KLEIMAN, D. (1977): Monogamy in mammals. O. Rev. Biol. 52, 39-69. LABUNETZ, N. F. (1968): The yellow steppe lemming (Lagurus luteus Eversmann) in south-east foothills of Khangai. Zool Zh. 47, 1265-1268 (in Russian). LOVEGROVE, B. G.; WisseEL, C. (1988): Sociality in molerats. Metabolic scaling and the role of risk sensi- tivity. Oecologia 74, 600-606. Lyon, S. A.; GETZ, L. L. (1993): Reproductive activation of virgin female prairie voles (Microtus ochro- gaster) by paired and unpaired males. Behav. Proc. 29, 191-200. MCARTHUR, R.H.; Wırson, E. ©. (1967): Theory of Island Biogeography. Princeton: Princeton Univ. Press. MCGURRE, B.; GETZ, L.L. (1991): Response of young female prairie voles (Microtus ochrogaster) to nonresident males: implications for population regulation. Can. J. Zool. 69, 1348-1355. MCGUunRE, B.; GETZ, L. L. (1995): Communal nesting in prairie voles (Microtus ochrogaster): an eva- luation of costs and benefits based on patterns of dispersal and settlement. Can. J. Zool. 73, 383- 398 McGusre, B.; PIZZUTo, T.; GETZ, L. L. (1990): Potential for social interactionin a natural population of prairie voles (Microtus ochrogaster), Can. J. Zool. 68, 391-398. MCGUulRE, B.; GETZ, L. L.; HOFMANN, J. E.; PIZZUTOo, T.; FRASE, B. (1993): Natal dispersal and philopatry The social organization of the Mandarine vole 355 in prairie voles (Microtus ochrogaster) in relation to population density, season, and natal social en- vironment. Behav. Ecol. Sociobiol. 32, 293-302. NADEAU, J. H. (1985): Ontogeny. In: Biology of New World Microtus. Ed. by R. H. Tamarın. Spec. Publ. N° 8 Amer. Soc. Mammal. Pp. 254-285. NAuMov, N. P. (1955): The Ecology of Animals. Moscow: Sovetskaya Nauka. (in Russian). NELSsoNn, J. (1987): Social organization in voles. Lund: Reprocentralen Lunds Universitet. NEvo, E. (1995): Mammalian evolution underground. The ecological-genetic-phenetic interfaces. Acta Theriol. Suppl. 3, 9-31. NIKOLSKI, A. A. (1979): Parallelisms of danger call in Rombomys opimus and Microtus brandti (Criceti- dae). Zool. Zh. 58, 1047-1058 (in Russian). ORLov, G. I. (1978): Biological peculiarities and possible origin of subterranean rodents. Zool. Zh. 57, 1706-1719 (in Russian). POKROVSKI, A. V. (1967): The seasonal variations in the age of sexual maturation in females of sagebrush voles and some other species of voles. Trudy Moskovskogo Obshchestva Ispytateley Prirody 25, 78- 81 (in Russian). PowELL, R. A.; FRIED, J. J. (1992): Helping by juvenile pine voles (Microtus pinetorum), growth and sur- vival of younger siblings, and the evolution of pine vole sociality. Behav. Ecol. 3, 325-333. ROTHSTEIN, B. E.; TAMARIN, R. H. (1977): Feeding behavior of the insular beach vole, Microtus breweri. J. Mammology 58, 84-85. SHADLER, M.; BUTTERSTEIN, G. M. (1979): Reproduction in the pine vole, Microtus pinetorum. J. Mam- mology 60, 841-844. SHCHIPANOV, N. A.; KASATKIN, M. V. (1996): Microtus socialis in changes landscapes of South Dagestan: population aspect of survival. Zool. Zh. 75, 1412-1425 (in Russian). SHUBIN, I. G. (1961): On the ecology of the mole-vole in Central Kazakhstan. Zool. Zh. 40, 1543-1551 (in Russian). SHUBIN, I. G. (1974): Ecology of the yellow steppe lemming in Zaisan hollow. Zool. Zh. 53, 272-277 (in Russian). SMORKATCHEVA, A. V. (1993): Population structure and reproduction of Lasiopodomys mandarinus in the Transbaical region. In: Rodents and Spatium IV. Mikolajkı. P. 101. SMORKATCHEVA, A. V.; AKSENOVA, T. G.; ZORENKO, T. A. (1990): The ecology of Lasiopodomys mandari- nus (Rodentia, Cricetidae) in the Transbaical area. Zool. Zh. 69, 115-124 (in Russian). SMORKATCHEVA, A. V.; ZAKHAROV, K. V.; TZYTZULINA, E. A. (1997): The behavioral and reproductive fea- tures of the mandarine vole Lasiopodomys mandarinus ın the context of the species’ social system. In: Rare Mammals of Russia and Contiguous Territories. Moscow. P. 90 (in Russian). SVIRIDENKO, P. A. (1934): Increase and decrease of population of mouse-like rodents. Trudy po zashchite rastenii. Ser. 4: Pozvonochnye. 3, 5-60 (in Russian). TAITT, M. J.; Kress, C. J. (1985): Population dynamics and cycles. In: Biology of New World Microtus. Ed. by R. H. TamaRrın. Spec. Publ. N’8 Amer. Soc. Mammal. Pp. 567-620. Turov, S. (1926): To the biology and spreading of Prometheomys schaposchnikovi Sat. Bull. Scı. Inst. Exploration Regional du Caucase du Nord 1, 23-30 (in Russian). WOLFF, J. O. (1985): Behavior. In: Biology of New World Microtus. Ed. by R. H. Tamarın. Spec. Publ. N’ 8 Amer. Soc. Mammal. Pp. 340-372. Xın-Rong, W.; MENG-Jun, W; WEN-Oin, ZH.; GuANG-HE, W. (1998): The social structure and mating system of the brandt vole (Microtus brandti). In: Intern. Conf. on Rodent Biology and Manage- ment. Beijing. Pp. 84-85. ZORENKO, T. A.; SMORKATCHEVA, A. V.; AKSENOVA, T. G. (1994): Reproduction and postnatal ontogenesis of mandarin vole Lasiopodomys mandarinus (Rodentia, Arvicolinae). Zool. Zh. 73, 120-129 (in Russian). ZORENKO, T. A.; JAKOBSONE, G. KH. (1986): The postnatal development of brant’s voles. (Lasiopodomys brandti Radde). In: Okhrana, ekologiya i etologiya zhivotnykh. Riga. Pp. 5-24 (in Russian). ZUBKO, Y. A.; OSTRYAKOV, C. I. (1961): On the reproduction of mole-voles (Ellobius talpinus Pal.) in the South of Ukraine. Zool. Zh. 40, 1577-1579 (in Russian). Author’s address: ANTONINA V. SMORKATCHEVA, Department of Vertebrate Zoology, St. Petersburg State University, Universitetskaya Nab. 7/9, 199034, St. Petersburg, Russia. Z. Säugetierkunde 64 (1999) 356-362 ZEITSCHRIFT © 1999 Urban & Fischer Verlag SÄUG EII E RKU N DE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Allozyme variation of Cottontail rabbits (Sy/vilagus) from Mexico By F. A. CERVANTES, J. P. RAMIREZ-SILVA, ARIADNA MARIN, and GLORIA L. PORTALES Departamento de Zoologia, Instituto de Biologia, Universidad Nacional Autonoma de Mexico, Mexico City, Mexico Receipt of Ms. 27. 11. 1998 Acceptance of Ms. 08. 06. 1999 Abstract We examined the allozyme variation of cottontail rabbits of the genus Sy/vilagus from Mexico, and described their genic relationships. Samples of kidney and heart were run in horizontal starch-gel elec- trophoresis to assess the variation of 23 presumptive loci, and the BIOSYS-1 software was used to compute estimates of genic variation. Results showed that 60.3% of the loci were polymorphic. 5. floridanus was the most genically variable rabbit as revealed by mean number of alleles per locus and percentage of polymorphic locı. The fixation index showed genetic differentiation among species. The smallest genetic distance was between S. floridanus and S. brasiliensis whereas the largest one was between S$. mansuetus and S. audubonii. A phenogram showed $. mansuetus branching out first, $. au- dubonii next, and finally S. floridanus and S. brasiliensis together. In conclusion, $. floridanus showed the largest genic variation, S. mansuetus was the most distinctive rabbit, and $. audubonii and S. brasi- liensis were the most closely related species. Key words: Sylvilagus, cottontail rabbits, allozymes, electrophoresis, Mexico Introduction Cottontail rabbits of the genus Syl/vilagus are a speciose group that occurs in the New World (HoFFMAnnN 1993). Unfortunately, genetic variation and species relationships within the genus have been barely examined (SCRIBNER and WARREN 1986). Reports in the litera- ture on these topics are limited to scarce and scattered references. This is remarkable if we consider the high species richness and population abundance of this genus in North America (CHAPMAN and CEBALLOS 1990). ) Just the Mexican territory, for instance, hosts eight species of Silvilagus, four being endemics (CERVANTES et al. 1994). The genetic varia- tion of Mexican cottontail rabbits still remains unexplored. Genetic distance estimates are available for few lagomorph species (GLOVER et al. 1977; GRILLITSCH et al. 1992). Only recently, HALAnycH and Rosınson (1997) provided useful information to gain insight into the evolutionary history of some cottontail rabbit species using sequence data from mtDNA. Sylvilagus floridanus, a cottontail species occurring in Mexico, is genetically variable in Maryland, U.S.A., due to its intensive introduction to that state (CHApMAN and MoRr- GAN 1973; MoRGAN and CHAPMAN 1981). Isolated populations of the same species in Texas, U. S. A., are genetically differentiated and characterized by periodical high disper- sion rates and low genetic flow due to agricultural land use (Van DEN BusscHeE et al. 1987). S. audubonii from Texas, another cottontail rabbit occurring in Mexico, displays a 0044-3468/99/64/06 - 356 $ 12.00/0 Allozymes of Mexican rabbits 3m high genetie similarıty to Texan populations of S. floridanus, although no gene flow be- tween them is expected to take place (SCRIBNER and WARREN 1986). These data thus suggest the presence of genic variation within and among species of Sylvilagus and may provide an estimation of relationship between species. Therefore, the purpose of this study is to examine the allozyme variation of selected species of Mexican Sylvilagus. Material and methods Rabbits were collected with a shotgun. Specimens of black-tailed jackrabbit (Lepus californicus) from a Mexican locality were included as outgroup. Samples of heart, kidney, and liver were removed and immediately frozen in liquid nitrogen. All specimens were preserved as standard museum vouchers and deposited in the mammalian collection (Coleccion Nacional de Mamiferos, CNMA, formerly IBUNAM) of Instituto de Biologia, Universidad Nacional Autonoma de Mexico, in Mexico City, Mexico. Localities are listed by species and sample sizes are indicated in parenthesis as follows. Sylvilagus floridanus: 14km W Villa de Arista, Municipio Moctezuma, San Luis Potosi, Mexico, 1620 m (2); 2kmW Santa Maria del Mar, Municipio Juchitan, Oaxaca, Sm (3); 10km NW+2kmE La Rosa Amarilla, Municipio La Manzanilla, Jalisco, Mexico, 2050 m (3); 11km E+1.5kmN San Jose de Gra- cia, Municipio Marcos Castellanos; Michoacan, Mexico, 2100 m (3). S. audubonii: 140 km NE Gomez Palacio, Municipio Mapimi, Durango, Mexico, 1189 m (3). S. mansuetus: Isla San Jose, Municipio La Paz, Baja California Sur, Mexico, 5m (3). S. brasiliensis: El Chajul, Municipio Ocosingo, Chiapas, Mexico (1); km 35 road Catemaco-Balzapote, Municipio Catemaco, Veracruz (1). L. californicus: same locality as that for $. auduboni (2). Homogenates of kidney and heart were analyzed for electrophoretically detectable protein varia- tion and prepared according to the methods of SELANDER et al. (1971). Procedures for horizontal starch gel electrophoresis also followed those of SELANDER et al. (1971). A total of 23 presumptive loci were examined as follows (abbreviations and IEC numbers follow HARRIS and Hopkınson 1976): buffer system tris-citrate I (pH 6.7-6.3) was used for malate dehydro- genase (MDH-1, MDH-2, 1.1.1.37), lactate dehydrogenase (LDH-1, LDH-2, LDH-3, 1.1.1.27), acid phosphatase (ACP, 3.1.3.2), glucose-phosphate isomerase (GPI, 5.3.1.9), 6-phosphosphogluconate de- hydrogenase (PGD, 1.1.1.44), glucose dehydrogenase (GDH-1, GDH-2, 1.1.1.47), purine nucleoside phosphorylase (NP, 2.4.2.1), and general proteins (GP); buffer system tris-citrate II (pH 8) for malic enzyme (ME-1, ME-2, 1.1.1.40), L-glutamate dehydrogenase (GLUD, 1.4.1.3); buffer system PGI-po- tassium phosphate (ph 6.7) for isocitrate dehydrogenase (ICD, 1.1.1.42), aldolase (ALD, 4.1.2.13), superoxide dismutase (SOD, 1.15.1.1), xantine dehydrogenase (XDH, 1.2.3.2); buffer system tris-ma- late EDTA (pH 7.4) for sorbitol dehydrogenase (SDH, 1.1.1.14), hexokinase (HK, 2.7.1.1); and buffer system litium hydroxide (A = 10 %, B = 90 %) for alcohol dehydrogenase (ADH, 1.1.1.1), and esterase (EST 3.2.1.1). Alleles at each locus were designated by mobility relative to the most common allele at that locus. Results were summarized in the form of individual genotypes by locus for each individual. Estimates of allelic frequencies, polymorphism, heterozygosity, WRIGHT’s (1965) F-statistics, coeff- cients of genetic distance (D) of Rogers (1972) and of unbiased distance (D) of Neı (1978) were com- puted using the BIOSYS-1 program of SwWOFFORD and SELANDER (1981). The coeffcients of genetic dis- tance were calculated with the inclusion of monomorphic loci. Clustering of distance matrices was performed using the unweighted pair-group method with arithmetic averages procedure (UPGMA,; SNEATH and SOoKAL 1973; SWOFFORD and SELANDER 1981). Results and discussion Of the 23 loci examined electrophoretically, 14 (60.3%) were polymorphic (Tab. 1), whereas GP, ME, GDH-I1, ADH, MDH-1, LDH-2, GPI, SOD, and ACP were mono- morphic. Rare variants (frequency of the most common allele as greater than 0.95) were 358 F. A. CERVANTES et al. Table 1. Alleles (a-d), allele frequencies (in parenthesis), sample size (n), average number of alleles per locus (AVER), percent of polymorphic locı (POLY), and expected average individual heterozyg- osity (HETE) for leporids (Sylvilagus mansuetus, S. floridanus, S. audubonii, S. brasiliensis and Lepus californicus) from Mexico. Only polymorphic locı (14 out of 23) are listed. Estimate of POLY includes only those locı for which dominant allele has a frequency less than 0.95. See text for loci abbrevia- tions. Sylvilagus Sylvilagus Sylvilagus Sylvilagus Lepus mansuetus audubonii brasiliensis floridanus californicus a (0.091) b (0.633) c (0.273) a (0.091) b (0.182) c (0.720) a (0.667) a b (0.333) b b a (0.333) b b (0.667) b b b a (0.125) b (0.875) b b b b (0.677) b (0.125) c (0.333) c (0.875) b b a a a a a a (0.667) Ce (0.333) 8 1.3 2, 0.084 b b not present. The locus with the highest number of alleles per locus (=4) was EST, fol- lowed by ALD, ICD, and GLUD (=3). All loci appeared as single banded homozy- gotes. Among the polymorphic loci, seven were polytypic among species of Sylvilagus whereas the other seven (GDH-2, ME-2, NP, LDH-3, XDH, HK, and SDH) were fixed for the same allele in all species of Sylvilagus (Tab. 1). All loci were monomorphic in S. brasiliensis and L. californicus. Sylvilagus floridanus showed a slightly larger average number of alleles per locus (Tab. 1). Proportions of polymorphic loci (95 % criterion) averaged 8.7 among Sylvilagus species and ranged from 0 to 21.7. $. floridanus also displayed the highest polymorphism value and alternate alleles at four loci (GLUD, ICD, LDH-1, EST). Therefore, Sylvilagus floridanus was the most genically variable rabbit. This is similar to what has been found in several localities of the United States of America for the same species (MoRGAN and CHAPMAN 1981; SCRIBNER and WARREN 1986; Van DEn BusschHe et al. 1987). For instance, populations of S. floridanus and 5. audubonii from Texas displayed 33 and 25 % of polymorphism, respectively (SCRIBNER and WARREN 1986). On the other hand, polymorphism recorded in brown hares (Lepus europaeus) Allozymes of Mexican rabbits 859 from Central Europe was 16.3 % (HARTL et al. 1990). In contrast, no polymorphic loci for S. brasiliensis was recorded herein. In our study no rabbit species showed heterozygote individuals. Similarly, heterozy- gote deficiencies were noted for populations of $. floridanus from Texas (Van DEN BusschHe et al. 1987), although previous studies of $. floridanus and S. audubonii revealed fair amounts of heterozygosity (MoRGAN and CHAPMAN 1981; SCRIBNER and WARREN 1986). Selected populations of the European wild rabbit (Oryctolagus cuniculus) from East Anglia, England, also showed heterozygote deficiencies at most loci (SURRIDGE et al. 1998). Moreover, low levels of heterozygosity were reported in pikas (Ochtotona prin- ceps) from Colorado, U.S.A. (GLovEr et al. 1977). The expected genic heterozygosity for S. floridanus calculated from Hardy-Weinberg assumptions was 8.4% (Tab. 1), whereas the mean value for vertebrate populations is 5- 6% (SELANDER and JoHnson 1973). In contrast, expected figures for $. mansuetus and $. audubonii were lower (4.6 and 2.3 %, respectively). Sylvilagus species occurring in Mexican territory thus show detectable protein varia- tion. The extent of variation is, however, comparatively low. Our estimates of genetic distance (Ner’s, 1978, unbiased distance) between species turned out to be relatively low compared to those of other vertebrate populations. Our results are comparable to those found for conspecific populations whose coefficients of similarity are generally at the 0.90’s level (SELANDER and JoHNnson 1973; HARTL et al. 1990). This is particularly true for genetic distances among the species pairs $. audubonii — S. brasiliensis, S. brasiliensis — S. floridanus, and $. audubonii — S. floridanus (Tab. 2). Po- pulations of S. audubonii from Texas also displayed high genetic similarity (ROGER’S simi- larity index = 0.884) to sympatric S. floridanus (SCRIBNER and WARREN 1986). S. floridanus from the same region displayed NEr’s genetic distances between populations ranging from 0.20 to 0.388 (Van DEN BusschHe et al. 1987). The lowest genetic distance recorded was between Sy/vilagus floridanus and S. brasi- liensis (Tab. 2), the only two cottontail rabbit species that occur as far south as the tempe- rate and tropical habitats of South America. They mostly are allopatric species although they also may be parapatric, seldomly sympatric. In contrast, the largest genetic distance recorded herein was between S. mansuetus and S. audubonii (Tab. 2), species adapted to xeric conditions and whose ranges are very close. Sylvilagus mansuetus, once thought to be a subspecies of 5. bachmani, turned out to be the most distinctive rabbit of the species sample examined (Fig. 1). This cottontail is restricted to a small island of 194 km? in the Gulf of California, Mexico. Unfortunately, other than the original description of the species, there are no reports on the relationships of this rabbit to other cottontail rabbits. The mean genetic distance (Ner’s, 1978 unbiased distance) between S. mansuetus and other Sylvilagus species was 0.057. This species dis- Table 2. Ner’s (1978) unbiased distances (above diagonal) and Rocers’ (1972) genetic distances (be- low diagonal) among leporids (Sy/vilagus mansuetus, S. floridanus, 5. audubonit, 5. brasiliensis, and Le- pus californicus from Mexico. Sylvilagus Sylvilagus Sylvilagus Sylvilagus Lepus mansuetus audubonü brasiliensis floridanus californicus Sylvilagus mansuetus Sylvilagus auduboniüi Sylvilagus brasiliensis Sylvilagus floridanus Lepus californicus 360 F. A. CERVANTES etal. played alternate alleles at two locı (6PGD, MDH-2) relative to S. audubonii, its nearest geographic sample examined in this study. In addition, these pair of species were fixed for one alternate allelle at the EST locus (Tab. 1). The values of genetic differentiation (Fst) among species were relatively high (Tab. 3), except for four locı (GLUD, ICD-1, 6PGD, and MDH-2). The mean Fst was high too considering all species examined (0.851). When the outgroup (1. californicus) is removed from the calculations, the average is lower (about half = 0.462; Tab. 3), but still indicative of substancial species differentiation. This fits that $. mansuetus, S. audubonii, S. florida- nus, and S. brasiliensis are also morphologically distinctive (CHAPMAN and CEBALLOS 1990). Sımilarly, populations of S. floridanus from Texas revealed a significant degree of genetic differentiation too (VAN DEN BusscHe et al. 1987). Table 3. Fixation index (WRIGHT’'s Fst) for polymorphic loci calculated among leporids (Sy/vilagus mansuetus, S. floridanus, 5. auduboniü, S. brasiliensis, and Lepus californicus) from Mexico, and among the same samples excluding Z[. californicus. All samples All samples exclusive of L. californicus 0.60 0.10 0.05 0.00 IX Sylvilagus mansuetus Sylvilagus audubonii Sylvilagus floridanus Sylvilagus brasiliensis Lepus californicus Fig1. UPGMA tree of leporids (Sylvilagus mansuetus, S. audubonii, S. brasiliensis, S. floridanus, and Lepus californicus) from Mexico based on NEr’s (1978) unbiased distances. The cophenetic correlation coefficient = 0.999. Allozymes of Mexican rabbits 361 The UPGMA procedure of a matrix of unbiased distances (NEı 1978) revealed three allozymic groups present within the Sy/vilagus group (Fig. 1). The first consisted of S. mansuetus branching out first (fixed for one allele at the EST locus), S. audbonii next, and S. floridanus and S. brasiliensis together, who were relatively closely allied, and sepa- rated by a genetic distance of 0.010. None of these samples was fixed for different electro- morphs relative to all of the other Sylvilagus samples. This arrangement may not reflect phylogenetic relationships, but allows to understand the overall genetic resemblance con- tained in the distance matrix computed. Assessement of the genic relationship of $. audubonii to S. floridanus herein fit find- ings on the systematic relationships among ten species of Sylvilagus based on diploid chromosomal numbers (CHAPMAN and CEBALLos 1990). That is, $. audubonii and S$. flori- danus also occur on separate branches of a dendrogram. Similarly, 12SrDNA data showed that $. floridanus and S. audubonii were not each other’s closest relatives in the species set examined (HALANnYcH and Rosgınson 1997). Results presented here are the first data set that outline genic relationships among se- lected species of Sylvilagus occurring in Mexico. It is shown that samples of Sylvilagus species occurring in Mexico display detectable protein variation, although lower than that reported for other leporid and vertebrate species. On the other hand, S. mansuetus is the most genetically distinctive cottontail rabbit examined, whereas S$. floridanus and S. brasi- liensis are the most closely related species pair. Although the four species examined are morphologically well differentiated the genetic distances among them are smaller than expected. Acknowledgements We thank C. LORENZO, J. VARGAS, F. X. GONZALEZ, A. GONZALEZ-R., L. HERNANDEZ, A.L. COLME- NARES, R. M. GONZALEZ, J. MARTINEZ, A. RoJAs, S. T. ALVAREZ, and P. CorTESs, for assistance in the field and laboratory. This project was supported in part by Universidad Nacional Autonoma de Mexi- co (grant IN-203793) and by a grant from the Sir Peter Scott Fund, through the Lagomorph Specialist Group of the Species Survival Commission, World Conservation Union. C. LORENZO provided useful suggestions to earlier drafts. Zusammenfassung Allozym-Variation von Wollschwanz-Kaninchen (Sylvilagus) aus Mexiko. Wir untersuchten die Allozym-Variation von ausgewählten Arten des Wollschwanz-Kaninchens der Gattung Sylvilagus aus Mexiko und beschrieben deren genetische Verwandtschaften. Proben von Niere und Herz wurden horizontaler Stärkegelelektrophorese unterzogen, um die Protein-Varlation von 23 vermuteten Loci zu bestimmen, und die Software BIOSYS-1 wurde benutzt, um Schätzungen der genetischen Variation zu berechnen. Die Ergebnisse zeigten, daß 60,5% der Loci polymorph waren. S. floridanus war das genetisch variabelste Kaninchen, wıe die mittlere Anzahl von Allelen pro Locus und der Prozentsatz von polymorphen Loci zeigten. Der „Fixations-Index“ zeigte gene- tische Differenzierung unter den Arten. Die geringste genetische Distanz bestand zwischen S. flori- danus und S. brasiliensis, die größte zwischen $. mansuetus und S. audubonii. Ein Phaenogram zeigte, daß S. mansuetus als erste Art abzweigt, S. audubonii als nächste und S$. floridanus und S. brasiliensis gemeinsam. Schlußfolgernd zeigte S. floridanus die größte genetische Variation, S. mansuetus war das unterschiedlichste Kaninchen und S$. floridanus und S. brasiliensis waren die am engsten verwandten Arten. 362 F. A. CERVANTES etaal. References CERVANTES, F. A.; CASTRO, A.; RAMIREZ-PULIDO, J. (1994): Mamiferos terrestres nativos de Mexico. Anal. Inst. Biol., Univ. Nal. Auton. Mex., Ser. Zool. 65, 117-190. CHAPMAN, J. A.; CEBALLOS, G. (1990): The cottontails. In: Rabbits, Hares and Pikas, Status Survey and Conservation Action Plan. Ed. by J. A. CHapman and E.C.J. Frux. Gland, Switzerland: IUCN. Pp. 95-120. CHAPMAN, J. A.; MORGAN, R. P. (1973): Systematic status of the cottontail rabbit in western Maryland and nearby West Virginia. Wildl. Monogr. 36, 1-54. GLOVER, D., G.; SMITH, M. H.; Ames, L.; JOULE, J.; DUBAcH, J. M. (1977): Genetic variation in pika popu- lations. Can. J. Zool. 55, 1841-1845. GRILLITSCH, M.; HARTL, G. B.; SUCHENTRUNK, F.; WILLING, R. (1992): Allozyme evolution and the mole- cular clock in the Lagomorpha. Acta Theriol. 37, 1-13. HALANYcCH, K. M.; RoBinson, T. J. (1997): Phylogenetic relationships of cottontails (Sy/vilagus, Lago- morpha): congruence of 12S rDNA and cytogenetic data. Mol. Phylogenet. Evol. 7, 294-302. HARTL, G. B.; MARKOWSKI, J.; KovAcs, G.; GRILLITSCH, M.; WirLinG, R. (1990): Biochemical variation and differentiation in the brown hare (Lepus europaeus) of Central Europe. Z. Säugetierkunde 55, 186-193. HARRIS, H.; HopkInson, D. A. (1976): Handbook of Enzyme Electrophoresis in Human Genetics. Ams- terdam, Holland: North-Holland Publ. Comp. HOFFMANN, R. S. (1993): Order Lagomorpha. In: Mammal Species of the World. A Taxonomic and Geo- graphic Reference. 2nd ed. Ed. by D.E. Wırson and D. M. REEDER. Washington, D. C., USA: Smithsonian Institution. Pp. 807-828. MORGAN, R. P.; CHAPMAN, J. A. (1981): Serum proteins of the Sy/vilagus complex. In: Proceedings of the World Lagomorph Conference. Ed. by K. Myers and C. D. MAcInnes. Guelph, Canada: University of Guelph. Pp. 64-72. Neır, M. (1978): Estimation of average heterozygosity and genetic distance from a small number of indi- viduals. Genetics 89, JSS3-5%. ROGERS, J. S. (1972): Measures of genetic similarıty and genetic distance. Austin, USA: Univ. Texas Publ. 7213, 145-153. SCRIBNER, K. T.; WARREN, J. (1986): Electrophoretic and morphologic comparisons of Sylvilagus florida- nus and S. audubonii in Texas. Southwest. Nat. 31, 65-71. SELANDER, R.K.; JOHNSON, W. E. (1973): Genetic variation among vertebrate species. Ann. Rev. Ecol. Syst. 4, 75-91. SELANDER, R.K.; SMITH, M.H.; Yang, S. Y.; JoHNSoN, W. E.; GENTRY, W. B. (1971): Biochemical poly- morphism and systematics in the genus Peromyscus. Variation in the old-field mouse (Peromyscus polionotus). Austin, USA: Univ. Texas Publ. 7103, 49-90. SNEATH, P.H. A.; SOKAL, R.R. (1973): Numerical taxonomy. San Francisco, USA: W. H. Freeman and Co. SURRIDGE, A.K.; BELL, D. J.; HEWITT, G. M. (1998): Using molecular tools to study biogeography of the European wild rabbit (Orytolagus cuniculus) in Britain. Game Wildl. 15, 65-74. SWOFFORD, D. L.; SELANDER, B. (1981): BIOSYS-1: A FORTRAN program for the comprehensive ana- lysis of electrophoretic data in population genetics and systematics. J. Heredity 72, 128-283. Van DEN BuUsscHE, R. A.; HAMILTON, M. J.; CHESSER, R. K.; SCRIBNER, K.T. (1987): Genetic differentia- tion among cottontails from isolated playa basins. Genetica 75, 153-157. WRIGHT, S. (1965): The interpretation of population structure by F-statistics with special regard to sys- tems of mating. Evolution 19, 395-420. Authors’ address: FERNANDO A. CERVANTES, J. PABLO RAMIREZ-SILVA, ARIADNA MARIN, and GLO- RIA L. PORTALES, Departamento de Zoologia, Instituto de Biologia, Universidad Nacional Autonoma de Mexico, Apartado Postal 70-153, Coyoacan, Mexico City, D. F. 04510, Mexico. Z. Säugetierkunde 64 (1999) 363-370 © 1999 Urban & Fischer Verlag SAUGETIERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Allosuckling behaviour in Ammotragus By J. CASSINELLO Estacion Experimental de Zona Aridas, Almeria and Departamento de Ecologia Evolutiva, Museo Na- cional de Ciencias Naturales, Madrid, Spain Receipt of Ms. 21. 12. 1998 Acceptance of Ms. 20. 08. 1999 Abstract Allosuckling behaviour is investigated in a captive population of Saharan arrui (Ammotragus lervia sahariensis). Allosuckling attempts are sporadic and usually unsuccessful (75%). The age of alien calves is strongly related to the age of allomother calves, and small differences in age between them are associated with successful allosuckling attempts. Young calves are also more successful when at- tempting allosuckling, although they undergo the most aggressive responses by the allomother. On the other hand, the two adoption instances observed took place in newborn calves. Neither the sex of the calf nor his/her coefficient of relationship with the allomother has any influence on the success of allosuckling. A series of hypotheses ıs discussed in the light of the results obtained; such as mistaken identification of the mother, kinship and crowded captive conditions. Key words: Ammotragus, ungulates, allosuckling, adoption, maternal investment Introduction Adoption has been reported in Ovis and Capra genera (HERSHER et al. 1963; GUBERNICK 1980) characterized by strong mother-infant bonds (e.g. SCHALLER 1977; GUBERNICK, 1981; ROMEYER et al. 1993). Non-offspring nursing has been observed occasionally in both domestic and wild sheep (Hass 1990; see also RowELL 1991); allosuckling behaviour re- ported in the bighorn, Ovis canadensis (Hass 1990) and roe deer, Dama dama (BiIRGERS- son et al. 1991); and both casual allosuckling and adoption are commonly considered to occur in the bison, Bison bison (McHuch 1958; ALTMAnN 1963; LotrT 1972), and water buffalo, Bubalus bubalis (TuLLocH 1979; MURPHEY et al. 1991). Whenever alloparental care refers to allosuckling, occasional or frequent, costs for the donor (allomother) rise considerably (see CLUTTON-Brock et al. 1989). Recently, PACKER et al. (1992) showed that non-offspring nursing is increased by captivity and in mamma- lian species that have large litters. These authors also pointed out that in monotocous taxa (those typically giving birth to only one infant) allosuckling is generally rare and females tend to be less tolerant (PAckER et al. 1992). The present study analyses both allosuckling and adoption in an African ungulate, the Saharan arrui (Ammotragus lervia sahariensis), a monotocous species wıth only one twin birth every 4.4 singles (CAssıneLLo and ArLApos 1996). The aim of this study is to clarify the incidence of allosuckling behaviour ın an ungulate social group, attempting to deter- mine whether it is related to mistaken identifications of the mother by the calf, or pro- moted by either kinship or crowded conditions. 0044-3468/99/64/06 - 363 $ 12.00/0 364 J. CASSINELLO Material and methods Data were collected in a captive population of Saharan arrui which is successfully breeding in captiv- ity at the Estaciön Experimental de Zonas Arıdas (EEZA), Almeria, south of Spain (CAssınELLo and ArApos 1996). This population originates from just one male and one female captured in 1975 in the western Sahara. It is suspected that the subspecies is already extinct in the wild (ALADos and VERICAD 1993). Sampling was carried out from 1990 to 1992 in a herd made up of 17 males and 26 females at the beginning of the study, and 33 males and 43 females at the culmination of the study. The total sur- face available to the animals was 950 m’. Detailed information on sampling method and routine can be found in CAssınELLo (1996), although a general description has been added here. The anımals were identified by means of coloured plastic ear tags, the differing position and shape of these tags determined a number for each individual. Birth date, parturition type (single or twin), sex, identity of father and mother, and inbreeding coefficient were known for each individual. Sam- pling was carried out during evenings, when females and calves were more active and the great major- ity of suckling events took place. Focal sampling was used to record mother and calf behaviour (ALT- MANN 1974; MARTIN and BATEson 1986), each sample being 20 min duration. A sampling period was the total number of focals carried out on a given day. Every female which gave birth during 1990 and 1991 was sampled four times a week during the calf’s first two months of life; during the remainder of the lactation period (for the weaning process, see CAssınELLO 1997a) sampling was carried out 1.5 times per week. All mother-calf interactions between them and the other group mates were re- corded. Suckling events were sampled ad libitum. A total of 26 mother-calf pairs was sampled; 8 calves shared their nursing with a sibling and 18 were single. Only allosuckling attempts directed towards lactating females were taken into account. For the sake of clarity, throughout this study a calf attempting a suckling event on a lactating female other than his/her actual mother is named an alien calf, and the lactating female an allomother. In relation to the females’ responses to the allosuckling attempts, four behaviours were distinguished: aggression, withdrawal, no response and adoption. No response means the allomother did not react when faced with an alien calf’s allosuckling attempt, but this lack of response did not imply a successful attempt, as it might fail due to other factors, such as the lack of alien calf’s determination while the allomother is engaged in other activities, ı.e. feeding; in addition on one occasion a wire fence prevented the alien calf (no. 192) from allosuckling. Also, successful attempts could be followed by allomother rejection (aggression or withdrawal) once she is aware of the alien calf’s presence, which might occur a few sec- onds later, enough time to score a successful event. The coefficients of relationship (c.r.) between alien calves and allomothers were calculated fol- lowing MACIEJOwskI and ZIEBA (1982). The social ranks were calculated for all the lactating females, and the rank given to a particular individual corresponded with the percentage of individuals with a lower dominance status (CAssınELLo 1995). When parametric tests were used, non-normal dependent variables were transformed according to, e.8., ZAR, (1984). Replication was tackled by means of the analysis of intra and intergroup variance, which showed for all the response variables that the inter- group variance was not greater than the intragroup variance, so that all the data were considered as independent. Results Allosuckling, carried out by 10 different calves, occurred sporadically in the population under study: only 20 events were registered during the whole sampling period, two of them leading to adoptions (Tab. 1). Taking into account the first suckling attempt which led to an adoption as an allosuckling one, the percentage of occurrence of allosuckling behaviour was as follows: 3.3% of the whole suckling behaviour (20 out of 611 events re- gistered), 1.4% of the successful suckling events (5 out of 357), and 5.9% of the unsuc- cessful ones (15 out of 254). Allosuckling took place at very different ages, from 0 to 463 days-old (Tab. 1), although the median is 19 days, and the mean 62 days (0 days refers to newborn individuals). A strong and positive relationship between the age of alien calves and the age of allo- mothers’ calves can be seen first (simple regression: n = 20, r” =0.51, P = 0.0004; Fig. 1). Allosuckling behaviour in Ammotragus 863 Table 1. Some characteristics of the allosuckling behaviour observed in the studied population. Re- sponse refers to the allomother’s reaction against allosuckling attempts. Key: F = female, M = male, S = successful, U = unsuccessful, Adop=adoption,. NR=no response, Agg= aggression, and W = withdrawal Calf number Calf sex Allosuckling Allomother no. Response ae la) Sa a a a Se ageeeeeegeog a aGeee %G Aa Age of alien calves SEE Oirzast 255 ES Sr Zar an TZam2 T2 25 25 Age of allomothers' calves Fig. 1. Relationship between the age of alien calves and the age of allomothers’ calves. Age shown is in days transformed into their logarithm to run the simple regression analysis. 366 J. CASSINELLO However, no relationship was found between the age of alien calves and their coefficient of relationship with the allomothers (n = 18, r* = 0.001, P = 0.89). The age of calves who successfully attempted allosuckling (25% of the total attempts) was lower than that of calves who performed unsuccessful allosuckling (successful calves: 8+4 days; unsuccessful calves: 80 +30 days; ANOVA: F(1.18) = 10.33, P = 0.005); also, successful allosuckling was observed on mothers who were nursing younger calves (4+2 days) than unsuccessful ones (77 # 18 days; F(1.18) = 11.00, P = 0.004). Furthermore, the success of allosuckling was greater when the difference in age between alien calves and allomothers’ calves was smaller (4+3vs 66+25 days; F(1.18)=5.48, P= 0.03). Neither the coefficient of relationship between alien calves and allomothers (F(1.18) = 0.02, P= 0.89) nor the allomothers’ social rank (F(1.18) = 1.66, P= 0.21) had any influence on the success of the allosuckling attemptes. Both the mean age of the alien calves and the age of the allomother were significantly different for the four allomother’s responses considered; the differences being particularly acute between calves actually adopted (newborn) and the remaining cases (see Tab. 2). The four types of responses showed similar values for the difference in age between alien calves and allomothers’ calves (F(3.16) = 0.88, P = 0.47), and the coeficients of relation- ship between alien calves and allomothers (F(3.16) = 2.93, P = 0.07). The allomothers’ type of response varıed depending on their social rank: withdrawals were carried out by low-ranking females (Tab. 3); on the contrary, neither the actual mother’s rank nor the rank difference between mother and allomother was related to the latter’s response (Tab. 3); although there is a tendency for allomothers of relatively low rank in relation to the actual mother’s rank to withdraw when faced with calf’s allosuck- ling attempts (Fisher’s post-hoc test: P = 0.06). A three-fourths majority of the calves who attempted allosuckling was females, but this number was not statistically significant (Fisher’s Exact Test: df=1, P= 0.19). Both males and females attempted allosuckling at the same average age (F(1.18) = 0.11, Table 2. Mean (+SE) age in days of both alıen and allomother’s calf in relation to the response of the allomother to the allosuckling attempt. ?ANOVA: F(3.16) = 7.15, P = 0.003; Fisher’s post-hoc test: aggression vs no response P=0.05, adoption vs the others P< 0.004. PANOVA: F(3.16) = 4.39, P = 0.02; Fisher’s post-hoc test: adoption vs aggression P = 0.008, adoption vs no response P = 0.007. Allomother’s response Alien calf age? Allomother’s calf age” Agression 28 72 +24 Withdrawal 98 3277 SD E22 No response 166 # 102.5 92 +43 Adoption 0 0908 Table 3. Mean (SE) mother’s rank, allomother’s rank, and ranks difference in relation to the re- sponse of the allomother to the allosuckling attempt. *ANOVA: F(3.16) = 0.91, ns. PANOVA: F(3.16) = 4.20, P = 0.02; Fisher’s post-hoc test: aggression vs wıthdrawal P = 0.007, wıthdrawal vs no re- sponse P = 0.02. "ANOVA: F(3.16) = 1.63, ns. Allomother’s response Mother’s rank (A)” Allomother’s rank (B)” Agression Withdrawal No response Adoption Allosuckling behaviour in Ammotragus 367 P = 0.74). No relationship was found between the sex of the alien calf and the coefficient of relationship with the allomother (F(1.18) = 0.17, P = 0.68). Finally, the success of the al- losuckling behaviour (Fisher’s Exact Test: df = 1, P= 0.27) and the allomother’s responses (Contingency table: df=3, x” = 1.56, P = 0.67) were not influenced by the sex of the alien calf. Only once did a permanent adoption event happen in the study population, where a female (no. 26) allosuckled an alien female calf (no. 149) from birth to her premature death. Just after birth, calf no. 153 was unusually adopted during one day by female no. 46; but eventually the allomother stopped nursing the calf which was then nursed by her actual mother (no. 64). Female no. 46 had deserted her twins (calves no. 148 and 149) just after birth, and one of them (no. 148) was killed by female no. 64. Both adoption events happened to female calves whose age was significantly lower than that of the calves which attempted allosuckling (F(1.18) = 14.93, P = 0.001). The coefficients of relationship with the allomothers were slightly higher in the adoption cases (0.57 + 0.005 vs 0.54 + 0.01), but the difference was not statistically significant (F(1.18) = 0.84, P= 0.37). Finally, social ranks of adoptive mothers did not differ statistically from those of non-adoptive mothers LS) =, FZUNSD))} Discussion Casual allosuckling has already been reported in Ammotraus by Katz (1949) and Haas (1959), and is considered to occur occasionally in ungulates (e.g. Hass 1990). In the study population, a strong relationship is found between the age of alien calves and the age of allomothers; successful allosuckling attempts depending on this relationship. Moreover, young alien calves are more successful when attempting allosuckling; this might be re- lated to an inefficient sort of “labelling” (sensu GUBERNICK 1980) or the mother’s ability to memorize their own calves’ signatures (PORTER et al. 1991; but see ROMEYER et al. 1993); thus, arrui mothers would recognize older calves more easily. Also the older calves’ behaviour might account for these results. Captivity conditions might promote a certain permissive behaviour towards alien in- fants, as the mothers have unlimited access to food and the costs associated with nursing are diminished (see PAckER et al. 1992); also overcrowding conditions may favour this sort of behaviour (RIEDMAN 1982). But, on the other hand, in social herbivores alien calves are usually rejected by females, a behaviour which is expected to be favoured by natural selection (GUBERNICK 1981; HARPER 1981) unless some sort of benefits came along with allomothering, such as those related to kinship selection (BERTRAM 1976). Furthermore, recent evidence on monotocous mammalian species, where most social herbivores are in- cluded, seems to show that they are poorly tolerant of allosuckling behaviour (PACKER et all y92)) Calves’ mistaken identification of their own mothers may be an explanation for allo- suckling to occur, although in some instances single alien calves (not used to share suckles) attempted allosuckling when the allomother’s own calves were already suckling. Also, allosuckling attempts might be part of calves’ learning behaviour, which may be im- plemented through negative responses. A rather speculative hypothesis which might explain the origin of this behaviour re- fers to young arrui calves seeking extra-milk intakes from lactating females other than their mothers, when young calves of similar ages can be mistaken by adult females; thus, the allosuckling behaviour may well be pursued by alien calves to obtain an additional source (MURPHEY et al. 1991); however, not enough data are available to test this hypo- thesis. 368 J. CASSINELLO On the other hand, allomothers’ responses against casual allosuckling were particu- larly aggressive when addressed to young calves; this may indicate that they represent a more serious threat to maternal resources than older calves, perhaps due to their more ef- ficient suckling behaviour (CAssınELLo 1996). Also, the hypothetical source of error com- mented above may provoke this strongly agonistic response. Allomothers who displayed the most submissive response (withdrawal) held the lowest social ranks; but the scarcity of data available does not permit any conclusion on this matter. The lack of any relationship found between the coefficient of relationship and the al- losuckling behaviour in the study population of Saharan arrui might be explained by the high inbreeding coefficients that characterize this population (e.g. CAassınEeLLo 1997b), which causes a low variance for this coefficient (c.r. range = 0.13; minimum value = 0.49; maximum value = 0.62). On the other hand, familiarity might also play an important role in the allomother’s tolerance towards alien calves (D’AmaArto 1993), but we have no data yet to support or refute such a hypothesis. It has been found in sheep that even with a high degree of familiarity (i.e. dizygotic twins), mothers can discriminate between lambs; and familiarity tended to reduce the rejection of a twin kept separated from its mother when compared with a totally alien lamb (RoMEYER et al. 1993). The low frequency of occurrence of casual allosuckling behaviour raises the question of costs and benefits to both alien calves and allomothers. It might be postulated that costs to the calf in terms of aggression, chasing, etc. tend to exceed the benefits obtained (milk intake); although no direct measure of such costs/benefits is available. As for the adoption events observed in the study population, the high degree of re- latedness between all of the individuals makes an explanation in terms of kinship selec- tion unsatisfactory. Hass (1990) reported alloparental behaviour in bighorn females which had lost their lambs 3-40 days after parturition, and RoweErr (1991) referred to a similar case in sheep. Probably these lactating females were physiologically and behav- iourally “primed” for lactation (OFTEDAL 1985). A painful, distended udder may be the impetus for nursing alien calves, although it cannot explain why these females continue to nurse calves for the entire lactation period (Hass 1990). Moreover, in sheep, as in other mammals (see GUBERNICK and KLoPrFEr 1981), there appears to be a sensitive peri- od following parturition during which ewes will be responsive to all lambs, since ewes separated from their lambs at birth will still show maternal behaviour towards any lamb within 4-8h postpartum (SımıtH et al. 1966; Poınpron et al. 1979). In addition, and although GUBERNIcK’S (1980) labelling hypothesis has been questioned recently (ROMEYER et al. 1993), maternal attachment in ungulates may take some time for the mothers to dis- criminate between their own and alien calves. In the studied Saharan arrui population, the adoptees were always newborn calves, while the allomothers had just given birth. Allomothers had not lost their calves before adoption took place, so OFTEDAL'S (1985) hypothesis cannot apply here. The adoptees were deserted by their own mothers a few minutes after birth, although in the one-day adoption this abandonment persisted for only one day. Recently, a new controversy has arisen concerning a hypothetical transmission of phe- notypic characters by means of adoption (AvıraL and JABLONKA 1994, 1996; FIANSEN 1996). In the Saharan arrui, no benefits, in terms of acquiring a higher social status, were observed (CAssınELLo 1995) during the actual adoption event; moreover, the allomother had to allocate her resources towards the adoptee and her own calf. In sum, the one-day adoption observed might be considered as a temporal allomother’s mistake or even toler- ance, and the long-lasting adoption might have been provoked by the overcrowding con- ditions (RIEDMAN 1982) or by the abandoned alien calf which managed to take advantage of a lactating female who had just given birth (SmitH et al. 1966). In conclusion, evidence of allosuckling behaviour in Ammotragus is provided, and possible adaptive interpretations postulated. Allosuckling behaviour in Ammotragus 369 Acknowledgements Thanks go to T. ABAIGAR, M. GOMENDIO, and two anonymous referees for constructive comments on earlier versions of the manuscript, and to JUAN MORENO for translating the abstract into German. Dur- ing data collection the author enjoyed a predoctoral grant from MEC (PG89 27511603), being cur- rently supported by DGES Project PB96-0880. Zusammenfassung Fremdsaugen bei Ammotragus In einer in Gefangenschaft gehaltenen Population des Mähnenspringers (Ammotragus lervia saharien- sis) wurde die Milchaufnahme von Kälbern bei fremden Müttern untersucht. Derartige Saugversuche sind selten und gewöhnlicherweise nicht erfolgreich (75%). Das Alter der fremden Kälber ist dem der eigenen Jungtiere ähnlich, doch geringe Altersunterschiede steigern den Erfolg bei Fremsaugaktivitä- ten. Wenn auch junge Kälber besonders aggressive Reaktionen der fremden Mutter hervorrufen, so sind sie doch bei ihren Versuchen besonders erfolgreich. Es wurden zwei Fälle von Adoption gerade bei neugeborenen Kälbern beobachtet. Weder das Geschlecht, noch der Grad der Verwandtschaft mit der Fremdmutter beeinflussen den Erfolg der Jungtiere bei der Nahrungsaufnahme an fremder Quelle. Im Lichte der erhobenen Befunde werden mehrere Hypothesen diskutiert: Fehl-Identifika- tion, Verwandtschaft und hohe Populationsdichte in der Gefangenschaft wurden berücksichtigt. References ALADOoSs, C. L.; VERICAD, J. R. (1993): Aoudad. Ammotragus lervia from Western-Sahara. International Studbook. Bol. Inst. Estud. Almerienses. Cienc. 11/12, 65-97. ALTMANN, M. (1963): Naturalistic studies of maternal care in moose and elk. In: Maternal Behaviour in Mammals. Ed. by H. L. RHEINGoLD. New York: John Wiley and Sons. Pp. 233-248. ALTMANN, J. (1974): Observational study of behaviour: sampling methods. Behaviour 49, 227-267. AVITAL, E.; JABLONKA, E. (1994): Social learning and the evolution of behaviour. Anim. Behav. 48, 1195-1199. AVITAL, E.; JABLONKA, E. (1996): Adoption, memes and the Oedipus complex: a reply to Hansen. Anım. Behav. 51, 476-477. BERTRAM, B. C.R. (1976): Kin selection in lions and in evolution. In: Growing Points in Ethology. Ed. by P.P.G. Bareson und R. A. HınDe. Cambridge: Cambridge Univ. Press. Pp. 281-301. BIRGERSSON, B.; EKVALL, K., TEMRIN, H. (1991): Allosuckling in fallow deer, Dama dama. Anım. Behav. 42, 326-327. CASSINELLO, J. (1995): Factors modifying female social ranks in Ammotragus. Appl. Anım. Behav. Scı. 45, 175-180. CASSINELLO, J. (1996). High ranking females bias their investment in favour of male calves in captive Ammotragus lervia. Behav. Ecol. Sociobiol. 38, 417—424. CASSINELLO, J. (1997 a): Mother-offspring conflict in the Saharan arrui. Relation to weaning and mother’s sexual activity. Ethology 103, 127-137. CASSINELLO, J. (1997 bb): High levels of inbreeding in captive Ammotragus lervia (Bovidae, Artiodacty- la): Effects on phenotypic variables. Can. J. Zool. 75, 1707-1713. CASSINELLO, J.; ALADOS, C. L. (1996): Female reproductive success in captive Ammotragus lervia (Bovi- dae, Artiodactyla). Study of its components and effects of hierarchy and inbreeding. J. Zool. (Lon- don) 239, 141-153. CLUTTON-BROocK, T. H.; ALBoN, S. D.; GuInnEss, F. E. (1989): Fitness costs of gestation and lactation in wild mammals. Nature 337, 260-262. D’Amaro, F.R. (1993): Effect of familiarity with the mother and kinship on infantıcıdal and alloparen- tal behaviour in virgin house mice. Behaviour 124, 313-326. GUBERNICK, D. J. (1980): Maternal “imprinting” or maternal “labelling” in goats? Anım. Behav. 28, 124-129. 370 J. CASSINELLO GUBERNICK, D. J. (1981): Parental and infant attachment in mammals. In: Parental Care in Mammals. Ed. by D. J. GUBERNICK und P. H. KLoprer. New York: Plenum. Pp. 243-289. GUBERNICK, D. J.; KLOPFER, P. H. (Eds.) (1981): Parental Care in Mammals. New York: Plenum. Haas, G. (1959): Untersuchungen über angeborene Verhaltensweisen bei Mähnenspringern (Ammotra- gus lervia Pallas). Z. Tierpsychol. 16, 218-242. HANnSsEN, T. F. (1996): Does adoption make evolutionary sense? Anim. Behav. 51, 474-475. HARPER, L. V. (1981): Offspring effects upon parents. In: Parental Care in Mammals. Ed. by D. J. Gu- BERNICK und P. H. KLoPFEr. New York: Plenum. Pp. 117-160. Hass, C. C. (1990): Alternative maternal-care patterns in two herds of bighorn sheep. J. Mammalogy 71, 24-35. HERSHER, L.; RICHMOND, J. B.; MoORE, A. U. (1963): Maternal behaviour in sheep and goats. In: Mater- nal Behaviour in Mammals. Ed. by H. L. RHEINGoLD. New York: John Wiley and Sons. Pp. 203-232. Karz, I. (1949): Behavioural interactions in a herd of Barbary Sheep (Ammotragus lervia). Zoologica 34, 9-18. LoTT, D. F. (1972): The way of the bison: fighting to dominance. In: The Marvels of Animal Behaviour. Ed. by P.R. MARLER. Washington, D.C.: Nat. Geographic Soc. Pp. 321-332. MACIEJOWSKI, J.; ZIEBA, J. (1982): Genetics and Animal Breeding. Part B. Stock Improvement Methods. Amsterdam: Elsevier Scientific Publ. Comp. MARTIN, P. BATESoNn, P. (1986): Measuring Behaviour. An Introductory Guide. Cambridge: Cambridge Univ. Press. McHug#, T. (1958): Social behaviour of the American buffalo (Bison bison). Zoologica 43, 1-40. MURPHEY, R. M.; PARANHOS DA Costa, M.J.R.; Lima, L. ©. S.; DUARTE, F. A.M. (1991): Communal suckling in water buffalo (Bubalus bubalis). Appl. Anım. Behav. Sci. 28, 341-352. ÖFTEDAL, O.T. (1985): Pregnancy and lactation. In: Bioenergetics of Wild Herbivores. Ed. by R. J. Hupson and R. G. WHITE. Boca Ratön, Florida: CRC Press, Pp. 215-238. PACKER, C.; LEWIS, S.; PUSEY, A. (1992): A comparative analysis of non-offspring nursing. Anim. Behav. 43, 265-281. POINDRON, P.; MARTIN, G. B.; HooLEy, R. D. (1979): Effects of lambing induction on the sensitive period for the establishment of maternal behaviour in sheep. Physiol. Behav. 23, 1081-1087. PORTER, R. H.; L£vy, F.; POINDRON, P.; LITTERIO, M.; SCHAAL, B.; BEYER, C. (1991): Individual olfactory signatures as major determinants of early maternal discrimination in sheep. Devel. Psychobiol. 24, 151-158. RIEDMAN, M.L. (1982): The evolution of alloparental care and adoption in mammals and birds. O. Rev. Biol. 57, 405-435. ROMEYER, A.; PORTER, R. H.; LEvYy, F.; NOWAK, R.; ORGEUR, P.; POINDRON, P. (1993): Maternal labelling is not necessary for the j establishment of discrimination between kids by recently parturient goats. Anım. Behav. 46, 705-712. Rowell, T. E. (1991): Till death us do part: long-lasting bonds between ewes and their daughters. Anım. Behav. 42, 681-682. Schaller, G. B. (1977): Mountain Monarchs. Wild Sheep and Goats of the Himalaya. Chicago: Univ. Chi- cago Press. SMITH, F. V.; vAN-TOLLEN, C.; Boys, T. (1966): The “critical period” in the attachment of lambs and ewes. Anım. Behav. 14, 120-125. TULLOcCH, D. G. (1979): The water buffalo, Bubalus bubalis, in Australia: reproductive and parent-off- spring behaviour. Aust. Wildl. Res. 6, 265-287. ZAR, J. H. (1984): Biostatistical Analysis. 2nd Ed. Englewood Cliffs, New Jersey: Prentice-Hall, Inc. Author’s address: JORGE CASSINELLO, Departamento de Ecologia Evolutiva, Museo Nacional de Ciencias Naturales, C/Jose Gutierrez Abascal 2, E-28006 Madrid, Spain GES: EEE, Sa Z. Säugetierkunde 64 (1999) 371-375 ZEITSCHRIFT FÜR © 1999 Urban & Fischer Verlag SAUGETIERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY WISSENSCHAFTLICHE KURZMITTEILUNGEN The maned rat, Lophiomys imhausii Milne-Edwards, 1867, in Djibouti, NE-Africa (Mammalia: Rodentia: Lophiomyinae) By D. Kock and T. KünzeL Forschungsinstitut Senckenberg, Frankfurt a. M. Receipt of Ms. 28. 05. 1999 Acceptance of Ms. 23. 08. 1999 Key words: Lophiomys, Djibouti, African distribution The long-haired fur with a contrasting black-and-white facial pattern renders the maned rat, Lophiomys imhausii Milne-Edwards, 1867, an unmistakable African rodent (WirL- LIAMS 1967; HALTENORTH and DiLLer 1977; Kınapon 1997). The genus is currently consid- ered to be monotypic, comprising a number of taxa described as species or subspecies (ELLERMAN 1940; MısonnE 1974; Musser and CARLETON 1993). On March 3rd, 1993, two L. imhausii were found (by T.K.) in the SW of Djibouti. They lay dead on road some 12 km SW of Ouea on the route from Djibouti to Ali Sabieh. The specimens (an adult and a juvenile or subadult) were not preserved, but both photo- graphed to document the presence of L. imhausii in Djibouti (Fig. 1). The habitat where the carcasses were found is a hilly area (ca. 400 m a.s.1.) with open thorn bush (Acacia mellifera, A. tortilis,; AUDRU et al. 1987) on stony ground. The biology of L. imhausii is so little known, that no convincing explanation can be offered for two individuals being acci- dentally killed close together. They may have been a female followed by an adolescent young. From distribution records available (see below) the finding of L. imhausti in Djibouti extends the species range to the west. However, the appearance of the species in Djibouti surely is not a recent event, but due to more intensively observing Djibouti wild lıfe (KünzeL and KünzeL 1998). The distribution range of L. imhausii was listed or mapped more or less accurately, to- tally or in part, by several authors. Its occurrence in Djibouti is not known (MISONNE 1974; SIMONEAU 1974; MUSSER and CARLETON 1993). However, it could possibly be that the brush-tailed porcupine, Aetherus africanus, a Central African rain forest dweller in- cluded in the Djibouti mammal fauna by SIMONEAU (1974), in reality was a /. imhausii. The type specimen of L. imhausii was bought alive in Aden (Yemen), but with the dis- covery of the species in Eritrea, the origin of the specimen was assumed to have been somewhere in “Somalia”. Until present the existence of the species in the region of the type locality Aden could not be proven. The discovery of skull fragments, dated to the 11th century, in a cave in the Judean Desert near the Dead Sea was related neither to a palaeontological nor an archaeological context (Dor 1966), and thus it cannot be ex- cluded to have been an imported animal. Further indications for the existence of L. im- hausii in Arabia (Kınapon 1990) cannot be substantiated at present. We present a detailed map of the species’ African range (Fig. 2), preferring either ori- ginal sources or reliable compilations. The species is actually known from: 0044-3468/99/64/06 - 371 $ 12.00/0 SL D. Kock and T. KÜnzEL Fig. 1. Lophiomys imhausiti, adult (above) and adolescent (bottom), dead on road, SE-Djibouti, 3. March 1993. Photos: TH. KÜNZEL. Sudan: PETERS (1867): Jebel Maman, 16.16. N-36.48.E, N of Kassala. GıGLioLi (1881): (Jebel) Eskanid, Red Sea Hills between Suakin and Sinkat. SETZER (1956): near Port Su- dan. Mısonne (1974) erroneously gives SE-Sudan (instead of NE) as the species range in this country. Records available for Eritrea and Ethiopia were mapped by YALDEN et al. (1976). Somalia: GiGLioLı (1881): Somali coast. (PEEL 1900: mentions “Sheikh” as a local- ity for L. imhausii, but does not indicate whether this is [Upper] Sheikh in Somalia or Sheikh [Hussein] in Ethiopia). DRAKE-BROCKMAN (1910): Burao; Upper Sheikh. SIMONET- TA (1963), RocHE (1976): Jesomma, 04.03. N-45.44. E, between Bulo Burti and EI Bur. Sı- Lophiomys imhausii in Djibouti, NE-Africa 373 MONETTA et al. (1978): Hargeisa and Berbera. Kenya: AnDErson and DE Wiınton (1902): Ravine Station, Mau Distr. THomas (1905): Elburgon, Mau Forest; between Londiani and Lumbwa, Mau Forest. Tuomas (1910): near Njoro, Mau Forest; Nakuru, Rift Valley; Mu- taragwa (= Ndaragwa, 00.07. S-36.37.E), 9000 ft, Aberdare Mts. DoLıman (1911): Solai, 00.07.N-36.12.E. HELLER (1912): Mt. Gargues (Uaragess), 6000 ft, Mathews Range. LÖNNBERG (1912): Mau Escarpment. KoLLMAnN (1913): Mt. Kenya, 2400 m. HOLLISTER (1919): Naivasha Escarpment; W-side Mt. Kenya, 8500 ft. GoLDFINcH (1923): Aberdare Lo L- --- - ----- -- Sudan De „Djibouti ı \ a, Fig. 2. Distribution of Lophiomys imhausii; for details see text. In the Kenyan range some symbols cov- er one to three neighbouring locality records. 374 D. Kock and T. KÜünzEL side of Nakuru. Ruxron (1926): Cherengani Hills. ARTHUR (1957): Nanyuki, W-side Mt. Kenya. Jounson (1960): Sabukia, 00.00.-36.14. E. WırLıams (1967): Mt. Elgon. GucGsis- BERG (1968): near Rongai, W of Nakuru; near Eldoret, Uasin Gishu Plateau. DELANY (1975): Trans Nzoia, 1900 m. WAHLERT (1984): Laikipia Escarpment, 0.28. N-36.07.E; Lai- kipia Forest; Nyeri; N-Abderdares; SW-Kenya. Jonnson et al. (1993): Muruku Subloca- tion, 0.35. N-36.15.E, Laikipia Distr. Uganda: TuomAs (1906) listed Z. imhausü for Ugan- da, which at that time comprised W-Kenya east to the Rift Valley, and is based on a specimen from Ravine Station, Mau Distr., collected by the then Governor of Uganda, F. J. JACKSON (see ANDERSoN and DE Wınton 1902). Kınapon (1974): Moroto, Karamoja; in map (:525) the Kidepo area in the NE is plotted [original source not traced]. Tanzania: Kınapon (1974): 525, map; 1997: 188, map: Mahali Mts., E of L. Tanganyika. As far as the fossil history of the genus Lophiomys is documented (ToPAcEVvsKI and SKORIK 1984; WAHLERT 1984; AGUILAR and MiıcHAUX 1990) it is an immigrant from Asia and its range became restricted to northeastern Africa. This could be attributed to geo- logical and climatic factors (erosion and vegetational changes). Modern records avail- able indicate that the recent species range appears to be fragmented, at least between Dibouti-Somalia and Ethiopia by the Danakil Desert, and between Ethiopia and Kenya by an extensive arid region. Furthermore, it seems that in the southern part of the spe- cies range denser forests are inhabited (Mahali Mts., Mt. Kenya, Aberdares, Mau) than in the north (Eritrea, Red Sea Hills). However, the collecting data equally indicates that more intensive search for L. imhausii may interconnect some of the known disjunct po- pulations. References AGUILAR, J. P; MicHAux, J. (1990): Un Lophiomys (Cricetidae, Rodentia) nouveau dans le Pliocene du Maroc; rapport avec les Lophiomyinae fossiles et actuels. Paleontologia i Evoluciö 23, 205-211. ANDERSON, J.,; DE WINTon, W. E. (1902): Zoology of Egypt. 2. Mammals of Egypt: Mammalia. Revised and completed by W. E. DE Winton. London: Hugh Rees. ARTHUR, D. R. (1957): The /xodes schillingsi group: ticks of Africa and Madagascar, parasitic on pri- mates, with description of two new species (Ixodoidea, Ixodidae). Parasitol. 47, 544-559. ÄUDRU, J.; CESAR, J.; FORGIARINI, G.; LEBRUN, J. P. (1987): La veg£tation et des potentiales de la Republi- que de Djibouti. Paris: Institut d’Elevage et de Medicine veterinaire des Pays Iropicaux. DELANY, M. J. (1975): The Rodents of Uganda. London: Trust. Brit. Mus. Nat. Hist. Dorıman, G. (1911): New and interesting mammals from East Africa. Ann. Mag. nat. Hist. (8) 8, 124- 132. Dor, M. (1966): Restes subfossiles de Lophiomys trouves en Israel. Mammalia 30, 199-200. DRAKE-BROCKMAN, R. E. (1910): The Mammals of Somaliland. London: Hurst and Blackett. ELLERMAN, J. R. (1940): The Families and Genera of Living Rodents. With a List of Named Forms (1758-1936) by R. W. Hayman and G. W. C. Horr. 1. Rodents other than Muridae. London: Trust. Brit. Mus. GiIGLIoLı, H. H. (1881): Lophiomys imhausi A. Milne-Edwards. Zool. Anz. 4, 45. GOLDFINCH, G. H. (1923): Notes on the African crested rat (Lophiomys imhausi). Proc. zool. Soc. Lond. 1923, 1091. GUGGISBERG, CH. (1968): My friendly pet - the rat. Kenya weekly News 2218, 23-35. HALTENORTH, TH.; DILLER, H. (1977): Säugetiere Afrikas und Madagaskars. München, Bern, Wien: Bayerischer Landwirtschaftsverlag. HELLER, E. (1912): New rodents from British East Africa. Smiths. Miscellan. Coll. 59, 1-20. HOLLISTER, N. (1919): East African mammals in the United States National Museum. H. Rodentia, La- gomorpha, and Tubulidentata. Bull. U. S. Nat. Mus. 99, 1-184. JoHnson, P.T. (1960): The Anoplura of African rodents and insectivores. Technical Bull. U.S. Dept. Agric. 1211, 1-116. JOHNSoN, R. N.; NGUMBI, P. M.; MWANYUMBA, J. P; ROBERTS, C. R. (1993): Host preference of Pleboto- mus guggisbergi, a vector of Leishmania tropica in Kenya. Med. Vet. Entomol. 7, 216-218. Lophiomys imhausii in Djibouti, NE-Africa 375 Kınsoon, J. (1974): East African Mammals: an Atlas of Evolution in Africa. 2B (Hares and Rodents). London, New York: Academic Press. Kınsoon, J. (1990): Arabian Mammals. A Natural History. Bahrain: Al Areen Wildlife Park and Re- serve. Kınsoon, J. (1997): The Kingdon Field Guide to African Mammals. San Diego, London: Academic Press. KoLLMAnN, M. (1913): Note sur les mammiferes rapportes de l’Afrique orientale par Mm. ALLUAUD et JEANNEL. Bull. Mus. natn. Hist. nat. 1913, 138-144. KünzerL, T.; KÜnZeL, S. (1998): An overlooked population of the beira antelopes Dorcatragus megalotis in Djibouti. Oryx 32, 75-80. LÖNNBERG, E. (1912): Mammals collected by the Swedish zoological expedition to British East Africa 1911. Kungl. svenska Vetensk. Akad. Handl. 48, 1-188. MILNE-EDWARDS, A. (1867): Memoire sur le type d’une nouvelle famille de l’ordre des rongeurs. Nouv. Arch. Mus. Hist nat. Paris 3, 81-116. MIsoNnNE, X. (1974): Order Rodentia. In: The Mammals of Africa: an Identification Manual. Ed. by J. MEESTER and H. W. SETZER. Washington, D. C.: Smithson. Inst. Press. Part 6, 1-39. MUSSER, G. G.; CARLETON, M. D. (1993): Family Muridae. In: Mammals Species of the World. 2nd ed. Ed. by D. E. Wırson and D. M. REEDER. Washington, London: Smithson. Inst. Press. Pp. 501-755. PEEL, C. V. A. (1900): Somaliland being an account of two expeditions into the far interior together with a complete list of every anımal and bird known to inhabit that country, and a list of the reptiles col- lected by the author. Reprint 1986, London: Dart Publishers. PETERS, W. (1867): Briefliche Mittheilung über den neuen Nager Phractomys aethiopicus gegen REI- CHERT. Z. gesam. Naturwiss. 29, 195. ROCHE, J. (1976): Capture de rongeurs appartenant aux genres Lemniscomys, Saccostomus, Steatomys et Lophiomys en Republique de Somalie. Monit. Zool. ital. (N. S.) Suppl. 8, 195-201. RuxTon, A.E. (1926): On mammals collected by Capt. C. R. S. PıTman, Game Warden, Entebbe, Ugan- da. Ann. Mag. nat. Hist. (9) 18, 28-33. SETZER, H. W. (1956): Mammals of the Anglo-Egyptian Sudan. Proc. U. S. Nat. Mus. 106, 447-587. SIMONEAU, E. L. (1974): Les animaux du Territoire Frangais des Afars et des Issas. Djibouti: E.L. Simo- neau. SIMONETTA, A.M. (1963): Ricerche sulla fauna della Somalia. Nota I. Relazione sulla prima fase delle ricerche. Ricerca sci., rendic. B (2) 3, 307-316. SIMONETTA, A.;, ROCHE, J.;, MAnnuccı, P. (1978): Primo tentativo di descrizione e di identificazione dei roditori della Republica Democratica Somala. Studi Rich. 2, 157-185. THoMmAs, O. (1905): New African mammals of the genera Glauconycteris, Lutra, Funisciurus, Arvi- canthis, Lophiomys, and Procavia. Ann. Mag. nat. Hist. (7) 15, 77-83. THomas, ©. (1906): List of mammals known to occur in the Uganda Protectorate. In: The Uganda Pro- tectorate. Ed. by H. Jonnston. London: Hutchinson. THoMmas, O. (1910): Notes on African Rodents. II. The species of Lophiomys. Ann. Mag. nat. Hist. (8) 6, 222-224. TOPACEVSKI, V. A.; SKORIK, A. F. (1984): The first find of a Lophiomyinae (Rodentia, Cricetidae) fossil representative. Vestnik Zool. 1984, 57-60 (in Russian). WAHLERT, J. H. (1984): Relationships of the extinct rodent Cricetops to Lophiomys and the Cricetinae (Rodentia, Cricetidae). Novit. Amer. Mus. nat. Hist. 2784, 1-15. WILLIAMS, J. G. (1967): A Field Guide to the National Parks of East Africa. London: Collins. YALDEN, D. W.; LARGEN, M. J.,; Kock, D. (1976): Catalogue of the mammals of Ethiopia. II. Insectivora and Rodentia. Monit. Zool. ital. (N. S.) Suppl. 8, 1-118. Authors’ addresses: Dr. DIETER Kock, Forschungsinstitut Senckenberg, Senckenberg-Anlage 25, D-60325 Frankfurt a. M.; THoMAs KÜNnZEL, Schrevenborner Weg 28, D-24226 Heikendorf Z. Säugetierkunde 64 (1999) 376-379 © 1999 Urban & Fischer Verlag SÄUG ETI KUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Faecal analysis of the edible dormouse (Glis glis) in the northwest Iberian Peninsula By A. GisirEY and J. M. Rey Departamento de Biologia Animal, Facultad de Biologla, Universidad de Santiago de Compostela, Santiago de Compostela, Spain Receipt of Ms. 15. 04. 1999 Acceptance of Ms. 14. 07. 1999 Key words: Glis glis, dıet, faecal analysis, Galicia, Spain The edible dormouse (Glis glis Linnaeus, 1766) is a tree-dwelling and exclusively crepus- cular/nocturnal rodent. To date, there have been few studies of its diet in optimal habitats, and all such studies have been based either on analysis of stomach contents (KAHMANN 1965; HoLısovA 1968; CASTROVIEJO et al. 1974; GIGIREY and Rey 1998) or on field observa- tions, which require considerable effort and are very difficult in view of the species’ habits (VIETINGHOFF-RIESCH 1960; RopoLrı 1994). The aim of this study is to investigate the diet of G. glis through faecal analysis. The study was carried out in the Parque Natural de Invernadeiro (Galicia, NW Spain). The study area (1000-1200 m) is a 4.1 ha area of mixed broad-leafed woodland, mainly of Ouercus robur, with a well-developed understorey. The diet studies were based on the analysis of 293 droppings collected in the study area in 1997, over the period June-October inclusively. All droppings were obtained in- side or on top of nest-boxes specifically designed for G. glis. The model is similar to that of Morris et al. (1990) for Muscardinus avellanarius, but larger in size. A total of 47 nest-boxes were distributed throughout the study area at regular intervals of about 25 m, at a height of 2-3 m above ground, and they were checked monthly. Since nest-boxes were not occupied until July, in May and June it was necessary to use Sherman traps, baited with apple and peanut butter, to obtain the droping. It should be stressed that this may have had some influence on the results. The method used for faecal analysis fol- lowed Warts (1968), Hansson (1970), and RıcHARDS et al. (1984). Samples were grouped into batches, each batch comprising all the droppings collected from a given nest-box or Sherman trap in a given month. A total of 21 batches of droppings was examined; each batch was pooled homogenized, and a total of 5 slides was prepared; within each slide, a to- tal of 100 fields of view was examined at 40x, recording the food remains present in each field. Food remains were identified with the aid of a reference collection. Pollen, spores, and remains of the bait were not recorded in droppings from Sherman traps. The results revealed a basically herbivorous diet (Tab. 1), with a marked variation over the activity period, as summarized in figure 1. The fleshy fruits detected were mainly blackberry and apple, and to lesser amounts bilberry and rowan-berry. Nuts included acorns and hazelnuts. Leaves identified were mostly from Rubus ulmifolius; other species were Quercus robur, Betula celtiberica, and Ilex aquifolium. Flowers were not identified at the species level. Animal-prey remains were exclusively insects (hymenoptera, coleop- 0044-3468/99/64/06 - 376 $ 12.00/0 Faecal analysis of Glis glis Sl Table 1. Diet of the edible dormouse in the Montes do Invernadeiro: pooled results for all 21 batches of droppings. N = total number of remains of the food type detected. %F = percentage frequency (N as a percentage of total N). Fleshy fruit Nuts Leaves/flowers Arthröpods Fungi Briophytes 100 Fl fruit M Nut N DO Fung Fig. 1. Changes in the frequency of consumption of the different food types over the activity period (June-October). Values shown are frequencies of occurrence (number of batches of droppings contain- ing that food type, as a percentage of the total number of batches in that month). tera, and hemioptera), and/or arachnida. Fungus remains were mostly ascomycetes of the genus Elaphomyces. The consumption of moss was probably accidental. These results are in accordance with previous reports. In Italy, KAHmAnn (1965) found that the June/July diet comprised plant remains, insects and Rubus flowers, whilst in July/ August seeds of hornbeam, nuts, and blackberies. In Czechoslovakia, HorısovA (1968) found that the early-summer diet comprised vegetative plant structures; in late-summer ve- getative plant structures, fungi, hazelnuts, and dogwood fruits; and in autumn principally dogwood fruits, hawthorn fruits and sycamore seeds. In the Iberian Peninsula, CASTROVIEJO et al. (1974) found that the June/July diet was comprised of insects, leaves, and fruits, the August diet leaves, fruit, and nuts, and the September/OÖctober diet fruit and nuts. During May we did not find faeces either in nest-boxes or in traps. This suggests that in our study area the activity period does not begin until June, as has been reported for other regions (KAHMANN 1965; GAISLER et al. 1977; RoDoLrı 1994). During September and particularly October (pre-hibernation period), the diet is dominated by nuts. This can be attributed to the need to accumulate the body fat re- quired to survive the winter; indeed, their abundance may be an important determinant of population density (VIETINGHOFF-RIESCH 1960; CASTROVIEJRO et al. 1974; STORCH 1978). 378 A, GIGIREY and J. M. Rey We have not found any evidence of storage of food in larders, previously reported by KoenıG (1960). However, food appears to have been P brought to and eaten in nest- boxes, which may be a predator-avoidance behaviour. Our results also indicate that fruits are eaten regardless of their degree of ripeness (bilberries in June and hazel nuts in Au- gust), as has been reported previously by SyKorA (1970) and Roporrı (1994). An evidence of food preference was provided by wild apple: only a single apple tree is present in the study area, but during August this food item constituted a major part of the diet, and in some cases we can infer that the dormouse moved 200 m to reach the tree. This preference has been noted in previous studies (THompson 1952; MORRIS and Hoopıess 1992). It seems obvious that dormice actively select apples as a source of car- bohydrates, once the reproductive period has finished. The large amounts of arthropod remains detected in July may be a response to the high energy demand over the sexual activity, in view of the fact that no energy-rich plant foods are available at this time, as previously suggested by Franco (1990). We did not de- tect any evidence of vertebrate prey, as has been cited in previous studies (VIETINGHOFF- RıEscH 1960; STORCH 1978; ROBEL and LEITENBACHER 1993). The genus Elaphomyces, comprises fungi with below-ground fruiting bodies, implying that dormice must have dug in the soil to reach this food source. It is possible that this re- flects deliberate searching; alternatively, and particularly during September and October, this food item may have been found during searching for suitable hibernation sites. Results from September and October are similar to those obtained by analysis of stomach contents of individuals captured in the same area (GiGIREY and Rey 1998); but faecal analysis gave lower nut estimates. This difference can be attributed to the effi- ciency of the edible dormouse in digesting nuts for conversion into body fat (GEBCZYNSKI et al. 1972), leading to under-estimation in faeces. Acknowledgements We thank FRANcISCO DE DIEGO CALONGE, of the Real Jardin Botänico in Madrid, for help with identi- fication of fungi, and the Servicio de Medio Ambiente of the Xunta de Galicia for facilitating our fieldwork. This work was financed by projects no. XUGA 20011B90 and XUGA 20011B96. References CASTROVIEJO, J.; GARZON, J.; PALACIOS, F.; CASTROVIEJO, S. (1974): Sobre el lirön gris (Glis glis pyrenaicus Cabrera, 1908) en Espana. Donana, Acta Vert. 1, 121-142. FRANco, D. (1990): Feeding habits of a dormouse population (Myoxus glis) of the Asiago Plateau (Ve- netian Prealps). Hystrix 2, 11-22. GAISLER, J., HoLas, V.; HoMoLKA, M. (1977): Ecology and reproduction of Gliridae (Mammalia) in northern Moravia. Folia Zool. 26, 213-228. GEBCZYNSKI, M.: GORECKI, A.; DROZDZ, A. (1972): Metabolism, food assimilation and bioenergetics of three species of dormice (Gliridae). Acta Theriol. 17, 271-294. GIGIREY, A.; Rey, J. M. (1998): Autumn diet of the edible dormouse in Galicia, Northwest Spain. Acta Theriol. 43, 325-328. Hansson, L. (1970). Methods of morphological diet micro-analysis in rodents. Oikos 21, 255-266. Horısova, V. ( 1968): Notes on the food of dormice (Gliridae). Zool. Listy 17, 109-114. KAHmann (1965): Le loir (Glis glis L. 1776) dans les Monts Gargano Italie (Apulie). Mammalia 29, 72- 94. Koenig, L. (1960): Das Aktionssystem des Siebenschläfers (Glis glis L.). Z. Tierpsychologie 17, 427- 505. MoRrrıs, P. A.; Hoopıess, A. (1992): Movements and hibernaculum site in the fat dormouse (Glis glis). J. Zool. (London) 228, 685-687. Faecal analysis of Glis glis 379 MORRIS, P. A.; BRIGHT, P. W.; Woops, D. (1990): Use of nestboxes by the dormouse Muscardinus avella- narius. Biol. Conserv. 51, 1-13. RICHARDS, C. G.; WHITE, A. C,; HURRELL, E.; PRICE, E. F. (1984): The food of the common dormouse, Muscardinus avellanarius, in South Devon. Mammal Rev. 14, 19-28. ROBEL, K.; LEITENBACHER, G. (1993): Der Einfluß des Siebenschläfers Glis glis auf die Höhlenbrüterpo- pulation in künstlichen Nisthöhlen am Surspeicher. Orn. Anz. 32, 59-63. RoporrI, G. (1994): Dormice, Glis glis, activity and hazelnut consumption. Acta Theriol. 39, 215-220. STORCH, G. (1978): Glis glis (Linnaeus 1776) — Edible dormouse (Siebenschläfer). In: Handbuch der Säugetiere Europas. Vol. 1/I, Ed. by J. NIETHAMMER und F. Krapp. Wiesbaden: Akad. Verlagsges. Pp. 243-258. SYKORA, B. (1970): Erfahrungen mit Methoden zum Nachweis des Siebenschläfers (Glis glis L.) in den Waldgebieten des Altenburger Landes. Abh. Ber. Naturk. Mus. Mauritianum 6, 227-233. THomPpson, H. (1952): The edible dormouse (Glis glis L.) in England, 1902-1951. Proc. Zool. Soc. Lond. 122, 1017-1024. VIETINGHOFF-RIESCH, A. (1960): Der Siebenschläfer (Glis glis L.). Monographien der Wildsäuge- tiere, 14. Jena: Gustav Fischer Verlag. WATTS, C. H. S. ( 1968): The foods eaten by wood mice (Apodemus sylvaticus) and bank voles (Clethrio- nomys glareolus) in Wytham Woods, Berkshire. J. Anim. Ecol. 37, 25-41. Authors’ address: ANTONIO GIGIREy and Jos£ M. Rey, Departamento de Biologia Animal, Facultad de Biologia, Universidad de Santiago de Compostela, Campus Sur s/n, 15706 San- tiago de Compostela, A Coruna, Spain. E-mail: batonho@usc.es. Z. Säugetierkunde 64 (1999) 380-382 Re = FÜ © 1999 Urban & Fischer Verlag SÄUGETI ERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY MITTEILUNG DER GESELLSCHAFT Protokoll über die Mitgliederversammlung der Deutschen Gesellschaft für Säugetierkunde e. V. am 27. September 1999 im Bad Dürkheimer Haus, Bad Dürkheim Der 1. Vorsitzende, Herr ERKERT, eröffnet die Versammlung um 16.10 Uhr. 1. Die Tagesordnung wird angenommen. 2. Herr SCHRÖPFER verliest den Bericht über das Jahr 1998. Auf Einladung der Herren Prof. Dr. S. C. KAREL STULIK, Prof. Dr. V. Hanak und Herrn Doz. Dr. LEO SIGMUND, Lehrstuhl für Zoologie, fand die 72. Jahrestagung der Deutschen Gesellschaft für Säugetierkunde e.V. vom 20.-24. September 1998 an der Karls-Universität zu Prag statt. In 81 Vorträgen und Poster-Präsentationen zu den Schwerpunktthemen „Biodi- versitätsforschung“, „Morphologie und Phylogenie“ und „Biologie der Insektivora und Chiroptera“ sowie zu freien Themen bekamen ca. 120 Teilnehmer ein inhalts- reiches Programm geboten. Eine wissenschaftliche Exkursion galt dem Zoo von Dvur Kralove, eine weitere dem Thaya-TIal. Beide waren eine sinnvolle und anre- gende Ergänzung des Vortragsprogramms. Herr SCHRÖPFER Spricht den Veranstaltern, ihren Mitarbeiterinnen und Mitarbeitern seinen herzlichen Dank für die Gast- freundschaft aus. Alle Teilnehmer werden diese Tagung, die im Rahmen des 650jäh- rigen Jubiläums der ehrwürdigen Karls-Universität stattfand, in guter Erinnerung be- halten. Die Preise des Poster-Wettbewerbs der Prager Tagung gingen an 1. C. WEBER, K. KUGELSCHAFTER, R. FRANK: Spätsommerliches Erkunden von Baum- höhlenwinterquartieren durch juvenile Abendsegler (Nyctalus noctula), 2. K. JoHANNESSON-GRoss: Zur Grabaktivität des Maulwurfs (Talpa europaea): Ab- messungen von Erdhügeln und ihre Größenverteilung im Jahresverlauf. 3. H. AnsorcGe: Biologische Daten des Marderhundes aus der Oberlausitz. Den Fritz-Frank-Förderpreis erhielt Dipl.-Biol. Dr. SIMONE SOMMER für ihre Forschun- gen über die Populationsökologie und -genetik von Hypogeomys antimena, einer en- demischen Nagerart im Trockenwald Westmadagaskars. Band 63 der Zeitschrift für Säugetierkunde erschien in sechs Heften mit insgesamt 384 Seiten. Er enthielt 30 wissenschaftliche Originalarbeiten, 16 wissenschaftliche Kurzmitteilungen, vier Mitteilungen der Gesellschaft und fünf Buchbesprechungen. Den beiden Schriftleitern und den Mitarbeitern des wissenschaftlichen Beirates wird ebenso gedankt wie dem Verlag. Außerdem wurde das Supplementheft herausgegeben, das über die wissenschaftlichen Beiträge der 72. Jahrestagung der Gesellschaft in Prag Auskunft gibt. Es enthält auf 69 Seiten 81 Kurzfassungen von 134 Autoren. Am 31. 12. 1998 hatte die Gesellschaft 601 Mitglieder. 0044-3468/99/64/06 - 380 $ 12.00/0 Mitteilung der Gesellschaft 381 Durch Tod verlor die Gesellschaft folgende Mitglieder: Prof. Dr. WALTER PFLUMM, Mitglied seit 1974 Herrn HENDRIK VAN DEN BERGH, Mitglied seit 1973, Herrn WALTER PoLZın, Mitglied seit 1930. (99) . Herr SCHRÖPFER erläutert den von Frau KÜnnrich vorgelegten Kassenbericht und dankt der Schatzmeisterin für ihre umsichtige Arbeit. 4. Die Herren BOHLKEN und SCHLIEMANN haben die Kontounterlagen, der Gesellschaft in Hamburg geprüft und für korrekt befunden. nn . Die Anträge auf Entlastung der Schatzmeisterin und des Vorstandes werden bei vier Enthaltungen angenommen. 6. Die Herren BOHLKEN und SCHLIEMANN werden als Kassenprüfer für das Jahr 1999 einstimmig wiedergewählt. Beide sind mit der Wahl einverstanden. —SI . Der Vorstand schlägt vor, die Mitgliedsbeiträge für das Jahr 2000 unverändert zu las- sen. Die Anwesenden beschließen dies mit großer Mehrheit. 0,0) . Die Mitgliederversammlung beschließt, die 74. Jahrestagung vom 24.-28. 9. 2000 ge- meinsam mit der holländischen Säugetiergesellschaft VZZ in den Niederlanden abzu- halten. Der Tagungsort wird rechtzeitig mitgeteilt werden. Als Schwerpunktthemen sind vorgesehen „Meeressäugetiere“, „Biologie und Schutz der Fledermäuse“, „So- ziobiologie, Sozioendokrinologie, Stress“. Für das Jahr 2001 laden Herr ZELLER und Herr FrÄDrich nach Berlin ein (75 Jahre DGS). Im Zusammenhang mit dem Tagungsinhalt wird die Frage erörtert, ob es sinnvoll ist, wie bisher drei Schwerpunkt- themen zu benennen oder künftig nur zwei, wodurch den Freien Themen ein größerer Raum zukommen würde. 9. Herr REHKÄMPER stellt sein Konzept für die Arbeit der Tierschutzkommission vor. Zur Zeit ist vorgesehen, folgende Themen zu behandeln: Tierversuche, Tierschutz und Grundgesetz, Zootiere, Haustiere. In der anschließenden Diskussion wird deut- lich, daß einige Mitglieder den Themenkreis erweitert sehen möchten, doch soll aus praktischen Gründen vorerst darauf verzichtet werden. Herr PeErz berichtet kurz über die Aktivitäten der Artenschutzkommission und weist auf die Notwendigkeit hin, den juristischen Status des Fischotters zu verändern. Herr SCHRÖPFER gibt die Bildung einer Mustelidengruppe bekannt. 10. Da für eine intensive Diskussion der Fragen, die den künftigen Titel, das Format und Layout sowie den Inhalt der Zeitschrift für Säugetierkunde betreffen (englischer Un- tertitel oder Titel, Internationalisierung der Zeitschrift) die Zeit nicht mehr ausreicht, wird beschlossen, die Sitzung zu unterbrechen und die weitere Behandlung von TOP 10 am 28. 9. 1999 um 12.00 Uhr fortzuführen. Die Sitzung endet um 18.50 Uhr. Am folgenden Tag eröffnet der 1. Vorsitzende, Herr Erkert, die Fortsetzung der Ver- sammlung um 12.15 Uhr. Nach intensiver Diskussion über die Gestaltung der Zeitschrift für Säugetierkunde empfiehlt die Mitgliederversammlung mehrheitlich (34 Ja-Stimmen, 0 Gegenstimmen, 2 Enthaltungen), auf der nächsten Frühjahrssitzung des erweiterten Gesamtvorstandes die notwendigen Änderungen in einem „Gesamtpaket“ einvernehmlich mit dem Verlag zu beschließen. 382 Mitteilung der Gesellschaft Die Mitgliederversammlung beschließt mit drei Enthaltungen, daß das Logo der Ge- sellschaft auf der Titelseite erscheinen soll. Die Titelseite soll eine Abbildung tragen (angenommen bei 4 Gegenstimmen und ei- ner Enthaltung). Das Format der Hefte soll nur dann geändert werden (zweispaltiger Druck), wenn die Änderung kostenneutral ausfällt (16 Ja-Stimmen, 12 Gegenstimmen, 7 Enthaltungen). Es wurde angeregt, im Hinblick auf eine Abbildungskartei für die Titelseite der Zeit- schrift im Jahr 2001 in Berlin eine Bilderausstellung durchzuführen. Die Sitzung endet um 13.18 Uhr Prof. Dr. H. Erkert Prof. Dr. R. Schröpfer Dr. H. Frädrich (1. Vorsitzender) (Geschäftsführer) (Schriftführer) HR = Z. Säugetierkunde 64 (1999) 383-384 ZEITSCHRIFT ®=* FÜR © 1999 Urban & Fischer Verlag | SÄUGETIERKUNDE http://www.urbanfischer.de/journals/saeugetier INTERNATIONAL JOURNAL OF MAMMALIAN BIOLOGY Buchbesprechungen WöJCcıK, J. M.; WoLsAan, M. (eds.) (1998): Evolution of Shrews. Bialowieza: Mammal Research Insti- tute, Polish Academy of Sciences. Hardcover, 458 pp. num. figs. and tabs. US $ 38.-. ISBN 83-907 521-0-7. The aim of this book is to review recent knowledge on the evolutionary biology of shrews and to point out current problems. This comprehensive multi-authored compilation is presented in altogether 13 chapters, well-ordered, dealing with several aspects, and written by internationally well-known ex- perts in their fields. Following introductory editorial remarks, including brief conclusions of the differ- ent themes, the first chapter (REUMER) deals with the classification of these special and species-rich in- sectivores. Here, in addition to older opinions, a new approach is presented. Accordingly, certain extinct genera with zygomatic arches and further skull characters, formerly ranked as subfamily, are now classified in the family Heterosoricidae, whereas the vast majority of fossil and recent shrews lacking zygomatic arches and attributed with other distinct skull peculiarities is supplemented under the family Soricidae with 5 subfamilies. The numerous genera and species of the subfamily Soricinae are further assigned to 7 tribes. Most of the other authors of this book follow strictly these conclu- sions. The next chapters deal with the history of shrews documenting the situation in Europe (RZEBIK- KowALsKA), Asia (STORCH, QIU, ZAZHIGIN), Africa (BUTLER), and North America (HARRIS). No fossils were recorded from South America and only a few Cryptotis-species are recognized from northern parts of this continent in recent distribution. Thus, late immigration from the north appears probable. According to the fossil documentation Heterosoricidae were obviously distributed from Middle and Late Eocene to Middle and Late Miocene in North America respectively Europe and from Early Olı- gocene to Late Miocene in Asia, whereas Soricidae were first recognized from Early Oligocene in Europe (with the Soricinae being “older” than Crocidurinae), from Late Oligocene in North America, and from Early Miocene to present in Asıa (with the Soricinae and Crocidurinae likewise since Mid- dle Miocene). The history in Africa is poorly documented for only some Soricidae from the Middle Miocene on. This is discussed in connection with the origin of shrews on land masses of the northern hemisphere and the fact that Tertiary Africa and Arabia were separated from Eurasia by the Tethys Sea. A land bridge connecting Arabia with southeastern Asia first emerged during Middle Miocene. Thus, since these eras immigrations of shrews and faunal exchanges between Africa and Eurasia are supposed. A further chapter (DAnnELID) deals with dentition, especially that of extant genera but data on extinct forms are included. Both ecological and phylogenetical conclusions are presented very convincingly in detail and in overview as well. Evolutionary and several convergent adaptive trends in Soricidae are focused concerning tooth reductions, tooth modification, and pigmentation. Two subse- quent chapters are devoted to the chromosomal configuration in shrews. At first (ZimA, LUKACOVA, MACHOLAn) a general overview documents present knowledge on 52 species of Crocidurinae with di- ploid numbers ranging from 22 to 60 and 45 species of Soricinae with a range from 20 to 68. These are commented on in their basic karyotypes and intra- as well as interspecific variations. The unusual heterosomal system known from 8 Sorex species is also stressed with males possessing XYıY>, while females have the normal XX. Then, a special chapter (SEARLE, WöjJcıK) is centered on the phenome- nal chromosomal variability of the Robertsonian type in Sorex araneus. This species varies remarkably in diploid number from 20 to 33 at a constant fundamental number of 40. Several karyotypic races are characterized and possible phylogenetic relationships between these are reconstructed with special emphasis on interracial hybrid zones. Two consecutive chapters are devoted to results obtained from molecular methods. Protein variation (RuEDI) is focused at the specific and generic level of some Cro- cidurinae and Soricinae, and mt DNA diversities (HAUSSER, FUMAGALLI, TABERLET) in shrews are dis- cussed as an additive tool for phylogenetic reconstruction among western European species and with- in karyotypic races of the S. araneus group. Physiological characteristics are also presented in a review (TAyYLoR) concerning evolutionary patterns and energetics. Here, the generally higher metabolic rate levels and strict homeothermy of soricines are contrasted with lower rates in crocidurine species and their ability to enter torpor. These generally different adaptations are discussed in connection with 0044-3468/99/64/06 - 383 $ 12.00/0 384 Buchbesprechungen geopraphical origin, phylogeny, environmental adaptation, reproductive traits, and other biological parameters. The two last chapters are devoted to ethology, e.g., social organisation (RyCHLIK) and mating biology (STOCKLEY, SEARLE) documented as results from observations in the field and in captiv- ity as well. Four social systems are described in extant species in relation to communication, mating systems, rearing of young, predation avoidance, habitat use, etc. The mating systems are documented as different adaptıve radıations concerning oestrus, mating behaviour, dispersal and spatial organiza- tion. Finally, an appendix (WoLsAn, HUTTERER) offers an updated list and systematics of the known 335 species in 23 genera with their common names, distribution, and brief habitat characterisation. A taxonomic index is also added. This book certainly deserves close attention not only by specialists but also mammalogists in gen- eral. It contains a very large amount of details on the biology of these mammals and many aspects to evaluate and understand phenomena and characteristics. It is also intended to stimulate further re- search. D. KruskAa, Kiel Kııma, M. (1999): Development of the Cetacean Nasal Skull. In: Advances in Anatomy, Embryology and Cell Biology, Vol. 149. Berlin, Heidelberg, New York: Springer-Verlag. Softcover, 143 pp., 68 illus- trations, 1 table. DM 186,- / US$ 119,-. ISBN 3-540-64996-4. The author of this book on the nasal skull of the Cetacea was able to investigate embryological mate- rial from seven species of the Odontoceti (toothed whales) and three species of the Mysticeti (baleen whales). Because of the general rarity of appropriate cetacean material in collections around the world, the reader has to agree with the author’s self-confident statement that his “contribution, even though incomplete, is probably the most thorough treatment of the topic for some time to come”. After a section dealing with the prenatal differentiation of the nasal skull, emphasis is paid to changes in position and form of structures in this part of the cranial skeleton, followed by an interspe- cific comparison of structures in embryonic and adult Cetacea. These sections are illustrated by very informative drawings and micrographs. Relevant structures are marked exclusively by abbreviations, which are listed and explained in a table at the beginning of the book. To appreciate the illustrations properly, the reader has to thumb through page upon page, a rather cumbersome procedure! In the subsequent section the author deals with the question why the nostrils in whales “lie in the highest point of the cetacean body”, but in the fossil ichthyosaurus the “nostrils were situated laterally on the skull, directly in front of the eyes”. This difference may be related to the ventro-dorsal move- ment of the cetacean fluke and the right-left motion of the ichthyosaurous tail. In a final section KLıma comments on the systematics within the order Cetacea. His findings do not contradict the modern view that sperm whales have to be classified in a separate subfamily (Phy- seteroidea) from the other toothed whales, the Delphinoidea, and the baleen whales, Balaenopteroi- dea. The illustrations for this important section are taken from previous papers (KLımA 1995 and Mı- LINKOVITCH, 1995). The editors should have marked homologous anatomical structures by clear signatures (Fig. 67) and should have removed redundant abbreviations (Fig. 68). P. LANGER, Gießen Instructions to Authors Submission and Acceptance of Manuscripts: Manuscripts for publication should be sent to the managing editor, Prof. Dr. D. Kruska, Institut für Haustierkunde, Christian-Albrechts-Universität, Olshausenstr. 40-60, D-24118 Kiel, Germany, e-mail: dkruska@ifh.uni-kiel.de. Acceptance of the manuscript follows the bylaws of the German Society for Mammalogy (Deutsche Gesellschaft für Säugetierkunde). Receipt of the manuscript will be confirmed immediately by mail, and as soon as the peer reviews are received the authors will be informed concern- ing the decision for acceptance. All correspondence dealing with details of production should be sent to: Zeitschrift für Säugetierkunde, Urban & Fischer Verlag, Löbdergraben 14a, D-07743 Jena, Germany. Form of the Manuscript: Each manuscript should be submitted in triplicate, one original and two copies, in English or German in DIN A4 format either from a typewriter or legible computer print with a left-hand margin of 4cm and true double spacing. The first page of the manuscript should contain the following information: title name(s) of author(s): full first name for female authors, first initials only for male authors department and university affiliation In case of long titles, a running title not exceeding 72 characters should be provided. Script type should be uniform throughout the manuscript. Use of bold print, italics and spaced-letters must be avoided. Page footnotes and appendices at the end of the text are not allowed. Names of authors cited in the text are to be underlined in pencil. Botanical and zoological generic and species names should be underlined in pencil with a wavy line, whereby the name of the scientist who first described the species is not underlined. Authors should indicate in the margins the approximate location in the text for illustrations and tables. At the end of the manuscript following References, the exact postal address(es) of the author(s) should be given. Content of the Manuscript: Manuscripts can be published as original investigations or short communications. a) Original investigations: In addition to the text, original investigations should include illustrations, tables and references, and must not exceed 30 typewritten, double-spaced pages. The text should be divided into: Abstract (in English), Introduction, Material and methods, Results, Discussion (or together as Results and Discussion), Acknowledgements, Zusammenfassung (in German), References. If written in German, an English title should precede the English abstract; if written in English, a German title should precede a German summary. b) Short communications: Short communications must not exceed 5 typewritten, double-spaced pages and do not require either an Abstract or Summary. If written in German, an English title should be given on page 1; if written in English, please give a German title on page 1. Short communications need not be headed with subtitles into In- troduction, Materials and methods, Results, and Discussion but should be organized according to this form. Key words: Please supply five key words following the Abstract. Ilustrations: Number and size of illustrations must be kept to an absolute minimum. Only well focussed and read- ily reproducible original prints showing optimal contrast can be accepted; colour plates cannot be processed. La- belling on figures should be uniform, and large and clear enough for eventual reduction in size. On the back of each illustration the following information should be given, using a soft pencil: figure number, name of the journal, first author. Captions to figures should be written on a separate sheet of paper. Tables: The number and size of tables must be reduced to a minimum. Each table should appear on a separate page including heading and running number. Comprehensive tables not fitting onto one page should follow on consecutive DIN A4 format sheets of paper. References: Each paper cited in the text must appear in alphabetical order at the end of the text. Only internation- ally available literature should be quoted. The names of the journals should be given in abbreviated form in conformity to international use. Examples a) From journals: KALkXo, E. K. V; HANDLEY, C. O. Jr. (1994): Evolution, biogeography, and description of a new species of Fruit-eat- ing bat, genus Artibeus Leach (1821), from Panamä. Z. Säugetierkunde 59, 257-273. b) From books: HERRE, W.; RÖHRs, M. (1990): Haustiere —- zoologisch gesehen. 2. Aufl. Stuttgart, New York: Gustav Fischer. c) From reference books: NIETHAMMER, J. (1982): Cricetus cricetus (Linnaeus, 1758) — Hamster (Feldhamster). In: Handbuch der Säugetiere Europas. Ed. by J. NIETHAMMER und F. Krapp. Wiesbaden: Akad. Verlagsges. Vol. 2/1, 7-28. Printing: The principal author will receive page proofs of the manuscript as well as a set of the illustrations and the original manuscript for proof reading. Only type-setting errors should be corrected. Any major revision undertaken by the author at this stage will result in costs to the authors themselves. The conventional proof-readers’ symbols should be used. The corrected page proofs should be signed by the author to indicate that the paper is “ready for print” and be sent to Prof. Dr. Peter Langer, Institut für Anatomie und Zellbiologie, Justus-Liebig-Universität Gies- sen, Aulweg 123, D-35385 Giessen, Germany, e-mail: peter.langer@anatomie.med.uni-giessen.de Reprints: 50 reprints of each paper are supplied free of charge. When returning the page proofs additional reprints can be ordered at the cost of the author(s). Vorbehalt aller Rechte Die in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Die dadurch begründeten Rechte, ins- besondere 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 gesetz- lichen Bestimmungen des Urheberrechtsgesetzes der Bundesrepublik Deutschland vom 9. September 1965 in der Fassung vom 24. Juni 1985 zulässig. Sie ist grundsätzlich vergütungspflichtig. Zuwiderhandlungen unterliegen den Strafbestimmun- gen des Urheberrechtsgesetzes. Gesetzlich zulässige Vervielfältigungen sind mit einem Vermerk über die Quelle und den Ver- vielfältiger 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 in- ternal 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. Cöpyright 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: Publishers Fee List” of the CCC. Publisher: Urban & Fischer Verlag GmbH & Co. KG, Branch Office Jena, P.O. Box No. 1005 37, D-07705 Jena, Germany. Phone ++49(0)36 41/6 26-3; Fax ++49(0)36 41/62 65 00; e-mail: journals@urbanfischer.de Type setting, printing and binding: druckhaus köthen GmbH Advertisement bookings: Enquiries should be directed to Urban & Fischer Verlag, Advertising Manager: Sabine Schröter, P.O. Box 10 05 37, D-07705 Jena, Germany. Phone ++49(0)36 41/62 64.45, Fax ++49(0)36 41/62 64 21 Advertising rates: The price list from January 1, 1999 is effective at present. Distribution and subscription agency: For customers in the German speaking countries: Please turn to your bookshop or to SFG-Servicecenter Fachverlage, Zeitschriftenvertrieb: Barbara Dressler, Löbdergraben 14a, D-07743 Jena, Germany. Phone ++49(0)36 41/62 64. 44, Fax ++49(0)36 41/62 64.43; For customers in all other countries: Please contact your local bookshop or Stockton Press., c/o Debbie Floyd, The Sub- scription Dept., Houndmills, Basingstoke, Hants RG 21 6XS, UK. Phone ++44(0)12 56/32 92 42. Fax ++44/1256/812358. Subscribers in the USA can call toll free on ++1/800/ 747 3187. Terms of delivery (1999): 1 volume consisting of 6 issues. Subscription rate (1999): Price per volume: DM 478.00*; UK Sterling 192.00 Single issue price: DM 95.00*; UK Sterling 39.00 * Suggested list prices. Prices are subject to change without notice. Reduced membership rates are available on request. Please turn to our publ. house. All prices mentioned above are exclusive postage. USdollar prices are converted at the current exchange rate. We accept the following credit cards: VISA Euro-/Mastercard/American Express, (Please note card-No. and expiry date). Banking connections: For customers in the German speaking countries: Deutsche Bank AG Jena, Konto-No. 6 284 707, BLZ 820 700 00; Postbank Stuttgart, Konto-No. 92 372 704, BLZ 600 100 70. For costumers in the other countries: UK Post Office Giro Account No. 5 192 455. Full details must accompany the payment. Subsciption information: If not terminated the subscription will be effected until recalled. If no cancellation is given until October, 31st the delivery of the journal will be continued. Copyright: The articles published in this journal are protected by copyright. All rights are reserved. No part of the journal may be reproduced in any form without the written permission of the publisher. Authorization to photocopy items for internal or personal use of specific clients, is granted by Urban & Fischer, for libraries and other users registered with the Copyright Clearance Center (CCC) Transaction Reporting Service, provided that the base fee of $12.00 per copy is paid directly to CCC, 222 Rosewood Drive, Danvers, Ma 01923, USA, 0044-3468/99/064/ $ 12.00/0 This consent does not extend to copying for general distribution, for advertising or promotional purposes, for creating new collective works or for resale and other enquiries. In the cases, specific written permission must be obtained from the pub- lisher Urban & Fischer. Printed in Germany 2 STOCKTON Printed on acid-free paper effective with vol. 61, no. 1, 1996. © 1999 Urban & Fischer Verlag New: For detailed journal information see our home page: http://www.urbanfischer.de/journals Abstracted/Indexed in Animal Breeding Abstracts; Current Contents Agriculture, Biology Environmental Sciences; Biological Abstracts; BIOSIS database; Current Advances in Ecological and Environmental Sciences; Dairy Science Abstracts; Elsevier BIOBASE/Cur- rent Awareness in Biological Sciences; Fisheries Review; Helminthological Abstracts; Index Veterinarius; South Pacific Periodicals Index; Veterinary Bulletin; Key Word Index to Wildlife Research; Wild Review (Fort Collins); Zoological Record MT re na nr Ka PNA RR Kuh ir: en Bene a FM Te { Wr 0 U, i Re Ba re HECKMAN BIND EIREN, INC. Bound-To-Please® MAY 00 N. MANCHESTER, INDIANA 46962 | | 2 | | | | | | 3 9088 009 ET una and mi Ener a una Ren Par, mann nnd Due IE zn « u EN EEE Fur nn ame . PB Impnre) u Ver an ae human ana) DL denke he name roh De N mangie" nr erg ar DEN kun er hans? Pe ig erdannnn DE Een un ZART han PERLE HER TE 3 RE ER ERHEN ET Eee IE RE um er we ne, er ran en DIRLBE IE In en Er en VRR" u ana nuremeng url) Er ee De Ei Va ch LT] PERPEPTRERFER Were en rare Mur ensure ne vn a VER ma urn wir ups n mumam nee DaY EAN NENNT Ten N DEE: parts mer Nunpnemphun, 07 a Een Dee wur FF en MER, pre NN PELTEIE en) R Kenn ee PIERRE PER un urapn >% DEV? rinnen ner EEE, rer PLA Ey LET Z L EeRr TE LET La union . werner PER ’ ia Kjrren ara ERPTETRRDHTNT wesen Burn 2 2 Er Eee Pre DR Euehan REED SU 974 N Me dBERTNEIT MANSON Konz bee ae walhin . ; An nF HpTn Ann PAR ERGRÄRREN Eh Keule reg n r Dt h 3, PP 7 Npahamı daran PDEDESTTD a auch DEZE Pe Er nn a en sin Furnseeng el ka U DL IFIDe I Prag TEE: Er en Dre DIVEEET IV A VE Ed LG 2077 ten nn non en ran m Er a Er neun ann RN NR wenn eh 7 SEO DR LYZ DIE van ag N ” BE ra nA NARM ug Va era RN ey ap Aria ne A A A, uusha an rue PEN EEG une a TINTE ETT SEN DR EL nv SEE DE 7 A 07 EBK EP RNET er an mehr run fi nmenı DR RE RT NETT i DER AT En N rn BE nr ER mrapEm me DIEITE, RE Ne LE ut STE on nn un eeen He) mn Inn nn ee DR new) DENT ORDNEN rnamaenmne m / wur nr un mag b La yr Sn; Da BRREN Dr I N wenda nn, BL EL ER Pay 6, ent nur urn, DD ln ua VIE RPBIERENZÄR LEI AD 2 17 PRO au EINEN Su LET # A Th MORTER 2 DAR Ihe) Kan ru Sn In FRI RN, ee 0 RE ee ea RT. © Ivan ar Re LE ERSTER VAR AREA EHE EA main aaa tree an RN mn" TERN URN Tara un N eh AR x BE ee al) mal np, nen DAHER NS Da nen Pa ih BEL REE AE " 7 ame Kan na pr Em A a man, LT Papa Anh En werde n re ee RAN Kruse nm AA ER AR ÄMTERN NE TA NT ae =) Böen BE on ie pn, en en, = . ’ D al un ze An NG N Aitn Deu Run 2 ne 1 hi un an hm, BIETER RER Fre KIA? 2 BR we ar an RR NT n REBEL BEN ZT 27 DIET egal el ER Er a ld r Pe ee RER AHA len] La Eu D De a Le A ERTL, nn nun ne a na aa NR I LE LE LUE De Da ne khan dann une hen Ai r a dr ER AT TETT Dane RETRIEVER NER FI AV NTT Ebene hans au a0 EEE E07, RR. 1 M LE wur“ mar Id Ba a 1 1 IR EZ IT ER Denen N DR EIER A en pn Als) a SL EEE PZTZT TEE, L N En a N Pa ee ZZ Er DPDErr, LIST A ma a al an EL ee a 7, wars ende ru ER RT / e" 1 Per} LAT a Ko LT nn 2 wi van, A Ta re TR en l fi > N unten 2 „ rn PN Sera be Aa Ara palen any r 2 ah £ a RA dt BZ un. VE Tem een na pwjmäsrisng ram mul EHRT, hd VAN RF Ahr Artur rar ana HI, VAR 777 mr hei ra TEE a EEE - ar 0 . hide a ar VD N U Ererae, ur NN : a ee ah nu aan ai md ultaın "78, n, ” RD Pal a hal Led 0 17 I s worst y ESEL TRLNG et Ta er? Den RETAIE, AI VERA ER hr BERN BR EEE AA A ro nn Nam « BE RE RE ee ALTEN HA pas el NER ER AA li a rn enpun a TE Beer Nahen Kata Ne Urn De a EL ER LEERE ir ep ARTE Ze Jr ans anne Ray np ap rare aan nr rn re Par men Im ER NT wre Ark RN ARENA RD PN RT NM N TRENNEN ET a ac Se PAPIER ARENA RN RER, . ARTE PER REIÄERRAIN EN PR RR en ent re a Be 1 ? Arte Zr a FE An JENE ORAL a WERTET P nie el Ei se ae iv PT VEPL I n BT ale ee Zr sr BÄNDER LER air Al RT AT" An onpe ah FURL AR VL RR RAU, aim nal nn a I En RT en ER Peru ner‘ onen. a RT Nnapyase A nz Ir a N > Pe, BEP DENE KR E79 > Be y Raum RE, TE IOK, TRATEN 77 ag re ee im m nrzen fern N RN AN ENENZ. TERROR TRETEN RT, a e dla 2 Bra ee), Rn ee en DA lehnen KARA PR Aa TRIP TO Sure pr rer a BRETT IRRE AUT UP WET NTL, SER nr EG NT? DIN DAN en em N er „ BEE LU AAN ET ruhe a, A aan no WIENER BE ELLE LA RO A770 Var Wr er BRD ENGE v e" head DER en ann) ig“ 2 las u PL En une he, ERDE AITETEN I ERENTAT E I WHAT TEE ENGE NR NEN nn a BEN Ka at > 4a N NORA R RER U NL Ze j n ee Da. 2 ar nen BERN Piskoneemleu Am Ara Anh Fb N NZ Ehe Aare UBER ED nn ana run MIA nr as 0 f si Ale en ERDE D Al La ieh LE U RN Er NEE a ruhe SARA h 4 ER A n ee Pr - een Ban DIRT EN EITIE RZ HRT ALM RR SEHE - EN ERTL Be han Ach a ee Te Te 29 Wehr Zar an hr ya mn EN uE oh Rn Een os DRS hr RA uam: ARATRA NN" DAR ENDE aan Aleanpall in DEREN un er FE ee DE RECHTE REIN eRE OA RIED PN MI nn ae name werten, ee rena KuAY wm hr 4 2 NEN AR ungoh hun pur Mrd" en ye mn 7 ur a a A Eon fi sp an Da M eV NEL EUZYZ 7 rm DV VE Nor ENT RAE NE AR! m nt Ar u u An) nt RT A RAN ar rare N ne ne Ir rn zent yR EA Tin ELTA ARZT N une kur wem VAR Knete nn 2 Sach Karllan n a EL are Ara ar ch an Kı En nng DR NR? Be AA 7 Pen) BIT Aue Wanne & SZENEN BRETTEN ARE MERAN EHINGEN AR HERREN n van surfen f ihr rannte Var en Annan Nr Nr BEL NE EN JAH RU KT VERRATEN ZORRAIR Tal TITLE NAH AR IMPHRE I ER Br Auen vorge ge WA nr ee mg en hi a rk ENTIRD EMEN aan A DA) Nr AVETEREULN wimahr? verein une Ah? En hr hrer Kaen PAR“ ir ERTANN AN